Strategies to increase production of betalain pigments and their use as SynBio tools Alfonso Timoneda Monfort Darwin College Department of Plant Sciences University of Cambridge This thesis is submitted for the degree of Doctor of Philosophy December, 2021 Declaration This thesis is the result of my own work and includes nothing which is the outcome of work done in collaboration except as declared in the preface and specified in the text. It is not substantially the same as any work that has already been submitted before for any degree or other qualification except as declared in the preface and specified in the text. It does not exceed the prescribed word limit of 60,000 for the Biology Degree Committee. Alfonso Timoneda 13th December 2021 Strategies to increase production of betalain pigments and their use as SynBio tools Alfonso Timoneda Monfort Betalains are plant pigments showing a wide range of potential applications in the pharmacological, biotechnological and commodity sectors. They are currently being used as natural food colorants, and intermediates of the biosynthetic pathway exhibit important pharmacological properties (e.g. ʟ-DOPA). Due to this, there is an increasing interest in improving existing plant sources and obtaining new bio-industrial methods for semi-synthetic production. Betalain biosynthesis is achieved in plants via a relatively short metabolic pathway which has been successfully engineered in a diversity of heterologous hosts including plants, bacteria and yeast. Here, we explore a variety of mechanisms to utilise betalain pigments and enhance their bioindustrial production. First, we generated a betalain-based reporter system to visualise arbuscular mycorrhiza symbiosis processes in in Medicago truncatula and Nicotiana benthamiana roots. By expressing the betalain biosynthetic genes under plant symbiosis specific promoters, we were able to effectively restrict betalain production to areas of the root engaging with fungal colonisation. We also showed that this is an efficient tool for the dynamic tracing of root colonisation in vivo over time. This presents an advancement from the current tools used for the visualisation of fungal structures by the symbiosis research community and demonstrates the potential of the betalain pathway for the efficient reporting of other plant physiological processes. Secondly, we explored the existing natural variation in betalain biosynthetic enzymes in order to identify variants exhibiting higher activity than those currently known that could be used to improve betalain production. Initially we focused on the arogenate dehydrogenase (ADH) enzymes in the Caryophyllales, which show relaxed sensitivity to feedback inhibition by product and can produce higher titres of tyrosine, the substrate of the betalain pathway. We interrogated the ADH phylogeny of flowering plants to identify putative ADH homologs with relaxed sensitivity, and found that this phenomenon is not restricted to ADH enzymes of the Caryophyllales, and has likely also occurred repeatedly in other phylogenetically distant clades. Thirdly, we explored the utility of enzymatic diversity held within different origins of betalain pigmentation. Purification and analysis of different ʟ-DOPA 4,5-dioxygenase (DODA) enzymes from different inferred origins in the Caryophyllales revealed distinct kinetic properties and differences in overall pigment production performances in E. coli cultures. Scale-up experiments with the DODAα1 enzyme from Carnegiea gigantea (Cactaceae) in 2 L and 30 L bioreactors allowed for the biotechnological production of unsurpassed titres of betalains. Finally we used ancestral sequence reconstruction and enzyme activity assays in yeast to identify the residues important for high DODA activity in the Globular Inclusion clade, containing the Cactaceae. With this approach, we also confirm that the Globular Inclusion clade represents an independent origin of betalain pigmentation in the Caryophyllales, distinct from other known betalain-producing taxa such as Beta vulgaris. Acknowledgements Thank you to my supervisor Sam Brockington, who saw potential and talent in me, and nurtured and promoted it till this very day. Thank you for all your support and guidance through this PhD and my professional journey. Thank you to Hester Sheehan, because without her I wouldn’t be the researcher that I am today. For all your help and guidance with experiments and writing. For being a great mentor, friend, and role model. To Sebastian Schornack and Fernando García, for their mentorship and guidance through our collaborations. Working with you has widely broadened my experimental skills and scientific thinking. I thank Alex Gavrin, Temur Yunusov and Nathanael Walker-Hale for their invaluable contributions to our publications and some of the results of this thesis. To the senior members of my lab Bo Xu, Ross Mounce, Samuel López Nieves, Boas Pucker, Sandra Guerrero and Loubab Zédane, that through the years have also been amazing mentors and colleagues. And to my fellow graduate students who day after day have filled the lab with joy and laughter, Tao Feng, Gabriela Doria, Greg Mellers, Marion Lémenager, Ellie Tiley, Jordan Ferria, Xiaolong Lyu, Rui Guo, Emily Servanté, Satish Bharathwaj and of course my Brock bros Nathanael Walker-Hale and Brett Wilson. Thank you to my wonderful family and friends, and to my loving husband for their unconditional support. Lastly, I am grateful to my funding body BBSRC and the Coca-Cola company for giving me the opportunity to perform this research. Table of contents Chapter 1. Introduction ................................................................................................ 1 1.1. Introduction to plant pigments ....................................................................................................... 1 1.1.1. Betalain pigments and their colour properties ............................................................................ 1 1.1.2. Fluorescence of betalain pigments ............................................................................................. 3 1.2. Roles of betalain pigments in plants ............................................................................................. 4 1.2.1. Visual signals for animals ........................................................................................................... 4 1.2.2. Photoprotection .......................................................................................................................... 5 1.2.3. Tolerance to drought and salinity stress ..................................................................................... 6 1.2.4. Antioxidant capacity .................................................................................................................... 7 1.2.5. Defence against biotic stress ...................................................................................................... 8 1.3. The betalain biosynthetic pathway ................................................................................................ 9 1.3.1. Arogenate dehydrogenase (ADH) ............................................................................................ 10 1.3.2. Tyrosine hydroxylase (CYP76AD1/5/6/15) ............................................................................... 11 1.3.3. ʟ-DOPA oxidase (CYP76AD1) ................................................................................................ 12 1.3.4. ʟ-DOPA 4,5-dioxygenase (DODA) .......................................................................................... 13 1.3.5. Betalain modifications .............................................................................................................. 14 1.3.6. Transcriptional regulation of the betalain pathway ................................................................... 15 1.3.7. Betalain localisation and transport ........................................................................................... 17 1.4. Evolutionary origins of betalain pigmentation ............................................................................ 18 1.4.1. Evolutionary history of the betalain biosynthetic pathway ........................................................ 18 1.4.1. The mutual exclusion of betalains and anthocyanins ............................................................... 20 1.5. Sources and applications of betalain pigments ......................................................................... 22 1.5.1. Commercial, medical and industrial applications of betalains .................................................. 22 1.5.2. Current sources of betalain pigments ....................................................................................... 24 1.5.3. Metabolic engineering of the betalain pathway for enhanced production ................................ 25 1.5.4. Betalains as reporters and biosensors ..................................................................................... 27 1.5.5. Betalain stability and degradation ............................................................................................ 31 1.6. Thesis objectives ........................................................................................................................... 32 Chapter 2. Betalain pigments as in vivo visual markers for arbuscular mycorrhizal colonisation of root systems ............................................................... 35 2.1. Introduction .................................................................................................................................... 35 2.2. Results ............................................................................................................................................ 37 2.2.1. Betacyanins can be used to visualise AM fungus colonisation in living Medicago truncatula roots .................................................................................................................................................... 37 2.2.2. Betaxanthins as AM fungus colonisation markers in living Medicago truncatula roots ............ 40 2.2.3. Identification and validation of AM symbiosis marker genes in Nicotiana benthamiana .......... 42 2.2.4. Betacyanins visualise AM fungus colonisation in living Nicotiana benthamiana roots ............. 44 2.2.5. Betaxanthins as AM fungus colonisation markers in living Nicotiana benthamiana roots ........ 46 2.2.6. Stable expression of NbPT5b-p1 and NbBCP1b-p1 can cause shoot developmental defects in Nicotiana benthamiana ....................................................................................................................... 47 2.2.7. Expression of all betalain synthesis genes under AM symbiosis specific promoters ameliorates defective phenotypes in Nicotiana benthamiana ................................................................................ 50 2.2.8. Promoter-controlled betalain biosynthesis allows for dynamic tracing of root colonisation processes ........................................................................................................................................... 55 2.3. Discussion ...................................................................................................................................... 57 2.4. Methods .......................................................................................................................................... 60 2.4.1. Plant material and growth conditions ....................................................................................... 60 2.4.2. Promoter isolation ..................................................................................................................... 61 2.4.3. Ligation into pBlueScript. .......................................................................................................... 61 2.4.4. Transformation of vectors into E. coli DH5a. ............................................................................ 62 2.4.5. Colony check by PCR and plasmid sequencing ....................................................................... 62 2.4.6. Generation of vectors using the MoClo Golden Gate technology ............................................ 63 2.4.7. Transformation of binary vectors in Agrobacterium tumefaciens and A. rhizogenes ............... 65 2.4.8. Hairy root transformation of M. truncatula ................................................................................ 66 2.4.9. Stable transformation of N. benthamiana ................................................................................. 66 2.4.10. Plant inoculation with R. irregularis ........................................................................................ 67 2.4.11. RT-PCR gene expression analysis of N. benthamiana stable lines ....................................... 67 2.4.12. Histochemical staining for GUS activity .................................................................................. 68 2.4.13. Staining and quantification of fungal structures ...................................................................... 68 2.4.14. Dissection and microscopy of roots for betalain visualisation ................................................ 70 2.4.15. Betalain extraction and detection using HPLC ....................................................................... 70 2.4.16. Rhizotron ................................................................................................................................ 71 Chapter 3. ADH variants with putative relaxed feedback inhibition outside of the order Caryophyllales .................................................................................................. 73 3.1. Introduction .................................................................................................................................... 73 3.2. Results ............................................................................................................................................ 75 3.2.1. Identifying ADH candidates with relaxed tyrosine feedback regulation .................................... 75 3.2.2. Deregulated ADH enzymes may exist in plant taxa outside of the order Caryophyllales ......... 77 3.2.3. A single amino acid is responsible for the increase in tyrosine production observed with S. lycopersicum ADH2 ............................................................................................................................ 79 3.2.4. Improved tyrosine production shown by SlADH2 and MeADH2 is not sufficient to boost betalain pigment production ............................................................................................................... 81 3.3. Discussion ...................................................................................................................................... 82 3.4. Methods .......................................................................................................................................... 84 3.4.1. Plant material and growth conditions. ...................................................................................... 84 3.4.2. ADH phylogeny ......................................................................................................................... 85 3.4.3. Structure prediction of ADH enzymes using AlphaFold2 ......................................................... 85 3.4.4. RNA extraction and cDNA synthesis ........................................................................................ 86 3.4.5. Gene isolation and synthesis ................................................................................................... 86 3.4.6. Vector construction ................................................................................................................... 86 3.4.7. Site-directed mutagenesis of ADH sequences ......................................................................... 87 3.4.8. Transformation of binary vectors in Agrobacterium tumefaciens ............................................. 88 3.4.9. Transient transformation of Nicotiana benthamiana ................................................................. 88 3.4.10. Tyrosine extraction and quantification using HPLC ................................................................ 89 3.4.11. Betalain extraction and quantification using HPLC ................................................................ 89 3.4.12. Betalain extraction and quantification using spectrophotometry ............................................ 89 Chapter 4. Bioproduction of betalains in Escherichia coli bioreactors and characterization of a highly active DODA enzyme in the Cactaceae ..................... 91 4.1. Introduction .................................................................................................................................... 91 4.2. Results ............................................................................................................................................ 93 4.2.1. Plant DODAα1 enzymes from different Caryophyllales taxa exhibit different performances in Escherichia coli cultures ..................................................................................................................... 93 4.2.2. Plant DODAα1 enzymes from different Caryophyllales taxa exhibit different kinetic parameters ............................................................................................................................................................ 97 4.2.3. Molecular characterisation of the DODAα1 enzyme in the Cactaceae .................................. 100 4.2.4. Identification of CgDODAα1’s metal cofactor ......................................................................... 102 4.2.5. Scaled-up production of betalains in E. coli bioreactors expressing plant DODAα1 enzymes104 4.3. Discussion .................................................................................................................................... 108 4.4. Methods ........................................................................................................................................ 111 4.4.1. Vector construction ................................................................................................................. 111 4.4.2. Transformation of E. coli BL21 strain ..................................................................................... 111 4.4.3. Enzyme expression in E. coli cultures .................................................................................... 112 4.4.4. HPLC analysis of metabolites ................................................................................................. 112 4.4.5. Electrospray ionization mass analysis of metabolites ............................................................ 112 4.4.6. Protein purification .................................................................................................................. 113 4.4.7. Bradford protein quantification ............................................................................................... 114 4.4.8. SDS-PAGE protein electrophoresis ........................................................................................ 114 4.4.9 Absorbance spectroscopy ........................................................................................................ 115 4.4.10. Characterisation of enzyme kinetics and optimal pH ........................................................... 115 4.4.11. MALDI-TOF MS protein analysis .......................................................................................... 116 4.4.12. Trypsin digestion .................................................................................................................. 117 4.4.13. Fast Protein Liquid Chromatography (FPLC) ....................................................................... 117 4.4.14. Metal analysis by ICP-MS .................................................................................................... 117 4.4.15. Scale-up expression of CgDODAα1 in E. coli bioreactors ................................................... 118 Chapter 5. Identification of residues responsible for initial acquisition of DODA activity in CgDODAα1 .............................................................................................. 119 5.1. Introduction .................................................................................................................................. 119 5.2. Results .......................................................................................................................................... 121 5.2.1. High DODA activity in the Globular Inclusion clade appeared in the DODAα1 lineage after separation from the DODAα2 lineage .............................................................................................. 121 5.2.3. Necessity test of DODAα1 ancestor residues reveals residues that could be important for initial acquisition of DODA activity in the Globular Inclusion clade ............................................................ 122 5.2.4. High DODA activity can be gained in non-active ancestral enzymes of the Globular Inclusion clade with the mutation of only three residues ................................................................................. 125 5.3. Discussion .................................................................................................................................... 127 5.4. Methods ........................................................................................................................................ 130 5.4.1. Reconstruction of ancestral sequences ................................................................................. 130 5.4.2. Vector construction ................................................................................................................. 130 5.4.3. Transformation of yeast .......................................................................................................... 131 5.4.4. Yeast colony PCR .................................................................................................................. 132 5.4.5. DODA expression analysis in yeast ....................................................................................... 132 Chapter 6. Conclusion and Future Perspectives ................................................... 133 Bibliography ........................................................................................................................................ 137 List of Figures and Tables Figure 1.1. Representation of betalain structures, absorption spectra and betalain producing species in the Caryophyllales. .............................................................................................................................................. 2 Figure 1.2. The betalain biosynthesis pathway. .......................................................................................... 10 Figure 1.3. The mechanism of action of the two Arogenate dehydrogenase (ADH) variants present in Caryophyllales. ............................................................................................................................................ 11 Figure 1.4. The phylogenetic history of major genes implicated in the betalain biosynthesis pathway that exhibit a Caryophyllales-specific gene duplication. ..................................................................................... 19 Figure 1.5. Betalains are only produced in the order Caryophyllales, where they display a homoplastic distribution. .................................................................................................................................................. 21 Figure 1.6. Betalain production can be engineered in model plants and crops. ......................................... 26 Figure 2.1. Betacyanins can be produced in Medicago truncatula roots as a response to AM fungi colonisation. ................................................................................................................................................ 38 Figure 2.2. Betacyanin accumulation in root tissues of Medicago truncatula. ............................................ 39 Figure 2.3. Expression pattern of MtPT4 and MtBCP1 in Medicago truncatula roots upon colonisation with Rhizophagus irregularis. .............................................................................................................................. 40 Figure 2.4. Betaxanthins can also be produced in Medicago truncatula roots as a response to AM fungi colonisation. ................................................................................................................................................ 41 Figure 2.5. Betaxanthin accumulation in root tissues of Medicago truncatula. ........................................... 42 Figure 2.6. Expression of NbPT5b and NbBCP1b genes is induced under colonisation with Rhizophagus irregularis in Nicotiana benthamiana. .......................................................................................................... 43 Figure 2.7. Betacyanins can be produced in Nicotiana benthamiana roots as a response to AM fungi colonisation. ................................................................................................................................................ 44 Figure 2.8. Nicotiana benthamiana roots expressing betalains contain more fungal structures than white roots. ........................................................................................................................................................... 45 Figure 2.9. Betaxanthins can be produced in Nicotiana benthamiana roots as a response to AM fungi colonisation. ................................................................................................................................................ 47 Figure 2.10. Expression of NbPT5b-p1 in Nicotiana benthamiana stable transformants leads to developmental defects. ............................................................................................................................... 48 Figure 2.11. One-week old Nicotiana benthamiana T1 seedlings from NbPT5b-p1 and NbBCP1b-p1 expressing lines. .......................................................................................................................................... 49 Figure 2.12. Betanin detection by HPLC performed on T1 pigmented leaf tissue of NbPT5b-p1 expressing Nicotiana benthamiana lines. ...................................................................................................................... 50 Figure 2.13. Betacyanin production in Nicotiana benthamiana roots as a response to AM fungi colonisation via expression of the entire betalain pathway under AM symbiosis specific promoters. ............................. 51 Figure 2.14. Quantification of root AM colonisation shows no differences between Nicotiana benthamiana WT and reporter lines. ................................................................................................................................. 52 Figure 2.15. Betacyanin production in Medicago truncatula roots as a response to AM fungi colonisation via expression of the entire betalain pathway under an AM symbiosis specific promoter. ......................... 53 Figure 2.16. Hairy roots of dmi3 mutant Medicago truncatula plants expressing MtPT4-p3 do not produce any perceptible betalain colouration 4-weeks after inoculation with Rhizophagus irregularis. .................... 54 Figure 2.17. Red pigment distribution in root systems of NbBCP1b-p3 and NbPT5b-p3 Nicotiana benthamiana plants colonised by Rhizophagus irregularis on rhyzotron setup. ......................................... 55 Figure 2.18. NbBCP1b promoter-controlled betalain biosynthesis allows for dynamic tracing of root colonisation processes. ............................................................................................................................... 56 Figure 2.19. NbPT5b promoter-controlled betalain biosynthesis allows for dynamic tracing of root colonisation processes in Nicotiana benthamiana. ..................................................................................... 56 Figure 2.20. Schematic of the use of betalains as markers for AM colonisation in plant roots. .................. 57 Figure 2.21. Schematic of the fungal visualisation and quantification process of Medicago truncatula hairy roots after inoculation with Rhizophagus irregularis. ................................................................................... 69 Figure 3.1. Tyrosine can be formed via two alternative routes at the end of the Shikimate pathway. ........ 73 Figure 3.2. ADH gene phylogeny in the flowering plants. ........................................................................... 76 Figure 3.3. Structure models predicted for ADH enzymes. ......................................................................... 77 Figure 3.4. Protein alignment of ADH variant pairs. .................................................................................... 78 Figure 3.5. Tyrosine quantification of Nicotiana benthamiana leaves expressing ADH enzymes. ............. 79 Figure 3.6. Tyrosine quantification of Nicotiana benthamiana leaves expressing mutated versions of ADH enzymes. ..................................................................................................................................................... 80 Figure 3.7. Transient expression of the betalain biosynthetic pathway alongside ADH variants. ............... 82 Figure 4.1. Betalain production of plant DODAα1 enzymes expressed in Escherichia coli cultures. ......... 94 Figure 4.2. Time evolution of reaction intermediates and products in Escherichia coli cultures expressing plant DODAα1 enzymes. ............................................................................................................................. 95 Figure 4.3. Time evolution of reaction intermediates and products in E. coli cultures expressing plant DODAα1 enzymes (second repeat). ........................................................................................................... 96 Figure 4.4. Characterisation of plant DODAα1’s ʟ-DOPA 4,5-dioxygenase activity. .................................. 98 Figure 4.5. ESI-MS identification of CgDODAα1 reaction products. ........................................................... 99 Figure 4.6. Kinetic parameters of DODAα1 enzymes. .............................................................................. 100 Figure 4.7. CgDODAα1 forms dimers in native conditions. ....................................................................... 102 Figure 4.8. Effect of metals on plant DODAα1 activity. ............................................................................. 104 Figure 4.9. Biotechnological production of dopaxanthin in E. coli bioreactors expressing CgDODAα1. .. 106 Figure 4.10. Biotechnological production of indoline-betacyanin in E. coli bioreactors expressing CgDODAα1. .............................................................................................................................................. 107 Figure 5.1. Ancestral reconstruction of DODA genes in the Globular Inclusion clade. ............................. 123 Figure 5.2. Necessity test of GiDODAα1 ancestor residues. .................................................................... 124 Figure 5.3. Combined mutation of N8, N9 and N10 residues does not have a significant impact in enzyme activity. ...................................................................................................................................................... 125 Figure 5.4. Mutation of three residues is sufficient for the acquisition of DODA activity in the GiDODAα1/α2 ancestor. .................................................................................................................................................... 126 Figure 5.5. Residues identified by authors to be important for gain of DODA function in the Caryophyllales. .................................................................................................................................................................. 129 Table 2.1. Quantification of fungal structures observed in roots of Nicotiana benthamiana expressing NbPT5b-p1 and NbBCP1-p1 4 wpi with Rhizophagus irregularis............................................................... 46 Table 2.2. Quantification of fungal structures observed in hairy roots of Medicago truncatula expressing MtPT4-p3 4 weeks post-inoculation with Rhizophagus irregularis.............................................................. 54 Table 4.1. Expression and purification of Carnegiea gigantea DODAα1.................................................... 98 Table 4.2 . Metal concentrations detected in CgDODAα1 extracts........................................................... 103 List of Abbreviations Act2 Actin 2 ADH Arogenate Dehydrogenase AltAll Alternative state at all ambiguous sites AM Arbuscular mycorrhiza APS Ammonium persulfate B5GT/B6GT Betanidin 5-O-glucosyltransferase / Betanidin 6-O-glucosyltransferase BCP1 Blue copper protein 1 BIAs Benzylisoquinoline alkaloids bHLH basic-Helix-Loop-Helix CaMV 35S Cauliflower Mosaic Virus 35S cDNA Complementary DNA cDOPA5GT Cyclo-DOPA 5-glucosyltransferase cTP Chloroplast transit peptide CYP76AD Cytochrome P450 76A dmi3 Doesn’t Make Infections 3 DNA Deoxyribonucleic acid DODA ʟ-DOPA 4,5-dioxygenase DPA Dipicolinic acid ESI Electrospray ionisation FPLC Fast protein liquid chromatography GFP Green Fluorescent Protein GST Glutathione-S-transferase GUS β-glucuronidase HCGT Hydroxycinnamate glucosyltransferases HEK293T Human embryonic kidney 293 cells containing the SV40 T-antigen HPLC High performance liquid chromatography hTH Human tyrosine hydroxylase ICP-MS Inductively coupled plasma mass spectrometry IPTG Isopropyl β-D-1-thiogalactopyranoside Km Michaelis-Menten constant Ki Inhibition constant LB Luria-Betrani (medium) LDL Low-density-lipoprotein ʟ-DOPA ʟ-3,4-dihydroxyphenylalanine LUC Luciferase MALDI Matrix-assisted laser desorption/ionization MAP Maximum a posteriori MATE Multidrug and toxic extrusion transporters MBW Protein complexes containing MYB, bHLH and WD40 repeat factors MCS Multi-cloning site MoClo Modular cloning MS Mass spectrometry (metabolite quantification) MS Murashige-Skoog (medium) MYB Transcription factors containing a domain first identified in Myeloblastosis virus Nos nopaline synthase Ocs Octopine synthase PAGE Polyacrylamide gel electrophoresis PCR Polymerase chain reaction PDH Prephenate dehydrogenase PT4 Phosphate transporter 4 QTL Quantitative trait loci Q-TOF MS Quadrupole time of flight mass spectrometer RNA Ribonucleic acid ROS Reactive oxygen species RT-PCR Reverse transcription PCR SD Synthetic defined (medium) SDS Sodium dodecyl sulphate SCPL Serine-carboxy-peptidase-like TEMED Tetramethylethylenediamine Ub10 Ubiquitin 10 UDP Uridine-5'-diphosphate wpi Weeks post-inoculation WT wild type X-Gal 5-bromo-4-chloro-3-indolyl-beta-D- galacto-pyranoside YPD Yeast Extract-Peptone-Dextrose (medium) Chapter 1. Introduction 1 Chapter 1. Introduction 1.1. Introduction to plant pigments Plants are estimated to produce over 1,000,000 different metabolites (Afendi et al., 2012), many of which are a result of specialised metabolism, and which facilitate adaptation to their environment. Of those compounds, plant pigments are the ones responsible for the colouring of plant tissues and organs. Plant pigments can be divided in four main groups: chlorophylls, carotenoids, flavonoids (including anthocyanins) and betalains. Chlorophylls are the green pigments present in plant chloroplasts essential for absorbing light in the photosynthetic process and the molecules responsible for the characteristic green colouration of plants (Delgado- Vargas & Paredes-Lopez, 2003). Carotenoids are yellow, orange and red pigments that assist chlorophylls with light absorption during photosynthesis, but which can be also present in fruits and flowers of some plant species (Delgado-Vargas & Paredes-Lopez, 2003). Anthocyanin colouration ranges between red, purple and blue, and is responsible for the colour of many flowers and fruits, as well as being involved in plant tolerance against several environmental stresses (Delgado-Vargas & Paredes-Lopez, 2003). Finally, betalains can range from yellow to deep purple, and are commonly produced in flowers, fruits and seeds, as well as in response to a range of biotic and abiotic stresses. 1.1.1. Betalain pigments and their colour properties Betalains were first discovered in plants, where they are unique to the flowering plant order Caryophyllales (Brockington et al., 2011), but have also been reported in the proteobacterium Gluconacetobacter diazotrophicus (Contreras-Llano et al., 2019) and the fungi Amanita and Hygrocybe (Musso, 1979; Babos et al., 2011). Caryophyllales is an extensive order of plants containing many agricultural and ornamentally relevant species, including crops such as beetroot (Beta vulgaris), spinach (Spinacia oleracea), amaranth (Amaranthus hypochondriacus), quinoa (Chenopodium quinoa) and decorative species such as the four o’clock flower (Mirabilis jalapa), cockscomb (Celosia argentea), bougainvillea (Bougainvillea glabra) and cactus (Cactaceae) (Figure 1.1d). A. Timoneda, PhD thesis, 2022 2 Figure 1.1. Representation of betalain structures, absorption spectra and betalain producing species in the Caryophyllales. (a) Betacyanins originate via betanidin following the condensation of betalamic acid with cyclo-DOPA and further modifications. Betaxanthins are formed by the condensation of betalamic acid with amino acids and other amines, such as ʟ-3,4-dihydroxyphenylalanine (ʟ-DOPA) to form dopaxanthin. Coloured in green the electron resonance system responsible for the colour and fluorescence of betalains. (b) Betacyanin absorption spectra (betanin). (c) Betaxanthin absorption spectra (e.g. dopaxanthin). Green dotted line represents fluorescence emission of betaxanthins. Spectral data adapted from Gandía-Herrero et al., (2005c). (d) Examples of betacyanin and betaxanthin structures following decoration proceses such as glycosylation, acylation and decarboxylation. Betalain molecules are traditionally named after the species they were originally discovered in but may not be species or organism-specific. Betacyanins can be divided in the betanin group (betanidin 5-O-β-glucoside), gomphrenin group (betanidin 6-O-β-glucoside), amaranthin group (5-O-β-glucuronylglucoside), and bougainvillein group (betanidin 5-O-β-sophoroside or betanidin 6-O-β- sophoroside). Glc, glucose; GlcA, glucuronic acid. Adapted from Polturak & Aharoni (2017). Chapter 1. Introduction 3 Betalains are water soluble, tyrosine-derived pigments which can be biochemically defined by their inclusion of betalamic acid as the central chromophore of the molecule (Stafford, 1994; Tanaka et al., 2008). Betalains comprise two classes of compounds, yellow betaxanthins and purple betacyanins. Betalain colour properties are given by the electron resonance system supported between the two nitrogen atoms in the molecule (Figure 1.1a). Betalamic acid and betalains with the simplest structural features around the nitrogen atom display yellow colour, that can shift to orange if aromatic structures are present that resonate with the nitrogen atom (Gandía-Herrero et al., 2010). This is the case for betaxanthins, which originate from the conjugation of betalamic acid with different amino acids and other amines, and display an absorbance spectrum with maximal wavelength centered at 480 nm (Figure 1.1a,c) (Gandía- Herrero & García-Carmona, 2013). Betacyanins derive from the condensation of betalamic acid with cyclo-dihydroxyphenylalanine (cyclo-DOPA) (Stafford, 1994; Tanaka et al., 2008). When the resonance system of betalamic acid is connected to another aromatic system through an intramolecular cycle such as the catechol substructure of cyclo-DOPA in betacyanins, the colour of the molecule shifts to purple (Figure 1.1a). Betacyanins can undergo further modifications and present an absorbance spectrum with typical wavelength maximums centered at 538 nm (Figure 1.1b). In petals and bracts, betacyanins and betaxanthins can be simultaneously produced at different ratios, leading to a broad palette of intermediate colours such as orange and pink (Kugler et al., 2007). 1.1.2. Fluorescence of betalain pigments A pigment’s colour is determined by the amount of light it can absorb and the resulting portion of the light spectrum it reflects. Humans can perceive colours from molecules reflecting or transmitting light between 380 and 730 nm, while other animals like insects can detect a broader spectrum of wavelengths (Tanaka et al., 2008). The efficiency with which a molecule can absorb light is known as its attenuation coefficient or extinction coefficient (Baker & Lavelle, 1984). The higher the extinction coefficient, the higher a molecule’s ability to absorb light at a given wavelength. The energy acquired from absorbing light can undergo different pathways. Fluorescence happens when the radiation absorbed by a molecule at a certain wavelength is re-emitted at a different wavelength, almost always at a longer wavelength (Marshall & Johnsen, 2017). In addition to the reflected light, betaxanthins (and to much lesser extent betacyanins) can also emit fluorescence (Gandía-Herrero et al., 2005a). Typically, fluorescent organic compounds are molecules with a high degree of conjugated unsaturated bonds and extended electron cloud structures. All betaxanthins are a result of the condensation of amino acids and other amines with betalamic acid, which presents conjugated double bonds (Figure 1.1). The electronic resonance in the betalamic acid moiety is responsible A. Timoneda, PhD thesis, 2022 4 for the fluorescence observed in betaxanthins. Betaxanthins exhibit spectra with excitation maxima between 463 and 474 nm and emission maxima between 509 and 512 nm (Figure 1.1), thus able to absorb blue light and emit green light (Gandía-Herrero et al., 2005a). Fluorescence can be enhanced with the presence of electron withdrawing structures in the vicinity of the betalamic acid moiety, like the extra carboxylic groups in dopaxanthin and portulacaxanthin II (Gandía-Herrero et al., 2005b). Electron density-donating structures, like hydroxyl groups, have an opposite effect over the fluorescence of the molecule. This is the case for betacyanins, where the extra electronic resonance provided by the aromatic ring of the cyclo-DOPA moiety after conjugation with betalamic acid drastically reduces their fluorescence. Betacyanins are only weakly fluorescent, and mainly in the presence of carboxylic groups and in the absence of hydroxyl groups (Gandía-Herrero et al., 2010). Maximum excitation wavelengths in betacyanins occur between 521 and 529 nm and emission spectra are centred around 570 and 575 nm (Guerrero-Rubio et al., 2020a), however betacyanin fluorescence displays a very short excited- state lifetime and is barely detectable (Wendel et al., 2015). 1.2. Roles of betalain pigments in plants Betalains and anthocyanins display similar optical properties and are normally produced in the same tissues and cell types, and under similar conditions (Jain & Gould, 2015b). However, betalains and anthocyanins have never been found to co-exist in the same extant plant species (Brockington et al., 2011). These factors have led authors to postulate that betalains and anthocyanins could be functional analogues, and that betalains may have functionally replaced anthocyanins in betalain-producing families in the Caryophyllales (Jain & Gould, 2015b). Betalain pigmentation occurs broadly in many different tissues in plants: flowers, fruits, seeds, leaves, shoots and roots. While some species accumulate betalains both in vegetative and reproductive tissues, in others, betalain synthesis is mainly restricted to reproductive organs such as flowers and fruits (Jain & Gould, 2015b). 1.2.1. Visual signals for animals Like anthocyanins, betalains produced in flowers and fruits play a major role as visual cues in attracting animal pollinators and frugivores for fertilisation and seed dispersal. Many species of the Nyctaginaceae, Aizoaceae and Cactaceae, exhibit a wide diversity of elaborate and colourful flower displays, often establishing parallel pollen feeding and nectar production from filament bases (Ehrendorfer, 1976; Clement & Mabry, 1996). In an early study, Ehrendorfer (1976) proposed betalains could have evolved as a means to re-establish zoophily from what he speculated could have been a common wind pollinated ancestor. Authors have also Chapter 1. Introduction 5 hypothesised betaxanthin fluorescence could be involved in enhancing flower visibility for pollinators (Gandía-Herrero et al., 2005a; Guerrero-Rubio et al., 2020a). Fluorescence emitted by betaxanthins has been detected and photographed in yellow flowers of Portulaca grandiflora, Lampranthus productus and Celosia argentea species belonging to the Portulacaceae, Aizoaceae and Amaranthaceae families, respectively (Gandía-Herrero et al., 2005a; Guerrero- Rubio et al., 2020a). Bees possess photoreceptors with sensitivities that peak in the ultraviolet, blue and green parts of the light spectrum, and their ability to detect coloured targets has been shown to correlate with colour brightness and intensity (Hempel de Ibarra, 2000). Bees have also been shown to exhibit attraction to fluorescence emission, although exclusively in the blue spectral region (420 - 460 nm) (Rao & Ostroverkhova, 2015; Ostroverkhova et al., 2018). Bats of the species Glossophaga soricina, who feed on nectar and act as pollinators for member species of the Cactaceae, are also able to detect green light with a sensitivity maxima at 510 nm, which corresponds with the maximum emission of betaxanthins (Winter et al., 2003). Phototaxis studies with bats of other species confirmed they were indeed attracted to green light with a wavelength of 520 nm in distances of up to 23 metres (Voigt et al., 2017). 1.2.2. Photoprotection Exposure to high light has been shown to upregulate betalain production in many betalain producing plant species. Studies on cotyledons of several species in the genus Amaranthus found amaranthin production was strongly enhanced after irradiation with white light (De Nicola et al., 1975; Kochhar et al., 1981). Similar effects were found in Mesembryanthemum crystallinum of the Aizoaceae, where high white light intensities translated in an increase in the accumulation of betacyanins in epidermal bladder cells of young leaves and epidermal layers of fully expanded leaves (Vogt et al., 1999b; Ibdah et al., 2002). Betalain production was reported to be more effectively induced under light of wavelengths between 390 and 460 nm in callus lines of P. grandiflora, which represent the blue light portion of the visible spectrum (Kishima et al., 1995). Later studies on intact plants of M. crystallinum found betalain accumulation could also be induced by ultraviolet (UV) radiation, and was mostly increased in the waveband between 305 and 320 nm (Ibdah et al., 2002). However, to date, no studies have shown expression of the betalain biosynthetic genes to be induced under high light and UV radiation. This is in contrast with results obtained for the anthocyanin pathway, where transcripts of the phenylalanine ammonia-lyase (PAL) and chalcone synthase (CHS) enzymes, are known to be induced by UV radiation (Logemann et al., 2000). Moreover, the initial enzyme of the shikimate pathway, 3-deoxy-D-arabino-heptulosonate 7-phosphate synthase (DHAPS), which ultimately leads to the production of aromatic amino acids like phenylalanine and tyrosine, precursor molecules for anthocyanins and betalains respectively, is also strongly stimulated by UV light in parsley (Logemann et al., 2000). A. Timoneda, PhD thesis, 2022 6 Betalain pigment induction under high light intensities implies a potential role in photoprotection of aerial plant tissues. Prolonged exposures to light intensities in excess to the photosynthetic requirements of the plant can lead to the production of reactive oxygen species (ROS) and damage to the photosystem II (PSII) in the photosynthetic apparatus (Takahashi & Badger, 2011). Under the ‘light screening’ hypothesis, anthocyanins and betalains are produced in epidermal and palisade mesophyll cells in order to help absorb the excess light that would otherwise damage chloroplasts and the antenna pigments (Jain & Gould, 2015b). A number of studies support this hypothesis for betalain pigments. Quantum efficiency of the PSII declined substantially less in betalain-coloured leaves versus the green leaves of Disphyma australe under white and green light, but showed similarly large reductions under monochromatic red light, which cannot be absorbed and is reflected by betacyanins (Jain et al., 2015). Similar patterns have been described for the betalain producing species Suaeda salsa and Amaranthus cruentus (Wang & Liu, 2007; Nakashima et al., 2011). Non-photochemical quenching and ROS production have also been reported to be greater in non-pigmented than pigmented tissues, suggesting leaves with less betalain accumulation have a higher need for energy dissipation (Jain et al., 2015). Altogether, these findings present strong evidence for a protective light- screening role by foliar betalains. 1.2.3. Tolerance to drought and salinity stress Betalain producing plants in the Caryophyllales are commonly known to inhabit arid and saline environments such as deserts, marshes and dunes, especially in the families Amaranthaceae Aizoaceae, Cactaceae and Portulacaceae (Jain & Gould, 2015b). While growing in these environments, such plant populations show variation in shoot colour often displaying red colouration (Wang et al., 2007b; Jain & Gould, 2015a). An example of this are the halophyte species of the genus Suaeda which grow in salt marshes and intertidal zones where soil salt content is often higher than 3% (Wang et al., 2007b; Hayakawa & Agarie, 2010). S. salsa is an economically important halophyte in China, which gains its value from its high oil content in seeds. Plants of S. salsa have been reported to appear red-violet when growing in the intertidal zone of the Yellow River delta, and green when growing at higher elevation or further from the seaside (Wang et al., 2007b). Extreme water deficit can also enhance salt and mineral stress, and many betalain producing species of the Caryophyllales also grow on arid areas which are commonly not only associated with extreme drought but also high light intensities, such as the Cactaceae and Aizoaceae. Betalain content has been reported to increase under drought and salinity stress conditions in a number of plant species such as S. salsa, Suaeda japonica, A. cruentus and D. australe (Wang et al., 2007b; Hayakawa & Agarie, 2010; Nakashima et al., 2011; Jain & Gould, 2015a; Jain et al., 2015). Betalain production could be therefore induced in plants as a tolerance mechanism to Chapter 1. Introduction 7 adapt to and minimise osmotic related physiological damage. Two main mechanisms are provided to explain the role of betalains in drought and salinity tolerance: photoprotection and defence against ROS. Drought and salinity compromise the capacity of chloroplasts to process light energy (Parida & Das, 2005). As mentioned before, photoprotection conferred by betalain pigments could be indirectly helping to increase plant fitness by shielding chloroplasts from excess incident light. Consistent with this, betacyanin producing leaves of A. cruentus displayed a lesser degree of photoinhibition than non-pigmented leaves under water deficit stress conditions (Nakashima et al., 2011). Similarly, salinity treatments of D. australe morphs containing betacyanin pigments experienced smaller decreases in the quantum efficiency of PSII than green morphs (Jain & Gould, 2015a). In a recent study, stable heterologous expression of betacyanins in Nicotiana tabacum plants increased seedling survivability and photochemical quantum yields of PSII under salt stress (Zhou et al., 2021). These plants also exhibited a faster photosynthetic recovery after saturating light treatments, and a slower rate of chlorophyll and carotenoid degradation (Zhou et al., 2021). Both drought and salt stress impose a water deficit in plants due to osmotic effects on a wide variety of metabolic activities which ultimately lead to the formation of ROS (Parida & Das, 2005). Plants produce a number of antioxidant enzymes and molecules to scavenge ROS and protect against their potential cytotoxic effects. Higher levels of antioxidants in plant cells translate to greater resistance to oxidative damage (Parida & Das, 2005). Increases in betalain accumulation in S. salsa seedlings has been found to correlate with increases in the activity of the vacuolar H+-ATPase involved in compartmentalising sodium in the vacuoles and the foliar and chloroplastic superoxide dismutases which have a central role in the antioxidant defence network by scavenging highly reactive O2- radicals (Wang et al., 2007b). 1.2.4. Antioxidant capacity As previously discussed, most environmental stresses ultimately lead to ROS production, and therefore, oxidative stress in the cell. For example, water deficit and salinity, especially in combination with high light intensity or other stresses, affect photosynthesis and increase photorespiration, altering cell homeostasis and ROS production (Miller et al., 2010). Initially, ROS are important components of the signal transduction pathway in plant response to stress. However, at high concentrations, ROS have the potential to disrupt cell and organelle membranes, damage DNA and denature proteins (Jain & Gould, 2015b). Betalains have a strong antioxidant capacity and can be used by plants to combat oxidative stress, a hypothesis that is supported by a growing number of reports (Kanner et al., 2001; Butera et al., 2002; Cai et al., 2003; Stintzing et al., 2005; Georgiev et al., 2010; Nakashima & Bastos, 2019). Betacyanin- containing leaves of betalain producing species such as S. salsa and Amaranthus tricolor produce less of the ROS generating compound, hydrogen peroxide (H2O2), than green leaves A. Timoneda, PhD thesis, 2022 8 under light and cold stress (Wang & Liu, 2007; Shu et al., 2009). In fact, watering S. salsa roots with H2O2 directly leads to betalain accumulation in leaves (Wang et al., 2007a). Betalains extracted from a number of species of Amaranthaceae demonstrate strong antioxidant activity in vitro, in some cases 3 to 4-fold higher than common antioxidants like ascorbic acid (vitamin C), ruthin and catechin (Cai et al., 2003). The free radical scavenging activity of multiple betalain molecules has been studied and compared (Cai et al., 2003). Betalain antioxidant activity has been mainly attributed to the resonant structure of the molecule (Figure 1.1a, in green) and can increase with the number of hydroxyl and imino group modifications, as well as decrease with glycosylation (Cai et al., 2003; Nakashima & Bastos, 2019). 1.2.5. Defence against biotic stress It has previously been suggested that betalains could have evolved as a result of their antiviral and antifungal properties (Mabry, 1977; Piatelli, 1981; Stafford, 1994). However, the number of published studies supporting this hypothesis is still scarce. A screening of biotic and abiotic elicitors in in vitro hairy root cultures of B. vulgaris found betalain biosynthesis could be induced by a diversity of purified microbial glycans, extracts of whole fungal and yeast cultures and their respective culture filtrates (Savitha et al., 2006). Maximum betalain induction in this study was achieved by using dried cell powder of the fungus Penicillium notatum and was about 3.6-fold higher over that of control cultures (Savitha et al., 2006). In a recent study, heterologous betalain expression in Nicotiana tabacum leaves significantly reduced the extent of infection of the phytopathogenic ‘gray mold’ fungus Botrytis cinerea, which could reach to up to a 90% reduction in the lesion area (Polturak et al., 2017). To date, this remains the most direct piece of evidence to support the antifungal role of betalains. Other in vitro studies also identified methyl jasmonate as an effective biotic elicitor for betalain induction in hairy root cultures of B. vulgaris, callus cultures of B. glabra and suspension cell cultures of P. grandiflora (Bhuiyan & Adachi, 2003; Suresh et al., 2004; Lakhotia et al., 2014). Methyl jasmonate is a volatile organic molecule produced by plants as a signal in response to many biotic and abiotic stresses and certain developmental stages, and also has a major role in response to herbivory and wounding. Methyl jasmonate production is induced by herbivory attack and can be rapidly propagated throughout the plant as a signal to synthesise various defence-related compounds such as toxins and anti-digestive enzymes (Kessler & Baldwin, 2002). Jasmonates are also known to be involved in flower development processes and methyl jasmonate has been shown to have an effect on flowering time, floral organ morphology and floral gene regulation (Pak et al., 2009). Flowers are one of the main betalain producing organs in many betalain producing members of the Caryophyllales, therefore jasmonates could be participating in the betalain synthesis regulation network regardless of a link to anti-herbivory. Chapter 1. Introduction 9 1.3. The betalain biosynthetic pathway Elements of this section represent an updated development of a literature review originally used in the Introductions of the author’s MPhil thesis1 and an already published peer reviewed paper2. 1 Timoneda, A. (2017) Re-directing primary metabolism to increase betalain production. MPhil thesis, University of Cambridge. 2 Timoneda, A., Feng, T., Sheehan, H., Walker‐Hale, N., Pucker, B., Lopez‐Nieves, S., Guo, R. and Brockington, S. (2019) The evolution of betalain biosynthesis in Caryophyllales. New Phytologist, 224: 71-85. https://doi.org/10.1111/nph.15980 Betalains are synthesized from tyrosine, which is derived from the shikimate pathway. In most plants, the precursor arogenate is decarboxylated into tyrosine, by the action of arogenate dehydrogenase (ADH) (Rippert et al., 2009). Following tyrosine synthesis, the betalain pathway comprises three main steps requiring enzymatic catalysis (Figure 1.2). Tyrosine is first converted to ʟ-3,4-dihydroxyphenylalanine (ʟ-DOPA) through a tyrosine hydroxylation reaction, which can be catalysed by a number of cytochrome P450 enzymes of the CYP76AD family. The cyclic ring within ʟ-DOPA is then cleaved in a ring-opening oxidation reaction by the enzyme ʟ- DOPA 4,5-dioxygenase (DODA) to produce the intermediate 4,5-seco-DOPA, which then undergoes spontaneous intramolecular condensation to produce betalamic acid. Alternatively, ʟ-DOPA can be oxidized to dopaquinone after which it cyclizes to form cyclo-DOPA, a step that can only be catalysed by one of the previously mentioned cytochrome P450 enzymes, CYP76AD1. Betalamic acid can spontaneously conjugate with the imino group of cyclo-DOPA, ultimately leading to the formation of red-violet betacyanins. Alternatively, betalamic acid can spontaneously condense with the amino group of amino acids and other amines to give yellow betaxanthins. In addition to these central enzymatic steps, additional moieties such as glucosyl or acyl groups can be enzymatically added to betacyanins. Glycosylation of betacyanins is common and can occur either on cyclo-DOPA before condensation with betalamic acid, or on betanidin after the condensation of cyclo-DOPA and betalamic acid. These reactions are catalysed by cyclo-DOPA 5-O-glucosyltransferase (cDOPA5GT) or betanidin glucosyl-transferases, respectively (Figure 1.2). In addition to glycosylation, betacyanins can undergo other enzymatically catalysed modifications, which can add a range of moieties, contributing to the structural diversity of betalains. In the following section, we sequentially discuss the molecular genetic characterization of the enzymes implicated in betalain biosynthesis. A. Timoneda, PhD thesis, 2022 10 Figure 1.2. The betalain biosynthesis pathway. Schematic includes the step of oxidative decarboxylation of arogenate catalysed by arogenate dehydrogenase (ADH), tyrosine hydroxylation catalysed by cytochrome P450 enzymes (CYP76AD1/5/6) to form ʟ-3,4-dihydroxyphenylalanine (ʟ-DOPA), the formation of betalamic acid from ʟ-DOPA through the action of ʟ-DOPA 4,5-dioxygenase (DODA), the conversion of ʟ-DOPA to cyclo-DOPA by cytochrome P450 (CYP76AD1), glucosylation of cyclo-DOPA via the action of cyclo-DOPA5-O-glucosyltransferase (cDOPA5GT), glucosylation of betanidin to form betanin by betanidin 5/6 glucosyltransferases (B5GT/B6GT), and the spontaneous condensation of betalamic acid with cyclo-DOPA or cyclo- DOPA 5-O-glucoside to form red/purple betacyanins (in pink), or with amino or amine groups to form yellow betaxanthins (in yellow). Dotted grey box indicates steps belonging to the Shikimate pathway. 1.3.1. Arogenate dehydrogenase (ADH) Recent research has emphasized a role for modulation of primary metabolism in the evolution of the betalain biosynthesis pathway (Lopez-Nieves et al., 2018). In plants, the pathway to tyrosine synthesis is usually highly regulated at ADH, which is strongly feedback-inhibited by tyrosine (Maeda & Dudareva, 2012). A recent study recognized that many species in Caryophyllales possess a canonical form of ADH (termed ADHβ) and an additional paralogue of the ADH enzyme (termed ADHα). Functional characterization of the novel ADHα isoform using both in vitro and heterologous transient assays in Nicotiana benthamiana established that, in Chapter 1. Introduction 11 contrast to ADHβ, ADHα has relaxed sensitivity to the negative feedback inhibition by tyrosine (Figure 1.3) (Lopez-Nieves et al., 2018). As a result, these deregulated ADHα enzymes are able to synthesize higher concentrations of tyrosine in in vitro assays relative to the canonical tyrosine-sensitive ADHβ, with 10-fold increases in tyrosine reported on transient assay in N. benthamiana (Lopez-Nieves et al., 2018). A subsequent study coupled the deregulated ADHα to a module comprising the complete betalain biosynthesis pathway to demonstrate that increased availability of tyrosine results in as much as a 7-fold increase in betalain pigmentation on transient assay in N. benthamiana (Timoneda et al., 2018). Together these studies imply an intimate link between the evolution of Caryophyllales specific deregulated ADHα, and Caryophyllales-specific betalain biosynthesis. Figure 1.3. The mechanism of action of the two Arogenate dehydrogenase (ADH) variants present in Caryophyllales. (a) The canonical ADHβ enzyme catalyses the decarboxylation reaction of arogenate to tyrosine, and displays product feedback inhibition. (b) The Caryophyllales-specific ADHα enzyme has relaxed sensitivity to tyrosine feedback inhibition. From Timoneda et al., 2018. 1.3.2. Tyrosine hydroxylase (CYP76AD1/5/6/15) Elucidation of the enzymes responsible for the initial tyrosine hydroxylation step of betalain biosynthesis was only recently achieved and was, chronologically, the last contribution to our understanding of the pathway. As will be discussed in the next section, BvCYP76AD1 had been previously identified to catalyse the latter ʟ-DOPA oxidation step of the pathway in beet (Hatlestad et al., 2012). However, heterologous expression of BvCYP76AD1 in yeast demonstrated its capacity not only to oxidize ʟ-DOPA, but also to catalyse the conversion of tyrosine to ʟ-DOPA, the first step of the betalain pathway (DeLoache et al., 2015). Tyrosine hydroxylase activity of BvCYP76AD1 in planta was then confirmed by gene silencing in B. vulgaris and recombinant expression assays in N. benthamiana (Polturak et al., 2016). Further to the discovery of the tyrosine hydroxylase activity of BvCYP76AD1, subsequent efforts uncovered additional enzymes responsible for catalysing the first step of the betalain pathway. Polturak et al. (2016) showed that tyrosine hydroxylation is also catalysed by a closely A. Timoneda, PhD thesis, 2022 12 related CYP76AD homologue, namely BvCYP76AD6, identified in B. vulgaris. Sunnadeniya et al. (2016) confirmed these findings and identified yet another CYP76AD homologue, BvCYP76AD5 from B. vulgaris, also with tyrosine hydroxylase activity. Expression of both BvCYP76AD5 and BvCYP76AD6 homologues in yeast led to the production of ʟ-DOPA (Sunnadeniya et al., 2016). Yellow beet cultivars containing a defective BvCYP76AD1 allele were also shown to display overall limited pigment production and increased tyrosine accumulation in planta, indicating that the lack of an active BvCYP76AD1 can substantially reduce the plant’s capacity to oxidize tyrosine into ʟ-DOPA (Wang et al., 2017). Although these tyrosine-hydroxylase-specific CYP76AD enzymes were initially identified in B. vulgaris, other analyses have identified a homologue in M. jalapa, MjCYP76AD15 (Sunnadeniya et al., 2016; Polturak et al., 2018), which exhibits similar tyrosine hydroxylase activity. Together these data confirm the existence of a conserved class of cytochrome P450 with tyrosine hydroxylase activity in lineages of the Caryophyllales. 1.3.3. ʟ-DOPA oxidase (CYP76AD1) Initially, BvCYP76AD1 from B. vulgaris was detected as a highly expressed gene in betalain- producing tissues (Hatlestad et al., 2012). On silencing of the BvCYP76AD1 locus in B. vulgaris, a loss of betacyanins was observed, while transgenic expression of BvCYP76AD1 recovered betacyanin pigmentation in ‘Golden Globe’, a commercial yellow variant of B. vulgaris. In the same study, analysis of a B. vulgaris cultivar C8689 that segregates for yellow and red hypocotyls found that a recessive insertional mutant allele of BvCYP76AD1 leading to an inactive protein segregated with the yellow phenotype (Hatlestad et al., 2012). Together these results confirmed that BvCYP76AD1 possesses ʟ-DOPA oxidase activity, providing the cyclo- DOPA moiety for betacyanin production. Transposon-mediated mutation in MjCYP76AD3, an orthologue of BvCYP76AD1 derived from M. jalapa, results in loss of betacyanin production, consistent with a role of MjCYP76AD3 in cyclo-DOPA production (Suzuki et al., 2014). Most recently, a mutant screen in C. quinoa has led to the identification of a CYP76AD1 orthologue which restores betalain pigmentation in a mutant deficient for betalains in its hypocotyl (Imamura et al., 2018). Taken together, these data provide evidence of a class of cytochrome P450 enzymes with ʟ-DOPA oxidase activity that are conserved across distantly related lineages in Caryophyllales. Efforts to obtain a CYP76AD1 mutant with an increased tyrosine hydroxylase activity by abolishing its ʟ-DOPA oxidase activity led to the identification of two key residues for the maintenance of ʟ-DOPA oxidase activity (DeLoache et al., 2015). Mutations W13L and F309L were shown to have a cumulative effect on reducing metabolic flux to betanin, whose production directly depends on the ʟ-DOPA oxidase step, increasing betaxanthin production up to 4.3-fold. While the W13L mutation seemed to generally improve tyrosine hydroxylase activity, mutation Chapter 1. Introduction 13 F309L was identified as responsible for the majority of the ʟ-DOPA oxidase reduction accounting for a 80% reduction in betanin compared to wild type CYP76AD1 (DeLoache et al., 2015). This residue was found to be conserved in CYP76AD1 orthologs from A. cruentus, M. jalapa and C. cristata, and equivalent mutations led to similar reductions in ʟ-DOPA oxidase activity (DeLoache et al., 2015). Unlike BvCYP76AD1, the later identified enzymes BvCYP76AD5 and BvCYP76AD6 only exhibit tyrosine hydroxylase activity and cannot perform the later ʟ-DOPA oxidase step. However, betanidin was detected by LC-MS at very low intensities, suggesting that these enzymes may have a weak ability to produce cyclo-DOPA (Sunnadeniya et al., 2016). The differential ability to produce cyclo-DOPA between BvCYP76AD1, BvCYP76AD5 and BvCYP76AD6 provides the genetic architecture for selective production of yellow or red betalains, as is commonly seen in agricultural cultivars. In fact, coexpression of BvCYP76AD1 and BvCYP76AD6 in tobacco has been shown to produce a combination of betaxanthins and betacyanins, resulting in an orange- pink colouration and suggesting that the ratio of expression of these enzymes may be responsible for the variety of hues observed in nature (Polturak et al., 2017). 1.3.4. ʟ-DOPA 4,5-dioxygenase (DODA) In plants, betalamic acid is formed via the extradiol cleavage of ʟ-DOPA between positions 4 and 5 of the aromatic ring. A LigB enzyme with DODA activity was first identified and purified in the betalain-pigmented fungi Amanita muscaria (Girod & Zryd, 1991a; Terradas & Wyler, 1991). Cloning and heterologous expression of the A. muscaria DODA gene in white petals of P. grandiflora (Caryophyllales) and Antirrhinum majus (Lamiales) result in the formation of yellow and purple spots containing betalains (Mueller et al., 1997; Hinz et al., 1997; Harris et al., 2012). Informed by earlier classical genetic work, Christinet et al. (2004) isolated the transcript of a LigB candidate gene encoding an enzyme with DODA activity within Caryophyllales, by constructing subtractive cDNA libraries between differently coloured phenotypes of isogenic lines of P. grandiflora. Biolistic transformation and genetic complementation in white petals of P. grandiflora confirmed the role of this LigB gene in betalain production (Christinet et al., 2004). Following characterization ofthis LigB homologue in P. grandiflora, DODA activity has been either characterized or implicated in multiple LigB homologues across betalain-pigmented Caryophyllales, including A. hypochondriacus (Casique-Arroyo et al., 2014), Hylocereus polyrhizus (Qingzhu et al., 2016), S. salsa (Zhao et al., 2011), B. vulgaris (Hatlestad et al., 2012; Gandía-Herrero & García-Carmona, 2012), C. quinoa (Imamura et al., 2018), Parakeelya mirabilis (Chung et al., 2015), M. jalapa and B. glabra (Sasaki et al., 2009). The introduction of P. grandiflora LigB in non-betalain backgrounds and feeding with its substrate, ʟ-DOPA, were also sufficient to trigger betalain production (Harris et al., 2012). A. Timoneda, PhD thesis, 2022 14 1.3.5. Betalain modifications 1.3.5.1. Betanidin glucosyltransferases (B5GT and B6GT) All betacyanins are composed of betanidin conjugated to different glycosyl moieties (Strack et al., 2003; Sasaki et al., 2005). By contrast, betaxanthins do not possess a glucose moiety and are not found to be glycosylated in nature (Strack et al., 2003; Sasaki et al., 2004). Betacyanins can be glycosylated at the cyclo-DOPA (Wyler et al., 1984; Sasaki et al., 2004, 2005) or betanidin intermediates of the pathway (Vogt et al., 1999a; Vogt, 2002). Two betanidin glucosyltransferases have been identified from Dorotheanthus bellidiformis, which vary in the position of the aromatic ring to which they add the glucose molecule. Betanidin 5-O-glucosyltransferase (B5GT) catalyses the transfer of glucose to the 5-hydroxyl group of betanidin (Vogt et al., 1999a), whereas betanidin 6-O-glucosyltransferase (B6GT) acts on the 6-hydroxyl group (Vogt, 2002). Transcriptomic analysis of differentially pigmented varieties of dragon fruit, also found a number of betalain glycosyltransferase candidate sequences in Hylocereus megalanthus, one of which resulted in a significant betalain reduction upon silencing, namely HmB5GT1 (Xie et al., 2020). A UDP-glucosyltransferase with high similarity to B5GT has also been isolated from B. vulgaris, BvGT, whose expression correlates with betalain pigmentation induced by abiotic and biotic stresses (Sepúlveda-Jiménez et al., 2005). Antisense knockdown of BvGT led to a reduction in betalain pigmentation, supporting its role in betanidin modification, although substrate specificity of BvGT has not yet been confirmed (Sepúlveda- Jiménez et al., 2005). 1.3.5.2. Cyclo-DOPA 5-O-glucosyltransferase (cDOPA5GT) An early publication showed high amounts of cyclo-DOPA glucoside in red beet plants (Wyler et al., 1984). Later studies demonstrated glycosylation at the cyclo-DOPA step could also be found in M. jalapa crude petal extracts (Sasaki et al., 2004). Thus, an alternative glycosylation scenario was proposed in which betanin and other glycosylated betacyanins can be synthesized via cyclo-DOPA modification. Later, cDNA encoding the enzyme cDOPA5GT was isolated and characterized (Sasaki et al., 2005). Increased expression of cDOPA5GT correlates with the accumulation of betanin during flower development in M. jalapa, supporting its role in betalain biosynthesis (Sasaki et al., 2005). An orthologue of cDOPA5GT was also characterized from the distantly related Celosia cristata (Sasaki et al., 2005), indicating that this class of enzyme is broadly conserved across Caryophyllales. Furthermore, M. jalapa cDOPA5GT has been routinely included in studies using multigene constructs containing the other betalain synthesis genes designed for heterologous accumulation of stable betacyanins, suggesting the importance of cDOPA5GT for betacyanin pigmentation (Polturak et al., 2016, 2017; Timoneda et al., 2018). Chapter 1. Introduction 15 1.3.5.3. Betalain acylation In addition to the structural complexity attained through glycosylation, betacyanins can undergo additional modification reactions. Acylated betacyanins have been reported from at least four different families within Caryophyllales, and can be further decorated with a variety of groups, including salicyl, malonyl, hydroxycinnamoyl and other moieties (Strack et al., 2003). Much less is understood about the enzymes responsible for the addition of these diversity of moieties, however, betacyanin acylation would probably be catalysed by either acyl-coenzyme A- dependent acyltransferases from the BAHD superfamily or serine-carboxy-peptidase-like (SCPL) acyltransferases (Tanaka et al., 2008). Hydroxycinnamoyl D-glucoses have been shown to serve as acyl donors for betalain acylation in cell cultures derived from betalain species, and their synthesis require the participation of hydroxycinnamate glucosyltransferases (HCGT) (Bokern & Strack, 1988; Bokern et al., 1992). Consistent with these observations, recent comparative transcriptomics studies have identified candidate hydroxycinnamate glucosyltransferases from M. jalapa (MjHGCT) and H. megalanthus (HmHGCT2) (Polturak et al., 2018; Xie et al., 2020). Subsequent co-expression of MjHGCT in N. benthamiana in conjunction with the betalain biosynthesis genes reportedly resulted in cinnamoyl-betanin, coumaroyl-betanin, caffeoyl-betanin and feruloyl-betanin (Polturak et al., 2018). 1.3.6. Transcriptional regulation of the betalain pathway 1.3.6.1. The role of MYB transcription factors (BvMYB1) In addition to resolving the structural genes of the betalain biosynthesis pathway, progress has been made in understanding their transcriptional regulation. MYB and basic helix–loop–helix (bHLH) transcription factors are found in all eukaryotes and are among the largest transcription factor families in plants (Xu et al., 2015). Plant MYB transcription factors are characterized by two or three imperfect repeats of the MYB DNA-binding motifs (R1, R2 and R3). Most plant MYBs belong to the R2R3 family, and normally interact with WD40 and bHLH proteins, forming what is called the MBW complex to regulate gene expression. Members of the R2R3 MYB group control anthocyanin biosynthesis as well as a number of other plant traits such as trichome and root hair formation (Ramsay & Glover, 2005; Xu et al., 2015). Several putative binding sites for bHLH and MYB transcription factors were identified in two DODA genes in Phytolacca americana (Takahashi et al., 2009). In light of this, and given the fact that betalains and anthocyanins assume similar physiological roles in planta, it was hypothesized that the same MBW complex containing MYB and bHLH transcription factors could also be participating in the regulation of betalain biosynthesis (Hatlestad et al., 2014). A BLAST search for homologues of R2R3 MYB genes implicated in anthocyanin pigmentation in a beet RNAseq A. Timoneda, PhD thesis, 2022 16 database identified a highly expressed R2R3 MYB gene, BvMYB1, which was also underexpressed in unpigmented varieties of B. vulgaris. Overexpression of BvMYB1 in white beet roots resulted in the emergence of red colouration, indicating that BvMYB1 can regulate betalain synthesis (Hatlestad et al., 2014). Consistent with this, virus-induced gene silencing of BvMYB1 in red varieties of B. vulgaris produced plants with betalain-free sectors. Finally, yeast- one-hybrid analysis showed that BvMYB1 could bind the promoters of both BvDODA and BvCYP76AD1, and that BvMYB1 can upregulate BvDODA in vivo under an inducible system. These results confirm that betalain biosynthesis is controlled by a MYB R2R3 transcription factor, BvMYB1 (Hatlestad et al., 2014). 1.3.6.2. Potential additional regulatory partners The involvement of a R2R3 MYB as a regulator of betalain structural genes is consistent with the MBW model of regulation of anthocyanin pigmentation, but BvMYB1 is unable to interact with known bHLH partners derived from anthocyanic model organisms. Furthermore, it lacks five of the seven conserved amino acids previously determined to be important for bHLH interaction (Grotewold et al., 2000; Zimmermann et al., 2004; Hatlestad et al., 2014). Resurrection of these missing bHLH interaction residues enables interaction of BvMYB1 with standard anthocyanic bHLH partners (Hatlestad et al., 2014). Unlike R2R3 MYB transcription factors controlling anthocyanic pigmentation, it is possible that BvMYB1 does not interact with bHLH partners in the regulation of betalain synthesis. However, it is equally possible that BvMYB1 interacts with an unknown bHLH protein that possesses compensatory mutations which circumvent the requirement for the canonical R2R3 MYB interaction residues (Hatlestad et al., 2014). In light of this possibility, it is interesting that a genetic linkage map for S. oleracea constructed to determine quantitative trait loci (QTLs) controlling leaf colour identified at least two bHLH transcription factors in the vicinity of a major QTL that accounts for 69.3% of colour variation (Cai et al., 2018). At least one of these bHLH genes, ORF11, is highly expressed in betacyanic tissue in S. oleracea, consistent with a role in betalain pigmentation (Cai et al., 2018). Finally, a recent study has shown that a WRKY transcription factor, HpWRKY, identified by a RNAseq screen in H. polyrhizus, can bind a W-box motif present in the promoter of a CYP76AD1 homologue from the same species (Cheng et al., 2017). WRKY transcription factors have more recently emerged as putative interacting partners with MBW complexes, controlling anthocyanin synthesis in both Arabidopsis thaliana and Petunia hybrida (reviewed in Lloyd et al., 2017). Despite the absence of canonical bHLH interacting residues in BvMYB1 therefore, these data may hint that a more canonical-type MBW complex is at work in the regulation of betalain biosynthesis. Chapter 1. Introduction 17 1.3.7. Betalain localisation and transport Other than in reproductive organs, betalain pigments have been found to be produced in a variety of other plant tissues such as the foliar epidermis and mesophyll cells (Lee & Collins, 2001). Studies in beet described red betalain colouration could be found in epidermis and parenchyma cells of petioles and hypocotyls, in cell layers surrounding foliar vasculature, and in all root tissues with the exception of the epidermis and conductive vessels (Petrus-Vancea et al., 2010). In many cacti species, betacyanins could also be detected in stems, accumulating in the hypodermis and outer layers of the chlorenchyma (Mosco, 2012). Within pigmented cells, betalains, like anthocyanins, are stored in vacuoles (Tanaka et al., 2008). In species of the genus Rebutia in the Cactaceae, betalains have also been observed to form intensely pigmented spherical bodies within the cells of coloured tepals (Iwashina et al., 1988). GFP tagging of the betalain biosynthetic enzymes (CYP76AD1, DODAα1, cDOPA5GT, B5GT and HCGT) showed they localise to both the cytoplasm and surprisingly the nucleus, suggesting betalains are synthesised in both these subcellular compartments and later transported to the vacuole (Chen et al., 2017b; Xie et al., 2020). Vacuoles can be the largest organelle in plant cells and serve in the osmotic regulation of cell volume and storage of ions and metabolites, including secondary metabolites. These functions are mediated by the vacuole’s membrane, the tonoplast, by controlling the transport of compounds in and out of the vacuole (Matile, 1978). To date, there are no reported studies on the molecular transport of betalain pigments, and our knowledge is limited to what is known for anthocyanins and other flavonoids. Flavonoid transport mechanisms have been previously divided in two main models: mass transport by vesicles from the endoplasmic reticulum (ER), and molecular transport via binding to glutathione-S-transferase (GST) transporter proteins (Grotewold & Davies, 2008). Both mechanisms require compounds to cross at least one membrane, either the vesicular membrane that will eventually merge with the vacuole or the tonoplast itself. Two classes of transport proteins have been implicated in flavonoid trans-membrane movement: multidrug and toxic extrusion (MATE) transporters and ATP-binding cassette (ABC) transporters (Grotewold & Davies, 2008). Transcriptome analysis of pigmented and non-pigmented tissues from a variety of betalain producing species identified a series of betalain-related candidate genes in M. jalapa, two of which were annotated to be ABC transporters (Polturak et al., 2018). Expression of the highest scoring annotated transporter was higher in mature petals than in immature petals and sepals (Polturak et al., 2018). Although functional analyses of the identified M. jalapa ABC transporters will be required to confirm their involvement in betalain transport, this remains to date the only report of betalain transporter candidates. A. Timoneda, PhD thesis, 2022 18 1.4. Evolutionary origins of betalain pigmentation 1.4.1. Evolutionary history of the betalain biosynthetic pathway The origin of betalains in the plant kingdom can be attributed to a number of lineage-specific gene duplications in Caryophyllales (Figure 1.4) (Yang et al., 2015; Brockington et al., 2015). First, gene duplications within the CYP76AD lineage followed by neofunctionalization have been clearly shown to give rise to two key enzymatic steps in the betalain pathway – tyrosine hydroxylase and ʟ-DOPA oxidase – both from the cytochrome P450 family (Brockington et al., 2015; Sunnadeniya et al., 2016; Polturak et al., 2016). The cytochrome P450 gene subfamily CYP76AD lineage is closely related to the CYP76T and CYP76C families of cytochrome P450 genes (Hatlestad et al., 2012). Phylogenetic analysis of the CYP76AD lineage indicates that this lineage underwent two relatively deep gene duplications within Caryophyllales, giving rise to three paralogous lineages: CYP76AD alpha (α), beta (β) and gamma (γ) (Figure 1.4b) (Brockington et al., 2015). The BvCYP76AD1 and MjCYP76AD3 genes, which possess both tyrosine hydroxylase and ʟ-DOPA oxidase activity, fall into the CYP76ADα lineage. The BvCYP76AD5, BvCYP76AD6 and MjCYP76AD15 homologues, which possess only tyrosine hydroxylase activity, are paralogues of BvCYP76AD1 and MjCYP76AD3 and belong to the β clade (Brockington et al., 2015; Sunnadeniya et al., 2016). No function has been ascribed to any homologues within the CYP76ADγ clade. Furthermore, within a similar evolutionary time-frame, phylogenetic analysis of the LigB gene lineage in Caryophyllales identified that a gene duplication occurred early in the evolution of the order, giving rise to two major clades of LigB genes, termed DODAα and DODAβ (Figure 1.4c) (Brockington et al., 2015). Consequently, all betalain-pigmented lineages of Caryophyllales contain at least two LigB genes, including one paralogue from the DODAα lineage and one paralogue from the DODAβ lineage. The DODAα lineage also exhibits numerous gene duplication events (Sheehan et al., 2020). The function of DODAβ is unknown, but a number of lines of evidence suggest that, following this duplication, neofunctionalization occurred within the DODAα lineage leading to the evolution of DODA activity: the gene duplication giving rise to the DODAα lineage occurred just before the earliest inferred origin of betalain pigmentation; all functionally characterized LigB homologues known to possess DODA activity fall into the DODAα lineage (as opposed to the DODAβ lineage); and anthocyanic lineages that arose after the origin of the DODAα lineage have generally lost or downregulated their representative of the DODAα lineage, emphasizing that DODAα is associated with betalain-specific activity (Brockington et al., 2015). More recently, functional characterization of different DODA enzymes from different betalain producing species found only one paralog in each species possesses the high ʟ-DOPA 4,5-dioxygenase activity required to catalyse the formation of betalamic acid for Chapter 1. Introduction 19 betalain production, DODAα1 (Sheehan et al., 2020). Some species, like beet, can possess up to four more DODAα paralogs. The function of the other DODAα copies remains unknown. Intriguingly, one of the betalain-specific DODAα paralogues sits in close proximity to CYP76AD1 on chromosome 2 of the B. vulgaris genome, indicating the possibility of a metabolic operon (Osbourn, 2010), and implicating genomic rearrangement in the evolution of the betalain biosynthesis pathway, at least in Amaranthaceae (Brockington et al., 2015). Figure 1.4. The phylogenetic history of major genes implicated in the betalain biosynthesis pathway that exhibit a Caryophyllales-specific gene duplication. (a) Arogenate dehydrogenase (ADH) (modified from Lopez-Nieves et al., 2018); (b) cytochrome P450 subfamily CYP76AD (modified from Brockington et al., 2015); (c) L-3,4-dihydroxyphenylalanine (ʟ-DOPA) 4,5-dioxygenase (DODA) (modified from Brockington et al., 2015). Purple lines are genes derived from betalain-pigmented species, and blue lines are genes derived from anthocyanin taxa. Grey circles highlight inferred gene duplication events. Red boxes refer to genes known to be involved with the betalain pathway, grey boxes refer to genes with unrelated or unknown functions. cDOPA, cyclo-dihydroxyphenylalanine. Figure from Timoneda et al., 2019 by Dr Sam Brockington. A. Timoneda, PhD thesis, 2022 20 All central enzymatic steps in the committed betalain pathway therefore arose by lineage- specific duplication in only two gene lineages within the core Caryophyllales before the earliest inferred origin of betalain pigmentation. But although the proximal mechanisms underlying the evolution of the betalain biosynthetic pathway are clarified by the detection of these lineage- specific duplications, the discovery of a Caryophyllales-specific ADH now provides a deeper causal explanation (Lopez-Nieves et al., 2018). The ADH isoform that shows relaxed sensitivity to negative feedback inhibition by tyrosine is the product of a gene duplication event specific to core Caryophyllales (Figure 1.4a). Consequently, most species in Caryophyllales possess at least two copies of ADH, the deregulated ADHα and the canonical tyrosine-sensitive ADHβ (Lopez-Nieves et al., 2018). Moreover, ADHα appears to be lost, downregulated or under relaxed selection in anthocyanic lineages in Caryophyllales (Lopez-Nieves et al., 2018). After duplication, ADHα acquired relaxed sensitivity to feedback inhibition by tyrosine, and consequently was more efficient in the production of tyrosine. In this context, the newly increased availability of tyrosine served as a precondition that facilitated the subsequent radiation of tyrosine-derived metabolic pathways. Within this new metabolic adaptive landscape, duplication and neofunctionalization within the CYP76AD lineage gave rise to tyrosine hydroxylase activity, leading to an increased availability of ʟ-DOPA, facilitating the later radiation of ʟ-DOPA-derived metabolic pathways (Brockington et al., 2015). Subsequent duplication and neofunctionalization within the same CYP76AD lineage gave rise to ʟ-DOPA oxidase activity, while a gene duplication and neofunctionalization in the LigB lineage gave rise to DODA activity (Brockington et al., 2015), leading to the production of betalains. 1.4.1. The mutual exclusion of betalains and anthocyanins The core Caryophyllales is a well-defined clade of angiosperms comprising c. 29 families and c. 9000 species (Bremer et al., 2003). Anthocyanins have never been detected within betalain- producing species (Bate-Smith, 1962; Clement & Mabry, 1996); however, other flavonoid- derived compounds, such as proanthocyanidins, can be found in the seed coat of some betalain-pigmented species (e.g. Spinacia) (Shimada et al., 2004, 2005). These observations imply that betalain pigmentation can substitute for the otherwise ubiquitous anthocyanic pigmentation (Bischoff, 1876; Clement & Mabry, 1996), which is the dominant form of pigmentation across land plants (Campanella et al., 2014) (Figure 1.5a). Most families within the core Caryophyllales are betalain-pigmented, with the exception of Caryophyllaceae, Molluginaceae sensu strictu, Kewaceae, Limeaceae, Macarthuriaceae and Simmondsiaceae, which have been reported to produce anthocyanins (Clement & Mabry, 1996; Thulin et al., 2016). These six anthocyanic lineages are scattered across the core Caryophyllales, either sister to or nested within betalain-pigmented lineages resulting in a homoplastic distribution of Chapter 1. Introduction 21 these two pigments (Brockington et al., 2011, 2015) (Figure 1.5b). Betalains have never been detected or reported in any of these six anthocyanic lineages, drawing a clear pattern of mutual exclusion between the two pigment types (Stafford, 1994; Clement & Mabry, 1996). Figure 1.5. Betalains are only produced in the order Caryophyllales, where they display a homoplastic distribution. (a) A generic-level phylogenetic tree of the land plants, illustrating the predominant flavonoid-based pigments (blue) and the origin of betalains (purple) to Caryophyllales. (b) A family-level tree of the Caryophyllales sensu lato, depicting the homoplastic distribution of betalains and anthocyanins within core Caryophyllales. Numbers represent the four inferred origins of betalain pigmentation: (1) Stegnospermataceae clade, (2) Amaranthaceae clade, (3) Raphide clade, (4) Portulacineae clade. From Timoneda et al., 2019 (license 5214271241328) and Sheehan et al., 2019. The distribution of anthocyanin and betalain-pigmented lineages is consistent with three possible scenarios: 1) multiple origins of betalain pigmentation, 2) a single origin of betalain pigmentation with multiple reversals to anthocyanin, or 3) a combination of the previous two. Phylogenetic reconstruction analyses using the known pigment status of extant species in Caryophyllales have been used to explore the homoplastic patterns of anthocyanin and betalain pigmentation (Brockington et al., 2011, 2015; Sheehan et al., 2020). First observations pointing at a Caryophyllales-specific duplication event in the DODA lineage, and the subsequent loss of DODAα loci in anthocyanic lineages, initially led authors to argue for a single origin of betalain pigmentation, with multiple reversals to anthocyanin pigmentation (Brockington et al., 2015). However, recent reconstructions using an updated Caryophyllales phylogeny (Walker et al., 2018) and new pigment data for families previously uncharacterized such as Limeaceae and A. Timoneda, PhD thesis, 2022 22 Simmondsiaceae (Thulin et al., 2016), currently infer up to four separate origins of betalain pigmentation in Caryophyllales: namely 1) the Stegnospermataceae clade, 2) the Amaranthaceae clade, 3) the Raphide clade, and 4) the Portulacineae clade (Figure 1.5b) (Sheehan et al., 2020).This hypothesis was further supported by the functional characterisation of DODA enzymes from a range of betalain and anthocyanin-pigmented species in the Caryophyllales, which found that repeated gene duplication events in the DODA gene lineage gave rise to polyphyletic occurrences of elevated DODA activity within the clade (Sheehan et al., 2020). 1.5. Sources and applications of betalain pigments Betalain pigments have long been valued by mankind and have been shown to be useful in a variety of industrial sectors. The wide scope of possible applications of betalains has led to a growing interest in the improvement of pigment production via the optimisation of current sources, development of alternative production methods, and the discovery of new pigment structures to enrich the betalain colour palette. 1.5.1. Commercial, medical and industrial applications of betalains The most prominent application of betalain pigments is their use as colourants in the food and cosmetic industries. Betanin is currently catalogued as a food additive under the codes E-162 in the European Commission, and 73.40 in the 21 Code of Federal Regulation (CFR) section of the U.S. Food and Drug Administration (EFSA, 2015; FDA, 2021). Betalain pigments are more soluble in water than anthocyanins and also possess up to three times more tinctorial strength (Stintzing & Carle, 2007). Their biggest advantage in the food sector, however, is their stability at lower pH ranges, between 3 and 7, which makes them more suitable for colouring low-acid and neutral foods such as meat, juices and dairy products (Stintzing & Carle, 2007). Betalains and some intermediate molecules of the betalain pathway also exhibit many important beneficial properties for human health. Documented pharmacological activities include antioxidant, anti-cancer, anti-lipidemic and anti-microbial (Gengatharan et al., 2015). Betalains extracted from a number of species of Amaranthaceae, demonstrate strong antioxidant activity in vitro, in some cases 3-4 fold higher than common antioxidants like ascorbic acid, ruthin and catechin (Cai et al., 2003). In vivo, beet extracts also correlated with an increase in antioxidant molecules and enzymes in the organism and have been shown to have a protective effect on the oxidation of LDL molecules (Tesoriere et al., 2004; Netzel et al., 2005; Zielińska-Przyjemska et al., 2009). Feeding dyslipidaemic rats with beetroot crisps prevented the rise in serum total Chapter 1. Introduction 23 cholesterol and triacylglycerol levels (Wroblewska et al., 2011). The intermediate molecule ʟ- DOPA also acts as precursor for the production of dopamine and benzylisoquinoline alkaloids (BIAs), which include the opioid analgesics, morphine and codeine, the antibiotics, sanguinarine and berberine, and other muscle relaxants and cough suppressants (DeLoache et al., 2015). Additionally, betalains have also shown anti-cancer properties in several studies. Extracts containing betalains were able to inhibit the growth of human ovarian cancer cells and cervical epithelium cancer cells (Zou et al., 2005). Betanin was also found to induce dose and time dependent apoptosis of the human chronic myeloid leukaemia cell line (K562) (Sreekanth et al., 2007). Betalains exhibit anti-bacterial, anti-malarial and anti-fungal properties. Betalain-rich extracts demonstrated a broad anti-microbial spectrum by inhibiting the growth of gram-positive and gram-negative bacteria including Salmonella thyphymurium, Staphylococcus aureus and Bacillus cereus (Velicanski et al., 2011). Anti-malarial activity of extracts from the two african folk medicine plants Amaranthus spinosus (Amaranthaceae) and Boerhavia erecta (Nyctaginaceae) was reported in parasitized mice models (Hilou et al., 2006). In addition, tobacco plants engineered to produce betalains exhibited significantly increased resistance towards grey mold (Botrytis cinerea) (Polturak et al., 2017). Betalains have also been tested as light-harvesting pigments for dye-sensitized solar cells (Zhang et al., 2008; Calogero et al., 2012). Here, betalains act as sensitizers in the conversion of visible light into electricity, by absorbing light and entering an electronically excited state that allows them to inject an electron into semiconductor nanoparticles (Calogero et al., 2012). Despite showing modest results, the use of betalains achieved higher maximum photocurrents and power conversion efficiencies than using anthocyanins (Zhang et al., 2008). The presence of carboxylic groups in betalain molecules as well as their higher oxidation potential served as an advantage for anchoring the dye to the nanocrystalline surface (Calogero et al., 2012). Pigment extracted from beet and prickly pear have also been assessed as textile dyes (Sivakumar et al., 2009; Guesmi et al., 2012; Ganesan & Karthik, 2017). There is a growing demand for environmentally friendly colourants derived from natural sources in the clothing industry, as an alternative to potentially harmful synthetic dyes. Several studies have evaluated colour strength, colour fastness and antibacterial properties of betalain dyes in modified acrylic, silk, leather fabrics and paper (Sivakumar et al., 2009; Guesmi et al., 2012; Ganesan & Karthik, 2017). A. Timoneda, PhD thesis, 2022 24 1.5.2. Current sources of betalain pigments Global demand for food colour has resulted in constant efforts to find new commercial sources of pigments and to improve the existing technologies for their extraction and processing. In 2009, the global market for food colourants experienced a growth of 2.1% reaching a value of US $1.45 billion (Khan & Giridhar, 2015). However, the commercial production of natural pigments has been lagging behind due to shortage of significant quantities of highly pigmented fresh plant tissues, lack of simple and efficient methods of extraction and purification of these products (Delgado-Vargas & Paredes-Lopez, 2003). Currently, red beet (B. vulgaris) extracts stand as the major source of betalains and the only commercially used source for their use as food colourants with an estimated contribution of 99.99% to the global betalain production (Gengatharan et al., 2015; Khan & Giridhar, 2015). Beet is widely cultivated in parts of Europe and North America, but only a small fraction of the land used to grow it is destined for pigment production (Polturak & Aharoni, 2017). However, although domesticated beet varieties have been bred for centuries to contain higher pigment concentrations, their betalain profile and colour palette is limited. This is the case of domesticated red beet, which is mostly composed of betanin (Polturak & Aharoni, 2017). In addition, beet contains geosmin and pyrazine derivatives which can affect overall product palatability and normally have to be removed with the addition of extra processing steps (Stintzing & Carle, 2007). These factors, together with the possible carry-over of contaminating microbes, have stimulated the identification of alternative betalain sources. Other commercially available sources of betalains consist of coloured swiss chard (B. vulgaris ssp. cicla), amaranth (Amaranthus sp.), quinoa (C. quinoa), prickly pear (Opuntia sp.) and dragon fruit (Hylocereus sp.), among others. Reported betalain contents in these edible sources range from 0.04-0.08 mg/g of fresh weight in swiss chard petioles to 1.99 mg/g of fresh weight in amaranth seeds, with red beet’s titers normally ranging between 0.4 and 20 mg/g of fresh weight (Khan & Giridhar, 2015). Hairy root, callus and cell suspension cultures of some betalain producing species have also been developed in an attempt to broaden pigment sources. Plant in vitro cultures offer a series of advantages including the ability to maintain aseptic, controlled and reproducible conditions, and the possibility of cultivation in bioreactors (Georgiev et al., 2008). Red beet has been the most widely used source for the establishment of in vitro cultures, achieving pigment titers of up to 25 and 28 µmol/g of dry weight in hairy roots and callus cultures of certain varieties (Girod & Zryd, 1991b; Pavlov et al., 2003). Many other species in the Caryophyllales have also been studied for pigment production in in vitro cultures, including among others, cell suspension cultures of C. argentea (Guadarrama-Flores et al., 2015), P. americana (Schliemann et al., 1996), P. grandiflora (Böhm et al., 1991), and Chenopodium rubrum yielding up to 55 µmol/g of dry weight (Berlin et al., 1986), and callus lines of coloured Chapter 1. Introduction 25 C. quinoa varieties (Henarejos-Escudero et al., 2018) and the cactus Mammillaria candida (Santos-Díaz et al., 2005). 1.5.3. Metabolic engineering of the betalain pathway for enhanced production The elucidation of the enzymes required for the core biosynthetic steps of the betalain pathway has allowed for the heterologous expression of these pigments in other plant and microbial systems. In plants, betalains have been successfully engineered in Arabidopsis, rice and members of the Solanaceae (Figure 1.6) (Polturak et al., 2016, 2017; Tian et al., 2019; He et al., 2020b). Stable heterologous expression of beet’s DODAa1 and CYP76AD1, and M. jalapa’s cDOPA5GT resulted in betalain accumulation in crops such as Solanum lycopersicum (tomato), Solanum tuberosum (potato) and Solanum melongena (aubergine), and other model systems like N. benthamiana, N. tabacum, P. hybrida and tobacco BY2 cells (Figure 1.6a,c) (Polturak et al., 2016, 2017). In N. benthamiana flowers, expression of BvCYP76AD1 or BvCYP76AD6 alone or in combination proved diversion of metabolic flux between the dividing branches of the pathway is key to obtain a bigger diversity of colouration and can be controlled through enzyme expression ratios (Figure 1.6d) (Polturak et al., 2017). Authors were also able to restrict betalain accumulation to certain organs such as tomato fruits or rice endosperm by the use of tissue specific promoters (Figure 1.6f) (Polturak et al., 2017; Tian et al., 2019). Betalain production has also been achieved in A. thaliana plants, by the expression of a multi-cistronic vector containing DODAα1, BvCYP76AD1 and cDOPA5GT linked by 2A peptides (Figure 1.6e) (He et al., 2020b). Previous to this, betacyanin production had only been possible in A. thaliana via feeding of plants with pathway substrates (Harris et al., 2012; Sunnadeniya et al., 2016). Arabidopsis seedlings expressing the DODAα1 and CYP76AD6 enzymes from beet, were able to produce betaxanthins and appeared yellow. However, when expressing DODAα1 with CYP76AD1 instead, betacyanin colouration was only observed after feeding seedlings with tyrosine (Sunnadeniya et al., 2016). It is possible that CYP76AD1’s tyrosine hydroxylase activity could be limiting in certain contexts or cell environments which are taxon specific. In fact, earlier to the identification of CYP76AD1 as the enzyme responsible for the tyrosine hydroxylase step in plants, betaxanthin production in Arabidopsis T87 cells was achieved by expression of a fungal tyrosinase from shiitake mushrooms (Lentinula edodes). Moreover, recent work in rice described betacyanin accumulation by using a tyrosinase from Aspergillus oryzae additionally to BvDODAα1 and BvCYP76AD1 (Tian et al., 2019). Due to the high antioxidant capacity of betalains and their potentially beneficial effect for consumer’s health, engineering of the betalain pathway in crops not only provides potential alternative sources of pigments but also leads to a nutritional enhancement of the resulting products, normally referred to as biofortification (Martin, 2013). Additionally, the production of betalains in non-native hosts has previously led to the A. Timoneda, PhD thesis, 2022 26 identification of novel pigment structures which arise from condensation with available compounds or the action of enzymes not present in native cell environments (Polturak et al., 2017). Figure 1.6. Betalain production can be engineered in model plants and crops. (a) Nicotiana tabacum plants, (b) Solanum melongena fruits (aubergine), and (c) Solanum tuberosum tubers (potato) constituively expressing betalains (right), versus WT (left). (d) Expression of different CYP76AD cytochromes in flowers of N. tabacum results in different ratios of betacyanins/betaxanthins and diverse colurations. From left to right: WT flower, flowers expressing CYP76AD1 only, expressing both CYP76AD1 and CYP76AD6, and expressing CYP76AD6 only. Bottom row, flowers under blue light. From Polturak et al., 2016 (license 5214270797962) and Polturak et al., 2017. (e) Arabidopsis thaliana plants expressing betalains. From He et al., 2020 (license under http://creativecommons.org/licenses/by/4.0/). (f) Specific betalain expression in Oryza sativa (rice) endosperm (right) versus WT (left). From Tian et al., 2019 (license 5214270970597). The use of microbial systems for metabolite production offers a number of advantages. Among others, microbial cultures have a fast growth rate and can be scaled up in bioreactors to volumes in accordance with industrial demand. Expression of the bacterial DODA from G. diazotrophicus in Escherichia coli and feeding with ʟ-DOPA allowed for the production of Chapter 1. Introduction 27 betalamic acid and an array of betalain structures through addition of different amines (Guerrero-Rubio et al., 2019). The use of bacteria for pigment production enabled the scale up of the reaction in 2L bioreactors which led to the production of up to 150 mg of dopaxanthin, four orders of magnitude higher than the values previously obtained by semi-synthesis methods (Gandía-Herrero et al., 2006). Betalains have also been engineered in yeast (Saccharomyces cerevisiae) via partial or complete expression of the betalain biosynthetic pathway (Hatlestad et al., 2012; DeLoache et al., 2015; Sunnadeniya et al., 2016; Polturak et al., 2016). Expression of MjDODAα1 along with a mutated version of BvCYP76AD1 with increased enzyme activity (DeLoache et al., 2015) and a number of different glucosyl-transferases resulted in betanin production without any need of feeding of cultures in yeast. The use of BvCYP76AD5 and feeding with a diversity of natural and synthetic amines allowed for the semi-synthesis of a suite of betaxanthins with a diverse colouration palette and the identification of the first betaxanthin displaying a blue colour (Grewal et al., 2018). In a later publication, authors also achieved the in vitro chemical synthesis of yet another blue betaxanthin, named BeetBlue, which was suitable for the dying of a wide variety of matrices such as yogurt, maltodextrin, cellulose, hair, cotton and silk, and showed no toxicity against zebra fish embryos and human hepatic and retinal pigment epithelial cells (Freitas-Dörr et al., 2020). The metabolic engineering of microorganisms to produce betalains is a promising approach for the industrial production of these pigments and is already being used in the biotechnological sector. This is the case of Phytolon, an Israeli start-up that engineers baker’s yeast with plant genes to produce a diverse suite of betalains for their use as natural food colourants (Phytolon, 2021). 1.5.4. Betalains as reporters and biosensors Betalain pigments have raised interest in the scientific community for their use as molecular reporters and biosensors. Here, betalains offer a series of advantages. The betalain biosynthetic pathway is relatively short and only requires transformation of two or three enzymes in order to produce colouration. Heterologous expression of betalains has already been proven possible in a wide range of phylogenetically distant heterologous hosts. And more importantly, betalain pigmentation is easily observable, traceable, and quantifiable via commonly employed laboratory techniques and equipment. Furthermore, small-molecule reporters like betalains exhibit some advantages over protein-based reporters, like fluorescent proteins, such as ease to diffuse across cell membranes and resistance to denaturing conditions, for example during cell fixation processes (Stücheli et al., 2020). A. Timoneda, PhD thesis, 2022 28 1.5.4.1. Betalains as reporters of enzyme activity One of the first works describing the development of betalain-based reporter systems involved the use of betaxanthin fluorescence as a means to follow enzyme activity (Gandía-Herrero et al., 2009). Miraxanthin V, also known as dopamine-betaxanthin, results from the condensation of betalamic acid with a molecule of dopamine, and exhibits fluorescence with maximum excitation and emission wavelengths of 465 nm and 512 nm, respectively (Cai et al., 2001). Oxidation of dopamine-betaxanthin molecules by the action of tyrosinase enzymes results in quinone compounds with a different fluorescence spectra (Gandía-Herrero et al., 2009). Authors were interested in developing a method for the continuous recording of enzymatic tyrosinase activity based on fluorescence spectroscopy and were able to use the decay in dopamine- betaxanthin fluorescence at its corresponding maximum wavelengths as a proxy for enzyme activity. This ultimately allowed for the determination of the kinetic values, Vmax and Km, of the tyrosinase enzyme (Gandía-Herrero et al., 2009). ʟ-DOPA is an important intermediate in the production of benzylisoquinoline alkaloids (BIAs) which include valuable metabolites such as morphine, codeine, hydrocodone and more (DeLoache et al., 2015). Efforts to engineer the BIA pathway in S. cerevisiae required the identification of enzymes with high tyrosine hydroxylase activity and led to the use of MjDODA to generate betaxanthin fluorescence-based biosensors for the detection of high ʟ-DOPA production (DeLoache et al., 2015). Coupling MjDODA to a library of BvCYP76AD1 mutants permitted the discovery and characterisation of two amino acid substitutions that improved enzyme performance as well as the tyrosine hydroxylase to cyclo-DOPA oxidase activity ratio, giving rise to a new CYP76AD1 mutated variant with increased ʟ-DOPA production (DeLoache et al., 2015). Similarly, betacyanin production in N. benthamiana through transient expression of the betalain biosynthetic enzymes coupled with different ADH variants was successfully used as a proxy for tyrosine production and helped characterise the highly active B. vulgaris ADHα enzyme variant in planta (Timoneda et al., 2018). 1.5.4.2. Betaxanthins as fluorescent protein dyes Structurally, all betaxanthins are the result of the condensation of betalamic acid with the amino group of amino acids and other amines. Therefore, authors hypothesised betaxanthin fluorescence could also be used as a protein marker (Cabanes et al., 2016). Condensation of betalamic acid with free amino groups present in protein chain residues, such as lysine residues, gave rise to what they referred to as protein-betaxanthin conjugates. Incubation of purified betalamic acid with bovin serum albumin (BSA), ovalbumin and trypsin, resulted in the formation of fluorescent protein conjugates (Cabanes et al., 2016). The possibility of synthesising protein-betaxanthins led to the development of methods for protein detection in Chapter 1. Introduction 29 electrophoresis gels by fluorescent staining. Incubation of protein gels after denaturing SDS- PAGE electrophoresis with purified betalamic acid, as well as commercial beet root extracts, allowed for the proper visualisation of protein bands under conventional fluorescence scanners (Cabanes et al., 2016). 1.5.4.3. Betalains in the detection of microorganisms The ability of betalamic acid to spontaneously condense with different amines has also been useful in the generation of probes for live-cell imaging of malaria processes in blood cells (Gonçalves et al., 2013). Malaria disease is caused by unicellular protozoan parasite species in the genus Plasmodium, which infect erythrocytes in the blood stream where they complete part of their life cycle (Hall et al., 2005). Condensation of betalamic acid with the fluorescent hydrophobic chromophore 7-amino-4-methylcoumarin (C120) resulted in the production of the fluorescent synthetic betalain named BtC, which was able to cross mice erythrocyte and parasite membranes, and specifically accumulate in infected cells at the point of parasite localisation (Gonçalves et al., 2013). Interestingly, no fluorescent stain was observed under similar incubations with the betaxanthin indicaxanthin, pointing at the importance of the C120 compound in parasite-specific accumulation and the diversity in molecular properties that can be accessed via this approach (Gonçalves et al., 2013). In a recent study, betalains were also successfully used for the detection of anthrax spores (Guerrero-Rubio et al., 2020c). Betalains are able to form complexes with metals like copper (Cu2+) and europium (Eu3+). However, complexation with metals results in pigment bleaching and a decrease in the absorbance and fluorescent values of the pigment solution (Guerrero-Rubio et al., 2020c). Betalain complexation can be reverted in the presence of chelating compounds, which are able to compete with betalains for the metal ions. This is the case of dipicolinic acid (DPA), which was shown to effectively release betalain pigments from Eu3+ and restore pigments’ spectrophotometric properties (Guerrero-Rubio et al., 2020c). DPA is an important component of bacterial spores of species of the genus Bacillus, such as B. anthracis, causing agent of the anthrax disease. Addition of B. anthracis spores to betalain-Eu3+ complexes produced a strong detectable colour and fluorescent signal as a result of pigment release that can be used as a sensitive, rapid and simple detection method (Guerrero-Rubio et al., 2020c). 1.5.4.4. Betalains as in vivo reporters of gene expression The placement of betalain genes under inducible transcriptional regulators and promoters also stands as a powerful approach for the metabolic engineering of biosensor and reporter platforms. These have successfully been developed in microorganisms, mammalian cells and A. Timoneda, PhD thesis, 2022 30 plants for the in vivo reporting of expression of specific genes. The CopSR regulatory system and the copQ promoter are specifically induced under the environmental presence of Cu2+ in the heavy metal resistant bacterium Cupriavidus metallidurans. The employment of this regulatory system to control the expression of MjDODA in C. metallidurans, derived in the generation of a bacterial biosensor able to produce fluorescence upon Cu2+ exposure (Chen et al., 2017a). MjDODA was also used to engineer an E. coli based dual biosensor able to detect dopamine, a central catecholamine neurotransmitter with important roles in cognitive functions (Lin & Yeh, 2017). Dopamine is formed via the decarboxylation of ʟ-DOPA and can also act as substrate for the DODA enzyme to give 6-decarboxylated betalamic acid, which ultimately derives in the spontaneous formation of fluorescent 6-decarboxylated betaxanthin pigments. By including the FeaR E. coli catecholamine-inducible transcriptional regulator system with MjDODA in the construct design, authors were able to induce cell fluorescence in the presence of dopamine (Lin & Yeh, 2017). In another example, researchers were able to develop an inducible fluorescent reporter system in human HEK293T cells by expressing the human tyrosine hydroxylase (hTH) gene under a doxycycline-driven promoter and the A. muscaria DODA (AmDODA) under a constitutive promoter (Stücheli et al., 2020). Betaxanthin production did not impair the viability of HEK293T cells and addition of doxycycline in the medium induced fluorescence production in HEK293T cells similarly to other used standard reporters (Stücheli et al., 2020). In plants, specific promoters have been successfully used to target betalain production in specific tissues and cell types (Polturak et al., 2017; Tian et al., 2019). Betalain production in tomato plants transformed with the betalain pathway genes was restricted to the fruits when the expression of CYP76AD1 gene was driven by the specific E8 promoter (Polturak et al., 2017). Similarly, in rice, expression of CYP76AD1, DODAa1 and cDOPA5GT under the control of the endosperm-specific rice globulin-1 promoter resulted in the production of betalains only in the grains (Tian et al., 2019). Betalains have also been effectively expressed in plants by creating a polycistronic transcript with the betalain genes linked by 2A peptide sequences into a single open reading frame (He et al., 2020b). With this approach, authors generated RUBY, a reporter that could non-invasively monitor gene expression in Arabidopsis and rice. Expressing the fusion protein under the Arabidopsis At2S3 seed specific promoter and YUC4 promoter involved in auxin biosynthesis resulted in specific pigment accumulation in seeds, and leaf tips and apical regions of the gynoecium, respectively (He et al., 2020b). Expression of RUBY under the auxin responsive DR5 promoter in rice calli also facilitated the selection of transformed calli in in vitro tissue cultures, and resulted in pigmentation patterns in roots correlating with typical DR5 expression (He et al., 2020b). Chapter 1. Introduction 31 1.5.5. Betalain stability and degradation Pigment stability is an important feature regarding their general applicability, especially during food processing and storage. Degradation occurs through reactions that modify the molecular structure of pigments leading to alterations in their colouration and/or functional properties. The most common modifications that result in betalain degradation consist of deglycosilation, hydrolysis (for example, betanin hydrolytic cleavage to betalamic acid and cyclo-DOPA), decarboxylation and dehydrogenation (Herbach et al., 2006a). Betalain stability is affected by a series of factors such as concentration and structure, temperature, pH, light, oxygen, water activity and metal ions. Betacyanin are generally more stable than betaxanthins both at room and high temperatures (Aronoff & Aronoff, 1948; Sapers & Hornstein, 1979; Herbach et al., 2006b). For example, betanin has a half-life value 11 times higher than vulgaxanthin under thermal treatment (Singer & von Elbe, 1980). Betalain glycosylation, acylation and esterification with aliphatic acids have also been reported to increase their stability and half-live value (Herbach et al., 2006a). Glycosylation of betanidin to betanin, lowers its reactivity with oxygen and increases its half-life value by 17 times (von Elbe & Attoe, 1985), however, stability is not always improved by further glycosylation. Pigment concentration also directly affects stability. Betalain degradation rates in prickly pear juice at high temperatures were dependent on pigment concentration, and appeared slower for higher concentrations (Merin et al., 1987). Temperature is one of the most important factors governing betalain stability, especially during food processing and storage, since food preservation normally includes thermal treatment steps to ensure food safety (Herbach et al., 2006a,b). Betalain degradation accelerates with increasing temperature and heating period, following first-order reaction kinetics, and considerably increases above 50 °C (Saguy et al., 1978). Betalain stability is maintained across a broad pH range, normally established between 3 and 7, but rapidly declines beyond those values (Herbach et al., 2006a). Alkaline treatment of beet juice was found to readily induce betanin hydrolysis to betalamic acid and cyclo-DOPA 5-O-glucoside, shifting juice colouration to yellow (Schwartz & von Elbe, 1983; Huang & von Elbe, 1987). Exposure to light also leads to betalain degradation following first-order reaction kinetics in an oxygen-dependent manner (von Elbe et al., 1974; Attoe & von Elbe, 1981; Cai et al., 1998). In addition, betalain stability also decreases linearly with oxygen concentration. Betanin degradation was reported to accelerate in the presence of oxygen and hydrogen peroxide (Pasch & von Elbe, 1979; Wasserman et al., 1984). In contrast, supplementation with antioxidants such as ascorbic acid is able to enhance betalain stability (Attoe & von Elbe, 1982). Betalain hydrolysis requires the presence of water molecules, therefore, moisture and water activity (aw) are other critical factors affecting pigment stability which can have a big impact in the storage process. Half-life value of Amaranthus dried pigment powders was found to be around 22 times higher than aqueous solutions, maintaining A. Timoneda, PhD thesis, 2022 32 colouration to up to 23 months in contrast to one and a half months (Cai et al., 1998). Finally, the presence of metal ions such as ferrous, copper, tin, aluminum and chromium cations, has been repeatedly reported to accelerate betalain degradation and the formation of metal complexes that shift the spectral properties of the pigment beyond their typical colouration (Kuusi et al., 1977; Attoe & von Elbe, 1984; Czapski, 1990). 1.6. Thesis objectives The primary purpose of this thesis is to explore innovative methods to produce betalain pigments with a special focus on maximizing production titers and the development of betalain- based reporter systems in plants. This can potentially allow for the optimization of industrial practices for pigment production in the food, pharmacological and biotechnological industry, and to provide the research community with new tools for the study of specific plant physiological processes. The aim of this PhD thesis can then be subdivided into the following two main objectives. (1) Development of an in vivo betalain-based reporter system for the visualization and tracing of arbuscular mycorrhiza symbiosis in plant roots (Chapter 2). To establish betalain pigments as in vivo reporter systems for the visualization of physiological processes we engineered the betalain pathway under the control of specific promoters that only activate gene expression under certain events or stimuli. To this end, we used the promoters of phosphate transporter (PT) and blue copper protein (BCP) genes in Medicago truncatula known to be only expressed in root cells during colonisation by arbuscular mycorrhiza to construct inducible multigene vectors for plant transformation. (2) Identification of molecular tools for the optimization of betalain pigment production methods. ( a) Exploring ADH diversity (Chapter 3) Previous work by collaborators from the University of Wisconsin identified aspartate at position 208 as the residue in the active site responsible for the relaxation in tyrosine feedback sensitivity in spinach’s ADHα (Lopez-Nieves et al., 2021). Site-directed mutagenesis of this residue to glutamate (D208E) reduced the activity of the enzyme almost by half compared to wild-type enzyme. At the same time, the inverse mutation Chapter 1. Introduction 33 from glutamate to aspartate was enough to induce relaxed sensitivity to tyrosine inhibition in plants outside of the order Caryophyllales. The directed mutation of the corresponding residue in the A. thaliana ADH (E179D) was enough to increase enzymatic activity by a 51%. Phylogenetic analysis of all available ADH sequences in flowering plants allowed us to identify species which naturally display enzyme variants with amino acids other than glutamate at the identified critical position. ADH variants from these species could also show relaxed sensitivity to tyrosine negative feedback. We select a representative number of these enzymes to assess their capacity to increase tyrosine availability for pigment production as previously done with B. vulgaris (Timoneda et al., 2018). (b) Kinetic characterisation of DODA enzymes (Chapter 4) The DODA enzyme catalyses a core step in the betalain biosynthesis pathway. The evolution of DODA is complex, and has been predicted to have occurred at least three times within the order Caryophyllales (Sheehan et al., 2020). Different DODA origins could also underly a diversity in their catalytic properties and mode of action. We assess these differences using purification and kinetic characterisation of DODA enzymes belonging to taxa representative of these independent origins with the ultimate goal of identifying enzymes exhibiting improved performance. These enzymes can then be employed for the optimization of betalain production in heterologous hosts and microbial bioreactors. (c) Molecular characterization of high activity DODA enzymes (Chapter 5) Identification of the residues responsible for the high DODA activity observed in certain Caryophyllales taxa is key to understanding the factors that lead to a more efficient catalysis and to guide the efforts to engineer novel high-performance enzymes. Ancestral sequence reconstruction of early nodes in the DODA phylogeny can be a powerful tool to help identify these residues. By comparing the sequences inferred from early low-activity ancestors and later high-activity nodes we can obtain a list of residue candidates for the establishment of high DODA activity. The contribution of these residues to the overall activity of the enzyme can then be assessed in yeast with necessity assays and confirmed through sufficiency tests. A. Timoneda, PhD thesis, 2022 34 Chapter 2 35 Chapter 2. Betalain pigments as in vivo visual markers for arbuscular mycorrhizal colonisation of root systems This chapter represents an updated development of an already published peer reviewed paper in PLOS Biology1 and was done as part of my PhD in collaboration with Dr Sebastian Schornack’s group in the Sainsbury Laboratory of Cambridge University. 1 Timoneda A, Yunusov T, Quan C, Gavrin A, Brockington SF, Schornack S (2021). MycoRed: Betalain pigments enable in vivo real-time visualisation of arbuscular mycorrhizal colonisation. PLoS Biol 19 (7): e3001326. https://doi.org/10.1371/journal.pbio.3001326 Contributions: WGA stainings, root sections and confocal microscopy of Medicago truncatula roots were performed by Dr Alex Gavrin. Stable transformation of Nicotiana benthamiana and rhizotron experiments were performed in collaboration with Dr Temur Yunusov. 2.1. Introduction Arbuscular mycorrhiza (AM) fungi of the subphylum Glomeromycotina are soil fungi that engage in symbiosis with land plants (Spatafora et al., 2016). Symbiotic associations with AM fungi date back to over 400 million years ago and can be formed by approximately 70% of extant land plant species (Parniske, 2008; Brundrett & Tedersoo, 2018; Genre et al., 2020). AM fungi are obligate biotrophs that receive all their carbon intake from the plant, which is estimated at up to 20% of the plant’s photosynthate (Bago et al., 2000). In exchange, the fungus assists the plant with the acquisition of mineral nutrients, mainly phosphorus, whose availability in soils is often a limiting factor for plant growth (Holford, 1997). Phosphorus contribution through the mycorrhizal pathway can be very high, and, in some instances, can account for the entire phosphorus consumption of a plant (Smith et al., 2003). During AM symbiosis, fungal hyphae form dichotomously branched structures, named arbuscules, within root cortex cells. Hyphal extension and arbuscule formation are accompanied by the de novo extension of a specialised plant cell membrane that separates the fungal hyphae from the plant cytoplasm (Harrison, 2005). In order to accommodate arbuscule formation, plant cells undergo a series of changes in gene expression to aid the establishment of symbiosis (Franken & Requena, 2001). Examples of such AM-induced genes include MtPT4 and MtBCP1 in the model legume Medicago truncatula. MtPT4 encodes a phosphate transporter, belonging to the phosphate transporter 1 (PHT1) subfamily, which is exclusively expressed in arbuscule-containing cells (Harrison et al., A. Timoneda, PhD thesis, 2022 36 2002). MtPT4 localises to the plant cell membrane surrounding arbuscules and participates in the acquisition of phosphate released by the fungus during symbiosis. MtBCP1 encodes a member of the blue copper protein (BCP) family and is also specifically expressed in regions of the root hosting arbuscule development during AM symbiosis (Hohnjec et al., 2005). MtBCP1 expression is strongest in arbuscule-containing cells but can additionally be observed in adjacent cortical cells (Hohnjec et al., 2005). The study of AM symbiotic processes involves the detection, visualisation and quantification of fungal colonisation. Current techniques rely on the specific staining of fungal cell walls through fast, simple and cost-effective procedures. Commonly employed methods include the use of trypan blue (Phillips & Hayman, 1970), cotton blue and Sudan IV (Nicolson, 1959), acid fuchsin (Gerdemann, 1955), ink-vinegar (Vierheilig et al., 1998) or fluorescein-labelled wheat germ agglutinin (WGA) (Bonfante-Fasolo et al., 1990). All of these methods are destructive, requiring the excision and chemical treatment of roots that are typically visualised with light or fluorescent microscopy. On the other hand, nondestructive methods for detection and visualisation of AM symbiosis offer a number of significant research opportunities but often rely on specialised equipment or are only available for certain species. For example, the foliar accumulation of blumenol-derived metabolites can function as a quantitative proxy for AM colonisation in a number of crop and model plants, with potential applications in field based QTL mapping of AM fungi-related genes (Wang et al., 2018). However, blumenol accumulation is not visible and its detection requires specialised extraction and quantification steps. Furthermore, some cereal crops and species of the Liliaceae and Fabaceae naturally produce apocarotenoid yellow pigments in roots upon mycorrhizal colonisation (Klingner et al., 1995b,a), which has been useful, for example, for the identification of maize mutants affected in symbiotic interaction (Paszkowski et al., 2006). To date, however, the use of natural pigments as visual markers of AM symbiosis is limited to select species within these families and has not been implemented in other crops and model plants. Elucidation of the core enzymatic steps of the betalain biosynthetic pathway has enabled the engineering of betalain production in a wide range of heterologous hosts, including microbes such as S. cerevisiae (Grewal et al., 2018); plant model organisms like A. thaliana, N. tabacum, and P. hybrida (Harris et al., 2012; Polturak et al., 2016, 2017); and a diversity of crops such as O. sativa (rice), S. lycopersicum (tomato), S. tuberosum (potato), and S. melongena (aubergine) (Polturak et al., 2017; Tian et al., 2019). Betalains have been used as biosensors in a number of heterologous contexts to report increased production of metabolites, such as tyrosine, dopamine, and ʟ-DOPA in E. coli and N. benthamiana (Lin & Yeh, 2017; Timoneda et al., 2018; Chou et al., 2019), to measure metabolic flux between competing pathways in the synthesis of benzylisoquinoline alkaloids (BIAs) in S. cerevisiae (DeLoache et al., 2015) and for the detection of copper by heavy metal–resistant bacteria in bioremediation processes (Chen et al., Chapter 2 37 2017a). In plants, specific promoters have been successfully used to target betalain production in specific tissues such as fruits and seed endosperm (Polturak et al., 2017; Tian et al., 2019; He et al., 2020b). Expressing the betalain biosynthetic genes under the control of the AtYUC4 promoter in A. thaliana resulted in pigment production in tissues likely to represent auxin biosynthesis activity (He et al., 2020b). Similarly, the use of the DR5 synthetic auxin-responsive promoter in O. sativa calli allowed for easier selection of transformed calli in in vitro transformation protocols (He et al., 2020b). The use of betalains as in vivo reporters offer a number of advantages: 1) the relative simplicity of betalain biosynthesis; 2) the potential for heterologous betalain expression in phylogenetically diverse hosts; and 3) the ease of betalain visualisation and quantification. Here, we present MycoRed, a betalain-based in vivo and noninvasive reporter system for the occurrence and progression of AM symbiosis in roots of the two model species M. truncatula and N. benthamiana. We leveraged known AM-responsive genes from M. truncatula to identify orthologous N. benthamiana promoters that are similarly responsive to AM fungal colonisation. Heterologous expression of betalain biosynthesis genes specifically driven by AM-responsive promoters effectively tracked AM colonisation dynamics in both species. Collectively, our work demonstrates the efficacy of betalain pigments as reliable in vivo visual markers for the previously inaccessible dynamic tracing of AM symbiosis within root systems, thereby providing a valuable resource for the plant–microbe research community. 2.2. Results 2.2.1. Betacyanins can be used to visualise AM fungus colonisation in living Medicago truncatula roots To establish a reporter system that allows for the non-invasive visualisation of AM fungal colonisation in roots, we explored the use of betalain biosynthesis genes under the control of AM symbiosis-specific plant promoters. In M. truncatula, PT4 and BCP1 are specifically expressed during root colonisation by AM fungi and are often used as markers to quantify the extent of AM symbiosis (Harrison et al., 2002; Hohnjec et al., 2005). We generated T-DNA constructs harbouring CYP76AD1, the enzyme catalysing the tyrosine hydroxylation and ʟ-DOPA oxidation steps of the betalain biosynthetic pathway (Figure 1.2) driven by either MtPT4 or MtBCP1 symbiotic specific promoters. The T-DNA also carried genes encoding the two additional enzymes needed to produce betacyanins: DODA and cDOPA5GT (Figure 1.2) under 35S and Ub10 constitutive promoters, respectively. We decided to use different promoters in A. Timoneda, PhD thesis, 2022 38 order to avoid transcriptional gene silencing induced by sequence repeat. We refer to these multi-gene vectors as MtPT4-p1 and MtBCP1-p1 (Figure 2.1a). Next, we generated M. truncatula composite plants expressing these constructs in hairy roots. Resultant transgenic roots did not display visible pigments prior to AM inoculation. After 4-weeks of colonisation with the AM fungus Rhizophagus irregularis, roots produced visibly red coloured pigments (Figure 2.1b-g). We did not observe any pigment production in mock conditions for MtPT4-p1 or MtBCP1-p1 composite roots. Figure 2.1. Betacyanins can be produced in Medicago truncatula roots as a response to AM fungi colonisation. (a) Schematic of the multi-gene vectors constructed for inducible betacyanin expression in M. truncatula roots where only the first gene of the betalain biosynthesis pathway is controlled by AM symbiosis specific promoters. Expression of MtPT4-p1 (b,e,h) and MtBCP1-p1 (c,f,i) in roots of M. truncatula 4 weeks after inoculation with R. irregularis. (d, g, j) Example of MtPT4-p1 expressing root system mock inoculated with autoclaved R. irregularis. (e, f, g) Root system images filtered for red coloring only. Scale bar (b,c,d), 1 cm; scale bar (h,i,j), 1.5 mm. Chapter 2 39 Dissection and microscopy of pigmented and non-pigmented root fragments revealed betacyanin production consistently colocalised or was adjacent to arbuscule containing cells both in MtPT4-p1 and MtBCP1-p1 roots (Figure 2.2). Pigment presence often extended beyond arbuscule containing cells in the inner cortex and was also observed in the endodermis, the pericycle, the stele and in some cases adjacent cortical cells, which represent root tissue layers that did not show any intracellular fungal structures (Figure 2.2). To confirm that the employed promoter fragments were indeed associated with intracellular fungal structures, we generated MtPT4 and MtBCP1 GUS fusions and expressed them in roots of M. truncatula composite plants (Figure 2.3). Staining pattern after fungal colonisation showed that prominent GUS substrate accumulation was limited to inner cortical cells colonised by the fungus, as previously reported (Harrison et al., 2002; Hohnjec et al., 2005). Therefore, red betacyanin distribution extends to cells beyond those with promoter activity but still specifically labels colonised tissues and is thus a promising candidate for the in vivo visualisation of AM symbiosis processes. Figure 2.2. Betacyanin accumulation in root tissues of Medicago truncatula. Betacyanin accumulation is observed at different intensities in different tissue layers and appears higher in the endodermal cells adjacent to arbuscule-containing cortical cells. Root sections of M. truncatula expressing MtPT4-p1 (a,c) and MtBCP1-p1 (b,d) 4 weeks after inoculation with Rhizophagus irregularis. (c,d) Left, betacyanin pigments are visible in red; Right, WGA-FITC staining of fungal structures in blue. Open arrows mark internal hyphae and filled arrows signal cells containing arbuscules. Root sections, WGA staining and imaging performed by Dr Alex Gavrin. Scale bar, 100 μm. A. Timoneda, PhD thesis, 2022 40 2.2.2. Betaxanthins as AM fungus colonisation markers in living Medicago truncatula roots We also assessed the usability of betaxanthin pigments as reporters for AM colonisation in roots of M. truncatula. Betaxanthin production can be engineered by using the DODA enzyme in combination with the CYP76AD5 or CYP76AD6 enzymes, which unlike CYP76AD1, are only able to catalise the first step of the betalain biosynthetic pathway (Figure 1.2). We generated multigene constructs containing CYP76AD6 driven by either MtPT4 or MtBCP1 symbiotic specific promoters and DODA under the 35S promoter. We refer to these multi-gene vectors as MtPT4-p2 and MtBCP1-p2 (Figure 2.4a). Betaxanthins are not typically glycosylated, therefore we did not need to include a glycosylation enzyme such as cDOPA5GT to the construct. M. truncatula composite plants expressing these constructs in hairy roots produced betaxanthin pigments after a 4-week colonisation with R. irregularis. However, yellow betaxanthin coloration in the roots was not easily discernible by the naked eye and became more evident under a dissecting microscope (Figure 2.4b-g). Betaxanthin pigments are also fluorescent, exhibiting spectra with excitation maxima between 463 - 474 nm and emission maxima between 509 - 512 GUS staining of M. truncatula roots expressing MtPT4::GUS (a,b) and MtBCP1::GUS (c,d). GUS staining confirms promoter expression is limited to single cells in typically arbusculated tissue layers (inner cortex). Scale bar, 100 μm. Figure 2.3. Expression pattern of MtPT4 and MtBCP1 in Medicago truncatula roots upon colonisation with Rhizophagus irregularis. Chapter 2 41 nm (Gandía-Herrero et al., 2005a). Fluorescence was detected in yellow pigmented root areas under fluorescence microscopy (Figure 2.4d,g). Dissection and microscopy of pigmented root fragments revealed betaxanthin production also colocalised or was adjacent to arbuscule containing cells both in MtPT4-p2 and MtBCP1-p2 roots (Figure 2.5). Betaxanthin’s yellow coloration and fluorescence extended beyond arbuscule containing cells in the inner cortex and, like red betacyanins, were more prominent in the endodermis, the pericycle and the stele (Figure 2.5). Betaxanthin fluorescence was easily detectable and able to mark single pigment-containing cells accurately. Therefore, yellow betaxanthin distribution also extends to cells beyond those with promoter activity but still specifically labels colonised tissues. Figure 2.4. Betaxanthins can also be produced in Medicago truncatula roots as a response to AM fungi colonisation. (a) Schematic of the multi-gene vectors constructed for inducible betaxanthin expression in M. truncatula roots where only the CYP76AD6 is controlled by AM symbiosis specific promoters. Expression of MtPT4-p2 (b-d) and MtBCP1-p2 (e-g) in roots of M. truncatula 4 weeks after inoculation with Rhizophagus irregularis. (d,g) Betaxanthin fluorescence colocalizes with yellow pigmented root areas. Scale bar (b,e), 1 cm; scale bar (c,d,f,g), 1.5 mm. A. Timoneda, PhD thesis, 2022 42 Figure 2.5. Betaxanthin accumulation in root tissues of Medicago truncatula. Betaxanthin accumulation is observed in different tissue layers and appears higher in the internal tissue layers adjacent to arbuscule-containing cortical cells. Root sections of M. truncatula expressing MtPT4-p2 (a-d) and MtBCP1-p2 (e-h) 4 weeks after inoculation with Rhizophagus irregularis. (a,e,g) Betaxanthin pigments are visible in yellow; (b,f,d,h) Betaxanthin fluorescence is shown in green; (c,d) Images taken under confocal microscopy. Arrows signal cells containing arbuscules. Root sections and fluorescence microscopy performed by Alex Gavrin. Scale bar, 100 μm. 2.2.3. Identification and validation of AM symbiosis marker genes in Nicotiana benthamiana Transiently transformed M. truncatula hairy roots display varying degrees of transgene expression that could impact reporter intensity and functionality. To avoid this, we sought to establish betalains as reporters of AM symbiosis in stable transgenics. Here, we explored the classic model species N. benthamiana, which has been extensively used to develop methods of heterologous betalain production (Polturak et al., 2016; Timoneda et al., 2018; Sheehan et al., 2020). Homologs of MtPT4 (Niben101scf02726g00004.1; NbPT5b) and MtBCP1 (Niben101Scf07438g04015.1; NbBCP1b) in N. benthamiana were previously identified by Dr Clément Quan’s (Quan, 2018). Expression of homologs NbPT5b and NbBCP1b was confirmed to increase with R. irregularis inoculation and showed significantly elevated transcript levels 3-4 Chapter 2 43 weeks post inoculation (Quan, 2018; Timoneda et al., 2021). We therefore cloned promoter regions up to -1068 bp and -1231 bp upstream of the start of NbPT5b and NbBCP1b coding sequences respectively (Appendix 1). To address tissue specificity of NbPT5b and NbBCP1b promoters, we generated promoter-GUS fusions for expression in N. benthamiana. Stable N. benthamiana transformants expressing NbPT5b-GUS and NbBCP1b-GUS fusions displayed a GUS staining pattern specific to root areas colonised by R. irregularis (Figure 2.6b-c). In both cases GUS staining was stronger in cells containing arbuscules but could also be observed in adjacent cells at varying intensities (Figure 2.6b-c). Non-colonised areas were consistently free of staining in NbPT5b-GUS and NbBCP1b-GUS roots. Altogether, the results support the use of NbPT5b and NbBCP1b promoters for the construction of a betalain-based AM symbiosis reporter system in N. benthamiana. Figure 2.6. Expression of NbPT5b and NbBCP1b genes is induced under colonisation with Rhizophagus irregularis in Nicotiana benthamiana. GUS staining of (a) NbPT5b-GUS and (b) NbBCP1b-GUS expressing N. benthamiana roots 4 weeks after inoculation with R. irregularis. GUS activity can only be observed in root areas colonised by R. irregularis and is more predominant in arbuscule containing cells (white arrows). Root sections and imaging performed by Dr Temur Yunusov. Scale bar, 500 μm A. Timoneda, PhD thesis, 2022 44 2.2.4. Betacyanins visualise AM fungus colonisation in living Nicotiana benthamiana roots We then generated multi-gene betalain reporter constructs for stable transformation in N. benthamiana. A first generation of vectors placed CYP76AD1 under the control of the NbPT5b or NbBCP1b promoters, with DODA and cDOPA5GT driven by 35S and Ub10 constitutive promoters, respectively. Hereafter we refer to these multi-gene vectors with one AM fungal-colonisation specific promoter as NbPT5b-p1 and NbBCP1b-p1 (Figure 2.7a). We generated 16 independent transgenic N. benthamiana NbPT5b-p1 lines and 12 independent NbBCP1b-p1 lines. NbPT5b-p1 and NbBCP1b-p1 plants produced betacyanins in the roots upon colonisation with R. irregularis (Figure 2.7b-c). Betacyanin production was not visible in mock conditions for NbPT5b-p1 or NbBCP1b-p1 plants inoculated with autoclaved R. irregularis. Dissection and microscopy of pigmented and non-pigmented root fragments revealed betacyanin presence co-localised with fungal colonisation structures (Figure 2.7d-g). Figure 2.7. Betacyanins can be produced in Nicotiana benthamiana roots as a response to AM fungi colonisation. (a) Schematic of the multi-gene vectors constructed for inducible betalain expression in N. benthamiana roots where only the first gene of the betalain biosynthesis pathway is controlled by AM symbiosis specific promoters. Expression of NbPT5b-p1 (b) and NbBCP1b-p1 (c) in whole root systems of N. benthamiana T1 plants 4 weeks after inoculation with Rhizophagus irregularis. Right, root images are filtered for red color. (d-g) Confocal microscopy of red NbPT5b-p1 (d,f) and NbBCP1b-p1 (f,g) transgenic root sections which were cut and stained with WGA-FITC to visualise fungal structures. (d,f) color-filtered for red; (e,g) WGA-FITC green fluorescence. Root sections and fluorescence imaging performed by Dr Temur Yunusov. Scale bar (b,c), 1 cm, scale bar (d-g), 100 μm. Chapter 2 45 To assess the extent to which red pigment co-occurs with intraradical fungal structures, we divided NbPT5b-p1 and NbBCP1b-p1 T1 plant root systems into red-pigmented and non- pigmented fragments and ink stained them separately for the presence of fungal structures (Figure 2.8). We then imaged the stained root fragments and divided each root image into sections of similar length for the scoring of fungal presence. For both NbPT5b-p1 and NbBCP1b-p1 root systems, we found that all of the betacyanin-positive root fragments contained fungal structures and were extensively colonised (Table 2.1). White root segments of the NbPT5b line were overall much less likely to harbour fungal structures (max. 22.9% intraradical hyphae) compared to NbBCP1b (88.2%). Furthermore, the extent of fungal structures within individual white root segments was also much lower in NbPT5b than in NbBCP1b roots (4.8% vs 46.4%) (Table 2.1). This suggests that NbPT5b-driven CYP76AD1 more reliably labels root segments containing fungal structures. Red root sections predominantly displayed colonisation by arbuscules and hyphae and to a lesser extent vesicles, suggesting that accumulation of pigment in N. benthamiana is also associated with fully developed AM symbiosis but may decline at late stages (Table 2.1). Figure 2.8. Nicotiana benthamiana roots expressing betalains contain more fungal structures than white roots. Ink staining of Nicotiana benthamiana roots expressing NbPT5b-p1 (a-d) and NbBCP1b-p1 (e-h) after inoculation with Rhizophagus irregularis. (a,b,e,f) Root fragments that displayed betalain colouration before ink staining are represented by a red ‘+’ sign. (c,d,g,h) Root fragments that displayed no colouration before ink staining are represented by a grey ‘-’ sign. Red betalain pigmentation is lost during the ink staining process. (b,d,f,h) are amplified images of the area delimited by the dashed squares. Scale bar, 1 mm. A. Timoneda, PhD thesis, 2022 46 Table 2.1. Quantification of fungal structures observed in roots of Nicotiana benthamiana expressing NbPT5b- p1 and NbBCP1-p1 4 wpi with Rhizophagus irregularis. We cut and divided roots into pigmented (red rows) and nonpigmented (noncoloured rows) fragments for ink staining. N refers to the total number of root fragments analysed for each condition. Table shows the number of root fragments containing fungal structures and the average extent of colonisation by these structures over the length of the root fragment. Error is shown as standard error. %AC, percentage of arbuscule colonisation; %IHC, percentage of internal hyphae colonisation; %VC, percentage of vesicle colonisation; A, arbuscules; IH, internal hyphae; V, vesicles; wpi, weeks postinoculation. 2.2.5. Betaxanthins as AM fungus colonisation markers in living Nicotiana benthamiana roots Given that betacyanin expression under the NbPT5b promoter better reports total root colonisation, we decided to also assess betaxanthin performance in N. benthamiana under the NbPT5b promoter. We therefore constructed a multi-gene vector harboring DODA under the constitutive 35S promoter and CYP76AD6 under the control of the NbPT5b promoter, which we here on refer to as NbPT5b-p2 (Figure 2.9a). We generated 17 independent N. benthamiana NbPT5b-p2 lines, which were able to produce betaxanthins in their roots after 4-week colonisation with R. irregularis. Yellow pigmentation was not easily distinguishable for the naked eye but was more evident under the dissecting microscope (Figure 2.9b-c), as observed for M. truncatula. We were also able to detect betaxanthin fluorescence in yellow pigmented areas, which was found to colocalise with AM fungal structures in N. benthamiana (Figure 2.9d-e). Therefore, betaxanthin pigments can also report presence of AM structures in plant roots, however their yellow color makes them hard to be easily distinguished without the use of fluorescent microscopy. Due to the advantages in the observability and detectability of red betacyanins over betaxanthins, we decided to discontinue the betaxanthin reporter lines and resume all further work with betacyanin producing constructs in N. benthamiana. Chapter 2 47 Figure 2.9. Betaxanthins can be produced in Nicotiana benthamiana roots as a response to AM fungi colonisation. (a) Schematic of the multi-gene vectors constructed for inducible betaxanthin expression in N. benthamiana roots where only CYP76AD6 is controlled by the NbPT5b AM symbiosis specific promoter. (b) Expression of NbPT5b-p2 in whole root systems of N. benthamiana T1 plants 4 weeks after inoculation with Rhizophagus irregularis. (c) Yellow pigmentation is distinguishable under higher magnification. (d-e) Confocal microscopy of yellow NbPT5b-p21 transgenic root sections: (d) bright field; (e) betaxanthin fluorescence is shown in green. Bright signal corresponding to AM hyphae can also be detected due to autofluorescence. Root section and fluorescence imaging performed by Dr Temur Yunusov. Scale bar (b), 1 cm; scale bar (c), 1.5 mm; scale bar (d-e), 100 μm. 2.2.6. Stable expression of NbPT5b-p1 and NbBCP1b-p1 can cause shoot developmental defects in Nicotiana benthamiana Surprisingly, many NbPT5b-p1, NbPT5b-p2 and NbBCP1b-p1 transgenic lines displayed vegetative developmental defects of varying severity (Figure 2.10). Specifically, 13 out of 16 NbPT5b-p1, 8 out of 17 NbPT5b-p2, and 11 out of 12 NbBCP1-p1 lines were affected with moderate to severe defects. Aberrant phenotypes were not visible in early stages of development but became prominent during maturation and flowering. Affected plants were smaller and displayed altered leaf morphology, ranging from curvy leaf edges to acute deformation of leaf shape and thickness (Figure 2.10d,f). We also observed defects in flower morphology, including stunted or absent perianth and in more severe cases missing floral reproductive organs (Figure 2.10e,g). Only plants with mild to no phenotypic alterations were able to set seed, although seed set was not abundant. To address whether these phenotypes were linked to the expression of constitutive or AM specific promoter-driven betalain genes, we performed semiquantitative RT-PCR on abnormal leaves of NbPT5b-p1 transgenic plants. As expected, we detected 35S promoter-driven DODA expression and cDOPA5GT in all analysed T0 lines, but unexpectedly DODA expression was increased in severely affected lines (Figure 2.10h). Surprisingly, we detected expression of AM responsive PT5b promoter-driven A. Timoneda, PhD thesis, 2022 48 CYP76AD1 in T0 lines with light and moderate developmental defects (Figure 2.10h). Expression intensity of CYP76AD1 appeared to decrease with phenotype severity and was barely detectable in lines displaying acute developmental defects. Figure 2.10. Expression of NbPT5b-p1 in Nicotiana benthamiana stable transformants leads to developmental defects. Figure shows examples of T0 plants which varied in the severity of observable developmental defects and were phenotypically classified as severe (S; a,d,e), moderate (M; b,f,g) and light (L; c). Only plants with a light phenotype were able to set seeds, although not abundantly. Defects affected overall plant size, and leaf and flower morphology. Severely affected plants displayed thicker leaves with strong deformation commonly showing a bifurcated shape (d) and flowers lacking reproductive organs (e). Moderately affected plants displayed altered leaf elongation with curvy edges (f), and flowers with a shorter (or lacking) perianth (g). Light phenotypes included slight leaf deformation, and flowers with shorter perianth but typical overall plant growth (c). (h) Semiquantitative RT-PCR analysis of betalain biosynthetic transgene expression in leaves of N. benthamiana T0 stable NbPT5b-p1 transformants. Line 18 displayed light developmental defects and was able to set seed, line 6 displayed moderate developmental defects, and lines 1 and 15 displayed severe developmental defects. NbEF1α is used as housekeeping control. Positive control (+) is NbPT5b-p1 plasmid DNA. Negative control (-) is water. -RT stands for minus reverse transcriptase control generated from line 18. Chapter 2 49 T1 seedlings from various NbPT5b-p1 transformant lines developed smaller leaves and shorter roots than WT seedlings (Figure 2.11), but were nevertheless able to grow to adult stage, flower and set seed. In addition to these defects, we also detected unanticipated accumulation of betanin by HPLC in leaves of T1 plants that descended from NbPT5b-p1 lines with mild to no phenotypic alterations (Figure 2.12). Leaf betanin accumulation arose in both inoculated and mock conditions. In summary, we found that expression constructs with two constitutive biosynthesis genes frequently resulted in both developmental aberrations and unexpected aerial betalain expression. Figure 2.11. One-week old Nicotiana benthamiana T1 seedlings from NbPT5b-p1 and NbBCP1b-p1 expressing lines. NbPT5b-p1 lines developed shorter roots and smaller leaves, which also appeared lightly pigmented in some cases. NbPT5b-p1 lines 18, 19 and 23, and the NbBCP1b-p1 line 32, were able to produce pigments upon colonisation with Rizhophagus irregularis (not shown). NbPT5b-p1 line 23 didn’t produce any pigments upon colonisation (not shown). A. Timoneda, PhD thesis, 2022 50 Figure 2.12. Betanin detection by HPLC performed on T1 pigmented leaf tissue of NbPT5b-p1 expressing Nicotiana benthamiana lines. (a) Some T1 plants show pigmentation in the leaves at varying degrees. (b) Liquid chromatography of pigmented leaves and controls. Pigmented leaves of NbPT5b-p1 expressing plants both in inoculated (myc) and non-inoculated (mock) conditions showed a peak at the expected retention time for betanin (red dotted line). Betanin standard is a commercially available Beta vulgaris extract and displays two peaks corresponding to the two betanin isomers commonly present in beet extracts. NbPT5b::GUS leaf tissue of same age T1 plants was used as negative control. 2.2.7. Expression of all betalain synthesis genes under AM symbiosis specific promoters ameliorates defective phenotypes in Nicotiana benthamiana To avoid the developmental defects associated with the constitutive expression of betalain genes in N. benthamiana, we sought to express all three betalain biosynthesis genes under AM symbiosis specific promoters. Here, a second generation of vectors placed CYP76AD1, DODA and cDOPA5GT under the control of either the NbPT5b or NbBCP1 promoters. Hereafter, we refer to these multi-gene vectors as NbPT5b-p3 and NbBCP1b-p3 (Figure 2.13a). To test their Chapter 2 51 performance we generated 11 independent transgenic NbPT5b-p3 lines and 7 NbBCP1b-p3 lines and grew them with R. irregularis spores. Root system pigmentation patterns of the NbPT5b-p3 and NbBCP1b-p3 reporter constructs performed similarly to NbPT5b-p1 and NbBCP1b-p1 (Figure 2.17). Colonisation extent and fungal structures frequency in inoculated root systems did not differ between WT and NbPT5b-p3 and NbBCP1b-p3 reporter lines (Figure 2.14). Once again, plants grown in mock conditions did not harbour any observable coloration despite being grown in non-sterile substrate, further suggesting reporter expression is not activated by the presence of other microorganisms. Figure 2.13. Betacyanin production in Nicotiana benthamiana roots as a response to AM fungi colonisation via expression of the entire betalain pathway under AM symbiosis specific promoters. (a) Schematic of the multi-gene vectors constructed for inducible betalain expression in N. benthamiana roots with the three betalain biosynthetic genes controlled by the NbPT5b or the NbBCP1 promoters. Expression of NbPT5b-p3 (b,c) and NbBCP1-p3 (d,e) in roots of N. benthamiana 4 weeks after inoculation with Rhizophagus irregularis. Betalain production was not visible in mock conditions for any NbPT5b-p3 or NbBCP1b-p3 roots. (c,e) are an amplification of the areas delimited by the dashed squares. Ar, arbuscules; en, endodermis; ep, epidermis; ih, internal hyphae; v, vesicle. Scale bar (b,d), 500 μm; scale bar (c,e), 100 μm. A. Timoneda, PhD thesis, 2022 52 Figure 2.14. Quantification of root AM colonisation shows no differences between Nicotiana benthamiana WT and reporter lines. N. benthamiana plants were inoculated with Rhizophagus irregularis, and ink stained 4 weeks after inoculation (wpi). (a) Colonisation extent in N. benthamiana WT versus lines expressing NbBCP1b-p3 (BCP1-19 and BCP1-24) and NbPT5b- p3 (PT5b-16 and PT5b-21). Colonisation extent is calculated as the percentage of root containing arbuscules, vesicles or internal hyphae over the total root system. (b) Frequency of arbuscules and vesicles recorded in colonised roots of N. benthamiana WT versus lines expressing NbBCP1b-p3 and NbPT5b-p3. Frequency is calculated as the percentage of root containing arbuscules or vesicles over the extent of root colonised by R. irregularis. Individual points represent individual plants. Error bars represent standard errors. Dissection and microscopy of pigmented and non-pigmented NbPT5b-p3 and NbBCP1b-p3 root fragments inoculated with R. irregularis revealed once more that betalain presence co-localised with fungal colonisation structures (Figure 2.13). Arbuscule formation could be observed in all cortex cell layers in N. benthamiana. In NbPT5b-p3 and NbBCP1b-p3 roots, betalain pigmentation was confined to single cells that were clearly distinguishable. Pigmentation was however not restricted to arbuscule-containing cells and could also be detected in adjacent cells with no apparent arbuscule structures including cells of the endodermis (Figure 2.13). Similarly, some arbuscule-containing cells in the proximity of red pigmented cells appeared unpigmented both in NbPT5b-p3 and NbBCP1b-p3 roots (Figure 2.13). Importantly, only one out of the 11 lines in the NbPT5b-p3 genotype, and 2 out of the 7 lines in the NbBCP1b-p3 genotype, displayed the developmental defects seen with NbPT5b-p1 and NbBCP1b-p1 reporter constructs. Finally, the T1 generations derived from NbPT5b-p3 and NbBCP1b-p3 lines did not exhibit any discernible pigmentation in shoot tissues. Therefore, the expression of all betalain biosynthetic pathway genes under AM-specific promoters ameliorates the incidence of developmental alterations and non-specific betalain production in N. benthamiana plants. We also confirmed the functionality of constructs where all betalain synthesis genes are driven by arbusculated cell specific promoters in M. truncatula. We used the MtPT4 promoter to drive the expression of betalain genes and included a constitutively expressed DsRed marker to facilitate selection of transformed hairy roots. Betalain colouration was produced in colonised roots and was associated with arbuscule presence (Figure 2.15). However, once again staining Chapter 2 53 was not restricted to arbuscule-containing cells of the inner cortex and could be mainly observed in internal tissue layers like the endodermis, pericycle and stele (Figure 2.15j). M. truncatula dmi3 mutants (Doesn’t Make Infections 3) fail to progress in intraradical AM fungal colonisation and are unable to engage in symbiosis (Lévy et al., 2004; Mitra et al., 2004). Therefore, reporter expression in dmi3 mutants should not harbor any coloration regardless of inoculum presence. Notably, red pigmentation was not observed in composite roots of M. truncatula dmi3 mutants (Figure 2.16). The efficiency of betalain accumulation as a marker of AM colonisation with MtPT4 was similar to that observed for NbPT5b in N. benthamiana plants (Table 2.2). Figure 2.15. Betacyanin production in Medicago truncatula roots as a response to AM fungi colonisation via expression of the entire betalain pathway under an AM symbiosis specific promoter. (a) Schematic of the multi-gene vector constructed for inducible betalain expression in M. truncatula roots with the three betalain biosynthetic genes controlled by the MtPT4 promoter. (b-e) Expression of MtPT4-p3 in roots 4 weeks after inoculation with Rhizophagus irregularis led to pigment accumulation in roots. (f-i) Mock inoculated roots expressing MtPT4-p3 didn’t show any pigment production. (b,f) Images taken under reflective light, (c,g) are filtered for red colouring only. (d,h) Close up root images in bright field, (e,i) and filtered to observe DSRed fluorescence. (j) Root section of M. truncatula expressing MtPT4-p3. Betalain accumulation correlates with arbuscule formation but can be observed mainly in the endodermal layer adjacent to arbuscule-containing cortical cells, pericycle and steele. Open arrows mark internal hyphae and filled arrows signal cells containing arbuscules. Root section performed by Dr Alex Gavrin. Scale bar (b,f), 1 cm; scale bar (e,i), 1.5 mm; scale bar (j), 150 μm. A. Timoneda, PhD thesis, 2022 54 Figure 2.16. Hairy roots of dmi3 mutant Medicago truncatula plants expressing MtPT4-p3 do not produce any perceptible betalain colouration 4-weeks after inoculation with Rhizophagus irregularis. (a-b) Example of an A17 M. truncatula root system expressing MtPT4-p3 able to produce betalains upon inoculation. (c- d) Example of dmi3 M. truncatula root system expressing MtPT4-p3. (a,c) Images taken under reflective light, (b,d) are filtered for red colouring only. Scale bar, 1 cm. Table 2.2. Quantification of fungal structures observed in hairy roots of Medicago truncatula expressing MtPT4- p3 4 weeks post-inoculation with Rhizophagus irregularis. We selected roots for DSRed fluorescence, and then cut and divided them in pigmented (red rows) and non-pigmented (non-coloured rows) fragments for ink staining. N refers to the total number of root fragments analysed for each condition. Table shows the number of root fragments containing fungal structures, and the average extent of colonisation by these structures over the length of the root fragment. IH, internal hyphae; A, arbuscules; V, vesicles; %IHC, percentage of internal hyphae colonisation; %AC, percentage of arbuscule colonisation; %VC, percentage of vesicle colonisation. Error shown as standard error. Chapter 2 55 2.2.8. Promoter-controlled betalain biosynthesis allows for dynamic tracing of root colonisation processes Expression of the betalain biosynthesis genes under the control of NbPT5b and NbBCP1b promoters in N. benthamiana also allowed for the non-invasive imaging of AM fungal root colonisation in root systems. We grew NbPT5b-p3 and NbBCP1b-p3 transgenic N. benthamiana lines in a vertical rhizotron-based setup with R. irregularis spores that maintained root system architecture by separating roots from the substrate with a porous black cloth. This also allowed for non-disruptive imaging on a flatbed scanner and subsequent image analysis using a selective color filter for red pigment. At 52 days post inoculation (dpi), plants had developed significant root stretches with red pigmentation in both NbPT5b-p3 and NbBCP1b-p3 lines that was easily discriminated by computer imaging (Figure 2.17). Figure 2.17. Red pigment distribution in root systems of NbBCP1b-p3 and NbPT5b-p3 Nicotiana benthamiana plants colonised by Rhizophagus irregularis on rhyzotron setup. Images were taken at 52 dpi. (a,c) are reflective light images, (b,d) are filtered for red coloring only, (e) high magnification image of (d) showing varying red colouration in parallel running roots. Rhizotron set up performed by Dr Temur Yunusov. Scale bar = 1.3 cm. To assess AM colonisation dynamics over time, we also acquired consecutive images of intact NbBCP1b-p3 and NbPT5b-p3 rhizotrons at different days. This revealed increasing red pigment development within the root systems (Figure 2.18, Figure 2.19). Betalain pigmentation served as a reliable and long-lasting indicator of AM colonisation, as we did not observe root stretches that displayed red pigmentation and later lost it. Together these data demonstrate that red A. Timoneda, PhD thesis, 2022 56 pigmentation effectively traces fungal colonisation over time allowing for the dynamic and non- invasive assessment of root sections where the AM responsive promoters have been activated. (a) 51 dpi 53 dpi 56 dpi 59 dpi 64 dpi 67 dpi 71 dpi (b) Figure 2.18. NbBCP1b promoter-controlled betalain biosynthesis allows for dynamic tracing of root colonisation processes. Transgenic NbBCP1-p3 plants grown in a rhizotron setup supplied with Rhizophagus irregularis spore inoculum imaged over time. (a) Reflective light images, (b,c) represent the same images filtered for red/magenta hues, (c) is an amplification of the area delimited by the dashed square. dpi, days post inoculation. Scale bar, 1 cm. Figure 2.19. NbPT5b promoter- controlled betalain biosynthesis allows for dynamic tracing of root colonisation processes in Nicotiana benthamiana. Transgenic NbPT5b-p3 plants grown in a rhizotron setup supplied with Rhizophagus irregularis spore inoculum imaged over time. (a) Reflective light images, (b) represent the same images filtered for red/magenta hues. dpi, days post inoculation. Set up and imaging performed by Dr Temur Yunusov. Scale bar, 1 cm. Chapter 2 57 2.3. Discussion Arbuscular mycorrhizal (AM) symbiosis is a fundamental and widespread trait in plants that greatly expands the root-surface area for nutrient uptake. AM symbiosis is consequently a key agronomic trait in the drive to enhance future crop yields through environmentally sustainable mechanisms. However, enhancing AM fungal association in crop species requires the tools to understand its fundamental dynamics across varying agronomic and ecological contexts. To this end, we demonstrate the use of betalain pigments as in vivo visual markers for the occurrence and distribution of AM fungal colonisation in the roots of M. truncatula and N. benthamiana. We have generated multi-gene vectors in which AM colonisation specific plant promoters control the expression of core betalain synthesis enzymes in the production of betalain pigments. We show that AM-specific promoter-controlled betalain pigmentation is a powerful macroscopic tool to report and trace fungal colonisation in vivo along the root (Figure 2.20). Figure 2.20. Schematic of the use of betalains as markers for AM colonisation in plant roots. Left, red betacyanin pigmentation is easily observable in whole plant root systems. Right, red pigmentation is most prominent in colonised tissues as well as in adjacent tissue layers. Macroscopically, identification and tracking of root colonisation with red betacyanins was much clearer and faster than when expressing yellow betaxanthins. Betacyanin colouration was specifically limited to regions of the root colonised by R. irregularis in both M. truncatula and N. benthamiana, as we detected few to no false positives. Furthermore, no pigmentation was detected in mutant lines of M. truncatula impaired in AM symbiosis. Quantification of fungal structures revealed that pigmented root fragments were extensively colonised, showing arbuscule-containing cells over the majority of the root length for all tested promoters. We also observed some fungal structures in a low percentage of non-pigmented root fragments. In these cases, arbuscule colonisation was restricted over the length of the analysed root fragments, with the exception of NbBCP1b expressing N. benthamiana roots that exhibited a greater degree of unpigmented yet colonised root fragments. Thus, reporter ability to document the totality of A. Timoneda, PhD thesis, 2022 58 colonisation events depends on the promoter chosen for vector construction. It appears that the NbBCP1b promoter-containing constructs are not equally activated in all colonisation events, or perhaps pigment accumulation becomes evident only at certain stages of colonisation. Promoter sequences of PT4 homologs such as NbPT5b were able to report the majority of colonisation events in the root system and are therefore preferred for reporting total root colonisation. A future solution to document total AM colonisation could involve the establishment of systems whereby the betalain biosynthesis genes are activated by trans-activators, which could be then driven by promoters that are active at early, main and late AM fungal colonisation stages (Takeda et al., 2009; Xue et al., 2018; Floss et al., 2017). We also conclude red betacyanins are preferred over yellow betaxanthins for a clearer and faster macroscopic identification and tracking of root colonisation. Overall, betalain pigmentation effectively reports fungal colonisation with very little to no error and allows for simple selection of colonised root areas. Microscopy of pigmented tissues revealed that betalain accumulation extended beyond arbusculated cells and into adjacent cell layers in both M. truncatula and N. benthamiana. In M. truncatula, betalain pigmentation was most prominent in the endodermis and pericycle cells adjoining arbusculated inner cortical cells (Figure 2.2). In N. benthamiana, we also observed betalain pigmentation in the endodermis and pericycle cells, yet pigmentation was more strongly retained in cortical cells (Figure 2.13). However, colouration in N. benthamiana was sometimes present in non-arbusculated cells adjacent to arbusculated cortical cells, consistent with previously observed GUS staining patterns. These extended patterns occur even when using constructs where all three biosynthesis enzymes are driven by AM specific promoters, and cannot therefore be attributed to constitutive gene expression artefacts. Such pigment accumulation in non-arbusculated cells may be the result of a former colonisation or of moderate degree of pigment migration. Betalains are water-soluble pigments and, in native betalain-pigmented species, are produced in the cytoplasm and then stored in vacuoles (Chen et al., 2017), although the mechanisms responsible for the intracellular transport of betalains are unknown. As small water-soluble compounds, betalains may have the potential to move symplastically through root plasmodesmata, although this remains unproven. Transient expression of betalains in leaves of N. benthamiana leads to macroscopically well delimited pigmented areas (Polturak et al., 2015; Timoneda et al., 2018; Sheehan et al., 2019), but cellular migration across the boundary tissue has not been studied. Unpigmented cells harbouring arbuscules could also be a result of mechanical disruption during the sectioning of root tissues for microscopy, where cells that previously contained betalains lose pigmentation after being sliced open. Nevertheless, the betalain reporter system remains a highly effective marker of AM colonisation, especially at the macroscopic level. Chapter 2 59 The betalain biosynthetic pathway has been previously constitutively expressed in several plant species, including plants of the Solanaceae, with no report of observable developmental defects (Polturak et al., 2015; Polturak et al., 2017). Yet, in our study, expression of betalain producing constructs where only CYP76AD1 was under the control of AM-specific promoters led to developmental defects in N. benthamiana lines. Affected plants displayed dwarf phenotypes, altered leaf and flower morphology, and were unable to set seed (Figure 2.10). Our findings could be explained by a mechanism whereby constitutive expression of DODA in absence of CYP76AD1 expression cause developmental abnormality. Four observations are consistent with this hypothesis: 1) expression of constructs containing all three biosynthesis enzymes under symbiosis specific promoters substantially decreased the number of affected plants; 2) DODA expression appeared enhanced in leaves of severely affected plants; 3) expression of CYP76AD1 was detected in leaves of plants with mild defects and seemed to decrease with phenotype severity; 4) defective phenotypes were also observed in a big proportion of betaxanthin producing transformants, which do not express cDOPA5GT. We speculate that, in absence of ʟ-DOPA, the DODA enzyme could be promiscuously acting on alternative N. benthamiana substrates, with negative developmental consequences. This effect is alleviated by the presence of CYP76AD1 which provides the DODA enzyme with an abundance of its native substrate, ʟ-DOPA, resulting in the production of inert betalain pigmentation. Constitutive DODA expression may therefore create a shoot developmental conflict that can be partially mitigated by compensatory expression of CYP76AD1. Our selection process could therefore have been biased towards T0 plants with a degree of escaped CYP76AD1 expression, which would also explain the presence of betanin in a number of T1 plants descending from CYP76AD1 shoot-expressing lines (Figure 2.12). Further experimentation is required to support this hypothesis, but in any case developmental defects and vegetative betanin expression can be avoided when all betalain biosynthesis enzymes are driven by AM symbiosis specific promoters. The most powerful application of our betalain-based AM reporter lies in its ability to non- invasively document colonisation in fully developed root systems over time (Figure 2.18, Figure 2.19). This will facilitate answering important questions in the field of AM symbiosis and its potential for agronomic improvement. These include understanding the dynamics of root system colonisation, including the time difference between lateral root emergence and its colonisation, or the differential colonisation susceptibility of different root orders (Gutjahr et al., 2009). Colonisation can be assessed in scenarios where plants compete with each other (Whiteside et al., 2019) or with shoot pests (Charters et al., 2020), different nutrient regimes, CO2 concentration, temperatures, soil structures (Rilling & Mummey, 2006) and other abiotic factors. A second important application is in the survey of induced plant genetic variation or fungal natural genetic variation that impacts on root system colonisation. Here its use is only limited by A. Timoneda, PhD thesis, 2022 60 the transformability of the plant species, and an important future step will be to test our approach in monocot crops, some of which have already been engineered to express betalain pigments in grains (Tian et al., 2020). Finally, a third application derives from the ability to bulk collect betalain-pigmented root sections produced under AM fungal colonisation, especially when there is very little overall colonisation as is common in some symbiosis signalling pathway mutants or when colonisation is restricted to specific roots. Here, stage-specific early or late promoters will unlock targeted transcriptomic, proteomic or metabolomic analysis of colonised roots without the dilution effect of non-colonised tissues. Such enrichment strategies may for example help in understanding communication mechanisms between AM fungi and plants. Targeted sampling of pigmented roots will also simplify low throughput microscopy such as cryo-electron microscopy and drastically improve the signal-to-noise ratio inherent in current sampling processes. Selective sampling at red-to-white transition zones may additionally aid time-lapse microscopy of expanding fungal colonisation arrays within roots. In summary, betalains are plant pigments with a strong potential as visual markers for the study of physiological and developmental processes in plants and microorganisms. Here, we have expanded the application of these versatile pigments to add to an ever increasing range of betalain-based technologies as reporters of AM fungi colonisation in plant roots. MycoRed complements currently used fungal visualisation techniques, and constitutes a powerful tool which will be of great value for the plant-microbe research community to advance knowledge in the field of AM symbiosis. 2.4. Methods 2.4.1. Plant material and growth conditions M. truncatula A17 and dmi3 lines used were described in Rey et al., (2015). M. truncatula plants were grown in dried industrial sand at 21°C day temperature, 19°C night temperature, 65% humidity and a 16 h photoperiod at 350 μmoles·m-2·s-1 light intensity. N. benthamiana plants belong to a standard laboratory line maintained by selfing. N. benthamiana plants were grown in F2 soil (Levington, Frimley, UK) in controlled greenhouse facilities maintained at 25 °C day temperature, 15°C night temperature, ambient humidity and a 16 h photoperiod with 125 w·m-2 of supplementary light. Watering was performed via dripping on capillary mats for 2-4 minutes every 6h. Chapter 2 61 2.4.2. Promoter isolation Leaf tissue from N. benthamiana and M. truncatula A17 plants was snap frozen in liquid nitrogen and stored at -80°C. Frozen tissue was ground to a fine powder using 5 mm glass beads and a Tissue Lyser II homogeniser (QIAgen, Hilden, Germany). Genomic DNA extraction was performed on up to 100 mg ground tissue using the QIAGEN DNeasy Plant Mini Kit (Hilden, Germany) according to the manufacturer’s specifications. Resulting genomic DNA was used to amplify 5’ flanking regions for M. truncatula’s PT4 and BCP1 gene, and N. benthamiana’s PT5b and BCP1b genes by PCR using Phusion High-Fidelity DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) with the conditions specified below and the oligonucleotide primers listed in Appendix 2 in a Mastercycler Nexus (Eppendorf, Hamburg, Germany). Primers were designed to include a region of 836 bp upstream the start codon for MtPT4, 1108 bp for MtBCP1, 1068 bp for NbPT5b (Niben101Scf02726g00004.1), and 1231 bp for NbBCP1b (Niben101Scf07438g04015.1). Amplification length for N. benthamiana promoter sequences was based on published promoter sizes in M. truncatula and suitable oligo binding sites (Harrison et al., 2002; Hohnjec et al., 2005). Master Mix for Phusion HF Reagent (stock concentration) Final Concentration Volume (in 20 μL) Phusion HF 0.4 U 0.2 μL Phusion HF Buffer 5X 1X 4 μL dNTP mix (10 mM) 0.2 mM 0.4 μL Forward primer (10 μM) 0.5 mM 1 μL Reverse primer (10 μM) 0.5 mM 1 μL DNA template 1 pg – 10 ng 1 μL MilliQ water - 12.4 μL 2.4.3. Ligation into pBlueScript. Phusion HF produces blunt ends, and therefore PCR products were directly cloned into the pBlueScript SK (-) cloning vector (previously cut in house with EcoRV) using the T4 DNA ligase (New England Biolabs, Hitchin, UK). Ligation reactions were performed in a total volume of 5 μL, with 0.5 μL of T4 DNA ligase, 0.5 μL of T4 DNA ligase buffer (10X), 0.5 μL of pBlueScript SK (25 ng/μL), and up to 3.5 μL of PCR product. The ligation reaction was performed at 16°C for 16 h. Temperature Time 98°C 30 sec 98°C 10 sec 35 cycles Annealing Temp. 10 sec 72°C ≥ 30 sec 72°C 5 min 4°C-10°C ∞ PCR conditions for Phusion HF A. Timoneda, PhD thesis, 2022 62 2.4.4. Transformation of vectors into E. coli DH5a. Plasmids were transformed into E. coli DH5α strains maintained in house. 5 μL of ligation was added to 50 μL of chemically competent DH5α cells, immediately after thawing. Transformation was performed through a heat shock of 1 min at 42°C, after which cells were placed on ice for an extra 1 min. 1 mL of liquid LB (10 g·L-1 tryptone, 10 g·L-1 NaCl, 5 g·L-1 yeast extract) was added to each sample and cells were left to recover 30-60 min at 37°C. Cells were then pelleted through a 1 min centrifugation at 5000 rpm (~ 1677 g), and 850 μL of the supernatant was discarded. Cells were then resuspended and plated in solid LB agar plates (10 g·L-1 Bacto Agar) supplemented with ampicillin (100 mg·L-1), IPTG (100 mM) and X-Gal (50 mg·mL-1). Two plates were seeded per sample to account for possible transformation variability, one with 50 μL and the other with 100 μL of resuspended cells. Plates were then incubated overnight at 37°C to allow for colony growth. pBlueScript SK (-) contains an ampicillin resistance marker, therefore cells which have successfully acquired the plasmid will be able to survive under selective media. Proper ligation of the insert in the plasmid was assessed via the white/blue colony selection method. While/blue selection is based on the principle of α-complementation of the β- galactosidase gene. The DH5α strain contains the lacZ ΔM15 mutation which encodes a non- functional β-galactosidase lacking the α-peptide needed for proper tetramerisation. pBlueScript SK (-) carries the lacZα sequence which encodes this α-peptide, and when expressed in DH5α cells allows for the formation of a functional β-galactosidase enzyme. The plasmid carries within the lacZα sequence a multi-cloning site (MCS), with a suite of restriction sites that can be used for cloning of the insert (including the one used in this thesis, EcoRV). When ligation occurs successfully, the lacZα sequence is disrupted and cannot be transcribed and expressed. After transformation of DH5α, IPTG present in the medium induces the lac operon, while X-Gal is the substrate of the β-galactosidase enzyme. Therefore, cells that are transformed with successfully ligated plasmids, are not able to produce a functional version of the β-galactosidase enzyme and develop white colonies, whereas cells that are transformed with plasmids that did not ligate successfully and the insert did not disrupt the lacZα site (i.e. an empty plasmid), appear blue. 2.4.5. Colony check by PCR and plasmid sequencing White E. coli colonies were subjected to PCR to confirm insert and plasmid integration using EcoTaq polymerase (produced and provided by the Department of Plant Sciences, Cambridge, UK) and M13 primers (sequences in Appendix 2) with the conditions described below. PCR amplicons were observed in 1.5% agarose gel electrophoresis. Confirmed colonies were grown overnight in 3 ml of liquid LB (10 g·L-1 tryptone, 10 g·L-1 NaCl, 5 g·L-1 yeast extract) supplemented with ampicillin (100 mg·L-1) in a shaker incubator at 180 rpm and 37°C. Glycerol Chapter 2 63 stocks of grown culture aliquots were performed at a final concentration of 25% glycerol (v/v) for all new plasmids generated and stored at -80°C for future re-stocking. Plasmids were extracted using the NucleoSpin Plasmid extraction kit (Macherey-Nagel, Dueren, Germany) according to manufacturer’s specifications. Plasmids were later sequenced to confirm gene identity using the M13F and R primers described in Appendix 2, and retrieved sequences were analysed using the Geneious R9.1.8 software (Biomatters, Auckland, NZ). Sequencing services were provided by Source BioScience (Nottingham, UK). Cloned sequences can be found in Appendix 1. Master Mix for EcoTaq Reagent (stock concentration) Final Concentration Volume (in 25 μL) EcoTaq polymerase - 0.1 μL EcoTaq buffer (10X) 1X 2.5 μL dNTP mix (10 mM) 0.2 mM 0.5 μL Forward primer (10 μM) 0.4 mM 1 μL Reverse primer (10 μM) 0.4 mM 1 μL DNA template Colony pick - MilliQ water - 18.9 μL 2.4.6. Generation of vectors using the MoClo Golden Gate technology The construction of multi-gene vectors containing the betalain biosynthetic genes under the control of AM specific promoters was carried out using Golden Gate cloning (Engler et al., 2008). Cloning components were acquired from the MoClo Tool Kit (Weber et al., 2011; Werner et al., 2012) and the MoClo Plant Parts Kit (Engler et al., 2014) (Addgene, Cambridge, MA, USA), except for the 1327 bp A. thaliana Ubiquitin 10 promoter (Ub10) which was provided by Dr Nicola Patron (Earlham Institute, Norwich, UK). MtPT4, MtBCP1, NbPT5b and NbBCP1b promoter regions previously cloned into pBlueScript SK (-) were amplified using primers containing overhangs that included BpiI restriction sites and the four-base pair MoClo nomenclature for Promoter + 5’UTR level 0 vectors described in Weber et al., 2011 (5’-GGAG-3’ in forward primers; 5’-AATG-3’ in reverse primers). Amplified promoter regions were then visualised with gel electrophoresis and extracted using the PureLink gel extraction kit (Invitrogen, Waltham, MA, USA) according to manufacturer’s specifications. Amplified fragments were then cloned into the pICH41295 level 0 accepting vector for promoter + 5’UTR modules. Cloning proceeded through Level 0, Level 1 and Level 2 modules using the one-pot, one-step type IIS restriction mediated cloning reaction described below. Quantity of insert was calculated as a 2:1 (vector:insert) ratio over 100 ng of acceptor vector with the formula also described below. Temperature Time 98°C 5 min 94°C 30 sec 35 cycles Annealing Temp. 30 sec 72°C ≥ 30 sec 72°C 3 min 4°C-10°C ∞ A. Timoneda, PhD thesis, 2022 64 Master Mix for MoClo ligation Reagent (stock concentration) Volume (in 25 μL) Acceptor vector (100 ng/ μL) 1 μL Insert (PCR product or vector) n μL T4 DNA ligase 0.5 μL T4 DNA ligase buffer 1.5 μL Bovin Serum Albumin (10X) 1.5 μL BpiI* (level 0 & 2) or BsaI-HFv2 (level 1) 0.5 μL MilliQ water up to 20 μL (*) BpiI acquired from Thermo Fisher Scientific (Waltham, MA, USA) (𝒏) 𝑛𝑔 𝑖𝑛𝑠𝑒𝑟𝑡 = 𝑛𝑔 𝑜𝑓 𝑎𝑐𝑐𝑒𝑝𝑡𝑜𝑟 𝑣𝑒𝑐𝑡𝑜𝑟 × 𝑠𝑖𝑧𝑒 𝑜𝑓 𝑖𝑛𝑠𝑒𝑟𝑡 (𝑘𝑏)𝑠𝑖𝑧𝑒 𝑜𝑓 𝑎𝑐𝑐𝑒𝑝𝑡𝑜𝑟 𝑣𝑒𝑐𝑡𝑜𝑟 (𝑘𝑏) × 21 GUS reporter vectors were constructed with an eGFP-GUS fusion gene placed under the control of the cloned arbuscular-responsive specific promoters and also included a Basta resistance gene under the nos promoter and terminator. Previously generated plasmids containing the betalain biosynthetic genes were used for the construction of betacyanin producing multi-gene vectors (Timoneda et al., 2018). DODAα1 and CYP76AD1 sequences were originally isolated from B. vulgaris cDNA libraries, whereas the cDOPA5GT sequence was obtained from flowers of M. jalapa. Gene sequences had been previously domesticated to remove BsaI and BpiI restriction enzyme sites considering codon usage in N. benthamiana. The first set of multi-gene vectors created for hairy root transformation of M. truncatula and stable transformation of N. benthamiana contained BvCYP76AD1 under the control of the arbuscular- responsive specific promoters (MtPT4, MtBCP1, NbPT5b or NbBCP1b), BvDODAα1 and MjcDOPA5GT under the 35S and Ub10 constitutive promoters, respectively, and included a Basta resistance gene under the nos promoter and terminator. We refer to these multigene vectors as MtPT4-p1, MtBCP1-p1, NbPT5b-p1 and NbBCP1b-p1 respectively (Figure 2.1a, Figure 2.7a). Next we constructed another set of multi-gene vectors for betaxanthin production in roots of M. truncatula and stable N. benthamiana transformants this time containing BvCYP76AD6 under the control of the arbuscular-responsive specific promoters and BvDODAα1 under constitutive promoters, also including a Basta resistance gene under the nos promoter and terminator. We refer to these multigene vectors as MtPT4-p2, MtBCP1-p2, NbPT5b-p2 (Figure 2.4a, Figure 2.9a). The third set of multi-gene vectors created for root transformation of M. truncatula and stable transformation of N. benthamiana contained BvDODAα1, BvCYP76AD1 and MjcDOPA5GT genes all under the control of the arbuscular- responsive specific promoters (MtPT4, NbPT5b and NbBCP1b), and included a DSRed Temperature Time 37°C 20 sec 37°C 3 min 26 cycles 16°C 4 min 50°C 30 sec 80°C 5 min 16°C ∞ One-pot, one-step cloning conditions Chapter 2 65 fluorescent marker in the M. truncatula vector or a Basta resistance gene in N. benthamiana vectors both under the nos promoter and terminator. We refer to these multigene vectors as MtPT4-p3, NbPT5b-p3 and NbBCP1b-p3 respectively (Figure 2.13a, Figure 2.15a). Vectors were transformed in E. coli DH5α cells as previously described in sections 2.4.4 and 2.4.5, with the difference that level 2 vectors containing a canthaxanthin producing operon that allows for orange/white screening of colonies without addition of IPTG or X-Gal. Level 0 and Level 1 vectors were confirmed by Sanger sequencing of the full inserts. Level 2 vectors were confirmed by Sanger sequencing of insert boundaries and by diagnostic digestion with restriction enzymes. Primers used for colony check and sequencing are detailed in Appendix 2. 2.4.7. Transformation of binary vectors in Agrobacterium tumefaciens and Agrobacterium rhizogenes The A. tumefaciens strain used for stable and transient transformation of N. benthamiana plants was GV3101, which included a pMP90 helper vector containing the vir genes for T-DNA integration in the plant genome and a pSOUP vector containing the molecular machinery required for replication of binary vectors. The GV3101 chromosome contains a rifampicin resistance gene, the pMP90 plasmid contains a gentamicin resistance gene and the pSOUP plasmid a tetracyclin resistance gene. Transformation of A. tumefaciens GV3101 with 1 μL of the constructed Level 2 vectors was performed via electroporation in Gene Pulser/MicroPulser 0.2 cm gap cuvettes (BioRad, Hercules, CA, USA). Cells were subjected to an electric pulse (2400 V, 25 uF, 200 W) using a BioRad GenePulser Xcell (BioRad, Hercules, CA, USA). After a regeneration period of 3 h in liquid LB medium at 30°C, cells were plated into 9 cm dishes containing LB agar medium supplemented with kanamycin (50 mg·L-1) and gentamycin (25 mg·L-1) and incubated during 2 days at 30°C. Resulting colonies were checked by PCR using gene specific primers and EcoTaq polymerase (produced and provided by the Department of Plant Sciences, Cambridge, UK) as described in 2.4.5. Before PCR, colonies were picked in NaOH 20 mM and pre-treated at 98°C for 5 minutes. The A. rhizogenes strain used for hairy root transformation of M. truncatula was ARqua1, which contains the pRiA4b virulence plasmid and is resistant to carbenicillin. Transformation of A. rhizogenes ARqua1 with ~ 10 ng of the constructed Level 2 vectors was also performed via electroporation with an electric pulse of 1500 V for 5 msec. After a regeneration period of 1 h in liquid LB medium at 28°C, cells were plated into 9 cm dishes containing LB agar medium supplemented with kanamycin (50 mg·L-1) and carbenicillin (50 mg·L-1) and incubated during 2 days at 30°C. Colonies were then grown in liquid LB media supplemented with kanamycin (50 mg·L-1) and carbenicillin (25 mg·L-1), and plasmid integration was checked by restriction enzyme digestion after extraction with the QIAprep Spin Miniprep Kit (Qiagen, Hilden, Germany). A. Timoneda, PhD thesis, 2022 66 2.4.8. Hairy root transformation of M. truncatula Hairy root transformation of M. truncatula was performed according to the previously described Agrobacterium rhizogenes-mediated method with modifications (Boisson-Dernier et al., 2001). Multi-gene constructs were transformed into the A. rhizogenes ARqua1 strain, and grown at 28°C in LB media supplemented with antibiotics until reaching an OD600 of approximately 1.5. Cultures were then brought to a final OD600 of approximately 1 in sterile distilled water. M. truncatula seeds were sterilised with 96% sulfuric acid for 5 minutes and pure sodium hypochlorite for 5 minutes, both steps followed by 5-10 washes with sterile distilled water. After a 2-day stratification process at 4°C, seeds were germinated upside down on square dishes containing wet whatman paper overnight in the dark at 20°C. Hypocotyl tips were then cut and seedlings were suspended in A. rhizogenes cultures with acetosyringone (0.2 μM). Seed coats were removed and seedlings were moved to solid mycorrhization media (potassium nitrate, 80 mg·L-1; magnesium sulfate, 731 mg·L-1; potassium chloride, 65 mg·L-1; monopotassium phosphate, 4.8 mg·L-1; calcium nitrate, 288 mg·L-1; manganese (II) chloride, 6 mg·L-1; boric acid, 1.5 mg·L-1; zinc sulfate, 2.65 mg·L-1; sodium molybdate, 0.0024 mg·L-1; copper sulfate, 0.13 mg·L-1; potassium iodide, 0.75 mg·L-1; ethylenediaminetetraacetic acid ferric sodium salt, 8 mg·L-1; glycine, 3 mg·L-1; myoinositol, 50 mg·L-1; nicotinic acid, 0.5 mg·L-1; pyridoxine hydrochloride, 0.1 mg·L-1; thiamine hydrochloride, 0.1 mg·L-1; phytagel, 4 g·L-1; pH 5.5) to grow vertically for 7 days at 20°C and indirect lighting. Newly formed roots were then sectioned in half and seedlings were transferred to mycorrhization media supplemented with cefotaxime (250 μg·mL-1) and covered with Whatman’s filter paper. Plants were grown vertically for 4 weeks at 20°C and indirect lighting to allow for transgenic root development. Plants were then transferred to sand and subjected to fungal colonisation with R. irregularis. For constructs containing DsRed, all non-transformed roots were cut off before acclimation and inoculation. DSRed fluorescence was detected with a Leica M165 FC fluorescence dissecting stereomicroscope (Leica Biosystems, Wetzlar, Germany) using a filter with excitation and emission wavelengths of 510-560 nm and 590-650 nm, respectively. 2.4.9. Stable transformation of N. benthamiana N. benthamiana stable transformation was performed according to the previously described A. tumefaciens-mediated method with modifications (Sparkes et al., 2006b). Fully developed leaves of 4-week old N. benthamiana plants were agro-infiltrated with A. tumefaciens carrying the constructs of interest according to previously described methods (Sparkes et al., 2006b). All constructs were transformed into the Agrobacterium tumefaciens GV3101 strain, and grown at 28°C in LB media supplemented with antibiotics until reaching an OD600 of approximately 1.5. Cultures were then brought to a final OD600 of approximately 0.3 in infiltration media (10 mM Chapter 2 67 MgCl2, 0.2 mM acetosyringone, 10 mM MES at pH 5.6). Infiltration was performed on the abaxial surface of young expanding leaves of 4-week old N. benthamiana plants. Three days after infiltration, full leaves were excised and cut into 1-2 cm2 leaf squares. Leaf explants were sterilised in 70% ethanol for 5 minutes, 20% sodium hypochlorite supplemented with Tween 20 (0.1%) for 20 minutes, and washed thoroughly with sterile water. Sterilised leaf explants were then transferred adaxial side up to fresh selection media (1X Murashige and Skoog basal salt mixture, 1X Gamborg’s B5 vitamins, 1% sucrose, 0.59 g·L-1 MES, 2.0 mg·L-1 BAP, 0.05 mg·L-1 NAA, 0.4% Agargel, pH 5.7) supplemented with cefotaxime (500 mg·L-1), timentin (320 mg·L-1) and phosphinothricin (2 mg·L-1). Explants were subcultured onto new fresh media weekly until the appearance of first shoots, which were excised and planted in rooting media (½ Murashige and Skoog basal salt mixture, 0.5% sucrose, 0.25% Gelrite, 0.05 mg·L-1 NAA, pH 5.8) supplemented with augmentin (500 mg·L-1), timentin (320 mg·L-1) and phosphinotricin (2mg·L-1). Approximately after 2 weeks, transgenic plantlets with growing root systems were transferred to jars with sterile peat blocks. Transgenic plants were transplanted to the glasshouse when fully developed, and allowed to grow until flowering and harvesting of seeds. 2.4.10. Plant inoculation with R. irregularis M. truncatula and N. benthamiana plants were transferred to sand and inoculated with a commercial R. irregularis crude inoculum (consisting of oil-dry substrate, pieces of colonised maize roots, and R. irregularis spores and external hyphae) at a concentration of 1:20 (v/v) inoculum:sand. Plants were watered three times per week with 1x Long-Ashton solution (20x concentrate: potassium nitrate, 40 mM; calcium nitrate, 40 mM; monosodium phosphate, 15 mM; magnesium sulfate, 30 mM; manganese sulfate, 0.1 mM; copper sulfate, 0.01 mM; zinc sulfate, 0.01 mM; boric acid, 0.5 mM; sodium chloride, 1 mM; ammonium heptamolybdate, 0.7 μM; ethylenediaminetetraacetic acid ferric sodium salt, 0.6 mM). 2.4.11. RT-PCR gene expression analysis of N. benthamiana stable lines N. benthamiana leaf tissue was snap frozen in liquid nitrogen and stored at -80°C. Frozen tissue was ground to a fine powder using 5 mm glass beads and a Tissue Lyser II homogeniser (QIAgen, Hilden, Germany). RNA extraction was performed using the Concert Plant RNA Reagent (Invitrogen, Carlsbad, CA, USA) followed by the TURBO DNA-free kit (Ambion, Carlsbad, CA, USA) to remove DNA. RNA concentration was quantified by Nanodrop and RNA integrity was assessed by agarose gel electrophoresis. cDNA libraries were prepared using BioScript Reverse Transcriptase (Bioline Reagents, London, UK) and oligo dT primers, according to manufacturer’s recommendations. RT-PCR was performed on a 1:10 cDNA A. Timoneda, PhD thesis, 2022 68 dilution, using the KAPA 2G Fast DNA polymerase kit (KAPA Biosystems, Wilmington, USA) and the oligonucleotides specified in Appendix 2, following the conditions described below. Master Mix for KAPA 2G Fast Reagent (stock concentration) Final Concentration Volume (in 25 μL) KAPA 2G Fast 0.5 U 0.1 μL Kapa 2G Fast Buffer A (5X) 1X 5 μL dNTP mix (10 mM) 0.2 mM 0.5 μL Forward primer (10 μM) 0.5 mM 1.25 μL Reverse primer (10 μM) 0.5 mM 1.25 μL DNA template ~ 5 ng 1 μL MilliQ water - 15.9 μL 2.4.12. Histochemical staining for GUS activity Transgenic M. truncatula and N. benthamiana roots carrying the MtPT4::eGFP:GUS, MtBCP1::eGFP:GUS, NbPT5b1::eGFP-GUS and NbBCP1b::eGFP:GUS sequences were harvested 4 weeks after inoculation and treated with 80% acetone at -20°C for 30 minutes. Roots were then washed with PBS for 5 minutes and incubated in a staining solution containing 100mM sodium phosphate pH 7.0, 5 mM potassium ferrocyanide, 5 mM potassium ferricyanide and 2 mM 5-bromo-4-chloro-3-indoxyl-β-D-glucuronid acid (X-gluc) for 3 h at 37°C. Roots were then washed twice with PBS, and prepared for sectioning. M. truncatula roots were cast in paraffin wax, sectioned to a 5 μm width with a HM 340E rotary microtome (Thermo Fisher Scientific, Waltham, MA, USA), and observed with an Olympus BX41microscope (Olympus, Tokyo, Japan). N. benthamiana roots were embedded in 4% low melting point agarose, sectioned to a 100 μm width with a Leica VT1200S vibratome (Leica Biosystems, Wetzlar, Germany) and imaged with a Keyence VHX-5000 microscope (Keyence, Osaka, Japan). 2.4.13. Staining and quantification of fungal structures Ink staining of fungal structures was performed in M. truncatula and N. benthamiana roots as previously described (Vierheilig et al., 1998). M. truncatula hairy roots is a transient transformation method, and therefore not all roots will be transgenic. To properly quantify fungal structures on roots expressing the betalain reporter, we aided ourselves with a DSRed fluorescent transformation marker (Figure 2.25). Roots were incubated with 10% potassium hydroxide at 95°C for 5 minutes and washed three times with sterile deionized water. Roots were then incubated in a solution of 5% pen ink (Sheaffer, Providence, RI, USA) and 5% acetic acid at 95°C for 2 minutes, washed with 5% acetic acid and then three times with sterile Temperature Time 95°C 1 min 95°C 10 sec 30 cycles Annealing Temp. 10 sec 72°C ≥ 2 sec 72°C 5 min 4°C-10°C ∞ PCR conditions for KAPA 2G Fast Chapter 2 69 deionized water. N. benthamiana roots were additionally incubated with ClearSee (Kurihara et al., 2015) for 20 seconds, and washed three times with sterile deionized water. Stained root fragments were then mounted in parallel on a glass slide and imaged with a Keyence VHX-5000 digital microscope (Keyence, Osaka, Japan). Each root image was divided into equally distant sections by digitally applying a 0.75 mm x 0.75 mm grid, and presence or absence of fungal structures (internal hyphae, arbuscules and vesicles) were recorded in each square for each root. The length of root fragment colonised was calculated as the percentage of the number of squares containing fungal structures from the total number of squares in each root. Results were averaged for each genotype and condition. For wheat germ agglutinin-FITC conjugate (WGA-FITC) staining in M. truncatula, roots were incubated in 1 mg/mL of WGA-FITC conjugate (Sigma-Aldrich, Saint Louis, MO, USA) in potassium phosphate buffer 0.1 M (pH 7.2) 10 min in the dark, rinsed several times in potassium phosphate buffer 0.1 M (pH 7.2) and mounted on glass slides. For staining in N. benthamiana, roots were incubated in WGA-FITC conjugate in potassium phosphate buffer 0.1 M (pH 7.2) supplemented with sucrose (3%) for 1 h in the dark. Figure 2.21. Schematic of the fungal visualisation and quantification process of Medicago truncatula hairy roots after inoculation with Rhizophagus irregularis. (a) 1, M. truncatula transgenic hairy roots express the DSRed marker and can be selected through detection of fluorescence under 510-560 nm; 2, DSRed positive roots are cut and divided into betalain producing and non-producing root fragments; 3, Pigmented and non-pigmented root fragments are ink stained separately (betalain colouration fades upon incubation with potassium hydroxide during the ink staining process); 4, Ink stained root fragments are mounted in glass slides and imaged for visualisation and quantification of fungal structures. (b) Example of MtPT4-p3 pigmented root fragments showing fungal colonisation. (c) Example of MtPT4-p3 non-pigmented uncolonised root fragments. Scale bar, 100 μm. A. Timoneda, PhD thesis, 2022 70 2.4.14. Dissection and microscopy of roots for betalain visualisation M. truncatula roots expressing MtPT4-p1, MtBCP1-p1 and MtPT4-p3 were harvested 4 weeks after inoculation. Roots were hand-sectioned and mounted in potassium phosphate buffer 0.1 M (pH 7.2) on glass slides. Sections were imaged using a Zeiss Axioimager microscope (Zeiss, Oberkochen, Germany) with brightfield and UV light. N. benthamiana roots expressing NbPT5b- p1 and NbBCP1b-p1, were harvested 4 weeks after inoculation, embedded in 4% agarose supplemented with sucrose (3%) and sectioned to a 50 μm width with a Leica VT1200S vibratome (Leica Biosystems, Wetzlar, Germany). Sections were then mounted on a glass slide and stained with WGA-FITC conjugate as previously described. Root sections were visualised with a Leica TCS SP8 confocal laser scanning microscope equipped with a Leica DFC 7000T camera and HyD detectors (Leica Biosystems, Wetzlar, Germany). N. benthamiana roots expressing NbPT5b-p3 and NbBCP1b-p3, were harvested 4 weeks after inoculation, embedded in 4% agarose and sectioned to a 100 μm width with a Leica VT1200S vibratome (Leica Biosystems, Wetzlar, Germany). Sections were then mounted on a glass slide and imaged with a VHX-7000 Keyence digital microscope (Keyence, Osaka, Japan). Root Images were loaded into GIMP 2.8.2. Red coloring was visualised by adjusting Hue/Saturation via Colors menu. Betaxanthin fluorescence presents an excitation maxima between 463 - 480 nm and emission maxima between 509 - 512 nm, and was measured with conventional GFP fluorescent filters. 2.4.15. Betalain extraction and detection using HPLC Extraction and liquid chromatography of betalains in N. benthamiana leaf samples were performed as previously described (Sheehan et al., 2020). In brief, leaf tissue samples were snap frozen in liquid nitrogen and grinded to a fine powder using mortar and pestle. Betalains were extracted overnight at 4°C in 80% aqueous methanol with 50 mM ascorbic acid with a volume of 1 mL extraction buffer per 50 mg leaf tissue. After extraction, samples were clarified twice by centrifuging at 12,000 g for 5 mins and collecting the supernatants. Liquid chromatography analysis was performed using a Thermo Fisher Scientific Accela HPLC autosampler (Thermo Fisher Scientific, Waltham, MA, USA) and pump system incorporating a photodiode array detector. Betalains were separated using a Luna Omega C18 column (100 Å, 5 μm, 4.6 x 150 mm) from Phenomenex (Torrance, CA, USA) under the following conditions: 3 min, 0% B; 3–19 min, 0–75% B; 7 min, 0% B where mobile phase A was 0.1% formic acid in 1% acetonitrile and solvent B was 100% acetonitrile, and at a flow rate of 500 μL/min. The betacyanin betanin was quantified since it has been shown to be the predominant pigment arising from the combined expression of DODA, CYP76AD1 and cDOPA5GT (Timoneda et al., 2018). Betanin was detected by UV/VIS absorbance at a wavelength of 540 nm. Identification and quantification of betanin was carried out using a commercially available B. vulgaris extract (Tokyo Chemical Industry UK Ltd, Oxford, UK). Chapter 2 71 2.4.16. Rhizotron The rhizotron set up was devised by Dr Temur Yunusov. It consists of a 10 cm x 10 cm square petri dish where one of the sides of the base of the petri dish was removed to allow shoots to grow out. Ten-day old N. benthamiana seedlings were placed on the base of the petri dish, covered with a coarse woven black cloth layer, followed by a layer of sand containing AM fungal spore inoculum. The lid was fixed using transparent adhesive tape. Rhizotrons were placed in black plastic bags and incubated in upright position over 52 days. For imaging, the plastic bag was removed and rhizotrons were placed on a flatbed scanner. Images were loaded into GIMP 2.8.2. Red coloring was visualised by adjusting Hue/Saturation via Colors menu. A. Timoneda, PhD thesis, 2022 72 Chapter 3 73 Chapter 3. ADH variants with putative relaxed feedback inhibition outside of the order Caryophyllales 3.1. Introduction ʟ-Tyrosine is an essential aromatic amino acid required for protein biosynthesis and is synthesised de novo only in bacteria, fungi and plants (Schenck & Maeda, 2018). Tyrosine is a product of the shikimate pathway and can be generated from prephenate via two alternative routes. In most plants, including the Caryophyllales, tyrosine is produced via the arogenate route, where prephenate is first transaminated to arogenate, and then decarboxylated to tyrosine (Figure 3.1). The oxidative decarboxylation of arogenate is performed by the arogenate dehydrogenase (ADH) enzyme (Schenck & Maeda, 2018). In microorganisms and other plants such as the Fabaceae (legumes), these reactions also occur in the inverse order and tyrosine can be formed first through the oxidative decarboxylation of prephenate and the subsequent transamination of 4-hydroxyphenylpyruvate (Schenck et al., 2015). Here, the oxidative decarboxylation of prephenate to HPP is performed by the prephenate dehydrogenase (PDH) enzyme. Figure 3.1. Tyrosine can be formed via two alternative routes at the end of the Shikimate pathway. In most plants, prephenate is first transaminated to arogenate by prephenate aminotransferase (PPA-AT), and then decarboxylated to tyrosine by arogenate dehydrogenase (ADH). In microorganisms and legumes, prephenate is first decarboxylated to 4-hydroxyphenylpyruvate (HPP) by prephenate dehydrogenase (PDH), and then transaminated to tyrosine by 4-hyxdroxyphenylpyruvate aminotransferase (HPP-AT). Both ADH and PDH are subjected to negative feedback regulation by tyrosine. A. Timoneda, PhD thesis, 2022 74 Canonical ADH and PDH enzymes are both inhibited by the reaction product tyrosine and therefore experience negative feedback regulation (Figure 3.1) (Rippert & Matringe, 2002a,b). Most plant enzymes involved in the aromatic amino acid pathways, like ADH, are localised within the plastids. However, some legume PDH enzymes localise in the cytosol and exhibit lack of inhibition under high tyrosine concentrations in vitro (Schenck et al., 2015). Elucidation of the X-ray crystal structure of the soybean (Glycine max) PDH and phylogenetic comparisons between PDH and ADH enzymes, permitted the identification of one residue position, aspartate at position 218 (D218) of the soybean’s ADH (GmADH) active site, which corresponds to asparagine at position 222 (N222) of the PDH active site, as the position responsible for the enzyme’s substrate specificity (Schenck et al., 2017). Mutation of the asparagine (N) in GmPDH to the corresponding aspartate (D) residue present in ADH (N222D) resulted in the loss of enzyme activity over prephenate and the gain of ADH activity. The reciprocal mutation in GmADH (D218N) also resulted in the loss of affinity for the arogenate substrate and the gain of PDH activity (Schenck et al., 2017). These results highlight the importance of the D residue in the ADH active site for arogenate specificity. Later, a Caryophyllales-specific ADH variant was also discovered exhibiting relaxed sensitivity to tyrosine feedback inhibition (Figure 1.3) (Lopez- Nieves et al., 2018). Authors refer to this new variant as ADHα, which was shown to be able to maintain higher activities at higher tyrosine concentrations than the canonical ADHβ in a wide range of Caryophyllales species (B. vulgaris, S. oleracea, M. jalapa, P. oleracea and Rivina humilis). Relaxed sensitivity to tyrosine is also observed in members of the non-betalain producing families of the Caryophyllales, suggesting it may have evolved prior to the establishment of the betalain biosynthetic pathway (Lopez-Nieves et al., 2021). Both ADHα and ADHβ enzymes in B. vulgaris also have a predicted N-terminal chloroplast transit peptide (cTP) and were also shown to be targeted to the plastids (Lopez-Nieves et al., 2018; Timoneda et al., 2018). Recent work identified another single residue in the Caryophyllales ADHα active site to be critical for tyrosine deregulation (Lopez-Nieves et al., 2021). This residue is a D at position 208 (D208) of the active site of the ADHα enzyme of the betalain-producing species S. oleracea, and is located opposite to the other D residue (D304) identified in legumes to be important for substrate specificity. In ADHβ sequences, D208 corresponds to a glutamate (E) residue. Substitution of D208 for E in SoADHα (D208E) reduced enzyme activity at higher concentrations of tyrosine, pointing at a loss of relaxed sensitivity to feedback inhibition by tyrosine (Lopez-Nieves et al., 2021). Conversely, the reciprocal mutation in typically tyrosine- regulated ADH enzymes even in species outside of the Caryophyllales, such as A. thaliana TyrA2, resulted in reduced tyrosine inhibition (Lopez-Nieves et al., 2021). As shown by these studies, the relaxed sensitivity to tyrosine regulation of ADHα enzymes means they can maintain higher activities at higher concentrations of tyrosine, and ultimately achieve higher yields. For example, overexpression of BvADHα in N. benthamiana plants resulted in over a 10-fold increase in tyrosine concentration relative to control vectors containing Chapter 3 75 GFP and BvADHβ (Lopez-Nieves et al., 2018). This increase in tyrosine can be used to redirect metabolic flux towards tyrosine-derived pathways and improve production of certain metabolites such as betalain pigments. Expressing BvADHα in constructs additionally containing the betalain biosynthesis genes, also allowed for a 7-fold increase in betalain yields in planta (Timoneda et al., 2018). Therefore, the identification of enzymes with a lower sensitivity to feedback regulation such as ADHα can be a promising strategy for the optimisation of betalain production in heterologous systems. In this chapter, we aim to identify additional deregulated ADH enzymes by exploring the natural diversity within plant lineages outside the Caryophyllales. Relaxed sensitivity to tyrosine in ADH enzymes may have convergently evolved multiple times to accommodate for the establishment of enriched tyrosine-derived metabolisms which could have resulted advantageous for plants in front of a diversity of selective pressures. Residue information previously obtained for SoADHα can be used for the identification of enzyme candidates, by screening the flowering plant ADH phylogeny for duplication events where one paralog presents a D residue in the active site, equivalent to the D208 in SoADHα (associated with deregulation) (Lopez-Nieves et al., 2021). We then test these candidates’ abilities to produce tyrosine in planta and assess their potential to be used in the context of heterologous betalain production. This approach can potentially lead to the discovery of novel ADH forms able to access higher tyrosine production yields than those achieved by currently known enzymes such as BvADHα, and that therefore could represent a powerful tool for the optimisation and boosting of betalain pigment production. 3.2. Results 3.2.1. Identifying ADH candidates with relaxed tyrosine feedback regulation Using an ADH phylogeny constructed with available flowering plant ADH sequences made by Nathanael Walker-Hale (PhD student, Brockington Lab), we looked for species in the tree that, like Caryophyllales: 1) had experienced a duplication event on the ADH lineage giving rise to two paralog copies, and 2) where one copy presented an E residue (like in most canonical ADH and Caryophyllales ADHβ enzymes), and the other copy presented an amino acid substitution of this E at the corresponding amino acid position identified by Lopez-Nieves et al. (2021) to be involved with relaxed sensitivity to tyrosine in S. oleracea and B. vulgaris (D208). We did not limit the search to exclusively include substitutions to D and extended it to any amino acid change. From the taxa we found we selected two for further characterisation: Solanum lycopersicum (tomato) from the Solanaceae, and Manihot esculenta (cassava) from the Euphorbiaceae (Figure 3.2). Both taxa have a canonical ADH copy with a E in positions 200 and 195, respectively, and an extra ADH copy where this residue has been substituted for a D A. Timoneda, PhD thesis, 2022 76 in position 197 in the case of M. esculenta, and for a threonine (T) in position 218 in the case of S. lycopersicum. From here onwards, we will refer to canonical sequences containing E as ADH1, and copies with the amino acid mutation to D or T as ADH2. Figure 3.2. ADH gene phylogeny in the flowering plants. (a) Phylogeny of the ADH gene in flowering plants. Branches in red mark taxa belonging to the Solanum and Manihot genera. (b) Close-up of branches showing ADH duplication in the Solanum genus. (c) Close-up of branches showing ADH duplication in the Manihot genus. Phylogeny constructed by Dr Sam Brockington. Structure prediction of these enzymes using AlphaFold confirmed these residues are also located near the active site in S. lycopersicum and M. esculenta (Figure 3.3). The models revealed these residues are located in a region forming a pocket, which also contains the D218 residue identified in GmADH to be important for substrate specificity (Schenck et al., 2017), and a conserved histidine (H) residue previously identified to be involved in the hydrogenase reaction by structural and biochemical studies of bacterial PDH proteins (Christendat & Turnbull, 1999; Legrand et al., 2006). Model quality was low for C-terminal and N-terminal regions of the protein where the structure could not be properly resolved, however, support values were very high for most of the core region of the protein including the pocket region (Figure 3.3). This acts as further confirmation that the E and D/T residues are in close proximity to the active site of S. lycopersicum and M. esculenta enzymes and supports their involvement in tyrosine production. Chapter 3 77 Figure 3.3. Structure models predicted for ADH enzymes. (a) BvADHβ; (b) BvADHα; (c) SlADH1; (d) SlADH2; (e) MeADH1; (f) MeADH2. Panels show active sites of the protein structure predictions for each enzyme using AlphaFold. Protein areas shown exhibited a very high (>90) score in their predicted Local Distance Difference Test (plDDT), which assesses local model quality (Mariani et al., 2013). In red, the position of the glutamate (E) residues in SlADH1, MeADH1 and BvADHβ proteins (a, c, e), aspartate (D) residues in MeADH2 and BvADHα proteins (b, e) and threonine (T) residue in the SlADH2 protein (d); in pink, the position of the histidine (H) residue known to be in the active site of bacterial ADH enzymes and involved in the hydrogenase reaction (Christendat & Turnbull, 1999); in blue, the position of the D residue corresponding to soybean’s ADH D218 found to be important for substrate specificity (Schenck et al., 2017). Structure prediction was performed using the ColabFold AlphaFold2 structure prediction tool (Mirdita et al., 2021) and visualized with PyMOL. 3.2.2. Deregulated ADH enzymes may exist in plant taxa outside of the order Caryophyllales In order to assess whether the selected ADH variant pairs show relaxed sensitivity to tyrosine inhibition we decided to start by testing their ability to produce tyrosine in planta. Deregulated ADH enzymes have been shown to be able to accumulate higher tyrosine yields in A. thaliana and N. benthamiana (Lopez-Nieves et al., 2018, 2021). To isolate candidate ADH genes, we constructed cDNA libraries from tomato seedlings and cassava leaf tissue. SlADH1 and SlADH2 transcript sequences were successfully amplified from cDNA libraries, whereas MeADH1 and MeADH2 had to be codon optimised for expression in N. benthamiana and synthesised by Genewiz, after repeated failures to amplify from M. esculenta cDNA libraries. Transcript sequence information for MeADH2 was incomplete at the N-terminal end, where ADH enzymes are predicted to contain a cTP sequence important for the correct subcellular localisation of the protein. Therefore we decided to synthesise a chimeric version of MeADH2 containing the cTP sequence from BvADHα (Figure 3.4) (Lopez-Nieves et al., 2018). Amplified and synthesised ADH sequences can be found in Appendix 1. A protein alignment of all ADH sequences used in this experiment can be found in Figure 3.4. Sequences were then cloned into expression vectors under the control of the CaMV 35S (35S) promoter and the A. tumefaciens octopine synthase (ocs) terminator. Con structed vectors were transformed into A. tumefaciens for A. Timoneda, PhD thesis, 2022 78 transient expression in N. benthamiana leaves. Three days after agroinfiltration, tyrosine was extracted from infiltrated leaf tissue and quantified by liquid chromatography. For both S. lycopersicum and M. esculenta, expression of ADH2 variants resulted in an average of a ~ 3-fold increase in tyrosine accumulation over ADH1 variants (Figure 3.5). However, this increase did not amount to the levels of tyrosine produced by BvADHα, which was still 4 and 3.6 times higher than those obtained for SlADH2 and MeADH2 respectively. Statistical analysis using Welch two sample t-test between ADH1 and ADH2 samples, showed a significant difference between SlADH1 and SlADH2 with a p-value of 0.018 (<0.05), but did not in the case of the M. esculenta enzymes, where the p-value was calculated at 0.114 (>0.05). Figure 3.4. Protein alignment of ADH variant pairs. Protein sequences belong to Spinacia oleracea, Beta vulgaris, Solanum lycopersicum and Manihot esculenta. Red rectangle signals the amino-acid found to be responsible for tyrosine relaxation in S. oleracea (Lopez-Nieves et al., 2021) which in this alignment falls at position 220. The canonical SoADHβ, BvADHβ, SlADH1 and MeADH1 proteins contain a glutamate (E) residue at that specific position. SoADHα, BvADHα, SlADH2 and MeADH2 proteins contain mutations to aspartate (D) or threonine (T). Pink residues represent the portion of the predicted BvADHα chloroplast transit peptide (cTP ) fused to the start of the partial MeADH2 sequence. Protein alignment performed using Multiple Sequence Comparison by Log-Expectation (MUSCLE) with the Geneious R9.1.8 software. Chapter 3 79 Figure 3.5. Tyrosine quantification of Nicotiana benthamiana leaves expressing ADH enzymes. HPLC measurements of tyrosine extractions from N. benthamiana leaves 3 days after agroinfiltration with binary vectors harbouring constitutive expression of SlADH1, SlADH2, MeADH1 and MeADH2. In red, ADH enzymes from Solanum lycopersicum; in brown, ADH enzymes from Manihot esculenta; in grey, empty vector (ev) and BvADHα are used as negative and positive control respectively. Error bars represent standard deviation of 3 biological replicates. Asterisk represents significant difference in Welch two sample t-test analysis between same species’ ADH1 and ADH2 enzymes (p-value: 0.01< * < 0.05). 3.2.3. A single amino acid is responsible for the increase in tyrosine production observed with S. lycopersicum ADH2 SlADH2 and MeADH2 enzymes showed, on average, a higher ability than the ADH1 enzymes to accumulate tyrosine in N. benthamiana transient expression assays. However, it remains unproven whether, as shown in Caryophyllales species (Lopez-Nieves et al., 2021) and as we hypothesised, this increase is due to the mutation of E to D/T in ADH2 enzymes. To confirm this, we performed PCR site-directed mutagenesis of this sites in ADH1 and ADH2 sequences to reverse the amino acid state to match that of their ADH1 or ADH2 counterparts. We generated ADH1 versions where the E was substituted by D or T (SlADH1 E200T, MeADH1 E195D), and ADH2 versions with the reciprocal mutations (SlADH2 T218E; MeADH2 D197E). In the case of S. lycopersicum, we additionally produced a SlADH1 mutated version where the E was mutated to a D instead of a T (SlADH1 E200D) to assess for possible residue specificity. As in previous experiments, newly mutated sequences were cloned into binary vectors for transient expression in N. benthamiana leaves under the control of the 35S promoter and ocs terminator. A. Timoneda, PhD thesis, 2022 80 Figure 3.6. Tyrosine quantification of Nicotiana benthamiana leaves expressing mutated versions of ADH enzymes. HPLC measurements of tyrosine extractions from N. benthamiana leaves 3 days after agroinfiltration with binary vectors harbouring constitutive expression of (a) Solanum lycopersicum and (b) Manihot esculenta ADH variants and their mutated versions. In red, ADH enzymes from S. lycopersicum; in brown, ADH enzymes from M. esculenta; in grey, empty vector (ev) and BvADHα are used as negative and positive control respectively. Error bars represent standard deviation of 3 biological replicates. Asterisks represent significant differences in Welch two sample t-test analysis (p- value: *** < 0.001 < **< 0.01< * < 0.05). In the case of S. lycopersicum, changing the E residue at position 200 of the SlADH1 enzyme to the T residue observed in SlADH2 was sufficient to increase tyrosine production 3.4-fold to match and even overperform the SlADH2 enzyme (Figure 3.6a). Mutation of this residue to a D instead, resulted in a slight 1.5-fold increase of tyrosine production which, although significant, was not sufficient to recover the SlADH2 phenotype. Inversely, SlADH2 enzymes where the T residue at position 218 had been mutated to an E, performed similarly to SlADH1 and showed a drop in tyrosine levels compared to the original SlADH2 enzyme (Figure 3.6a). Statistical analysis using ANOVA and Welch two sample t-tests between SlADH1 and SlADH1 E200T, SlADH1 and SlADH1 E200D, and SlADH2 and SlADH2 T218E samples, showed significant differences for all comparisons (p-values of 0.0003, 0.035 and 0.0013, respectively). Therefore, the change from E to T in S. lycopersicum ADH enzymes appears to be largely responsible for the increase in tyrosine production observed in the ADH2 enzyme and is likely involved in the relaxation of enzyme regulation by tyrosine. In the case of M. esculenta, we were not able to find significant differences between original and mutated samples (Figure 3.6b). Only MeADH2 and MeADH2 D197E showed significant differences in Welch two sample t-test analysis (p-value of 0.04). Chapter 3 81 3.2.4. Improved tyrosine production shown by SlADH2 and MeADH2 is not sufficient to boost betalain pigment production As shown in Timoneda et al. (2018), an increase in tyrosine production in the context of the betalain biosynthetic pathway can translate into higher betalain pigment titers. In this study, expression of BvADHα coupled to the betalain biosynthetic genes in N. benthamiana leaves resulted in a 7-fold increase in pigment production compared to that obtained with BvADHβ (Timoneda et al., 2018). We hypothesised that the 3-fold increase in tyrosine production observed by ADH2 enzymes from S. lycopersicum and M. esculenta could also be used to increase betalain production. In order to test this, we constructed multi-gene vectors containing S. lycopersicum and M. esculenta ADH variants along with BvDODAa1, BvCYP76AD1 and MjcDOPA5GT, core genes of the betalain biosynthetic pathway (Figure 3.7a). Our vectors also included the Photinus pyralis luciferase (LUC) gene in order to account for differences in transformation efficiency. Transient expression of these constructs in N. benthamiana leaves resulted in the production of pigments within the infiltration spots, however, no clear visual differences in color intensity were observed between constructs containing ADH1 or ADH2 variants (Figure 3.7b). Spectrophotometric quantification of betacyanin pigments at 540 nm showed marginal differences between ADH1 and ADH2-containing constructs (Figure 3.7c). Absorbance values were corrected for the effect of chlorophyll a absorbance and normalised to luciferase expression. Constructs expressing MeADH2 showed a slight average increase of betacyanin production of 1.3-fold over constructs expressing MeADH1. However, in the case of S. lycopersicum enzymes, SlADH2-containing constructs had a lower performance than those containing SlADH1, which showed on average a 1.2-fold increase in pigment production over SlADH2 constructs (Figure 3.7c). Even though sample distribution seemed to overlap considerably between populations, differences observed were considered significant by Welch two-sample t-test analysis (p-values of 0.018 and 0.001 for S. lycopersicum and M. esculenta samples, respectively). Further repetitions of this experiment yielded similar results. We speculated that the lack of resolution between sample populations could be due to experimental limitations in using spectrophotometric measures of absorbance as a proxy for pigment production. Therefore, we decided to use liquid chromatography to quantify betalain content. Quantification of betanin via HPLC of infiltrated leaf extractions yielded similar results as previous analyses Figure 3.7d). MeADH2-expressing constructs showed a marginal increase of 1.2-fold over MeADH1-expressing constructs, whereas SlADH2-expressing constructs still underperformed SlADH1 constructs by a similar margin. However, these differences were not statistically significant according to Welch two-sample t-test analysis. Altogether, the use of SlADH2 and MeADH2 to increase the pool of tyrosine and increase metabolic flux towards the betalain metabolic pathway does not appear as a promising strategy to improve pigment A. Timoneda, PhD thesis, 2022 82 production, and cannot compete with the yield improvement previously observed with BvADHα (Timoneda et al., 2018). Figure 3.7. Transient expression of the betalain biosynthetic pathway alongside ADH variants. (a) Schematic of multi-gene vectors containing the betalain biosynthesis genes (DODA, CYP76AD1 and cDOPA5GT) and ADH1 or ADH2 variants, under constitutive promoters. (b) Purple pigmentation induced by transient expression of multi-gene vectors containing SlADH1, SlADH2, MeADH1 and MeADH2 (from left to right) in Nicotiana benthamiana leaves. (c) Spectrophotometric quantification of betacyanins at 540 nm from infiltrated leaf tissue. Absorbance data shown is relative to chlorophyl a absorbance at 660 nm and normalised to luminescence produced by the LUC gene. Each box plot represents 10 biological replicates. (d) Quantification of betanin by HPLC from infiltrated leaf tissue. Error bars represent the standard deviation calculated for 6 biological replicates. Asterisks represent significant differences in Welch two sample t-test analysis (p-value: 0.001 < **< 0.01< * < 0.05). 3.3. Discussion The identification of natural enzyme variants that can outperform currently known enzymes is a powerful strategy for the optimisation of heterologous production of valuable compounds (Owen et al., 2017). In previous work, we identified a variant of ADH in B. vulgaris containing an amino acid mutation that conferred relaxation of the enzyme’s tyrosine feedback inhibition resulting in higher catalytic efficiencies (Lopez-Nieves et al., 2018). Using this system as a model, we identified two taxa outside of the Caryophyllales, S. lycopersicum and M. esculenta, where duplications of the ADH gene gave rise to new ADH variants containing changes in the same amino acid. We hypothesised that these could represent new independent origins of relaxation of tyrosine feedback inhibition. Transient expression of these variants in N. benthamiana leaves showed in both cases that variants containing D or T at the studied position were able to produce approximately 3 times more tyrosine than the canonical variants containing E (Figure 3.5). Nevertheless, this increase was less pronounced than that obtained for the deregulated Chapter 3 83 B. vulgaris variant, BvADHα, which was able to produce approximately a 10-fold increase in tyrosine concentration under similar experimental conditions (Lopez-Nieves et al., 2018). Structure analysis using AlphaFold2 (Mirdita et al., 2021) confirmed that, as observed in spinach ADH enzymes (Lopez-Nieves et al., 2021), the E and D/T residues at the studied position in S. lycopersicum and M. esculenta enzymes also localised to the vicinity of the active site, in a protein pocket along with residues previously identified to be involved in reaction catalysis (Figure 3.3) (Christendat & Turnbull, 1999; Legrand et al., 2006; Schenck et al., 2017). This serves as further evidence that these residues could be involved in enzyme-substrate interaction and that similar molecular mechanisms could be convergently evolving in phylogenetically distant plant lineages to give rise to ADH enzymes with relaxed sensitivity to tyrosine feedback inhibition. To fully confirm this, enzyme purification and in vitro activity assays at different tyrosine concentrations should be performed in the future. To assess whether these residue changes are indeed responsible for the differences observed in tyrosine production between ADH1 and ADH2 variants, site-directed mutagenesis was performed to change their status to match the residues present in their variant counterparts. In the case of S. lycopersicum, mutation of this residue in ADH1 and ADH2 sequences was able to completely revert the tyrosine production phenotype (Figure 3.6). Therefore, the change from E to T in tomato ADH sequences is responsible for the observed increase in tyrosine concentration. This is the same residue change responsible for the increase of tyrosine production in S. oleracea ADHα enzymes (Lopez-Nieves et al., 2021), which further confirms the existence of similar molecular mechanisms for the convergent evolution of over-performing ADH enzymes in both taxa. S. lycopersicum and S. oleracea are phylogenetically distant species belonging to different orders of flowering plants: Solanales and Caryophyllales, respectively. In Caryophyllales, the evolution of enzymes with higher ability to produce tyrosine can be justified by the general abundance of betalain pigments and other tyrosine-derived compounds within these families. In this scenario, an increase in the metabolic flux towards tyrosine would support the appearance and establishment of tyrosine-derived metabolic pathways such as the betalain biosynthetic pathway (Timoneda et al., 2019). Plants in the Solanaceae do not produce betalain pigments, but have also been shown to contain high amounts of the tyrosine-derived compound tyramine (Kang et al., 2009). Tyramine is a key precursor for the production of hydroxycinnamic acid amides (HCAA) such as feruloyltyramine and 4-coumaroyltyramine, which have been widely described in solanaceous plants and play an important role in plant defence against pathogen infection and wounding (Pearce et al., 1998; Hahlbrock & Scheel, 2003; Campos et al., 2014). Lastly, we assessed the ability of the ADH2 enzymes from S. lycopersicum and M. esculenta to redirect metabolic flux into tyrosine-derived pathways by coupling ADH sequences to the betalain biosynthetic genes in multi-gene constructs. In our previous studies, this approach A. Timoneda, PhD thesis, 2022 84 resulted in a 7-fold increase in pigment production when using the de-regulated ADHα enzyme from B. vulgaris over the canonical ADHβ enzyme (Timoneda et al., 2018). In these experiments, transient expression of constructs containing SlADH2 and MeADH2 along with the betalain biosynthetic genes in N. benthamiana did not result in any significant differences in betalain yields compared to those expressing SlADH1 and MeADH1. This means, under our experimental conditions, the 3-fold increase in tyrosine levels observed when expressing SlADH2 and MeADH2 enzymes alone cannot translate into an increase in tyrosine-derived pigments when expressed in conjunction with downstream betalain genes. The reason for this is unclear. It is possible that tyrosine is being preferentially diverted towards other pre-existing metabolic routes in N. benthamiana, and that only after a bigger yield is obtained can this excess tyrosine feed the betalain biosynthetic pathway. Betalain biosynthetic enzymes could also be optimised for the use of tyrosine at the concentrations normally generated by the Caryophyllales ADHα, which are generally higher. Spatial separation of the different components of the engineered pathway could also make it difficult to achieve efficient flux towards pigment production. ADH enzymes localise to the plastids (Timoneda et al., 2018), whereas the rest of the betalain biosynthetic enzymes are expected to be expressed and translated in the cytosol (Chen et al., 2017b). Therefore, tyrosine has to be transported from the plastid stroma into the cytosol through the plastid membranes, step that will be limited by native transporter efficacy and availability. Overexpression of tyrosine transporters and the use of protein scaffolds that would anchor the betalain biosynthetic enzymes in close proximity to the plastid outer membrane could be a promising strategy to efficiently direct metabolic flux towards the betalain producing enzymes and increase pigment production. In conclusion, unlike BvADHα, SlADH2 and MeADH2 are unfortunately not good tools for increasing heterologous production of betalain pigments, however they are moderate evidence that ADH is a target for the modulation of primary metabolism outside of Caryophyllales. 3.4. Methods 3.4.1. Plant material and growth conditions. S. lycopersicum sp. ‘the amateur’ and M. esculenta were obtained from the Cambridge University Botanic Garden collection where they were grown in natural conditions (accession number 20150683*A and 19832811*E, respectively). N. benthamiana is a standard laboratory line maintained by selfing. N. benthamiana plants were grown in Levington M3 compost premixed with thiacloprid insecticide under the following controlled conditions: 20 °C, 60% humidity, 300 umol·m-2·s-1 light, and photoperiod of 16 h (16 h : 8 h, light : dark). Chapter 3 85 3.4.2. ADH phylogeny ADH phylogeny was constructed by Nathanael Walker-Hale (PhD student, Brockington Lab). In brief, ADH sequences from across flowing plants were collected from the 1,000 Plant transcriptome sequencing initiative (Carpenter et al., 2019) following the methods in Lopez- Nieves et al. (2018). Briefly, bait sequences were collected by searching for annotated ADHs in Phytozome v12 (https://phytozome-next.jgi.doe.gov/). Redundant transcripts were clustered using CD-HIT v4.8.1 (Fu et al., 2012) (-c 0.995 -n 5). Bait sequences were used to search each transcriptome with SWIPE v2.1.0 (Rognes, 2011), keeping the top 5 hits and removing any hit with a bitscore < 30. The resulting sequences were aligned with MAFFT v7.453 (Katoh & Standley, 2013) (--auto), the alignment was cleaned with pxclsq from the phyx package (Brown et al., 2017) (-p 0.1) and a tree was inferred with FastTree v2.1.10 (Price et al., 2010) (-wag -nosupport). Because non-homologous sequences are likely to fall on long branches, we trimmed any tip longer than 1.5 substitutions per site, and any tip longer than 1 substitution per site and > 10x the length of its sister. Further, because monophyletic transcripts originating from the same taxon may represent true duplications or splicing isoforms, we performed monophyletic masking, retaining in each case the longest transcript in the cleaned alignment. Finally, we cut subtrees at internal branches longer than 1 substitution per site, retaining subtrees containing bait sequences. Sequences corresponding to subtrees were extracted and alignment, tree inference and cleaning were conducted a further two times, as above. A final tree was inferred from the resulting sequences by aligning with MAFFT (--genafpair -- maxiterate 1000) and cleaning with pxclsq as above, then running RAxML v8.2.12 (Stamatakis, 2014) (-m PROTGAMMAAUTO). Sequences were selected for testing by investigating variation across the alignment at sites previously identified as important for feedback inhibition by Lopez-Nieves et al., (2021). 3.4.3. Structure prediction of ADH enzymes using AlphaFold2 Protein structure was predicted using the ColabFold tool (Mirdita et al., 2021), which integrates the AlphaFold2 protein structure and complex prediction (Jumper et al., 2021; Evans et al., 2021) using multiple sequence alignments generated through MMseqs2. Protein structure was later visualised using the PyMOL Molecular Graphics System, Version 2.0 (Schrödinger LLC, New York, USA). A. Timoneda, PhD thesis, 2022 86 3.4.4. RNA extraction and cDNA synthesis Seedling tissue from S. lycopersicum and leaf tissue from M. esculenta were snap frozen and ground to a fine powder with a mortar and pestle in liquid nitrogen. RNA extraction was performed on 100 mg ground tissue using the Concert Plant RNA Reagent (Invitrogen, Carlsbad, CA, USA) followed by the TURBO DNA-free kit (Ambion, Carlsbad, CA, USA) to remove DNA according to the manufacturer’s specifications. RNA concentration was quantified by Nanodrop and RNA integrity was assessed by 1.5% agarose gel electrophoresis. cDNA libraries were prepared using BioScript Reverse Transcriptase (Bioline Reagents, London, UK) and an oligo dT primer, according to the manufacturer’s recommendations. Oligo dT primer sequence can be found in Appendix 2. 3.4.5. Gene isolation and synthesis The B. vulgaris ADHα and ADHβ sequences used were those previously isolated in Timoneda et al., (2018). S. lycopersicum ADH1 and ADH2 sequences were amplified by PCR from cDNA libraries using the Phusion High-Fidelity DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) as described in section 2.4.2. Primers used are listed in Appendix 2. PCR products were visualised by gel electrophoresis and purified using the Invitrogen PureLink Quick Gel Extraction Kit (Invitrogen, Carlsbad, CA, USA), according to manufacturer’s recommendations. M. esculenta ADH1 and ADH2 sequences were codon optimised for N. benthamiana, domesticated to remove BsaI and BpiI restriction enzyme sites and synthesised by Genewiz (Bishop’s Stortford, UK). Synthesised sequences included… BsaI. The MeADH2 template sequence retrieved from transcriptomic data was a partial sequence lacking the predicted cTP at the N-terminal end of the sequence. The synthesised MeADH2 is a chimeric sequence containing 204 bp corresponding to the BvADHα cTP at the start of the sequence. Full amplified and synthesised sequences can be found in Appendix 1. 3.4.6. Vector construction Cloning into expression vectors and construction of multi-gene vectors containing the betalain biosynthetic genes were carried out using the MoClo Golden Gate cloning technology (Weber et al., 2011; Werner et al., 2012) as described in section 2.4.6. A new set of primers were designed to re-amplify ADH sequences from previous PCR products and include overhangs containing BpiI restriction sites and the four base pair MoClo syntax required for cloning into the pICH41308 Level 0 acceptor vector for CDS1 parts (5’-AATG-3’ in forward primers; 5’-GCTT-3’ in reverse primers). Primer sequences are listed in Appendix 2. For vectors expressing only Chapter 3 87 ADH variants, cloning was performed directly from Level 0 to the pICH86988 Level 2 acceptor for single gene expression under the CaMV 35S (35S) promoter and A. tumefaciens octopine synthase (ocs) terminator via BsaI. For the construction of multi-gene vectors containing ADH variants coupled with the betalain biosynthetic genes previously constructed Level 1 vectors were used (Timoneda et al., 2018). These consisted of B. vulgaris DODAα1 under control of the long 35S promoter and the A. tumefaciens nopaline synthase (nos) terminator; B. vulgaris CYP76AD1 under control of the long 35S promoter and the A. thaliana actin 2 (act2) terminator; and M. jalapa cDOPA5GT under control of the Ub10 promoter and 35S terminator. New Level 1 vectors were constructed for the newly cloned S. lycopersicum and M. esculenta ADH1 and ADH2 variants under the control of the Ub10 promoter and 35S terminator. Each multigene vector also included the firefly Photinus pyralis luciferase (LUC) gene under the nos promoter and terminator to adjust for differences in transformation efficiency and within leaf variation (Bashandy et al., 2015). The luciferase gene was obtained from the plasmid pNWA62 provided by Dr Nick Albert (Plant and Food Research, Palmerston North, New Zealand) and had previously been modified to include an intron and codon optimised in order to enhance translation (Albert, 2015). Level 2 binary vectors were comprised of transcriptional units in the following order: PpLUC in position 1 reverse orientation, BvDODAα1 in position 2 forward orientation, BvCYP76AD1 in position 3 forward orientation, MjcDOPA5GT in position 4 forward orientation, and ADH variants in position 5 forward orientation. For control constructs containing BvADHα, previously made vectors were used (Lopez-Nieves et al., 2018). 3.4.7. Site-directed mutagenesis of ADH sequences To generate point mutations in the selected codons of ADH variants we used PCR site-directed mutagenesis and the MoClo Golden Gate cloning technology (Weber et al., 2011; Werner et al., 2012). We generated the following mutated sequences: SlADH1 E200T; SlADH2 T218E; MeADH1 E195T; and MeADH2 D197E. The generation of ADH1 mutated versions was performed via conventional PCR of the whole plasmid with overlapping primers that including the base pair change. Primer sequences are listed in Appendix 2. Plasmid amplification was performed on already existing Level 0 vectors containing SlADH sequences with the Phusion High-Fidelity DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) as described in section 2.4.2. Amplicons were then visualised on agarose electrophoresis and bands extracted with the Invitrogen PureLink Quick Gel Extraction Kit (Invitrogen, Carlsbad, CA, USA). Gel extractions were then digested with DpnI for 30 min at 37°C. DpnI only cuts methylated sites and it is used to eliminate template plasmid from the sample, which unlike the newly amplified plasmids, is of bacterial origin and is methylated. The resulting samples were then used for transformation of E. coli DH5α cells as described in section 2.4.4. For the generation of ADH2 A. Timoneda, PhD thesis, 2022 88 mutated versions, primers were designed to amplify two regions per sequence: one from the start codon to the base pair point mutation, and another from the base pair point mutation to the stop codon. Primers included overhangs containing BpiI restriction sites and the four base pair MoClo syntax according to their position: 5’-AATG-3’, for forward primers at the start of the sequence; 5’-GCTT-3’, for reverse primers at the end of the sequence; and newly generated syntax including the base pair point mutation for the forward and reverse primers overlapping at the mutation site. Primer sequences are listed in Appendix 2. PCR was performed over already existing Level 0 vectors containing MeADH sequences with the Phusion High-Fidelity DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) as described in section 2.4.2. Amplicons were visualised by gel electrophoresis and purified using the Invitrogen PureLink Quick Gel Extraction Kit (Invitrogen, Carlsbad, CA, USA). Cloning of both amplicons into the pICH41308 Level 0 acceptor vector for CDS1 parts was performed simultaneously using the long one-pot, one-step type IIS mediated cloning reaction (Engler et al., 2008). Cloning proceeded through Level 0, Level 1 and Level 2 modules as described in section 2.4.6. 3.4.8. Transformation of binary vectors in Agrobacterium tumefaciens The A. tumefaciens strain used for transient transformation of N. benthamiana plants was GV3101. Transformation of GV3101 with constructed vectors was performed as described in section 2.4.7. 3.4.9. Transient transformation of Nicotiana benthamiana Transient gene expression assays in N. benthamiana were performed according to the previously described agroinfiltration method (Sparkes et al., 2006a) with some modifications as published in Timoneda et al., 2018. In brief, A. tumefaciens GV3101 colonies carrying the constructed binary vectors were grown at 28°C in LB media supplemented with kanamycin (50 mg·L-1), gentamycin (25 mg·L-1) and rifampicin (50 mg·L-1) until reaching an OD600 of approximately 1.5. Cultures were then brought to a final OD600 of approximately 0.5 in infiltration media (10 mM MgCl2, 0.1 mM acetosyringone, 10 mM MES at pH 5.6). Cultures were left at room temperature for 2–3 h before infiltration. Infiltration was performed on 5-week old N. benthamiana plants and individual plants represented independent biological replicates. Young, expanding leaves were chosen to infiltrate and infiltration was facilitated by first generating a small nick on the adaxial leaf surface. For expression of tyrosine producing constructs whole leaves were infiltrated. For expression of betalain producing constructs two opposing infiltration spots were made per leaf. Chapter 3 89 3.4.10. Tyrosine extraction and quantification using HPLC Infiltrated leaves were sampled 3 days post-infiltration and freeze dried overnight using a Super Modulyo freeze dryer (Edwards Vacuum, Crawley, UK). Dried leaf tissue was then broken down to smaller fragments. Approximately 10 mg of dried tissue were weighed in a 2 mL tube and grinded with 2 glass beads and 400 μL of methanol:chloroform (2:1) extraction buffer in a Tissue Lyser II (QIAgen, Hilden, Germany) for 2 minutes at 30 Hz. After this, 125 μL of chloroform and 300 μL of MilliQ water were added and mixed by vortexing. Samples were centrifuged 10 min at 12,100 g and 400 μL of the aqueous upper layer was retrieved and dried in vacuo overnight on a EZ-2 SpeedVac (GeneVac, Ipswich, UK) at 47°C. Dried samples were then resuspended in 150 μL of previously filtered MilliQ water and sonicated for 10 min. After sonication, samples were clarified by centrifugation at 12,100 g for 5 min. Tyrosine was quantified with a Thermo Fisher Scientific Accela HPLC autosampler and pump system incorporating a photodiode array detector as previously described (Wang et al., 2017). 10 μL of the samples were separated using an Atlantis T3 column (100Å, 3 μm, 2.1 x 150 mm) from Waters (Milford, Massachussets, USA) under the following conditions: 10 min, 0.1% B; 15 min, 0.1-76% B; 2 min, 76-0.1% B; 10 min 0.1% B; where mobile phase A was 1% formic acid and solvent B was 100% acetonitrile, at a flow rate of 0.4 mL·min-1. To detect tyrosine, the fluorescence detector was set at 274 nm and 303 nm for excitation and emission wavelengths respectively. Quantification was based on standard curves generated by injecting serial dilutions of tyrosine (Acros Organics, ThermoFisher Scientific, Geel, Belgium) at 50, 100, 200 and 400 μM. 3.4.11. Betalain extraction and quantification using HPLC Betalain extraction and quantification using liquid chromatography was performed as described in section 2.4.15. 3.4.12. Betalain extraction and quantification using spectrophotometry Betalain extraction and quantification using spectrophotometry was performed as previously described in Timoneda et al., 2018, with some modifications. Infiltration spots were sampled three days post-infiltration piercing with a leaf corer (9 mm diameter) per triplicate and snap frozen in liquid nitrogen in 2 ml tubes with a 5 mm glass bead. Ten biological replicates were sampled and measured per construct. Sampled leaf tissue was ground frozen using a Tissue A. Timoneda, PhD thesis, 2022 90 Lyser II homogeniser (QIAgen, Hilden, Germany). Homogenised samples were resuspended in 900 μL of SPB extraction buffer (50 mM sodium phosphate buffer, 2 mM dithiothreitol, 10% v/v glycerol, 1% v/v Triton X-100) (Girin et al., 2010) and mixed by vortexing. Samples were then centrifuged at 12,100 g for 10 min, and 100 μL of each supernatant was transferred to individual wells of a black µCLEAR 96-well microplate (Greiner Bio-One, Kremsmünster, Austria). Three technical replicates were performed per sample. Luminescence and absorbance were measured for each well with a CLARIOstar microplate reader (BMG Labtech, Aylesbury, UK). To measure luminescence, 100 μL of SteadyGlo Luciferase Assay Substrate (Promega, Madison, WI, US) was added to each well before plate reading and luminescence was measured at 25 °C using standard settings with no filter. To ensure that luminescence levels fell within a linear range, a luciferase standard curve was also produced using five 1:10 serial dilutions of QuantiLum Recombinant Luciferase (Promega, Madison, WI, US) in SPB supplemented with 1 mM bovine serum albumin (BSA). Absorbance values ranging from 400– 700 nm were measured with a resolution of 5 nm. Betacyanin relative concentration was calculated as A540 − (0.1 × A660) where A540 and A660 are the absorbance values for betacyanins and chlorophyll a at 540 nm and 660 nm respectively, corrected to the average SPB buffer values for each wavelength. Obtained betacyanin values were then normalised to the luciferase luminescence measured for each well and corrected to the average luminescence units recorded for the SPB buffer alone. Chapter 4 91 Chapter 4. Bioproduction of betalains in Escherichia coli bioreactors and characterization of a highly active DODA enzyme in the Cactaceae 4.1. Introduction Betalain pigments have accumulated industrial interest due to their bioactive properties and their potential application in the formulation of functional foods. Betalains are powerful antioxidant molecules with free radical-scavenging activities that can go up to 3 to 7 times higher than some of the most effective natural antioxidants known such as ascorbic acid (vitamin C) (Cai et al., 2003; Gengatharan et al., 2015). Betalains and betalain-rich extracts also exhibit chemopreventive (Zou et al., 2005), lipid-lowering (Tesoriere et al., 2004; Wroblewska et al., 2011; Clemente & Desai, 2011) and antimicrobial properties (Čanadanović-Brunet et al., 2011; Velicanski et al., 2011; Tenore et al., 2012). However, the most obvious and extensive use of betalains in the industry sector is their use as natural colorants. Betanin extracted from beetroot is one of the most widely used food colorants approved by the European Food Safety Administration (EFSA) and the U.S. Food and Drug Administration (FDA) (Delgado-Vargas & Paredes-Lopez, 2003; EFSA, 2015; FDA, 2021). It is used in the colouring of many dairy products, beverages, candies and fermented sausages (Georgiev et al., 2010). More recently, betalains have also started to gain an important role in the formulation of plant-based meat alternatives, a promising and growing industry sector. To mimic real meat, plant-based meat requires a red meaty colour that has to disappear and shift to a browner coloration upon cooking. Betalains are unstable and degrade at high temperatures, which makes them an ideal candidate for alternative meat formulations (He et al., 2020a). Currently, beet extracts represent the biggest source of betalains for food coloring applications (Polturak & Aharoni, 2017). Breeding efforts and improvements in the pigment extraction processes have been made in order to enhance yields (Polturak & Aharoni, 2017; Ciriminna et al., 2018), but increasing scientific interest has been directed towards developing alternative biotechnological production methods. Cell suspension cultures and hairy root cultures of a number of Caryophyllales species can be used for betalain production in bioreactor systems (Georgiev et al., 2008; Polturak & Aharoni, 2017). Over the years, different bioreactor setups such as fluidised bed bioreactors (Khlebnikov et al., 1995), stirred tank bioreactors (Rodríguez- A. Timoneda, PhD thesis, 2022 92 Monroy & Galindo, 1999), airlift bioreactors (Shin et al., 2002), column bioreactors (Kino-Oka et al., 1995), electrophoretic tubular membrane reactors (Yang et al., 2003), and nutrient mist bioreactors (Dilorio et al., 1992) have been investigated for this purpose. However, recent advances in elucidating the genetic components of the betalain biosynthetic pathway have enabled the metabolic engineering of the pathway in other plant species and microorganisms that could be more suitable for industrial processes. Recently, production of a variety of betaxanthins and betacyanins has been achieved in E. coli bioreactors expressing a novel bacterial DODA enzyme from G. diazotrophicus (Guerrero-Rubio et al., 2019). These cultures, however, had to be directly fed with ʟ-DOPA, the substrate of the DODA enzyme, a specialised metabolite whose manufacture would increase the overall production cost. Further advances in expressing more components of the pathway in E. coli would be required to make this system cost-efficient. Reconstruction of the whole biosynthetic pathway has been achieved in S. cerevisiae which also allowed for the production of a diverse range of betalains with different colour properties (Grewal et al., 2018). Here, authors achieved betanin yields of ~17 mg·L-1. Although production in microbial systems offer a series of advantages over pigment extraction from crops and is already being adopted by the biotechnological start-up sector (Phytolon, 2021), this titers are still much lower than the betacyanin content of 500-800 mg·L-1 reported in red beet juice (Wruss et al., 2015). Consequently, there is a need to improve heterologous production methods. Identification of enzymes with enhanced activity from those currently used has the potential to improve betalain production yields. In this context, characterisation of enzyme kinetics can help highlight differences in the activity and mode of action of enzymes from different species. Purification and characterisation of DODA enzymes from fungal (A. muscaria), plant (B. vulgaris) and bacterial and cyanobacterial species (G. diazotrophicus and A. cylindrica) unveiled differences in the substrate affinity and catalytic speed of each enzyme (Girod & Zryd, 1991a; Gandía-Herrero & García-Carmona, 2012; Contreras-Llano et al., 2019; Guerrero-Rubio et al., 2020b). In plants, betalain biosynthesis has been inferred to have evolved up to four times in the order Caryophyllales: 1) in the Stegnospermataceae, 2) Amaranthaceae, 3) the Raphide clade, and 4) the Portulacineae clade (Figure 1.5) (Sheehan et al., 2020). All betalain- producing species in the order contain at least one DODAβ and one DODAα paralog product of an early gene duplication in the clade, however most species contain multiple paralogues of DODAα. Studies in DODAβ and DODAα genes from representative species belonging to the four different inferred origins of betalain pigmentation (Stegnosperma halimifolium, B. vulgaris, M. crystallinum, and Carnegiea gigantea, in respective order) showed that only one of the DODAα paralogs (named DODAα1) encodes a protein which exhibits high levels of ʟ-DOPA 4,5-dioxygenase activity (Sheehan et al., 2020). This supports the hypothesis that DODA activity, and therefore betalain pigmentation, convergently evolved independently in each of these clades. It is possible, therefore, that the differences in the molecular pathways that led to Chapter 4 93 high DODA activity in each clade, could have also given rise to DODA enzymes with different kinetic characteristics and different overall performance profiles. In this chapter, activity differences between four DODAα1 enzymes from Caryophyllales taxa representing different inferred origins of betalain pigmentation are explored in both E. coli cultures and in vitro. We also perform a deeper kinetic and molecular study of the C. gigantea DODAα1 from the Cactaceae, and ultimately investigate its potential for pigment production in industrial grade bioreactors. 4.2. Results 4.2.1. Plant DODAα1 enzymes from different Caryophyllales taxa exhibit different performances in Escherichia coli cultures DODAα1 sequences were selected from species belonging to taxa in the Caryophyllales previously inferred to represent independent origins of betalain pigmentation (Sheehan et al., 2020). The chosen DODAα1 sequences belonged to B. vulgaris in the Chenopodiaceae (BvDODAα1), C. gigantea in the Cactaceae (CgDODAα1), M. crystallinum in the Aizoaceae (McDODAα1), and S. halimifolium in the Stegnospermataceae (ShDODAα1). Sequences were codon optimised and synthesised for expression in E. coli, and introduced in the pET28a vector for inducible bacterial expression with a N-terminal 6xHis tag. Vectors containing DODAα1 sequences were transformed into competent E. coli BL21 cells for expression of recombinant proteins. Protein expression was induced overnight with 1 mM IPTG after which cultures were fed with 7.6 mM ʟ-DOPA, substrate molecule of the DODAα1 enzyme (Figure 4.1a). We used 7.6 mM of ʟ-DOPA as it had been identified as optimal concentration for similar cultures in previous studies (Guerrero-Rubio et al., 2019). Ascorbic acid was added to prevent spontaneous pigment oxidation. Yellow color production was observed after addition of the reaction substrate, further confirming DODAα1 enzymes were being produced and were functional (Figure 4.1b). Coloration was observed to develop in cultures at different speeds, but was faster in the culture producing CgDODAα1, where yellow coloration started appearing immediately after supplementation of ʟ-DOPA. HPLC analysis revealed yellow coloration in all cultures was due to the presence of betalamic acid and dopaxanthin in the media. Betalamic acid was detected at a maximum wavelength λm = 405 nm with a retention time of 13.8 min (Figure 4.1d). Dopaxanthin was detected at λm = 480 nm and a retention time of 13.1 min (Figure 4.1e). Parameters for their identification through liquid chromatography had previously A. Timoneda, PhD thesis, 2022 94 been established with real standards (Gandía-Herrero et al., 2005c, 2009). The intermediate compound 4,5-seco-DOPA was also detected at λm =360 nm at a retention time of 13.8 min (Figure 4.1c). Figure 4.1. Betalain production of plant DODAα1 enzymes expressed in Escherichia coli cultures. (a) Schematic of the pathway leading to dopaxanthin in the presence of ʟ-DOPA and the DODAα1 enzyme. S*, spontaneous reaction. (b) E. coli cultures expressing plant DODAα1 enzymes produce betalains after addition of ʟ- DOPA. Culture yellow coloration is due to the production of betalamic acid and dopaxanthin and increases over time. (c) HPLC detection of 4,5-seco-DOPA, (d) betalamic acid and (e) dopaxanthin in E. coli cultures expressing plant DODAα1 enzymes. Example shown corresponds to cultures expressing CgDODAα1. Bv, Beta vulgaris; Cg, Carnegiea gigantea; Mc, Mesembryanthemum crystallinum; Sh, Stegnosperma halimifolium. Production of 4,5-seco-DOPA, betalamic acid and dopaxanthin was followed over time using liquid chromatography of culture media samples at different time points (Figure 4.2, ¡Error! No se encuentra el origen de la referencia.). All products were monitored for 154 h after addition of substrate. As the direct product of the ʟ-DOPA 4,5-dioxygenase reaction, 4,5-seco-DOPA was the first compound to accumulate and reached maximum concentration at 9 h, after which it rapidly decreased (Figure 4.2a). 4,5-seco-DOPA is a short-lived intermediate and its accumulation is known to be followed by a fast decay as it cyclises into betalamic acid (Terradas & Wyler, 1991). Next, a swift increase in betalamic acid production was observed that peaked at 32 h, consistent with previous studies using a newly discovered bacterial DODA in E. coli bioreactors (Guerrero-Rubio et al., 2019). Chapter 4 95 Dopaxanthin was the last compound to accumulate as it is generated by spontaneous condensation of betalamic acid with ʟ-DOPA molecules present in the media (Figure 4.1a). Dopaxanthin concentration increased progressively until it reached maximum production at 120 h (Figure 4.2). This differs slightly from previous studies, that situate the maximum of dopaxanthin production at 96 h (Guerrero-Rubio et al., 2019). A subsequent repeat of this experiment showed very similar results but placed dopaxanthin’s maximum of concentration at 96 h, indicating some degree of variation may exist between experiments under these specific conditions. Figure 4.2. Time evolution of reaction intermediates and products in Escherichia coli cultures expressing plant DODAα1 enzymes. (a) HPLC analysis of the evolution of 4,5-seco-DOPA (grey), betalamic acid (yellow) and dopaxanthin (orange) over time. Results are expressed as concentration of compound (left) and peak area normalised to the maximum peak value (right). Results shown are represented by cultures expressing BvDODAα1. Cultures expressing CgDODAα1, McDODAα1 and ShDODAα1 can be found in Figure 4.3. (b) Comparison of enzyme performance in the production of 4,5-seco-DOPA (left), betalamic acid (center), and dopaxanthin (right). 4,5-seco-DOPA is an unstable compound for which a calibration curve cannot be produced, thus concentration cannot be calculated from peak areas and is not shown. When compared between themselves, cultures expressing different DODAα1 enzymes showed different performances. Cultures expressing the CgDODAα1 enzyme exhibited a higher accumulation of all three compounds at all studied time points (Figure 4.2b). CgDODAα1 performance was followed by McDODAα1, ShDODAα1 and BvDODAα1 in descending order. Intriguingly, cultures expressing BvDODAα1 exhibited overall the lowest production titers of the A. Timoneda, PhD thesis, 2022 96 four, despite it being one of the most studied and employed enzymes in the metabolic engineering of the betalain pathway (Polturak et al., 2016, 2017; Timoneda et al., 2018; Tian et al., 2019). CgDODAα1 cultures produced 4.3 times more 4,5-seco-DOPA, 9.6 times more betalamic acid and 7.8 times more dopaxanthin than BvDODAα1 cultures at their respective time points of maximum concentration (Figure 4.2b). Figure 4.3. Time evolution of reaction intermediates and products in E. coli cultures expressing plant DODAα1 enzymes (second repeat). HPLC analysis of the evolution of 4,5-seco-DOPA (grey), betalamic acid (yellow) and dopaxanthin (orange) produced by BvDODAα1, CgDODAα1, McDODAα1 and ShDODAα1-expressing E. coli cultures over time. Results are expressed as (A) total chromatogram peak area, (B) peak area normalised to the maximum peak value, and (C) concentration of Chapter 4 97 compound. 4,5-seco-DOPA is an unstable compound for which a calibration curve cannot be produced, thus concentration cannot be calculated from peak areas and is not shown. 4.2.2. Plant DODAα1 enzymes from different Caryophyllales taxa exhibit different kinetic parameters To further analyse the differences between the selected DODAα1 enzymes, we purified them and compared their kinetic properties in vitro. Recombinant protein purification was performed by nickel-chelating affinity chromatography on crude BL21 E. coli culture extracts. Purity of protein extracts was assessed through SDS-PAGE protein electrophoresis in denaturing conditions. CgDODAα1 samples showed a pure band slightly below 34 kDa (Figure 4.4a), which corresponded well with the molecular weight calculated in silico from the protein sequence (30.1 kDa). This band was also obvious in culture samples after addition of IPTG but was not present in samples taken before addition of IPTG, indicating a correct inducible expression of the protein (Figure 4.4a). Electrophoresis of BvDODAα1, McDODAα1 and ShDODAα1 protein extracts did not show single pure bands and instead revealed multiple banding patterns indicating low degrees of protein purity (gel image not shown). Our previous experiments served as confirmation of enzyme presence, given that cultures expressing BvDODAα1, McDODAα1 and ShDODAα1 were able to perform the ʟ-DOPA 4,5-dioxygenase reaction and produce the yellow pigments, betalamic acid and dopaxanthin (Figure 4.1b). Multiple attempts at optimizing the purification protocol did not result in any noticeable improvements. These included changing all reagents anew, increasing the number of washing steps of the His-select nickel affinity beads, using nickel affinity beads from different suppliers, using cobalt-based resin columns, decreasing imidazole concentration in washing steps, and decreasing salt concentration in buffers. Activities of the DODAα1 enzymes were characterised spectrophotometrically by addition of the protein extracts to reaction media containing ʟ-DOPA. In all cases, addition of the enzyme yielded a yellow coloration with a λmax centered at 432 nm (Figure 4.4b). Identity of the reaction products formed was verified via ESI-MS (Figure 4.5), and further confirmed the above HPLC identification. We identified the optimal pH for each enzyme by monitoring their ability to produce betalamic acid in reaction media at pHs ranging from 3 to 8.5. The optimal pH for the ʟ-DOPA 4,5-dioxygenase activity was determined to be pH 6.5 in the case of BvDODAα1, McDODAα1 and ShDODAα1, and pH 6 for CgDODAα1 (Figure 4.4c). These pH values were used to test enzyme activity at increasing substrate concentrations and characterise their kinetic parameters (Figure 4.4d). Results obtained for BvDODAα1, McDODAα1 and ShDODAα1 should be interpreted carefully, since the lack of purity in these extractions will have had an overestimation effect on enzyme concentration calculations and ultimately in the determination of their kinetic parameters. Detailed information on the purification of CgDODAα1 can be found A. Timoneda, PhD thesis, 2022 98 in Table 4.1. Figure 4.4. Characterisation of plant DODAα1’s ʟ-DOPA 4,5-dioxygenase activity. (a) Electrophoretic analysis for the expression and purification of recombinant CgDODAα1 from BL21 E. coli culture. Lane 1, soluble protein content of cells harvested prior to IPTG induction; lane 2, soluble protein content of cells harvested 20 h after IPTG induction; lane 3, eluted protein after affinity chromatography purification. Molecular weights are in kDa. Arrows point at purified CgDODAα1 band. (b) Spectral evolution of the transformation of ʟ-DOPA (2.5 mM) by the addition of pure CgDODAα1 enzyme to the reaction medium. Spectra were recorded at 10-min intervals for 180 min, using a scanning speed of 2,000 nm · min–1. AU, arbitrary units. (c) Effect of pH on DODAα1 enzyme activity. (d) Enzyme activity dependence on ʟ-DOPA concentration measured in 50 mM sodium phosphate buffer at each enzyme’s optimal pH. Bv, Beta vulgaris; Cg, Carnegiea gigantea; Mc, Mesembryanthemum crystallinum; Sh, Stegnosperma halimifolium. Table 4.1. Expression and purification of Carnegiea gigantea DODAα1. Step Volume (mL) Protein (mg·mL1) Total protein (mg) Activitya (μM·min-1) Specific activityb (μmol·min-1·mg-1) Purification fold Yieldc (%) Crude extractd 6.0 10.5 63.0 3.010 0.086 1,0 100 Ni2+ chromatography 7.0 4.3 30.1 4.220 0.294 3.4 163.6 Chapter 4 99 a Activity was determined using 1 μl of protein extract under the assay conditions described in Material and Methods Section 4.4.10. b Specific activity of crude extract was calculated using total protein quantification values of the extract. c Yield is calculated as the ratio between the activity of the total volume of extraction to that of the total volume of crude extract. d Crude extract was obtained from a cellular paste harvested from a 0.5-liter culture. Figure 4.5. ESI-MS identification of CgDODAα1 reaction products. Chromatograms from HPLC-ESI-TOF-MS analysis. Reaction was initiated with 4.3 μg of purified CgDODAα1 in reaction media supplemented with 2.5 mM ʟ-DOPA and 10 mM sodium ascorbate approximately 6 hours before measuring. Betalamic acid was followed at EIC 212.0553 m/z, 4,5-seco-DOPA was followed at EIC 230.0659 m/z and dopaxanthin was followed at EIC 391.1136 m/z. In all cases, we observed activity patterns typical of enzymes which experience substrate inhibition. Substrate inhibition is the most common deviation from Michaelis-Menten kinetics and happens when instead of reaching a steady-state equilibrium at the maximum reaction speed, the velocity of a reaction rises to a maximum and then descends as substrate concentration continues increasing. This phenomenon can be observed in all DODAα1 kinetic curves to different degrees but seemed less pronounced in the ShDODAα1 and CgDODAα1 enzymes (Figure 4.4d), meaning these enzymes can maintain higher rates of activity at higher ʟ-DOPA concentrations. Enzyme kinetic parameters were estimated using non-linear regression with the substrate inhibition rate equation previously described by Haldane and shown in Figure 4.6a (Haldane, 1930). The Michaelis-Menten constant (Km) represents the concentration of substrate at which the reaction velocity is half the Vmax and is an inverse measure of the enzyme’s affinity to the substrate. Out of the four DODAα1 enzymes studied, BvDODAα1 presents the lowest Km (0.19 A. Timoneda, PhD thesis, 2022 100 ± 0.04 mM) and therefore exhibits the highest affinity to the substrate. McDODAα1 shows the lowest affinity with a Km of 4.76 ± 8.1 mM (Figure 4.6b). The substrate inhibition constant (Ki) is the concentration of substrate required to produce half the maximum of inhibition and is an inverse measure of how potently the substrate inhibits the enzyme. In our results, McDODAα1 shows the strongest substrate inhibition with a Ki estimated at 0.19 ± 0.33 mM. ShDODAα1 experienced the lowest substrate inhibition at a Ki of 4.78 ± 1.41 mM (Figure 4.6.b), as expected from the kinetic curve shapes. The highest maximum velocity (Vmax) was by far achieved by CgDODAα1 at 860.21 ± 190.13 μM·min-1, which was almost 20 times higher than the next fastest enzyme, McDODAα1 (Figure 4.6b). This big difference, however, is likely due to the overestimation of BvDODAα1, McDODAα1 and ShDODAα1’s protein concentration as a result of poor extract purity when compared to CgDODAα1. Nevertheless, this result is consistent with our previous observations of enzyme performance in E. coli cultures which altogether accentuate the potential of CgDODAα1 for its use in pigment production optimisation. We therefore decided to focus on further studying the CgDODAα1 enzyme. Figure 4.6. Kinetic parameters of DODAα1 enzymes. (a) Schematic representation and rate equation for an enzymatic system with inhibition by excess of substrate as described by Haldane et al., 1965. (b) Table including kinetic parameters for each DODAα1 enzyme, inferred from non- linear regression using the above rate equation and data obtained from enzyme activity assays at different ʟ-DOPA concentrations. E, Enzyme; S, Substrate; P, Product; Km, Michaelis-Menten constant; Ki, Substrate inhibition constant; vmax, maximum velocity achieved by the system; Bv, Beta vulgaris; Cg, Carnegiea gigantea; Mc, Mesembryanthemum crystallinum; Sh, Stegnosperma halimifolium. Chapter 4 101 4.2.3. Molecular characterisation of the DODAα1 enzyme in the Cactaceae Purified CgDODAα1 protein was analysed by HPLC-ESI Q-TOF MS in order to determine its accurate molecular mass. Mass spectra showed a single peak with a molecular mass of 32.004 kDa, in accordance with the band size obtained by SDS-PAGE (Figure 4.4a). To fully characterise the protein, its peptide mass fingerprint was determined by MALDI-TOF analysis after trypsin digestion. The main peptides identified correspond to the masses 749.86 m/z [(R)TWDHGTLGYASYK(F)], 630.28 m/z [(R)YDDVNNYQTK(A)], 1168.13 m/z [(K)IAHPIPEHFLPLHVAMGAAGEK(S)], 791.04 m/z [(K)ETFYVSHGNPAMLADESFIAR(N)] and 1506.06 m/z [(R)GFDHSSWVPLSLMYPEADIPVCQLSVQPHLSASHHFDVGR(A)]. Fast protein liquid chromatography (FPLC) was used to determine if CgDODAα1 enzyme is a monomer or whether it forms oligomers under native conditions. We observed two peaks eluting between 13-14.5 min and 14.5-16 min, with maximum concentrations being achieved at 13.5 min and 15 min for each peak (Figure 4.7a). The estimated molecular mass of the later peak was approximately 27.1 kDa. This is similar to the results obtained under denaturing conditions in SDS-PAGE (Figure 4.4a) and to the aforementioned calculated mass of CgDODAα1, and therefore likely corresponds to a protein monomer. The molecular mass of the earlier peak was estimated at 57.4 kDa, which is nearly double the value of the second peak. Therefore, the observed results obtained by gel filtration show that the CgDODAα1 protein exists both in monomer and dimer form under native conditions. We did not observe any difference in the activity of the elution fractions at the corresponding volumes for each individual monomer and dimer peak separately. To further determine the factors that can affect dimerisation, we studied monomer to dimer ratios under different conditions (Figure 4.7b). Ratios were calculated from chromatograms as the percentage of monomer or dimer peak area over the total area of both peaks. Samples were incubated for 30 minutes in each condition to allow for the monomer-dimer system to reach equilibrium, although incubation time did not show any effect over dimer formation (Figure 4.7c). Presence of salts has been previously reported to affect dimer formation (Shima et al., 1998; Sakurai et al., 2001; Purohit et al., 2017). Thus, we studied the effect of sodium chloride (NaCl) and calcium chloride (CaCl2) on the dimerisation of CgDODAα1. Dimer formation increased with increasing NaCl concentrations (Figure 4.7b). We observed a dimer proportion of 25.61% when proteins were incubated at 150 mM NaCl and 53.34% when incubated at 1 M NaCl. In contrast, a dimer proportion of 7.07% was observed when incubated with no salts. On the contrary, dimer formation decreased to a proportion of 1.6% when CgDODAα1 was incubated at 150 mM CaCl2 (Figure 4.7b). Protein concentration also had an effect on dimerisation ratios (Figure 4.7b). Dimerisation levels increased with protein concentration when incubated without salts and was A. Timoneda, PhD thesis, 2022 102 20 times higher in samples with 10 times more protein. This trend was more erratic and did not seem to follow any clear pattern when incubated with NaCl (Figure 4.7c). Figure 4.7. CgDODAα1 forms dimers in native conditions. (a) Gel filtration profile of undiluted CgDODAα1 in phosphate buffer containing 150 mM NaCl. Two peaks corresponding to the dimer and the monomer were observed. Protein dimers elute earlier due to their lower retention ability in the porous Superdex 200 column. (b, c) Monomer and dimer ratios at different salt concentrations, protein dilutions and incubation times. Ratios were calculated from chromatograms as the percentage of monomer or dimer peak area over the total area of both peaks. Dimer formation is favored at higher NaCl and protein concentrations. Monomer, striped bar; dimer, solid bar. 4.2.4. Identification of CgDODAα1’s metal cofactor DODA enzymes belong to the class of extradiol dioxygenase enzymes. These enzymes are known to use iron as a metal cofactor (Lipscomb, 2008). We used Inductively Coupled Plasma Mass Spectrometry (ICP-MS) to look for the presence of metal atoms in purified samples of CgDODAα1. The array of metals screened included iron (Fe), titanium (Ti), chromium (Cr), manganese (Mn), cobalt (Co), nickel (Ni), copper (Cu), zinc (Zn), arsenic (As), cadmium (Cd), barium (Ba), thallium (Tl), lead (Pb) and uranium (U). Of these, the metals found at the highest Chapter 4 103 concentration were Ni, at 680 parts per billion (ppb), Fe at 310 ppb, and Zn at 180 ppb (Table 4.2). The high nickel levels observed in the sample are likely carry-over from the nickel affinity beads used during the purification process. We therefore hypothesised that CgDODAα1 uses Fe as a metal cofactor. To gain more evidence for this hypothesis, we analysed the activity of CgDODAα1 in the presence of Fe, Ni and Zn (Figure 4.8a). Despite their low purity, we also tested the effect of these metals on BvDODAα1, McDODAα1 and ShDODAα1’s activity. In all cases, incubation of the DODAα1 enzymes with NiCl2 and ZnCl2 resulted in the abolition of enzyme activity. However, when incubated with FeCl2, activity increased for all tested proteins. CgDODAα1 activity was 9.4 times higher in the presence of Fe2+ than without it. In the case of BvDODAα1, McDODAα1 and ShDODAα1, activity was 4.8, 6.1 and 4.4 times higher respectively. Altogether, these results indicate that Fe2+ is likely to be the metal cofactor used by CgDODAa1, and possibly the rest of the Caryophyllales DODAα1 enzymes. Since supplementation with Fe2+ resulted in an increase of enzyme activity in vitro, we decided to study whether this could also translate into an increase of pigment production in E. coli cultures expressing the DODAα1 enzymes. Fe2+ supplementation was tested at two different stages: in the LB media used during the earlier cell growth phase, and/or in the water solution containing the reaction substrate used during the later pigment production phase. Fe2+ supplementation did not result in any significant yield improvement in any of the conditions assayed (Figure 4.8c). Although 4,5-seco-DOPA levels were higher when adding Fe2+ during the pigment production phase, this did not translate into a significantly higher accumulation of betalamic acid or dopaxanthin. In fact, maximum dopaxanthin levels were achieved in cultures where Fe2+ was not added at any point of the process (Figure 4.8c). Moreover, cultures developed a darker brown color immediately after addition of Fe2+ to the water solution used Concentration (ppb) Ti 58.39 Cr 0.36 Mn 54.28 Fe 310 Co < 0.000 Ni 680.25 Cu 2.07 Zn 180.6 As < 0.000 Cd < 0.000 Ba 6.44 Tl < 0.000 Pb < 0.000 U < 0.000 Table 4.2. Metal concentrations detected in CgDODAα1 extracts. Quantification of metal atoms by Inductively Coupled Plasma Mass Spectrometry (ICP-MS) in purified samples of CgDODAα1. The array of metals screened included iron (Fe), titanium (Ti), chromium (Cr), manganese (Mn), cobalt (Co), nickel (Ni), copper (Cu), zinc (Zn), arsenic (As), cadmium (Cd), barium (Ba), thallium (Tl), lead (Pb) and uranium (U). A. Timoneda, PhD thesis, 2022 104 during the pigment production phase, suggesting Fe2+ could be interacting with the ʟ-DOPA molecules in the media (Figure 4.8b). This is an undesirable byproduct for the industrial production of pure colors, and thus supplementation of Fe2+ should be avoided in media where the reaction is taking place. The use of Fe2+ in the LB medium at earlier stages did not suppose any improvement in betalamic acid or dopaxanthin production over the control cultures (Figure 4.8c). Therefore, Fe2+ supplementation in betalain producing E. coli cultures is not an effective strategy for the improvement of pigment production. Figure 4.8. Effect of metals on plant DODAα1 activity. (a) Fold change in ʟ-DOPA dioxygenase activity of purified plant DODAα1 enzymes in vitro in reaction media containing 50 μM FeCl2 (+Fe), 50 μM NiCl2 (+Ni) and 50 μM ZnCl2 (+Zn). Values shown are relative to a control reaction in absence of metals (not shown, value = 1). (b) Betalain production of E. coli cultures expressing the CgDODAα1 enzyme after addition of ʟ-DOPA in different conditions of supplementation with FeCl2. On the x axis, first ‘Fe’ refers to the presence (+Fe) or absence (-Fe) of FeCl2 in the LB medium used in the earlier cell growth phase; second ‘Fe’ refers to the presence (+Fe) or absence (-Fe) of FeCl2 in the water solution containing the reaction substrate used during the later pigment production phase. Yellow culture coloration is due to the production of betalamic acid and dopaxanthin. Unusual dark coloration is observed in the cultures supplemented with FeCl2 at the later pigment production stage. (c) Comparison of CgDODAα1 enzyme performance in the production of 4,5-seco-DOPA (left), betalamic acid (center), and dopaxanthin (right) in different conditions of culture supplementation with FeCl2. 4.2.5. Scaled-up production of betalains in E. coli bioreactors expressing plant DODAα1 enzymes Given the observed higher performance of CgDODAα1 in E. coli cultures, we decided to explore their biotechnological potential for pigment production in bioreactors of higher volumes. E. coli BL21 cells expressing CgDODAα1 were grown in 2 L bioreactors with controlled mechanical Chapter 4 105 agitation and temperature regulation. Protein expression was induced with 1 mM IPTG for 16h after which cultures were retrieved, pelleted and transferred back to the bioreactor in water supplemented with 15 mM ascorbic acid and 7.6 mM ʟ-DOPA. Yellow coloration then appeared and intensified over time (Figure 4.9b). Once again, we observed a first increase in betalamic acid later followed by accumulation of dopaxanthin (Figure 4.9a). Maximum production of betalamic acid was achieved between 48 and 72 h, at which point the calculated concentration was approximately 3 mM. Dopaxanthin’s maximum titers were achieved between 72 and 96 h, and were equivalent to an approximate concentration of 130 µM of compound. For a 2 L reactor, these values equate to a total of 1.25 g of betalamic acid and 100 mg of dopaxanthin. Previous publications using E. coli strains expressing a bacterial DODA enzyme in similar 2 L bioreactor settings reported 148.25 mg of dopaxanthin purified with a 92.70% compound recovery (Guerrero-Rubio et al., 2019). Further scaling-up of this system in a 30 L bioreactor allowed for a total production of approximately 2.5 g of dopaxanthin (Figure 4.9d). This is to our knowledge the highest betalain yield produced in a bioreactor to date. Available enzyme kinetic data has the potential to guide and inform the optimization process of industrial compound production. 7.6 mM is the solubility limit of ʟ-DOPA and, until this point, the concentration of substrate we used in our experiments following previous publications (Contreras-Llano et al., 2019; Guerrero-Rubio et al., 2019, 2020b). However, the kinetic model obtained for CgDODAα1 shows enzyme activity reaches its maximum at the range between 1 and 2 mM of substrate, after which it experiences a decay (Figure 4.4d). In fact, according to this model, activity at 7.6 mM of ʟ-DOPA is expected to be less than half of the maximum. In order to test whether optimisation of substrate concentration can have an effect on pigment production, we supplied an initial concentration of 2 mM to the CgDODAα1-expressing 2 L bioreactors (Figure 4.9c). Early production of 4,5-seco-DOPA was twice as high in the bioreactor fed with 2mM of ʟ-DOPA compared to the bioreactor fed with 7.6 mM of ʟ-DOPA, indicative of an increase in CgDODAα1’s turnover rate. Accumulation of betalamic acid was also improved when using 2mM of substrate (Figure 4.9c). Betalamic acid titers were generally higher, with a maximum concentration of 4.7 mM (a 1.6-fold increase compared to the previous 7.6 mM-fed bioreactor), and were also maintained at higher levels for longer (Figure 4.9c). Dopaxanthin production, however, experienced a decrease and was half that obtained when using 7.6 mM of substrate (65 µM). This is likely a result of the increased activity of the CgDODAα1 enzyme. To make dopaxanthin, two molecules of ʟ-DOPA are employed: one is used by the enzyme to produce betalamic acid, and a second one spontaneously condenses with the betalamic acid to form dopaxanthin (Figure 4.1a). If CgDODAα1 is indeed acting more efficiently, and thus consuming ʟ-DOPA at a faster rate, this could result in a reduction of available ʟ-DOPA to condense with betalamic acid, a step already known to be slower (Contreras-Llano et al., 2019), and ultimately a lower production of dopaxanthin. Therefore, A. Timoneda, PhD thesis, 2022 106 enzyme kinetic data can help us optimise substrate concentration to increase production of betalamic acid, but not dopaxanthin pigment which represents the desired final stable product. Figure 4.9. Biotechnological production of dopaxanthin in E. coli bioreactors expressing CgDODAα1. (a) HPLC analysis of the evolution of 4,5-seco-DOPA (grey), betalamic acid (yellow) and dopaxanthin (orange) production over time in 2 L bioreactors fed with 7.6 mM of ʟ-DOPA. Results are expressed as total chromatogram peak area values. (b) Dopaxanthin production in 2 L bioreactors at (left to right) 0, 1, 24 and 96 h. (c) Time course comparison of the evolution of 4,5-seco-DOPA (left), betalamic acid (center) and dopaxanthin (right) production in bioreactors fed with 7.6 mM of ʟ-DOPA (dashed line) and 2 mM of ʟ-DOPA (solid line). Results are expressed as total chromatogram peak area for 4,5-seco-DOPA and compound concentration for betalamic acid and dopaxanthin. (d) Dopaxanthin production in 30 L bioreactor (left), culture sample (right). We hypothesised that the addition of a new amine to the reaction media could alleviate this situation by providing a new pool of compounds to condense with the high levels of available betalamic acid (Figure 4.10b). A new 2 L bioreactor system was set up in which indoline was also added to the reaction media alongside ʟ-DOPA at 2 mM. Indoline is an aromatic bicyclic Chapter 4 107 amine which in condensation with betalamic acid shifts its spectral properties and produces the red indoline-betacyanin. Addition of ʟ-DOPA and indoline to the reaction media containing CgDODAα1-expressing E. coli cells produced a pink-red color that intensified over time (Figure 4.10d). In this system, betalamic acid levels peaked earlier at 48 h and then rapidly decayed as the concentration of indoline-betacyanin steadily increased (Figure 4.10a,c). The maximum concentration achieved by betalamic acid was of approximately 0.38 mM, 8 and 12 times lower than that obtained in previous bioreactors only fed with 7.6 mM and 2 mM ʟ-DOPA respectively, suggesting a fast and efficient condensation with indoline. Dopaxanthin titers in this case remained very low for the entire duration of the experiment, showing betalamic acid preferentially condenses with indoline molecules (Figure 4.10a,c). Indoline-betacyanin levels continued increasing until achieving a maximum concentration of 0.21 mM at 120 h and a total 135 mg of compound in the reactor. This represents a 1.6-fold increase from the maximum concentration of dopaxanthin produced by the first bioreactor fed with 7.6 mM of ʟ-DOPA and corresponds with the 1.6-fold increase in betalamic acid production achieved in the second bioreactor fed with 2 mM of ʟ-DOPA. Figure 4.10. Biotechnological production of indoline-betacyanin in E. coli bioreactors expressing CgDODAα1. (a, c) HPLC analysis of the evolution of 4,5-seco-DOPA (grey), betalamic acid (yellow), dopaxanthin (orange) and indoline-betacyanin (pink) production over time in 2 L bioreactors fed with 7.6 mM of ʟ-DOPA. Results are expressed as (a) concentration of compound and (c) total chromatogram peak area values. (b) Schematic of the pathway leading to dopaxanthin and indoline-betacyanin in the presence of ʟ-DOPA, indoline and the DODAα1 enzyme. S*, spontaneous reaction. (d) Indoline-betacyanin production in 2 L bioreactors at (left to right) 1, 3, 5 and 48 h. A. Timoneda, PhD thesis, 2022 108 4.3. Discussion Four DODAα1 sequences from different Caryophyllales taxa inferred to represent different independent origins of betalain pigmentation were chosen for analysis and comparison of their in vitro properties. Recombinant E. coli cultures expressing the four DODAa1 enzymes, and supplemented with the substrate, L-DOPA, produced yellow coloration that increased over time as a result of betalamic acid and dopaxanthin production in the media (Figure 4.1b). Cultures expressing the DODAα1 from C. gigantea in the Cactaceae outperformed the cultures expressing the other three Caryophyllales DODAα1 enzymes in the production of reaction intermediates and products (Figure 4.2b). CgDODAα1-expressing cultures exhibited 9.6 and 7.8 times more accumulation of betalamic acid and dopaxanthin respectively than the lowest performing culture, which expressed the DODAα1 enzyme from B. vulgaris. This was surprising since BvDODAα1 is one of the most studied and employed enzymes in the metabolic engineering of the betalain pathway (Gandía-Herrero & García-Carmona, 2012; Polturak et al., 2016, 2017; Timoneda et al., 2018; Tian et al., 2019). In previous experiments performed by our group, transient expression of CgDODAα1 along with the CYP76AD1 from B. vulgaris and the cDOPA5GT from M. jalapa in a plant host (N. benthamiana leaves) did not result in a higher pigment accumulation compared to BvDODAα1, McDODAα1 and ShDODAα1 (Sheehan et al., 2020). In fact, in this study, CgDODAα1 showed the lowest pigment accumulation of the four. This difference could be explained by the different concentrations of substrate available to the DODAα1 enzymes in each experiment and their effect on enzyme activity. Initial ʟ-DOPA concentration in our E. coli cultures was set to 7.6 mM following previous publications (Contreras-Llano et al., 2019; Guerrero-Rubio et al., 2019, 2020b), however, this concentration is likely to be different from the one achieved in N. benthamiana leaves after transformation, which will be determined by the activity of BvCYP76AD1 and the amount of available tyrosine in the tissue. When compared to previous pigment yields reported using a bacterial DODA enzyme from G. diazotrophicus (GdDODA) in E. coli cultures under similar set-up and feeding conditions (Guerrero-Rubio et al., 2019), CgDODAα1 produced 128 times more betalamic acid and 15 times more dopaxanthin than GdDODA, which highlights the strong benefits of using plant enzymes over bacterial enzymes for the biotechnological production of betalain pigments. For a more robust comparison both enzymes should be tested simultaneously, since differences in experimental conditions may affect the overall productivity of the system. Purification and in vitro kinetic characterization of the DODAα1 enzymes revealed kinetic curve shapes typical of enzymes that experience inhibition by excess of substrate (Figure 4.4d). Chapter 4 109 Substrate inhibition is known to be involved in the control of many physiological processes, but has also been claimed to be an artefact caused by the use of artificially high substrate concentrations in enzymatic assays (Kokkonen et al., 2021). However, CgDODAα1 and ShDODAα1 were able to maintain higher activities at higher substrate concentrations, while BvDODAα1 and McDODAα1 experienced a sharper decay in activity as substrate concentration increased after reaching a maximum around 0.5 - 1 mM of ʟ-DOPA. This could explain why CgDODAα1 can outperform other DODAα1 enzymes in 7.6 mM ʟ-DOPA cultures, but not in heterologous expression assays in N. benthamiana where substrate concentration could be lower. This hypothesis could be tested in the future by establishing a similar time course comparison in reaction media supplemented with lower ʟ-DOPA concentrations. Despite its lower performance in E. coli cultures, BvDODAα1 exhibits the lowest Km (0.19 ± 0.04 mM) and therefore has the highest affinity to the substrate of the tested enzymes (Figure 4.5). Km values obtained for CgDODAα1, McDODAα1 and ShDODAα1 were 0.9 ± 0.3, 4.76 ± 8.1, and 0.6 ± 0.14 respectively. All these, with the exception of McDODAα1, are lower than the Km values previously calculated for non-plant species such as A. muscaria (3.9 M) (Girod & Zryd, 1991a), G. diazotrophicus (1.36 mM) (Guerrero-Rubio et al., 2020b) and A. cylindrica (53 M) (Guerrero-Rubio et al., 2020b), suggesting that plant DODAα1 enzymes generally show a stronger substrate affinity. Another previous publication reported a Km of 6.9 ± 0.9 mM for the DODA enzyme in B. vulgaris (Gandía-Herrero & García-Carmona, 2012), however we were able to identify the sequence used in the study as belonging to the BvDODAα2 clade, whose function and involvement in pigment production still remains unclear (Chung et al., 2015; Brockington et al., 2015; Sheehan et al., 2020). Of our tested enzymes, McDODAα1 experiences the strongest substrate inhibition with a Ki estimated at 0.19 ± 0.33 mM (Figure 4.5). The highest Vmax was achieved by CgDODAα1 (860.21 ± 190.13 μM·min-1), which was many orders of magnitude higher than the other tested enzymes (Figure 4.5). These parameters, however, should be interpreted with caution. Only CgDODAα1 showed an adequate level of purity in SDS-PAGE gel electrophoresis (Figure 4.4a). Despite numerous optimization attempts, multiple banding patterns were observed in BvDODAα1, McDODAα1 and ShDODAα1 purifications. This excess of unwanted protein translates in an overestimation of the quantity of DODAα1 since the Bradford quantification method is non-specific and will take into account all proteins present in the sample. Enzyme activity is determined in relation to the concentration of protein present in the reaction, thus an overestimation of enzyme quantity results in lower activity values such as the ones observed for BvDODAα1, McDODAα1 and ShDODAα1. This mainly affects the value of the Vmax, since it directly depends on the protein concentration. Affinity and inhibition constants are unlikely to be strongly affected and can still be considered informative. Tagging of the DODAα1 enzymes in C-terminal or with tags of different nature, such as self-cleavable intein tags (Cooper et al., 2018), is recommended in the A. Timoneda, PhD thesis, 2022 110 future to obtain purer protein extracts. Altogether, the observed differences in kinetic curves and parameters between the four tested DODAα1 enzymes could underly a diversity of specialisation mechanisms in the optimisation of betalain production occurring after different independent establishments of the pathway. Counterintuitively, purification of CgDODAα1 resulted in a higher yield than the crude extract (395%; Table 4.1). This is technically not possible since a fraction of protein is routinely lost during the purification process. However, yield data is calculated using protein activity values, which means a higher yield value after purification could be achieved if the purified enzyme is more active than the enzyme present in the crude extract. There could be several explanations for this phenomenon: 1) proteins or compounds present in the crude extract inhibit enzyme activity; 2) enzyme oligomerisation enhances activity and is favoured under purified conditions. Further molecular characterisation of CgDODAα1 revealed it can form dimers in native conditions (Figure 4.6). Dimerisation ratios were observed to increase with NaCl and protein concentration. Betalains have long been proposed to be involved in plant response to salinity stress (Jain & Gould, 2015b). Many families in the Caryophyllales include halophyte species able to grow in salt marshes and coastal zones with high soil salinity. Red morphotypes of the betalain-producing Disphyma australe are more commonly found in intertidal zones than on higher land, where the green morphotype predominates (Jain & Gould, 2015a). In this context, an increase in dimer formation with increasing salt concentrations could be part of a physiological mechanism evolved to build up and maintain high levels of betalains as a defense against salinity stress. We did not find any difference in activity in the FPLC eluted fractions corresponding to the monomer and dimer peaks separately, but this could have been due to the monomer-dimer equilibrium being re-established in the time needed to set up the experiment. In the future, in vitro activity assays under different salt concentrations could be used to compare dimer and monomer efficiency in the production of betalamic acid. Given the observed high performance of CgDODAα1 in E. coli cultures, we decided to explore its industrial applicability by scaling pigment production up to 2 L and 30 L bioreactors. We were able to obtain an overall total yield of 2.5 g of dopaxanthin in 30 L bioreactors, which is to our knowledge the highest amount of betalains produced in a bioreactor to date. Enzyme kinetic parameters were successfully used to optimise substrate concentration for the maximisation of betalamic acid production. Bioreactors fed with 2 mM of ʟ-DOPA produced 1.6 times more betalamic acid and maintained higher concentrations for longer than the same bioreactors fed with 7.6 mM (Figure 4.9). This represents a double-fold improvement on the overall cost effectiveness of the system by accessing higher product titers at the same time that initial substrate costs are reduced. In the specific case of dopaxanthin this did not result in a parallel increase of the final desired product due to the fact that, to form dopaxanthin, the same pool of ʟ-DOPA is needed both as enzyme substrate and in the later condensation step of the pathway Chapter 4 111 (Figure 4.1a). However, we show that including other amines that condense with betalamic acid to the system, such as indoline, alleviates the pressure over ʟ-DOPA that can then be preferentially used by DODAα1 to feed the pathway. Altogether, our kinetic, molecular and scale-up findings indicate plant DODAα1 enzymes have a substantial potential for the biotechnological production of betalain pigments and their use for this purpose should be preferred over bacterial or fungal enzymes. 4.4. Methods 4.4.1. Vector construction DODAα1 sequences from B. vulgaris, C. gigantea, M. crystallinum and S. halimifolium were codon optimised for E. coli, domesticated for NdeI and NotI and synthesized by Twist Biosciences (San Francisco, CA, USA). Overhangs including restriction sites for NdeI and NotI were added before (5’-GAAGTGCCATTCCGCCTGACCTCAT-3’) the start codon and after (5’- GCGGCCGCAGGCTAGGTGGAGGCTCAGTG-3’) the stop codon. Synthesised sequences can be found in Appendix 1. Genes were cloned into the pET28a bacterial vector for expression of N-terminally 6xHis-tagged proteins by restriction digestion with NdeI and NotI (New England Biolabs, Ipswich, MA, USA), gel extraction and ligation with T4 DNA ligase (New England Biolabs, Ipswich, MA, USA). Ligated vectors were transformed into the lab-maintained competent DH5α E. coli strain as described in section 2.4.4. Resulting colonies were checked by PCR to confirm plasmid integration using EcoTaq polymerase (produced and provided by the Department of Plant Sciences, Cambridge, UK) and pET-F/R primers as described in section 2.4.5. Primer sequences can be found in Appendix 2. PCR amplicons were observed in 1.5% agarose gel electrophoresis. Confirmed colonies were grown overnight in 3 mL of liquid LB supplemented with kanamycin (100 mg·L-1) in a shaker incubator at 180 rpm and 37°C. Plasmid extractions were performed using the QIAprep Spin Miniprep kit (QIAgen, Hilden, Germany). Generated plasmids were checked by diagnostic digestion and by sequencing with the pET-F primer (Source Bioscience, Cambridge, UK). Sequencing results were analysed using the Geneious R9.1.8 software (Biomatters, Auckland, NZ). 4.4.2. Transformation of E. coli BL21 strain Plasmids were transformed into the One Shot BL21(DE3) chemically competent E. coli strain for expression of recombinant proteins (Invitrogen, Carlsbad, CA, USA) by a 30 second heat shock at 42°C. Cells were left to recover in the provided SOC media for 1 h at 37°C and plated on LB agar supplemented with kanamycin (100 mg·L-1). Resulting colonies were checked by PCR with A. Timoneda, PhD thesis, 2022 112 the Taq DNA polymerase with Thermo Pol Buffer (New England Biolabs, Ipswich, MA, USA), T7F/R primers, and an annealing temperature of 44°C. Positive colonies were grown overnight in liquid LB supplemented with kanamycin (100 mg·L-1) in a shaker incubator at 150 rpm at 37°C to generate glycerol stocks. 4.4.3. Enzyme expression in E. coli cultures Starter cultures were generated from single colonies in 10 mL of liquid LB supplemented with kanamycin (100 mg·L-1) and grown overnight at 37°C in a shaker incubator at 150 rpm. Larger cultures were reseeded by transferring 500 μL of starter culture into 50 mL of liquid LB supplemented with kanamycin (100 mg·L-1) in 500 mL conical flasks and grown for ~ 3 h at 37°C and 150 rpm until reaching an OD600 between 0.8 - 1.2, at which point protein expression was induced with 1mM of isopropyl-1-thio-β-D-galactopyranoside (IPTG). Cultures were then grown overnight at 20°C in a shaker incubator at 150 rpm. Cells were harvested and resuspended in a 50 mL solution of 7.6 mM ʟ-DOPA and 15 mM ascorbic acid in autoclaved milliQ water to initiate the reaction. Pellet volumes were calculated to achieve an initial OD600 of 1.9 in all cultures. Samples were taken over time at 0, 2, 4, 6, 8, 24, 32, 48, 56, 72, 96, 120 and 154 h for pigment quantification through HPLC. For iron supplementation assays, FeCl2 was added to the LB media and/or the reaction solution at a final concentration of 50 μM. Product concentrations were calculated using standard curves previously generated for each compound. 4.4.4. HPLC analysis of metabolites Liquid chromatography of compounds produced in bacterial liquid cultures was performed on the supernatants obtained after centrifugation of the culture samples. Previous studies did not find accumulation of betalamic acid or dopaxanthin in culture pellets (Guerrero-Rubio et al., 2019). A Shimadzu LC-20AD apparatus equipped with an SPD-M20A photodiode array detector (Shimadzu, Kyoto, Japan) and a 250 mm by 4.6 mm Kromasil 100 C18 column packed with 5 μm particles (Teknokroma, Barcelona, Spain) was used for analytical HPLC detection of metabolites. Separation of compounds was achieved by linear gradient performed for 24 min from 0% of solvent B to 35% of solvent B, where mobile phase A was 0.05% trifluoroacetic acid (TFA) in water and mobile phase B was 0.05% TFA in acetonitrile, with a flux of 1 mL·min-1 at 30°C (Gandía-Herrero et al., 2005b). 4,5-seco-DOPA, betalamic acid, dopaxanthin and indoline- betacyanin were detected by UV/VIS absorbance at 360 nm, 405 nm, 480 nm, 524 nm and retention times of 13.8 min, 13.8 min, 13.1 min and 22.1 min respectively. Samples were centrifuged 1 min at 13000 rpm before injection. Injection volume was 50 μL. Chapter 4 113 4.4.5. Electrospray ionization mass analysis of metabolites HPLC-ESI MS analyses of reaction products were performed by the Research Support Service (SAI) of the University of Murcia (Murcia, Spain) as previously described (Contreras-Llano et al., 2019), using an Agilent VL 1100 apparatus (Agilent Technologies, Santa Clara, CA, USA) with an LC mass selective detector (MSD) Trap. Elution conditions were analogous to those described above, using the same column. The vaporizer temperature was 350°C, and the voltage was 3.5 kV. Nitrogen at a pressure of 45 lb·in-2 was used as the sheath gas. Samples were ionized in positive mode. The ion monitoring mode was full scan in the range m/z 50 to 600. The electron multiplier voltage for detection was 1,350 V. A quantitative TOF (Q-TOF) Agilent 6220 MS equipped with a dual ESI-atmospheric pressure chemical ionization (APCI) interface (Agilent Technologies, Santa Clara, CA, USA) was used for accurate mass determinations. Samples were ionized in positive mode, using a capillary voltage of 3.5 kV. Nitrogen was used as the drying gas, the gas temperature was 350°C, flux was set at 11 L·min- 1, and the nebulizer pressure was 40 lb·in-2. All data were processed through the MassHunter software (Agilent Technologies, Santa Clara, CA, USA). 4.4.6. Protein purification Starter cultures were generated from single colonies in 10 mL of liquid LB supplemented with kanamycin (100 mg·L-1) and grown overnight at 37°C in a shaker incubator at 150 rpm. Bigger cultures were reseeded by transferring 5 mL of starter culture into 500 mL of liquid LB supplemented with kanamycin (100 mg·L-1) in 2.5 L baffled Tunair shake flasks (Merck, Darmstadt, Germany) and grown for ~ 3 h at 37°C and 150 rpm until reaching an OD600 between 0.8 - 1.2, at which point protein expression was induced with 1mM of isopropyl-1-thio- β-D-galactopyranoside (IPTG). Cultures were then grown for 15-20 h at 20°C in a shaker incubator at 150 rpm. Cells were harvested by centrifugation for 10 min at 5000 rpm and 10°C and resuspended on ice in 1.5 mL of sodium phosphate buffer (50 mM phosphate buffer, pH 8) with 0.3 M sodium chloride. Cell lysis was performed by 5 sonication rounds at a 20% amplitude and maximum temperature of 40°C in 10 second intervals in a Cole-Parmer 4710 series ultrasonic homogenizer (Chicago, IL, USA). Tubes were then centrifuged for 15 min at 13000 g and supernatants were recovered and mixed with 2 mL of binding buffer. Crude extracts were then added to 2 mL of HIS-Select Nickel Affinity Gel beads (Sigma-Aldrich, St. Louis, MO, USA) previously washed with binding buffer and incubated for 15 min at 20°C in a shaker incubator at 150 rpm to allow for binding of the tagged proteins to the beads. Beads were then washed 3 times with 10 mL of binding buffer through 5 min centrifugations at 5000 rpm and discarding of the supernatant. After this, bound proteins were eluted from beads by a 5 min incubation with A. Timoneda, PhD thesis, 2022 114 elution buffer supplemented with imidazole (0.3 M NaCl, 50 mM phosphate buffer, 250 mM imidazole, pH 8) at 20°C in a shaker incubator at 150 rpm. Samples were then centrifuged 5 min at 5000 rpm, and eluted proteins were separated from imidazole by filtering supernatants through PD-10 desalting columns (Sigma-Aldrich, St. Louis, MO, USA) previously washed with distilled water and elution buffer (20 mM phosphate buffer, pH 8.5) and discarding of the first elution product. Purified proteins were then recovered from PD-10 columns by eluting with 3.5 mL of elution buffer and maintained at 4°C for immediate pH and kinetic analysis. Samples were taken at different stages for verification of protein induction, binding and elution. For protein purification troubleshooting and optimization we increased the washing steps to five washes, used Nickel HisPur NiNTA Resin Beads and HisPur Cobalt Superflow Agarose (Thermo Scientific, Waltham, MA, USA), decreased imidazole concentration in washing buffer to 25 mM, and decreased salt concentration of binding, washing and elution buffers to 25 mM NaCl. 4.4.7. Bradford protein quantification Protein was quantified using the Bradford protein assay (Bio-Rad, Hercules, CA, USA) (Bradford, 1976). Protein extracts were mixed with 980 μL of Bradford reagent and absorbance was measured at a λ of 595 nm in a Jasco V-630 spectrophotometer (Jasco Corporation, Tokyo, Japan). Bovine serum albumin was used as standard to obtain a calibration curve. 4.4.8. SDS-PAGE protein electrophoresis Samples were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS- PAGE). Protein gels were made with 375 mM Tris-HCl buffer (pH 8.8), 0.1% sodium dodecyl sulfate (SDS), 0.05% ammonium persulfate (APS), 0.1% tetramethylethylenediamine (TEMED), and 12% acrylamide for separating gels or 4% acrylamide for concentrating gels. Samples taken from bacterial cultures before and after IPTG induction were treated with the BugBuster protein extraction reagent (Novagen, San Diego, CA, USA) for 20 min at room temperature and centrifuged 20 min at 13000 rpm before loading. Samples were mixed 1:1 with a 2X loading buffer (250 mM Tris-HCl buffer (pH 6.8), 350 mM SDS, 0.008% bromophenol blue, 40% glycerol and 20% β-mercaptoethanol) and incubated at 95°C for 5 min. Samples were then loaded on the gel and run in running buffer (25 mM Tris, 0.192 M glycine and 0.1% SDS) at 200 V for 50 min. The protein ladder used was the EZ-Run, pre-stained Rec Protein Ladder from Fisher Bioreagents (BP3603-500, Thermo Fisher Scientific, Waltham, MA, USA). Gels were then stained using a standard Coomassie blue solution made with 40% ethanol, 10% glacial acetic acid and 10% Coomassie Brilliant Blue R-250 (0.1% w/v) (Sigma-Aldrich, St. Louis, MO, USA), Chapter 4 115 for around 1.5 h and de-stained overnight at room temperature and gentle shaking with 30% ethanol and 10% acetic acid. 4.4.9 Absorbance spectroscopy Enzyme activity was determined using a continuous spectrophotometric method by measuring the accumulation of betalamic acid at λ = 414 nm (Girod & Zryd, 1991a; Gandía-Herrero & García-Carmona, 2012). Measures were carried out in a quartz QS 10.00 mm cuvette by a JASCO V-630 spectrophotometer (Jasco Corporation, Tokyo, Japan) through a wavelength scan from 200 to 700 nm every 10 min for 3 h at 25°C and a scanning speed of 2000 nm/min. The reaction medium contained 50 mM sodium phosphate buffer, 2.5 mM ʟ-DOPA, 10 mM sodium ascorbate, and approximately 15 ng of CgDODAa1 enzyme and was set at optimal pH. 4.4.10. Characterisation of enzyme kinetics and optimal pH Enzyme activities were determined using a continuous spectrophotometric method by measuring the absorbance due to betalamic acid appearance at λ = 414 nm as previously described (Girod & Zryd, 1991a; Gandía-Herrero & García-Carmona, 2012). Unless otherwise stated, the reaction medium contained 50 mM sodium phosphate buffer, 2.5 mM ʟ-DOPA, and 10 mM sodium ascorbate at pH 6.5 in a final sample volume of 300 μL. Measurements were performed at 25 °C in 96–well plates in a Synergy HT plate reader (Bio-Tek Instruments, Winooski, USA) at 21-second intervals for 1.5 h. For determination of optimal pH, enzymatic reactions were performed at different increasing pHs in 50 mM acetate buffer (pH 3, 3.5, 4, 4.5, 5, 5,5) or phosphate buffer (pH 5.5, 6, 6.5, 7, 7.5, 8, 8.5). Measurements were also taken for the same pH conditions without ʟ-DOPA to normalize for absorbance of protein precipitates at certain pHs. For determination of enzyme kinetic curves, reactions were performed at the determined optimal pH with increasing concentrations of ʟ-DOPA (0 mM, 0.63 mM, 1.26 mM, 1.9 mM, 2.53 mM, 3.16 mM, 3.8 mM). Enzyme amounts were assessed individually for each protein according to speed of color production to avoid saturation of equipment detection limits (100 μL for BvDODAα1, 1 μL for CgDODAα1, 50 μL for McDODAα1 and 100 μL for ShDODAα1). Reactions were initiated by addition of purified enzyme immediately before measuring. Activity on crude extracts was also assessed at optimal pH. Activity measurements in the presence of metals was performed in reaction media containing 50 μM of FeCl2, NiCl2 and ZnCl2 after a 10 min incubation. For the quantification of betalamic acid, the molar extinction coefficient at 424 nm, ε = 24,000 M-1·cm-1 was taken (Trezzini & Zrÿd, 1991). The plate reader detector signal was calibrated with betalamic acid solutions of known concentration. In the absence of enzyme, no detectable evolution was recorded. Measurements were performed per duplicate and mean values and standard deviations were plotted. In each case, errors associated with the results provided correspond to the residual standard deviations. The steady- A. Timoneda, PhD thesis, 2022 116 state rate was defined as the slope of the linear zone of the product accumulation curve. Kinetic data analysis was carried out by using non-linear regression fitting (Marquardt, 2006), using the SigmaPlot Scientific Graphing for Windows version 10.0 (Systat Software, San Jose, CA, USA). 4.4.11. MALDI-TOF MS protein analysis Protein molecular weight was determined by the Research Support Service (SAI) of the University of Murcia (Murcia, Spain) as previously described (Contreras-Llano et al., 2019). The matrix solution for peptide analyses was α-cyano-4-hydroxycinnamic acid (20 mg·mL-1) in acetonitrile/water/trifluoroacetic acid (TFA) (70:30:0.1). The peptide sample was dissolved in 0.1% TFA and mixed with the matrix solution. One microliter of this mixture was applied to the atmospheric-pressure matrix-assisted laser desorption ionization (AP-MALDI) target plate and allowed to dry. Experiments were carried out with an Agilent TOF mass spectrometer (Agilent Technologies, Santa Clara, CA, USA), equipped with an AP-MALDI ion source with an N2 laser (337 nm). Samples were measured in reflectron mode to identify molecular formulas based on precise mass measurements in positive mode. External calibration of the spectrometer was performed with standard peptides from the ProteoMass Peptide MALDI-MS calibration kit (Sigma-Aldrich, St. Louis, MO, USA). Data were recorded and processed with Agilent MassHunter Workstation software (Agilent Technologies, Santa Clara, CA, USA). Peptide mass fingerprint determination was carried out using Agilent Spectrum Mill software (Agilent Technologies, Santa Clara, CA, USA). Determination of protein absolute molecular mass was carried out using an HPLC-ESI-MS TOF system. This system comprises an HPLC Agilent VL 1100 apparatus equipped with an autosampler μ-well plate and a capillary pump connected to an Agilent 6100 TOF MS (Agilent Technologies, Santa Clara, CA, USA), and an electrospray ionization interface was used. The column employed was a Zorbax Poroshell 300SB-C18 column (1 by 75 mm with 5 μm diameter; Agilent Technologies, Santa Clara, CA, USA). The column was operated at 60°C, and the samples were injected with a flux of 0.2 mL·min-1. The protein was eluted with a linear gradient using water-CAN-formic acid (95:4.9:0.1) as solvent A, and water-CAN-formic acid (10:89.9:0.1) as solvent B. A linear gradient from 0% to 90% of solvent B was performed for 30 min. Protein separation was monitored at 210 and 280 nm using a multiple-wavelength detector. The mass spectrometer was operated in positive mode in the range of 100 to 2,200 m/z, using a capillary voltage of 3.5 kV. Nebulizer gas pressure was 30 lb·in-2, and drying gas flux was 8 L·min-1 at a temperature of 350°C. External spectrometer calibration was carried out using the ProteoMass peptide MALDI-MS calibration kit (Sigma- Aldrich, St. Louis, MO, USA). Two different peptides were used as controls (cytochrome c and carbonic anhydrase; Sigma-Aldrich, St. Louis, MO, USA). All data were recorded and processed through the Agilent MassHunter Workstation Qualitative Analysis software (Agilent Chapter 4 117 Technologies, Santa Clara, CA, USA), and the intact molecular weight of the protein was obtained using the deconvolution algorithm from this software. 4.4.12. Trypsin digestion Peptide mass fingerprinting was performed through trypsin digestion of proteins by the Research Support Service (SAI) of the University of Murcia (Murcia, Spain) as previously described (Guerrero-Rubio et al., 2020b). The protein sample was prepared in 100 μl of buffer NH4HCO3 50 mM, pH 8.0, with 0.02% ProteaseMAXTM Surfactant (Promega, Madison, WI, USA). Then, the sample was reduced with DTT 10 mM for 20 min at 56 °C and alkylated with iodoacetamide 50 mM at room temperature in the dark for 20 min. One microgram of proteomics grade trypsin (Promega) was added and the sample was incubated for 4 h at 37 °C. After that, the sample was centrifuged at 15000 g for 1 min to collect the condensate and the digestion was stopped by adding 0.5% TFA. Peptides were cleaned up with C18 Zip-Tips (Millipore) and evaporated using an Eppendorf vacuum concentrator model 5301. 4.4.13. Fast Protein Liquid Chromatography (FPLC) Samples of purified protein were applied to a Superdex 200 10/300 GL column and equilibrated with sodium phosphate buffer (50 mM, pH 7.5) with 0 mM, 150 mM, 450 mM, 1M and 1.5 M NaCl, or Tris-HCl buffer (50 mM, pH 7.5) with 150 mM CaCl2. The protein was eluted with the same buffer at a flow rate of 0.5 ml·min-1. Elutions were performed in an Äkta purifier apparatus (General Electric Healthcare, Chicago, IL, USA) and monitored at 280 nm. Column calibration was performed with the following protein markers (Sigma): cytochrome c (12.4 kDa), carbonic anhydrase (29 kDa), albumin (66 kDa), alcohol dehydrogenase (150 kDa), and β-amylase (200 kDa). 4.4.14. Metal analysis by ICP-MS A 0.5 mL sample of purified enzyme was loaded into the Agilent 7900 ICP-MS (Agilent Technologies, Santa Clara, CA, USA). Equipment tuning was performed with a 1% HNO3 solution containing 10 µg·L-1 of Li, Y, Ce, Tl and Co. Calibration was performed with an external multielemental solution. Sample was mixed with an internal standard solution containing 400 µg·L-1 of Sc, Ga, Ge, Rh and Ir before nebulisation. All elements were measured using an Octopolo system collision/reaction cell to minimise the presence of polyatomic species. Measured isotopes were 47Ti, 52Cr, 55Mn, 56Fe, 59Co, 60Ni, 63Cu, 66Zn, 75As, 111Cd, 137Ba, 205Tl, 208Pb and 238U. Obtained calibration curves for all elements were made with 8 data points and A. Timoneda, PhD thesis, 2022 118 were linear in all measured concentration ranges. The R-squared value was above 0.999 for all curves. 4.4.15. Scale-up expression of CgDODAα1 in E. coli bioreactors An EZ-Control system (Applikon Biotechnology, Delft, Netherlands) and a Biostat C30-2 system (Bbi-biotech, Berlin, Germany) with a DCU-3 control unit were used to power and monitor 2 L and 30 L bioreactors respectively. Cells were grown in bioreactors with controlled mechanical agitation and temperature regulation. Bioreactors were inoculated with starter cultures of E.coli BL21 cells harbouring pET28a::CgDODAα1 at a ratio of 1:100 in LB supplemented with kanamycin (100 mg·L-1) and grown at 37°C until reaching an OD600 of 0.8–1.0. Temperature was then dropped to 20°C and enzyme production was induced with IPTG 1 mM. Approximately 15-20 hours later, culture medium was retrieved from the reactors and centrifuged to pellet cells. Culture medium was then replaced with sterile Milli-Q water supplemented with 15 mM ascorbic acid and ʟ-DOPA at different concentrations (2 mM and 7.6 mM) and reintroduced in the bioreactor. Indoline-betacyanin producing bioreactors were also supplemented with 0.83 mM of indoline (Guerrero-Rubio et al., 2019). Pigment production was performed in batch and substrate was only added once to the reactor. Cultures were thereafter kept in the dark and shaken at 50 rpm and 20°C with 50-65% of dissolved oxygen. Dissolved oxygen was kept under control through an air inlet assisted by the oxygen sensor AppliSens DO2 (Applikon Biotechnology, Delft, The Netherlands). Chapter 4 119 A. Timoneda, PhD thesis, 2022 120 Chapter 5. Identification of residues responsible for initial acquisition of DODA activity in CgDODAα1 5.1. Introduction Identification of residues responsible for high enzymatic activity can help engineer new enzymes with increased performance for a desired product (Li & Cirino, 2014). Already multiple targets within the betalain pathway have been enhanced through a residue level understanding of their catalytic activity. For example, efforts to increase the tyrosine hydroxylase activity of BvCYP76AD1 for boosting production of the BIA intermediate (S)-reticuline resulted in the discovery of two mutations (W13L and F309L) that combined improved by 2.8-fold the amount of ʟ-DOPA produced by the enzyme (DeLoache et al., 2015). Authors found that the mutation F309L reduced the DOPA oxidase activity responsible for the generation of cyclo-DOPA from ʟ- DOPA in the production of red betacyanins. This allowed for the generation of new BvCYP76AD1 enzyme versions, one with increased overall activity (BvCYP76AD1W13L) and others that preferentially catalyse the tyrosine hydroxylase step to form ʟ-DOPA (BvCYP76AD1 F309L and BvCYP76AD1W13L F309L). BvCYP76AD1W13L F309L was then used in the metabolic engineering of the BIA pathway in yeast (DeLoache et al., 2015), whereas BvCYP76AD1W13L was employed in a later study for the production of a diverse betalain color palette in yeast (Grewal et al., 2018). Similarly, introduction of two mutations identified to be involved with Caryophyllales ADH feedback inhibition deregulation in the Arabidopsis ADH enzyme (E179D and D241N) was also sufficient to boost tyrosine production by 7.5-fold in planta (Lopez-Nieves et al., 2021). Although other enzymes have been the target of residue-level manipulation, the residue basis for the catalytic properties of the DODA enzyme are still largely unknown, despite being the central enzyme in the pathway responsible for producing the core chromophore, betalamic acid. Identification of key residues for the catalysis of the ʟ-DOPA 4,5-dioxygenase reaction could also be used for the engineering of DODA enzymes with improved performance. Early analysis of available LigB sequences from all land plants identified the amino acid motif P-(S,A)-(N,D)-x- T-P (P, proline; S, serine; A, alanine; N, asparagine; D, aspartate; T, threonine) to be conserved within betalain producing species of the Caryophyllales (Christinet et al., 2004). This motif is located immediately after a histidine residue (His177 in P. grandiflora) considered essential for the enzyme’s catalytic activity (Sugimoto et al., 1999; Christinet et al., 2004). The sequences Chapter 5 121 analysed belonged to P. grandiflora, B. vulgaris and M. crystallinum, member species of the Portulacaceae, Chenopodiaceae and Aizoaceae, respectively, which currently represent two different independent origins of DODA activity (Sheehan et al., 2020). Evolution of DODAα1 enzymes with high ʟ-DOPA 4,5-dioxygenase activity is predicted to have occurred up to four times in the order Caryophyllales (Sheehan et al., 2020). Early duplication in the clade of the DODA gene gave rise to the DODAβ and DODAα gene lineages which are present in all betalain-producing Caryophyllales. However, the DODAα lineage later experienced multiple duplication events along the tree (Brockington et al., 2015). Out of the resulting copies, only DODAα1 paralogs have been reported to present high ʟ-DOPA 4,5-dioxygenase activity in betalain-producing taxa, supporting the hypothesis of multiple independent origins of betalain pigmentation (Sheehan et al., 2020). A later study took a horizontal approach by comparing DODAα1 and DODAα2 (paralog with low to no ʟ-DOPA 4,5-dioxygenase activity) from betalain producing species in the Caryophyllales and found a different set of residues required for high activity acquisition in DODAα1 enzymes (Bean et al., 2018). Mutation of these residues in the B. vulgaris DODAα2 background, namely R74D, G75D, F76Y, K152N, N178D, G179E and T203I (R, arginine; G, glycine; F, phenylalanine; Y, tyrosine; K, lysine, E, glutamate; I, isoleucine), was sufficient to enable pigment production from ʟ-DOPA in yeast cultures (Bean et al., 2018). Evolution of high DODA activity in clades representing different independent origins may all have followed a similar molecular path or could have resulted from changes in a different set of residues. Results from Chapter 4 show that DODAα1 enzymes representative of inferred independent origins of high DODA activity exhibit differences both in their kinetic characteristics and in their ability to produce pigments in bacterial cultures. We identified CgDODAα1 as the highest performing enzyme in E. coli assays. In contrast, studies in N. benthamiana pointed at McDODAα1 as the highest performing enzyme in planta (Sheehan et al., 2020). C. gigantea is a member of the Cactaceae, and represents the independent origin inferred in the Portulacineae clade, whereas M. crystallinum belongs to the Aizoaceae and represents the Raphide clade (Figure 1.5b). Together, these two clades form what we hereafter call the Globular Inclusion clade (Gi). In this chapter, we aim to identify the molecular basis underlying the evolution of the high DODA activity observed in the Globular Inclusion clade. Previous horizontal approaches used by Christinet et al. (2004) and Bean et al. (2018), which compared extant high and low activity DODA homologs to identify key residues changes for function acquisition, suffer from several caveats: 1) they reflect all the changes occurred along the lineages from the last common ancestor, which may or may not be involved in the functional difference of interest; and 2) they underestimate the effect of epistasis, where additional residues are important to permit or prevent functional differences. Here, we use vertical comparative analysis of reconstructed ancestral sequences and test these using yeast activity assays in order to elucidate the A. Timoneda, PhD thesis, 2022 122 ancestral node at which DODA activity was gained in the clade. Then we compare the residue changes occurred in the branch leading to this node and test their individual effects on the enzyme’s function. In other words we attempt to reconstruct ancestral DODA sequences at different evolutionary time points to infer the exact location of the gain of activity event in the tree and further analyse the residue changes present on those phylogenetic branches leading to notable increases in DODA activity. 5.2. Results 5.2.1. High DODA activity in the Globular Inclusion clade appeared in the DODAα1 lineage after separation from the DODAα2 lineage To assess the molecular mechanisms underlying the high performance exhibited by the DODAα1 enzymes in the Globular Inclusion clade we attempted to identify and locate the evolutionary trajectory leading to high DODA activity. According to the most recent angiosperm phylogeny, the Globular Inclusion clade can be subdivided in two sub-clades: the Raphide clade, consisting of the families Lophiocarpaceae, Kewaceae, Barbeuiaceae, Aizoaceae, Gisekiaceae, Sarcobataceae, Phytolaccaceae, Petiveriaceae, and Nyctaginaceae; and the Portullugo clade, which includes the families Molluginaceae, Montiaceae, Halophytaceae, Didieraceae, Basellaceae, Talinaceae, Anacampserotaceae, Portulacaceae and Cactaceae (Figure 1.5b) (Stevens, 2021). We were unable to recover any DODAα1 sequences from transcriptomes belonging to species in the Molluginaceae, so we refer to this clade as the Portulacineae clade. C. gigantea (Cactaceae) and M. crystallinum (Aizoaceae) are representative species within the Portulacineae and Raphide clades, respectively. Ancestral reconstruction of DODA sequences at nodes along the DODA tree in the Globular Inclusion clade was performed by Nathanael Walker-Hale (PhD student, Brockington lab), and will be included as part of his forthcoming dissertation. In brief, ancestral sequences were inferred by sequence alignments of all available DODAα transcript sequences from the Globular Inclusion clade. For each node of interest we extracted the Maximum A Posteriori (MAP) sequence, the sequence with the highest probability state at each site. To characterise uncertainty, we also calculated the ‘AltAll’ sequence, which replaces the MAP state at each ambiguous site in the inferred sequence with the second highest probability state. We tested activity of these ancestral enzymes in yeast strains already containing a codon optimised version of BvCYP76AD6 for betaxanthin production (Sunnadeniya et al., 2016; Polturak et al., 2016). Yeast cultures expressing BvCYP76AD6 and a functional DODAα1 gene will therefore produce yellow coloration due to the production of betaxanthin pigments. Yeast has extensively Chapter 5 123 been used as a system to test enzyme performance in the production of betalains (Hatlestad et al., 2012; DeLoache et al., 2015; Sunnadeniya et al., 2016; Polturak et al., 2016; Bean et al., 2018; Grewal et al., 2018). Here, the use of yeast as a model system to test enzyme activity offers a number of advantages over plant systems: 1) it is faster to grow, 2) permits high- throughput analysis of a high number of enzymes simultaneously, and 3) facilitates the measure of absorbance and fluorescence without additional extraction protocols. Transformation of yeast strains containing BvCYP76AD6 with the extant CgDODAα1 and McDODAα1 sequences and subsequent feeding with tyrosine resulted in obvious yellow pigmentation (nodes 7 and 8 in Figure 5.1a). Similarly, no pigmentation was observed in yeast strains expressing extant DODAα2 paralogs, represented by CgDODAα2 (node 3 in Figure 5.1a). Obvious betaxanthin pigmentation was also observed in all MAP and AltAll ancestral sequences belonging to the DODAα1 branch (nodes 4, 5 and 6 in Figure 5.1a), but was not prominent in the common ancestor between DODAα1 and DODAα2, and the later DODAα2 ancestor (node 1 and 2 in Figure 5.1a). Quantification of cell fluorescence as a proxy for betaxanthin production, and ultimately DODA activity, again revealed low values for the common ancestor between DODAα1 and DODAα2, and DODAα2 ancestor, and showed a clear gradual increase through the DODAα1 branch after the α1/α2 duplication node (Figure 5.1b,c,d). In light of these results, the most likely scenario for the evolution of high DODA activity in the Globular Inclusion clade can be described as follows: first, the earliest common ancestor of the DODA enzymes in the clade had low or no DODA activity; then, two DODA copies appeared after a duplication event that gave rise to the DODAα1 and DODAα2 lineages present in all species of this clade; after that, DODA activity was gained along the branch leading to the common ancestor of the DODAα1 lineage; lastly, higher DODA activity was gradually and separately acquired twice, once along the branch leading to the Portulacineae, and once along the branch leading to the Raphide clades, with each clade containing extant high activity CgDODAα1 and McDODAα1 enzymes respectively (Figure 5.1). 5.2.3. Necessity test of DODAα1 ancestor residues reveals residues that could be important for initial acquisition of DODA activity in the Globular Inclusion clade The previous ancestral reconstruction analysis shows that DODA activity in the Globular Inclusion clade first evolved in the DODAa1 lineage sometime between the common DODAα1/α2 (GiDODAα1/α2; node 1 in Figure 5.1) and the DODAα1 ancestor (GiDODAα1; node 4 in Figure 5.1). Pairwise sequence comparison of the reconstructed ancestral sequences A. Timoneda, PhD thesis, 2022 124 at node 1 and node 4 revealed a total of 13 residue changes between the MAP versions of these two sequences (Figure 5.2a). Hereafter, we will be using the sequences reconstructed Figure 5.1. Ancestral reconstruction of DODA genes in the Globular Inclusion clade. (a) Betaxanthin production in yeast strains expressing reconstructed sequences for each ancestral node of the DODA phylogeny in the Globular Inclusion clade. Gain of high DODA activity is observed in the branch leading from the DODAα1 and DODAα2 ancestor to the DODAα1 ancestor. At each node, left culture represents the MAP sequence, Chapter 5 125 right culture represents the AltAll sequence. (b, c, d) Betaxanthin fluorescence as a proxy of DODA activity at each of the reconstructed ancestral nodes. In black, MAP sequences; in grey, Alt All sequences. (c, d) Evolution of increasing DODA activity along tree branches leading to CgDODAα1 and McDODAα1, respectively. Gi, Globular Inclusion clade; Port, Portulacineae clade; Raph, Raphide clade. with the MAP method for our analysis. In order to assess which of these residues are important for the acquisition of initial DODA activity in this lineage, we performed a necessity test in which residues on the GiDODAα1 ancestor background would be mutated one by one to match those of the GiDODAα1/α2 ancestor sequence (Figure 5.2b). In this assay, mutation of residues important for the initial levels of DODA activity should have a detrimental impact over the activity of the enzyme. Thirteen DODAα1 ancestor sequences containing each of the point mutations were synthesised and cloned into yeast expression vectors. Transformation of yeast and subsequent feeding with tyrosine revealed four residue mutations that significantly reduced the activity of the GiDODAα1 ancestor (Fig 5.3c). These residue changes were F116L (N7), P142S (N8), G171S (N9) and N177D (N10). However, mutation N7 showed lower values of significance in Dunnett’s multiple comparison tests and did not appear significant in additional experimental repeats (not shown). Similarly, mutation N9 did not appear significant in later experimental repeats (not shown). A. Timoneda, PhD thesis, 2022 126 Figure 5.2. Necessity test of GiDODAα1 ancestor residues. (a) Alignment between the GiDODAα1 and GiDODAα1/α2 ancestor reveals thirteen residue changes. In black, conserved residues; in white or light gray, mutated residues. (b) Schematic table summing up the residue changes present between the DODAα1 and DODAα1/α2 ancestors. (c) Necessity test of the GiDODAα1 ancestor, where each identified residue is mutated to match that of the GiDODAα1/α2 ancestor. Asterisks represent significant differences in a Dunnett’s multiple comparison test against the GiDODAα1 ancestor. t-values: 0.01-0.05 (*); < 0.001 (***). Intriguingly, none of these residue mutations were able to dramatically reduce or completely abolish enzyme activity by themselves. Therefore, to test if the effect of these mutations is additive, we performed multiple residue mutations in the GiDODAα1 ancestor background. We generated new expression vectors containing all double and triple mutation combinations between N8, N9 and N10 in the GiDODAα1 ancestor background. Vectors were expressed in yeast and the resulting strains were then fed with tyrosine. Surprisingly, no mutation combination resulted in any significant reduction in enzyme activity when compared to the GiDODAα1 ancestor (Figure 5.3). This result was confirmed twice. Figure 5.3. Combined mutation of N8, N9 and N10 residues does not have a significant impact in enzyme activity. Betaxanthin fluorescence and coloration produced by yeast strains expressing GiDODAα1 ancestor enzymes with double and triple mutation combinations of P142S (N8), G171S (N9) and N177D (N10). Statistical significance was assessed with a Dunnett’s multiple comparison test against the GiDODAα1 ancestor. 5.2.4. High DODA activity can be gained in non-active ancestral enzymes of the Globular Inclusion clade with the mutation of only three residues Despite the ambiguous results obtained between the single and multiple mutation necessity tests, we decided to assess whether the identified residues, namely F116, P142, G171, and Chapter 5 127 D177 (corresponding to mutations N7, N8, N9 and N10, respectively), are sufficient to induce high DODA activity in the GiDODAα1/α2 ancestor. In the context of epistatic interactions, a residue that is necessary for enzymatic function is not always sufficient to induce the gain of this function in a non-active background, as it may require the participation of other additional residues. To test this, we synthesised a version of the GiDODAα1/α2 ancestor sequence with the complementary mutations for the four significant positions identified in the first necessity test (Figure 5.2c), namely L116F, S142P, S171G, D177N (corresponding to N7, N8, N9 and N10, respectively), and another version with only the latter three mutations (N8, N9, N10), which appeared to have a more significant effect (Figure 5.2c). We hereafter refer to these two enzyme variants as S1 and S2, respectively. Surprisingly, expression of these enzymes in yeast and subsequent feeding with tyrosine, resulted in the production of yellow coloration both in S1 and S2 strains (Figure 5.4). Quantification of betaxanthin fluorescence revealed that in both cases enzyme activity was as high as that obtained for the GiDODAα1 ancestor enzyme (Figure 5.4). Therefore, mutation in only three sites (S142P, S171G, and D177N) is enough to confer DODA activity to the otherwise non- or low-active GiDODAα1/α2 ancestor background. It remains to be assessed, however, whether the totality of these three residues is required for gain of function, or if mutation at only one or two of these sites could still be sufficient for the acquisition of DODA activity. Figure 5.4. Mutation of three residues is sufficient for the acquisition of DODA activity in the GiDODAα1/α2 ancestor. Betaxanthin fluorescence and coloration produced by yeast strains expressing GiDODAα1/α2 ancestor enzymes with triple and quadruple reverse mutations of N7 (L116F), N8 (S142P), N9 (S171G), and N10 (D177N). S1 refers to the GiDODAα1/α2 ancestor enzyme with all four mutations, and S2 refers to the GiDODAα1/α2 ancestor enzyme with mutations of N8, N9 and N10. Statistical significance was assessed with a Dunnett’s multiple comparison test against the GiDODAα1/α2 ancestor. A. Timoneda, PhD thesis, 2022 128 5.3. Discussion In this chapter we have interrogated the evolutionary history of the DODA lineage in the Globular Inclusion clade with the aim of identifying the molecular mechanisms by which DODA activity evolved in this context and led to the high activity observed in extant enzymes such as CgDODAα1 and McDODAα1. Reconstruction of the ancestral sequences at key nodes along the phylogenetic tree and subsequent characterization of their activity in yeast cultures was a useful tool to locate the approximate point at which DODA activity evolved in the clade. With this approach, we found that DODA activity was gained in the ancestor of the DODAα1 enzymes after a duplication event that gave rise to both DODAα1 and DODAα2 lineages (Figure 5.1a). Prior to that duplication, the earliest ancestor of the DODA enzymes in the clade presented no or very low DODA activity. These findings are in line with previous studies by our group proposing multiple independent origins of betalain pigmentation in the order Caryophyllales, and serve as strong evidence for this scenario (Sheehan et al., 2020). Similar analysis of clades in the Caryophyllales representing other independent origins, such as the Amaranthaceae clade, are already being performed and should provide further evidence to confirm this hypothesis. It is yet to be discovered what the function of the extant DODAα2 enzymes is, however, it can be speculated that this function may be a retention of the ancestral function that could have been present in the earliest DODA ancestor of the clade. A duplication event in this gene, could have then relaxed the selective pressure on one of the gene copies and allowed for neofunctionalisation towards a new enzymatic function in the production of betalamic acid for betalain biosynthesis. The presence of trace DODA activity in the ancestor of the DODAα1 and DODAα2 lineages (as well as the later DODAα2 ancestor) could also point at the possibility of a subfunctionalisation event, where the ancestor enzyme would have presented both DODA activity and the unknown activity of the DODAα2 enzymes. After duplication, each new enzyme would have specialised in one of these two functions and given rise to the DODAα1 and DODAα2 clades. Our findings also suggest that DODA activity originated once in the Globular Inclusion clade, but later separately gained even higher activity via two independent routes in the Portulacineae clade and the Raphide clade (Figure 5.1c,d). This could explain the differences observed in enzyme kinetics reported in Chapter 4. Identification of the molecular changes that led to these separate increases in activity, is already being done with similar analyses in residue changes in branches spanning from the GiDODAα1 common ancestor to the Portulacineae and the Raphide specific DODAα1 ancestors, and will provide a better understanding of the Chapter 5 129 mechanisms involved in second-stage enhancing DODA activity. Contrary to our observations in E. coli, expression of CgDODAα1 in yeast did not outperform McDODAα1 in the production of betaxanthins, but instead exhibited slightly lower activity (Figure 5.1b). This is in line with results obtained in transient expression of DODAα1 enzymes in planta (Sheehan et al., 2020). Here, like in our yeast assays, the levels of ʟ-DOPA available to be used by the DODA enzyme is set by the cytochromes (CYP76AD1 and CYP76AD6, respectively), which is likely to be much lower than the 7.6 mM of ʟ-DOPA supplemented externally in our E. coli experiments and will favour enzymes, like CgDODAα1, that can maintain higher activities at higher substrate concentrations (Figure 4.4c). Clearly there are discrepancies in the relative performance of these enzymes in plants, bacteria and yeast, which we are yet to fully understand. If DODA activity has independently evolved in different clades of the order Caryophyllales, then did this gain of activity occur via the same molecular mechanisms and residue changes, or do different pathways exist to achieve it? And if so, can these differences explain the diversity in the DODAα1 enzymatic performances observed in E. coli assays? To identify the residues involved in the gain of DODA function in the Globular Inclusion clade, we performed a necessity test of all residue changes present between the earliest DODAα1/α2 ancestor and the active DODAα1 ancestor (Figure 5.2). Single mutation of four of these residues significantly reduced the activity of the DODAα1 ancestor (F116, P142, G171, and D177), but didn’t completely abolish it. We argued that these residues could be showing an additive effect in activity reduction, but simultaneous mutation of the three most significant residues (P142, G171, and D177) did not result in any statistical reduction. The reason behind this observation is still unclear, but it could be related to epistatic interactions with other amino acids that did not mutate in the branch but are somewhat important for catalysis. Interestingly, addition of these same three residues to the non-active DODAα1/α2 ancestor background was enough to confer DODA activity at levels comparable to those exhibited by the DODAα1 ancestor enzyme (Figure 5.4). Again this finding is difficult to reconcile with the necessity tests and triple reversal of these residues. Despite having confirmed these results in duplicate, some variability was later observed for the statistical significance of the mutation effect of residues F116 and G171 when performing additional experiments. Future experiments should focus on assessing if the addition of P142 and D177 alone or combined is sufficient to induce gain of DODA function in the DODAα1/α2 ancestor background. Previous publications have suggested a series of residues and domains to be important for the acquisition of DODA function in DODAα1 enzymes of species of the Caryophyllales (Christinet et al., 2004; Bean et al., 2018). Interestingly, one of the residues identified in our experiments, D177, is also present both in the residue list proposed by Bean et al., (2018) for B. vulgaris enzymes, and in the six amino acid domain pointed by Christinet et al., (2004) following the highly conserved histidine (H177) of the catalytic site in all Caryophyllales (Figure 5.5) A. Timoneda, PhD thesis, 2022 130 (Sugimoto et al., 1999). It is, therefore, highly likely that the D177 residue is directly involved with gain of DODA function, and that this mutation has occurred independently at least twice in the Caryophyllales: both in the Amaranthaceae clade and the Globular Inclusion clade. This convergence in the evolutionary mechanisms that lead to enzyme neofunctionalisation manifests the importance of D177 as a favoured evolutionary step towards production of betalamic acid from ʟ-DOPA in the establishment of the betalain pigmentation pathway. Deeper analysis of all the residues across both phylogenetic branches leading to high DODA activity will likely reveal additional convergent residues. Figure 5.5. Residues identified by authors to be important for gain of DODA function in the Caryophyllales. Multiple alignment of CgDODAα1, CgDODAα2 and reconstructed sequences of the DODAα1 and DODAα1/α2 ancestors in the Globular inclusion clade. In red, residues identified in this study to significantly reduce the activity of the GiDODAα1 ancestor upon mutation: N7 (F116), N8 (P142), N9 (G171), N10 (D177)). In green, residues proposed by Bean et al., (2018) in BvDODAα1 to be necessary for the gain of DODA activity in BvDODAα2. In blue, domain identified by Christinet et al., (2004) to be present in DODA enzymes of all betalain producing species. This domain is located right after the H177 residue known to locate to the catalytic site of the enzyme and be directly involved with reaction catalysis (Sugimoto et al., 1999). Chapter 5 131 5.4. Methods 5.4.1. Reconstruction of ancestral sequences Ancestral reconstruction of DODA sequences was performed by Nathanael Walker-Hale (PhD student, Brockington lab), and will be included as part of his forthcoming PhD dissertation. In brief, we began from the most inclusive sampling of DODAα sequences from the Globular Inclusion clade in Sheehan et al. (2019). Firstly, an alignment of coding sequences was inferred with PRANK v.170427 (Löytynoja & Goldman, 2005) (-codon -iterate=5). Columns with more than 90% missing data were removed with pxlcsq from the phyx package (Brown et al., 2017) (-p 0.1). A tree was inferred from the cleaned alignment with IQ-TREE v1.6.12 under the GTR+G model with four rate categories (Nguyen et al., 2015) (-m GTR+F+G). Because the full sequence set included sequences with a large proportion of gaps which could prove problematic for ancestral sequence inference, we winnowed the alignment by removing any sequence with long terminal INDELs. We also removed any sequence containing an internal INDEL longer than two amino acids that was not shared in at least two sequences. We pruned the tree to match the reduced sequence sample. The reduced tree was used as the guide tree for a second round of alignment on the reduced sequence set with PRANK, trusting insertions (-t=tree -codon -once -F). The resulting alignment was translated to amino acids and used to optimise branch lengths on the reduced tree under the best-fitting model of amino acid sequence evolution (JTT+G) with IQ-TREE (-m MFP -te tree) (Kalyaanamoorthy et al., 2017). Ancestral sequences were inferred from the reduced alignment and amino acid branch length tree using FastML v3.11 (Pupko et al., 2000, 2002). We used marginal inference and coded and reconstructed gaps as binary characters with stochastic mapping using the inbuilt approach in FastML, preferring gaps over characters with a Posterior Probability (PP) > 0.5. For each node of interest we extracted the Maximum A Posteriori (MAP) sequence, the sequence with the highest PP state at each site. To characterise uncertainty, we also calculated the ‘AltAll’ sequence (Eick et al., 2017)), which replaces the MAP state at each ambiguous site (PP of MAP state < 0.8) in the inferred sequence with the second highest PP state, if PP > 0.2. We left reconstructed gap states fixed between MAP and AltAll sequences. 5.4.2. Vector construction Reconstructed and extant DODA sequences were codon optimised for yeast, domesticated for BsmBI, BsaI and NotI and synthesized by Twist Biosciences (San Francisco, CA, USA). A. Timoneda, PhD thesis, 2022 132 Synthesised sequences can be found in Appendix 1. Overhangs were added before the start codon (5’-GCATCGTCTCATCGGTCTCAT-3’) and after the stop codon (5’-ATGCCGTCTCAGGTCTCAGGAT-3’) both including restriction sites for BsmBI and BsaI (New England Biolabs, Ipswich, MA, USA) for cloning into MoClo Entry level vectors (equivalent to Level 0) or directly into MoClo Cassette level vectors (equivalent to Level 1), respectively. DODA versions containing point mutations for necessity and sufficiency tests were also synthesized by Twist Biosciences as described above. Synthesised genes were cloned using the MoClo Golden Gate cloning technology previously described in section 2.4.6, first via BsmBI into the pYTK001 Entry level vector, and later via BsaI into the pTMP137 cassette vector along with a ScCCW12 promoter and ScADH1 terminator for constitutive expression in yeast. The pTMP137 vector is a gift from Prof. John Dueber (University of California, Berkeley, CA, USA) and contains homology sites for transgene integration in the LEU2 locus of the yeast genome via NotI. Constructed vectors were transformed and stored in the lab-maintained competent DH5α E. coli strain as described in section 2.4.4. The pTMP137 vector also contains a GFP marker within the cloning region, which gets excised out of the plasmid upon proper ligation with the insert. Therefore, positive colonies will exhibit no fluorescence under blue light. Non- fluorescent colonies were checked by PCR with the primers ConLS_F and ConRE_R as described in 2.4.5. Constructed vectors were further checked by diagnostic digestion and by sequencing with the ConLS_F and ConRE_R primers targeting plasmid sites flanking the transgene (Source Bioscience, Cambridge, UK). Primer sequences can be found in Appendix 2. 5.4.3. Transformation of yeast Plasmids were transformed into the yHS023 S. cerevisiae strain previously generated by Dr Hester Sheehan. yHS023 is a result of the integration of BvCYP76AD6 and the URA3 marker gene in the genome of the yWCD230 mother strain gifted by Prof. John Dueber (University of California, Berkeley, CA, USA). yWCD230 is BY4741 (MATa his3Δ0 leu2Δ0 met15Δ0 ura3Δ0). Therefore, yHS023 can grow without uracil supplementation. Fresh yHS023 colonies were grown overnight in liquid YPD media at 30°C and 200 rpm. Cultures were then started in liquid 2xYPD at an OD600 of 0.5 and a volume of 5 mL per transformation, and grown at 30°C and 200 rpm until reaching an OD600 of 2 (approximately 4 h). 1 μg of each plasmid was digested with NotI at 37°C for at least 3 h. Cultures where then washed 3 times with sterile DI water. For each transformation, 100 μL of culture was separated and its supernatant was replaced by 340 μL of transformation buffer consisting of Polyethylene glycol Mn 3,350 (35.3%; w/v), lithium acetate (LiOAc) 0.1 M, and single-stranded carrier DNA (0.3 mg/mL) (all from Sigma-Aldrich, St. Louis, MO, USA). Digested plasmid was then added to each transformation mix. Cells were then vortexed vigorously and incubated at 42°C for 30 min. Cells were then pelleted, resuspended in sterile DI water by vortexing, and plated in solid minimal SD media lacking leucine and uracil (SD/-U/-L). Genome integration of the DODA genes is done alongside the LEU2 marker, Chapter 5 133 therefore, resulting colonies will be able to grow without both leucine and uracil (as per their mother strain). Plates were incubated at 30°C for 2-3 days. 5.4.4. Yeast colony PCR Single colonies were picked and lysed by resuspending in 75 μL of NaOH 10 mM and boiling at 98°C for 15 min. Cell debris was then pelleted, and 1 μL of supernatant used for PCR. Gene amplification was performed with the Phusion High-Fidelity DNA polymerase (Thermo Fisher Scientific, Waltham, MA, USA) and the ConLS_F and ConRE_R primers in a Mastercycler Nexus (Eppendorf, Hamburg, Germany), as described in section 2.4.2, with an annealing temperature of 60°C and elongation time of 1 minute. Amplicons were then visualised through 1.5% gel electrophoresis. Positive colonies present two bands corresponding to the transformed DODA gene and the BvCYP76AD6 gene already present in the mother strain yHS023 at their corresponding sizes. 5.4.5. DODA expression analysis in yeast Positive colonies were grown in 2 mL of liquid SD minimal media without Uracil (SD/-U) and supplemented with tyrosine 1 mM (Acros Organics, Geel, Belgium) and ascorbic acid 10 mM (Sigma-Aldrich, St. Louis, MO, USA) at 30°C and 200 rpm for 2 days. We did not grow colonies in SD/-U/-L to facilitate experimentation, since the assay also contains the yHS023 strain as a negative control and transformation through genomic integration prevents loss of the transgene, regardless of the selective pressure. After 2 days, 500 μL of liquid SD/-U supplemented with tyrosine 1 mM and ascorbic acid 10 mM were inoculated with 1:50 dilution of starter culture in 96 deep-well plates (Eppendorf, Hamburg, Germany) and grown for 24 h at 30°C and 1000 rpm. Cultures were vortexed before inoculation. Plate set up was randomised and accounted for 4 replicates per strain, and 4 uninoculated wells for background normalisation. Plates were sealed with a gas permeable adhesive seal (Abgene, Portsmouth, NH, USA). After 24 h, 200 μL of cultures were trasnfered into a new 96-well PCR plate and washed twice by pelleting for 2 min at 4000 rpm in a Heraeus Megafuge 8 centrifuge with plate adaptors (Thermo Scientific, Waltham, MA, USA) and resuspending with PBS 1X. 100 μL of resulting samples were transferred to a black µCLEAR 96-well microplate (Greiner Bio-One, Kremsmünster, Austria) for cell fluorescence quantification in a CLARIOstar microplate reader (BMG Labtech, Aylesbury, UK). Betaxanthin fluorescence was measured with an excitation wavelength of 460-480 nm and at an emission wavelength of 500-520 nm. OD600 was also measured in each well to account for cell density. The average fluorescence and OD600 values at uninoculated wells were subtracted from all samples. Fluorescence values were normalised to the OD600 at each well, and the A. Timoneda, PhD thesis, 2022 134 average normalised fluorescence of the yHS023 negative control values was also subtracted from all samples. Chapter 6 135 Chapter 6. Conclusion and Future Perspectives Betalain pigments are attracting growing interest due to their applications in the food and commodity sector, and because of their potential use as reporter systems in biological research. Advances in the diverse application of betalains to different fields and disciplines has been driven by recent developments in the metabolic engineering of the betalain biosynthetic pathway, which is both relatively simple and easy to manipulate. Within the four year duration of my PhD program, engineering of the betalain pathway has been reported in a wide range of heterologous hosts from bacteria, yeast and human cell lines to a diversity of plants including commercially important crops like tomato, potato, aubergine and rice, and plant model systems like A. thaliana, N. benthamiana, N. tabacum and P. hybrida (Polturak et al., 2017; Grewal et al., 2018; Tian et al., 2019; Stücheli et al., 2020; Sheehan et al., 2020; He et al., 2020b). Through my own work, we have extended this list to add the legume species M. truncatula (Timoneda et al., 2021). Given the flexibility of the system in terms of species amenability, together with the colour and fluorescence properties of betalains, we expect this list of organisms to grow further in the future, with an even broader range of applications. During this work, we have expanded the existing knowledge of the enzymes that produce betalains and how they originated in the Caryophyllales, and validated the use of betalain pigments as reporters for plants. We have done this via two different investigatory strands: (1) Establishing betalains as tools for synthetic biology. The establishment of betalains as reporters in plants offers a series of important benefits over other existing reporter systems. In many cases, previous extraction or chemical treatment steps that kill the plant tissue are required, as well as the use of specialised microscopy for visualisation of the reporter signal. Betalain colour is easy to detect on living specimens even with the naked eye, and to measure through simple spectroscopy. Through this work, we have proven the application of betalains as reporter systems for arbuscular mycorrhizal symbiotic processes in roots. By expressing the betalain biosynthetic enzymes under plant symbiotic specific promoters, we were able to restrict betalain pigment production to root areas colonized by arbuscular mycorrhiza. This approach will be very useful, for example, in the high-throughput screening of high numbers of plants for fungal colonisation, since betalain red pigmentation is immediately visible after the removal of plant roots from the soil and the reporter system does not require any additional sampling and staining steps. These benefits are likely to translate A. Timoneda, PhD thesis, 2022 136 into many other research areas, and we expect betalain-based reporters to be developed for a wide range of other plant physiological processes. Recently, another publication has shown the use of betalains for non-invasively monitoring gene expression in plants (He et al., 2020b). Here, authors created RUBY, a reporter system where the betalain biosynthetic genes are fused in a polycistronic-like transcript and separated by 2A peptides for post- translational cleavage. This approach simplifies vector construction and avoids problems that could rise from differences in expression ratios or transgene silencing due to repetitive elements. We predict that RUBY will be adopted by many researchers in the near future to study the spatial and temporal expression patterns of genes in a diverse range of plant physiological processes. (2) Using natural variation to enhance production of betalain pigments. We have found that relaxed sensitivity to product feedback inhibition in the ADH gene could have evolved multiple times in flowering plants, but to date, none of the studied ADH enzymes have matched the production of tyrosine levels as seen with B. vulgaris ADHα. More promisingly, we have also found that DODAα1 enzymes from species belonging to clades inferred to represent different independent origins of betalain pigmentation in the Caryophyllales exhibit different kinetic properties in vitro. Although all of them present kinetic curves characteristic of enzymes that suffer inhibition by substrate, enzymes from C. gigantea and S. halimifolium are able to maintain higher activities at higher substrate concentrations. This could explain why CgDODAα1 enzymes outperformed other enzymes in E. coli cultures, where the substrate was added externally at high concentrations. Therefore this research has the potential to further optimise the bioindustrial conditions for betalain production. Furthermore, despite exhibiting stronger substrate inhibition, B. vulgaris DODAα1 showed the highest affinity for the substrate ʟ-DOPA. Clearly, creating a new DODAα1 enzyme that can integrate these two characteristics, namely a high affinity for the substrate and lower inhibition at high substrate concentrations, could result in the creation of a super-performing enzyme that could be used for the enhancement of pigment production. To do this, key residues for enzyme activity, substrate affinity and substrate inhibition in each clade will have to be characterised. Previously, identification of key residues for the relaxation of ADH activity in Caryophyllales has been useful in the generation of ADH variants exhibiting enhanced tyrosine production capabilities (Lopez- Nieves et al., 2021). Here, we have been able to initially identify a number of residues important for the acquisition of DODA activity in the Globular Inclusion clade, which encompasses CgDODAα1 and McDODAα1, and represents a separate inferred origin of betalain pigmentation from the Amaranthaceae clade containing BvDODAα1. Finally, if different DODAα1 enzymes from clades inferred to represent independent origins of Chapter 6 137 betalain pigmentation exhibit different kinetic characteristics, it is possible that, similarly, other enzymes involved in betalain biosynthesis from these same origins could also exhibit differences in their catalytic properties. Tapping the natural diversity within these different evolutionary trajectories to betalain pigmentation in the Caryophyllales could also help with the discovery of novel high performing enzymes in other gene lineages that could be used to increase betalain production, diversify the enzymatic toolkit, and expand the colour range and chemical diversity of bioindustrially synthesised betalains. Collectively my assays on the application of betalain pigments to arbuscular mycorrhiza symbiosis, and the harnessing of evolutionary enzymatic diversity to boost betalain titres, herald an exciting future for both pure and applied betalain research. A. Timoneda, PhD thesis, 2022 138 Bibliography 139 Bibliography Afendi FM, Okada T, Yamazaki M, Hirai-Morita A, Nakamura Y, Nakamura K, Ikeda S, Takahashi H, Altaf-Ul-Amin M, Darusman LK, et al. 2012. KNApSAcK family databases: integrated metabolite–plant species databases for multifaceted plant research. Plant and Cell Physiology 53: e1–e1. Albert NW. 2015. Subspecialization of R2R3-MYB repressors for anthocyanin and proanthocyanidin regulation in forage legumes. Frontiers in Plant Science 6: 1165. Aronoff S, Aronoff EM. 1948. Thermal degradation of dehydrated beets II. Chromatographic separation of red beet-root pigments. Journal of Food Science 13: 59–65. Attoe EL, von Elbe JH. 1981. Photochemial degradation of betanine and selected anthocyanins. Journal of Food Science 46: 1934–1937. Attoe EL, von Elbe JH. 1982. Degradation kinetics of betanine in solutions as influenced by oxygen. Journal of Agricultural and Food Chemistry 30: 708–712. Attoe EL, von Elbe JH. 1984. Oxygen involvement in betanin degradation - Oxygen uptake and influence of metal ions. Zeitschrift für Lebensmittel-Untersuchung und Forschung 179: 232–236. Babos M, Halász K, Zagyva T, Zöld-Balogh Á, Szego D, Bratek Z. 2011. Preliminary notes on dual relevance of ITS sequences and pigments in Hygrocybe taxonomy. Persoonia 26: 99–107. Bago B, Pfeffer PE, Shachar-Hill Y. 2000. Carbon metabolism and transport in arbuscular mycorrhizas. Plant Physiology 124: 949–957. Baker ET, Lavelle JW. 1984. The effect of particle size on the light attenuation coefficient of natural suspensions. Journal of Geophysical Research 89: 8197–8203. Bashandy H, Jalkanen S, Teeri TH. 2015. Within leaf variation is the largest source of variation in agroinfiltration of Nicotiana benthamiana. Plant Methods 11: 47. Bate-Smith E. 1962. The phenolic constituents of plants and their taxonomic significance. I. Dicotyledons. Botanical Journal of the Linnean Society 58: 95–173. Bean A, Sunnadeniya R, Akhavan N, Campbell A, Brown M, Lloyd A, Lloyd A. 2018. Gain-of-function mutations in beet DODA2 identify key residues for betalain pigment evolution. New Phytologist 219: 287- 296. Berlin J, Sieg S, Strack D, Bokern M, Harms H. 1986. Production of betalains by suspension cultures of Chenopodium rubrum L. Plant Cell, Tissue and Organ Culture 5: 163–174. Bhuiyan MNH, Adachi T. 2003. Stimulation of betacyanin synthesis through exogenous methyl jasmonate and other elicitors in suspension-cultured cells of Portulaca. Journal of Plant Physiology 160: 1117–1124. Bischoff H. 1876. Das Caryophyllinenroth. Inaugural dissertation, University of Tubingen, Tubingen, Germany. Böhm H, Böhm L, Rink E. 1991. Establishment and characterization of a betaxanthin-producing cell culture from Portulaca grandiflora. Plant Cell, Tissue and Organ Culture 26: 75–82. Boisson-Dernier A, Chabaud M, Garcia F, Bécard G, Rosenberg C, Barker DG. 2001. Agrobacterium rhizogenes -transformed roots of Medicago truncatula for the study of nitrogen-fixing and endomycorrhizal symbiotic associations. Molecular Plant-Microbe Interactions 14: 695–700. Bokern M, Heuer S, Strack D. 1992. Hydroxycinnamic acid transferases in the biosynthesis of acylated betacyanins: purification and characterization from cell cultures of Chenopodium rubrum and occurrence in A. Timoneda, PhD thesis, 2022 140 some other members of the Caryophyllales. Botanica Acta 105: 146–151. Bokern M, Strack D. 1988. Synthesis of hydroxycinnamic acid esters of betacyanins via 1-O-acylglucosides of hydroxycinnamic acids by protein preparations from cell suspension cultures of Chenopodium rubrum and petals of Lampranthus sociorum. Planta 174: 101–105. Bonfante-Fasolo P, Faccio A, Perotto S, Schubert A. 1990. Correlation between chitin distribution and cell wall morphology in the mycorrhizal fungus Glomus versiforme. Mycological Research 94: 157–165. Bradford MM. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72: 248–254. Bremer B, Bremer K, Chase MW, Reveal JL, Solits DE, Soltis PS, Stevens PF, Anderberg AA, Fay MF, Goldblatt P, et al. 2003. An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG II. Botanical Journal of the Linnean Society 141: 399–436. Brockington SF, Walker RH, Glover BJ, Soltis PS, Soltis DE. 2011. Complex pigment evolution in the Caryophyllales. New Phytologist 190: 854–864. Brockington SF, Yang Y, Gandia-Herrero F, Covshoff S, Hibberd JM, Sage RF, Wong GKS, Moore MJ, Smith SA. 2015. Lineage-specific gene radiations underlie the evolution of novel betalain pigmentation in Caryophyllales. The New phytologist 207: 1170–1180. Brown JW, Walker JF, Smith SA. 2017. Phyx: phylogenetic tools for unix. Bioinformatics 33: 1886–1888. Brundrett MC, Tedersoo L. 2018. Evolutionary history of mycorrhizal symbioses and global host plant diversity. New Phytologist 220: 1108–1115. Butera D, Tesoriere L, Di Gaudio F, Bongiorno A, Allegra M, Pintaudi AM, Kohen R, Livrea MA. 2002. Antioxidant activities of sicilian prickly pear (Opuntia ficus-indica) fruit extracts and reducing properties of its betalains: Betanin and indicaxanthin. Journal of Agricultural and Food Chemistry 50: 6895–6901. Cabanes J, Gandía-Herrero F, Escribano J, García-Carmona F, Jiménez-Atiénzar M. 2016. Fluorescent bioinspired protein labeling with betalamic acid. Derivatization and characterization of novel protein-betaxanthins. Dyes and Pigments 133: 458–466. Cai Y, Mei Sun A, Corke H. 2003. Antioxidant activity of betalains from plants of the Amaranthaceae. Journal of Agricultural and Food Chemistry 51: 2288−2294. Cai Y, Sun M, Corke H. 1998. Colorant properties and stability of Amaranthus betacyanin pigments. Journal of Agricultural and Food Chemistry 46: 4491–4495. Cai Y, Sun M, Schliemann W, Corke H. 2001. Chemical stability and colorant properties of betaxanthin pigments from Celosia argentea. Journal of Agricultural and Food Chemistry 49: 4429–4435. Cai X, Xu C, Wang X, Wang S, Zhang Z, Fei Z, Wang Q. 2018. Construction of genetic linkage map using genotyping-by-sequencing and identification of QTLs associated with leaf color in spinach. Euphytica 214: 229. Calogero G, Yum JH, Sinopoli A, Di Marco G, Grätzel M, Nazeeruddin MK. 2012. Anthocyanins and betalains as light-harvesting pigments for dye-sensitized solar cells. Solar Energy 86: 1563–1575. Campanella JJ, Smalley J V., Dempsey ME. 2014. A phylogenetic examination of the primary anthocyanin production pathway of the Plantae. Botanical Studies 55: 10. Campos L, Lisón P, López-Gresa MP, Rodrigo I, Zacarés L, Conejero V, Bellés JM. 2014. Transgenic tomato plants overexpressing tyramine N-hydroxycinnamoyltransferase exhibit elevated hydroxycinnamic acid amide levels and enhanced resistance to Pseudomonas syringae. International Society for Molecular Plant-Microbe Interactions 27: 1159–1169. Bibliography 141 Čanadanović-Brunet JM, Savatović SS, Ćetković GS, Vulić JJ, Djilas SM, Markov SL, Cvetković DD. 2011. Antioxidant and antimicrobial activities of beet root pomace extracts. 29: 575–585. Carpenter EJ, Matasci N, Ayyampalayam S, Wu S, Sun J, Yu J, Jimenez Vieira FR, Bowler C, Dorrell RG, Gitzendanner MA, et al. 2019. Access to RNA-sequencing data from 1,173 plant species: The 1000 Plant transcriptomes initiative (1KP). GigaScience 8 (10): giz126. Casique-Arroyo G, Martínez-Gallardo N, González de la Vara L, Délano-Frier JP. 2014. Betacyanin biosynthetic genes and enzymes are differentially induced by (a)biotic stress in Amaranthus hypochondriacus. PLoS ONE 9: e99012. Chen PH, Lin C, Guo KH, Yeh YC. 2017a. Development of a pigment-based whole-cell biosensor for the analysis of environmental copper. RSC Advances 7: 29302–29305. Chen N, Yu Z-H, Xiao X-G. 2017b. Cytosolic and nuclear co-localization of betalain biosynthetic enzymes in tobacco suggests that betalains are synthesized in the cytoplasm and/or nucleus of betalainic plant cells. Frontiers in Plant Science 8: 831. Cheng MN, Huang ZJ, Hua QZ, Shan W, Kuang JF, Lu WJ, Qin YH, Chen JY. 2017. The WRKY transcription factor HpWRKY44 regulates CytP450-like1 expression in red pitaya fruit (Hylocereus polyrhizus). Horticulture Research 4: 17039. Chou YC, Shih CI, Chiang CC, Hsu CH, Yeh YC. 2019. Reagent-free DOPA-dioxygenase colorimetric biosensor for selective detection of L-DOPA. Sensors and Actuators, B: Chemical 297: 126717. Christendat D, Turnbull J. 1999. Identifying groups involved in the binding of prephenate to prephenate dehydrogenase from Escherichia coli. Biochemistry 38: 4782–4793. Christinet L, Burdet FX, Zaiko M, Hinz U, Zrÿd J-P. 2004. Characterization and functional identification of a novel plant 4,5-extradiol dioxygenase involved in betalain pigment biosynthesis in Portulaca grandiflora. Plant physiology 134: 265–274. Chung H-H, Schwinn KE, Ngo HM, Lewis DH, Massey B, Calcott KE, Crowhurst R, Joyce DC, Gould KS, Davies KM, et al. 2015. Characterisation of betalain biosynthesis in Parakeelya flowers identifies the key biosynthetic gene DOD as belonging to an expanded LigB gene family that is conserved in betalain- producing species. Frontiers in Plant Science 6: 499. Ciriminna R, Fidalgo A, Danzì C, Timpanaro G, Ilharco LM, Pagliaro M. 2018. Betanin: A bioeconomy insight into a valued betacyanin. ACS Sustainable Chemistry and Engineering 6: 2860-2865. Clement JS, Mabry TJ. 1996. Pigment evolution in the Caryophyllales: a systematic overview. Botanica Acta 109: 360–367. Clemente A, Desai P. 2011. Evaluation of the hematological, hypoglycemic, hypolipidemic and antioxidant properties of Amaranthus tricolor leaf extract in rat. Tropical Journal of Pharmaceutical Research 10: 595– 602. Contreras-Llano LE, Guerrero-Rubio MA, Lozada-Ramírez JD, García-Carmona F, Gandía-Herrero F. 2019b. First betalain-producing bacteria break the exclusive presence of the pigments in the plant kingdom. mBio 10 (2). Cooper MA, Taris JE, Shi C, Wood DW. 2018. A convenient split-intein tag method for the purification of tagless target proteins. Current Protocols in Protein Science 91: 5.29.1-5.29.23. Czapski J. 1990. Heat stability of betacyanins in red beet juice and in betanin solutions. Zeitschrift für Lebensmittel-Untersuchung und Forschung 191: 275–278. Delgado-Vargas F, Paredes-Lopez O. 2003. Natural colorants for food and nutraceutical uses. CRC Press, Florida, USA. A. Timoneda, PhD thesis, 2022 142 DeLoache WC, Russ ZN, Narcross L, Gonzales AM, Martin VJJ, Dueber JE. 2015. An enzyme-coupled biosensor enables (S)-reticuline production in yeast from glucose. Nature Chemical Biology 11: 465–471. Dilorio AA, Cheetham RD, Weathers PJ. 1992. Growth of transformed roots in a nutrient mist bioreactor: reactor performance and evaluation. Applied Microbiology and Biotechnology 37: 457–462. EFSA. 2015. Scientific opinion on the re-evaluation of beetroot red (E 162) as a food additive. EFSA Journal 13 (12): 4318. Ehrendorfer F. 1976. Closing remarks: Systematics and evolution of centrospermous families. Plant Systematics and Evolution 1976 126:1 126: 99–106. Eick GN, Bridgham JT, Anderson DP, Harms MJ, Thornton JW. 2017. Robustness of reconstructed ancestral protein functions to statistical uncertainty. Molecular Biology and Evolution 34: 247–261. von Elbe JH, Attoe EL. 1985. Oxygen involvement in betanin degradation - Measurement of active oxygen species and oxidation reduction potentials. Food Chemistry 16: 49–67. von Elbe JH, Maing IY, Amundson CH. 1974. Color stability of betanin. Journal of Food Science 39: 334–337. Engler C, Kandzia R, Marillonnet S. 2008. A one pot, one step, precision cloning method with high throughput capability. PloS One 3: e3647. Engler C, Youles M, Gruetzner R, Ehnert T-M, Werner S, Jones JDG, Patron NJ, Marillonnet S. 2014. A golden gate modular cloning toolbox for plants. ACS Synthetic Biology 3: 839–43. Evans R, O’Neill M, Pritzel A, Antropova N, Senior A, Green T, Žídek A, Bates R, Blackwell S, Yim J, et al. 2021. Protein complex prediction with AlphaFold-Multimer. bioRxiv: 2021.10.04.463034. FDA. 2021. CFR - Code of Federal Regulations Title 21. Chapter I. Part 73. Subpart A. Section 73.40. U.S. Food and Drug Administration. Franken P, Requena N. 2001. Analysis of gene expression in arbuscular mycorrhizas: new approaches and challenges. New Phytologist 150: 517–523. Freitas-Dörr BC, Machado CO, Pinheiro AC, Fernandes AB, Dörr FA, Pinto E, Lopes-Ferreira M, Abdellah M, Sá J, Russo LC, et al. 2020. A metal-free blue chromophore derived from plant pigments. Science Advances 6: eaaz0421. Fu L, Niu B, Zhu Z, Wu S, Li W. 2012. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics 28: 3150–3152. Gandía-Herrero F, Escribano J, García-Carmona F. 2005a. Betaxanthins as pigments responsible for visible fluorescence in flowers. Planta 222: 586–593. Gandía-Herrero F, Escribano J, García-Carmona F. 2010. Structural implications on color, fluorescence, and antiradical activity in betalains. Planta 232: 449–60. Gandía-Herrero F, García-Carmona F. 2012. Characterization of recombinant Beta vulgaris 4,5-DOPA- extradiol-dioxygenase active in the biosynthesis of betalains. Planta 236: 91–100. Gandía-Herrero F, García-Carmona F. 2013. Biosynthesis of betalains: yellow and violet plant pigments. Trends in Plant Science 18: 334–43. Gandía-Herrero F, García-Carmona F, Escribano J. 2005b. A novel method using high-performance liquid chromatography with fluorescence detection for the determination of betaxanthins. Journal of Chromatography A 1078: 83–89. Gandía-Herrero F, García-Carmona F, Escribano J. 2005c. A novel method using high-performance liquid chromatography with fluorescence detection for the determination of betaxanthins. Journal of Chromatography A 1078: 83–89. Bibliography 143 Gandía-Herrero F, García-Carmona F, Escribano J. 2005d. Floral fluorescence effect. Nature 437: 334–334. Gandía-Herrero F, García-Carmona F, Escribano J. 2006. Development of a protocol for the semi- synthesis and purification of betaxanthins. Phytochemical Analysis 17: 262–269. Gandía-Herrero F, Jiménez-Atiénzar M, Cabanes J, Escribano J, García-Carmona F. 2009. Fluorescence detection of tyrosinase activity on dopamine-betaxanthin purified from Portulaca oleracea (common purslane) flowers. Journal of Agricultural and Food Chemistry 57: 2523–2528. Ganesan P, Karthik T. 2017. Analysis of colour strength, colour fastness and antimicrobial properties of silk fabric dyed with natural dye from red prickly pear fruit. Journal of the Textile Institute 108: 1173–1179. Gengatharan A, Dykes GA, Choo WS. 2015. Betalains: Natural plant pigments with potential application in functional foods. LWT - Food Science and Technology 64: 645–649. Genre A, Lanfranco L, Perotto S, Bonfante P. 2020. Unique and common traits in mycorrhizal symbioses. Nature Reviews Microbiology 18: 649–660. Georgiev V, Ilieva M, Bley T, Pavlov A. 2008. Betalain production in plant in vitro systems. Acta Physiologiae Plantarum 30: 581–593. Georgiev VG, Weber J, Kneschke EM, Denev PN, Bley T, Pavlov AI. 2010. Antioxidant activity and phenolic content of betalain extracts from intact plants and hairy root cultures of the red beetroot Beta vulgaris cv. Detroit Dark Red. Plant Foods for Human Nutrition 65: 105–111. Gerdemann JW. 1955. Relation of a large soil-borne spore to phycomycetous mycorrhizal infections. Mycologia 47: 619–632. Girin T, El-Kafafi E-S, Widiez T, Erban A, Hubberten H-M, Kopka J, Hoefgen R, Gojon A, Lepetit M. 2010. Identification of Arabidopsis mutants impaired in the systemic regulation of root nitrate uptake by the nitrogen status of the plant. Plant Physiology 153: 1250–1260. Girod P-A, Zryd J-P. 1991a. Biogenesis of betalains: Purification and partial characterization of dopa 4,5- dioxygenase from Amanita muscaria. Phytochemistry 30: 169–174. Girod PA, Zryd JP. 1991b. Secondary metabolism in cultured red beet (Beta vulgaris L.) cells: Differential regulation of betaxanthin and betacyanin biosynthesis. Plant Cell, Tissue and Organ Culture 25: 1–12. Gonçalves LCP, Tonelli RR, Bagnaresi P, Mortara RA, Ferreira AG, Bastos EL. 2013. A nature- inspired betalainic probe for live-cell imaging of Plasmodium-infected erythrocytes. PLoS ONE 8: e53874. Grewal PS, Modavi C, Russ ZN, Harris NC, Dueber JE. 2018. Bioproduction of a betalain color palette in Saccharomyces cerevisiae. Metabolic Engineering 45: 180–188. Grotewold E, Davies K. 2008. Trafficking and sequestration of anthocyanins. Natural Product Communications 3: 1251–1258. Grotewold E, Sainz MB, Tagliani L, Hernandez JM, Bowen B, Chandler VL. 2000. Identification of the residues in the MYB domain of maize C1 that specify the interaction with the bHLH cofactor R. Proceedings of the National Academy of Sciences of the United States of America 97: 13579–13584. Guadarrama-Flores B, Rodriguez-Monroy M, Cruz-Sosa F, Garcia-Carmona F, Gandia-Herrero F. 2015. Production of dihydroxylated betalains and dopamine in cell suspension cultures of Celosia argentea var. plumosa. Journal of Agricultural and Food Chemistry 63: 2741–2749. Guerrero-Rubio MA, Escribano J, García-Carmona F, Gandía-Herrero F. 2020a. Light emission in betalains: from fluorescent flowers to biotechnological applications. Trends in Plant Science 25: 159–175. A. Timoneda, PhD thesis, 2022 144 Guerrero-Rubio MA, García-Carmona F, Gandía-Herrero F. 2020b. First description of betalains biosynthesis in an aquatic organism: characterization of 4,5-DOPA-extradiol-dioxygenase activity in the cyanobacteria Anabaena cylindrica. Microbial Biotechnology 13: 1948– 1959. Guerrero-Rubio MA, López-Llorca R, Henarejos-Escudero P, García-Carmona F, Gandía-Herrero F. 2019. Scaled-up biotechnological production of individual betalains in a microbial system. Microbial Biotechnology 12: 993–1002. Guerrero-Rubio MA, Martínez-Zapata J, Henarejos-Escudero P, García-Carmona F, Gandía-Herrero F. 2020c. Reversible bleaching of betalains induced by metals and application to the fluorescent determination of anthrax biomarker. Dyes and Pigments 180: 108493. Guesmi A, Ladhari N, Hamadi N Ben, Sakli F. 2012. Isolation, identification and dyeing studies of betanin on modified acrylic fabrics. Industrial Crops and Products 37: 342–346. Hahlbrock K, Scheel D. 2003. Physiology and molecular biology of phenylpropanoid metabolism. Annual Review of Plant Physiology and Plant Molecular Biology 40: 347–369. Haldane J. 1930. Enzymes. Longmans, Green and Co. New York, USA. Hall N, Karras M, Raine JD, Carlton JM, Kooij TWA, Berriman M, Florens L, Janssen CS, Pain A, Christophides GK, et al. 2005. A comprehensive survey of the Plasmodium life cycle by genomic, transcriptomic, and proteomic analyses. Science 307: 82–86. Harris NN, Javellana J, Davies KM, Lewis DH, Jameson PE, Deroles SC, Calcott KE, Gould KS, Schwinn KE. 2012. Betalain production is possible in anthocyanin-producing plant species given the presence of DOPA-dioxygenase and L-DOPA. BMC Plant Biology 12: 34. Harrison MJ. 2005. Signaling in the arbuscular mycorrhizal symbiosis. Annual Review of Microbiology 59: 19–42. Harrison MJ, Dewbre GR, Liu J. 2002. A phosphate transporter from Medicago truncatula involved in the acquisition of phosphate released by arbuscular mycorrhizal fungi. The Plant cell 14: 2413–2429. Hatlestad GJ, Akhavan N a, Sunnadeniya RM, Elam L, Cargile S, Hembd A, Gonzalez A, McGrath JM, Lloyd AM. 2014. The beet Y locus encodes an anthocyanin MYB-like protein that activates the betalain red pigment pathway. Nature Genetics 47: 92–96. Hatlestad GJ, Sunnadeniya RM, Akhavan N a, Gonzalez A, Goldman IL, McGrath JM, Lloyd AM. 2012. The beet R locus encodes a new cytochrome P450 required for red betalain production. Nature Genetics 44: 816–820. Hayakawa K, Agarie S. 2010. Physiological roles of betacyanin in a halophyte, Suaeda japonica Makino. Plant Production Science 13: 351–359. He J, Evans NM, Liu H, Shao S. 2020a. A review of research on plant-based meat alternatives: Driving forces, history, manufacturing, and consumer attitudes. Comprehensive Reviews in Food Science and Food Safety 19: 2639–2656. He Y, Zhang T, Sun H, Zhan H, Zhao Y. 2020b. A reporter for noninvasively monitoring gene expression and plant transformation. Horticulture Research 7: 152. Hempel de Ibarra N. 2000. Detection of bright and dim colours by honeybees. The Journal of Experimental Biology 203: 3289–3298. Henarejos-Escudero P, Guadarrama-Flores B, Guerrero-Rubio MA, Gómez-Pando LR, García- Carmona F, Gandía-Herrero F. 2018. Development of betalain producing callus lines from colored quinoa varieties (Chenopodium quinoa Willd). Journal of Agricultural and Food Chemistry 66: 467–474. Herbach KM, Stintzing FC, Carle R. 2006a. Betalain stability and degradation - Structural and chromatic aspects. Journal of Food Science 71: R41–R50. Bibliography 145 Herbach KM, Stintzing FC, Carle R. 2006b. Impact of thermal treatment on color and pigment pattern of red beet (Beta vulgaris L.) preparations. Journal of Food Science 69: C491–C498. Hilou A, Nacoulma OG, Guiguemde TR. 2006. In vivo antimalarial activities of extracts from Amaranthus spinosus L. and Boerhaavia erecta L. in mice. Journal of Ethnopharmacology 103: 236–240. Hinz UG, Fivaz J, Girod PA, Zyrd JP. 1997. The gene coding for the DOPA dioxygenase involved in betalain biosynthesis in Amanita muscaria and its regulation. Molecular and General Genetics 256: 1–6. Hohnjec N, Vieweg MF, Pühler A, Becker A, Küster H. 2005. Overlaps in the transcriptional profiles of Medicago truncatula roots inoculated with two different Glomus fungi provide insights into the genetic program activated during arbuscular mycorrhiza. Plant physiology 137: 1283–301. Holford ICR. 1997. Soil phosphorus: Its measurement, and its uptake by plants. Australian Journal of Soil Research 35: 227–239. Huang AS, von Elbe JH. 1987. Effect of pH on the degradation and regeneration of betanine. Journal of Food Science 52: 1689–1693. Ibdah M, Krins A, Seidlitz HK, Heller W, Strack D, Vogt T. 2002. Spectral dependence of flavonol and betacyanin accumulation in Mesembryanthemum crystallinum under enhanced ultraviolet radiation. Plant, Cell and Environment 25: 1145–1154. Imamura T, Takagi H, Miyazato A, Ohki S, Mizukoshi H, Mori M. 2018. Isolation and characterization of the betalain biosynthesis gene involved in hypocotyl pigmentation of the allotetraploid Chenopodium quinoa. Biochemical and Biophysical Research Communications 496: 280–286. Iwashina T, Ootani S, Hayashi K. 1988. On the pigmented spherical bodies and crystals in tepals of cactaceous species in reference to the nature of betalains or flavonols. The Botanical Magazine Tokyo 101: 175–184. Jain G, Gould KS. 2015a. Functional significance of betalain biosynthesis in leaves of Disphyma australe under salinity stress. Environmental and Experimental Botany 109: 131–140. Jain G, Gould KS. 2015b. Are betalain pigments the functional homologues of anthocyanins in plants? Environmental and Experimental Botany 119: 48–53. Jain G, Schwinn KE, Gould KS. 2015. Betalain induction by L-DOPA application confers photoprotection to saline-exposed leaves of Disphyma australe. The New phytologist 207: 1075–1083. Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, Tunyasuvunakool K, Bates R, Žídek A, Potapenko A, et al. 2021. Highly accurate protein structure prediction with AlphaFold. Nature 596: 583–589. Kalyaanamoorthy S, Minh BQ, Wong TKF, Von Haeseler A, Jermiin LS. 2017. ModelFinder: fast model selection for accurate phylogenetic estimates. Nature Methods 14: 587–589. Kang K, Lee K, Sohn SO, Park S, Lee S, Kim SY, Kim YS, Back K. 2009. Ectopic expression of serotonin N-hydroxycinnamoyltransferase and differential production of phenylpropanoid amides in transgenic tomato tissues. Scientia Horticulturae 120: 504–510. Kanner J, Harel S, Granit R. 2001. Betalains - A new class of dietary cationized antioxidants. Journal of Agricultural and Food Chemistry 49: 5178–5185. Katoh K, Standley DM. 2013. MAFFT Multiple Sequence Alignment Software Version 7: Improvements in performance and usability. Molecular Biology and Evolution 30: 772–780. Kessler A, Baldwin IT. 2002. Plant response to insect herbivory: The emerging molecular analysis. Annual Review of Plant Biology 53: 299–328. Khan MI, Giridhar P. 2015. Plant betalains: Chemistry and biochemistry. Phytochemistry 117: 267–295. A. Timoneda, PhD thesis, 2022 146 Khlebnikov A, Dubuis B, Kut OM, Prenosil JE. 1995. Growth and productivity of Beta vulgaris cell culture in fluidized bed reactors. Bioprocess Engineering 14: 51–56. Kino-Oka M, Taya M, Tone S. 1995. Culture of red beet hairy roots in a column-type reactor associated with pigment release. Plant tissue culture letters 12: 201–204. Kishima Y, Shimaya A, Adachi T. 1995. Evidence that blue light induces betalain pigmentation in Portulaca callus. Plant Cell, Tissue and Organ Culture 43: 67–70. Klingner A, Bothe H, Wray V, Marner FJ. 1995a. Identification of a yellow pigment formed in maize roots upon mycorrhizal colonization. Phytochemistry 38: 53–55. Klingner A, Hundeshagen B, Kernebeek H, Bothe H. 1995b. Localization of the yellow pigment formed in roots of gramineous plants colonized by arbuscular fungi. Protoplasma 185: 50-57. Kochhar VK, Kochhar S, Mohr H. 1981. An analysis of the action of light on betalain synthesis in the seedling of Amaranthus caudatus, var. viridis. Planta 151: 81–87. Kokkonen P, Beier A, Mazurenko S, Damborsky J, Bednar D, Prokop Z. 2021. Substrate inhibition by the blockage of product release and its control by tunnel engineering. RSC Chemical Biology 2: 645–655. Kugler F, Stintzing FC, Carle R. 2007. Characterisation of betalain patterns of differently coloured inflorescences from Gomphrena globosa L. and Bougainvillea sp. by HPLC-DAD-ESI-MS. Analytical and Bioanalytical Chemistry 387: 637–648. Kurihara D, Mizuta Y, Sato Y, Higashiyama T. 2015. ClearSee: A rapid optical clearing reagent for whole-plant fluorescence imaging. Development 142: 4168–4179. Kuusi T, Pyysalo H, Pippuri A. 1977. The effect of iron, tin, aluminium, and chromium on fading, discoloration, and precipitation in berry and red beet juices. Zeitschrift für Lebensmittel-Untersuchung und Forschung 163: 196–202. Lakhotia P, Singh KP, Singh SK, Singh MC, Prasad K V., Swaroop K. 2014. Influence of biotic and abiotic elicitors on production of betalain pigments in Bougainvillea callus cultures. Indian Journal of Horticulture 71: 373–378. Lee DW, Collins TM. 2001. Phylogenetic and ontogenetic influences on the distribution of anthocyanins and betacyanins in leaves of tropical plants. International Journal of Plant Sciences 162: 1141–1153. Legrand P, Dumas R, Seux M, Rippert P, Ravelli R, Ferrer JL, Matringe M. 2006. Biochemical characterization and crystal structure of Synechocystis arogenate dehydrogenase provide insights into catalytic reaction. Structure 14: 767–776. Li Y, Cirino PC. 2014. Recent advances in engineering proteins for biocatalysis. Biotechnology and Bioengineering 111: 1273–1287. Lin YK, Yeh YC. 2017. Dual-signal microbial biosensor for the detection of dopamine without inference from other catecholamine neurotransmitters. Analytical Chemistry 89: 11178–11182. Lipscomb JD. 2008. Mechanism of extradiol aromatic ring-cleaving dioxygenases. Current Opinion in Structural Biology 18: 644-649. Lloyd A, Brockman A, Aguirre L, Campbell A, Bean A, Cantero A, Gonzalez A. 2017. Advances in the MYB–bHLH–WD Repeat (MBW) pigment regulatory model: Addition of a WRKY factor and co-option of an anthocyanin MYB for betalain regulation. Plant and Cell Physiology 58: 1431–1441. Logemann E, Tavernaro A, Schulz W, Somssich IE, Hahlbrock K. 2000. UV light selectively coinduces supply pathways from primary metabolism and flavonoid secondary product formation in parsley. Proceedings of the National Academy of Sciences of the United States of America 97: 1903–1907. Bibliography 147 Lopez-Nieves S, El-Azaz J, Men Y, Holland CK, Feng T, Brockington SF, Jez JM, Maeda HA. 2021. Two independently evolved natural mutations additively deregulate TyrA enzymes and boost tyrosine production in planta. The Plant Journal, 109 (4): 844-855. Lopez-Nieves S, Yang Y, Timoneda A, Wang M, Feng T, Smith SA, Brockington SF, Maeda HA. 2018. Relaxation of tyrosine pathway regulation underlies the evolution of betalain pigmentation in Caryophyllales. New Phytologist 217: 896-908. Löytynoja A, Goldman N. 2005. An algorithm for progressive multiple alignment of sequences with insertions. Proceedings of the National Academy of Sciences 102: 10557–10562. Mabry TJ. 1977. The Order Centrospermae. Annals of the Missouri Botanical Garden 64: 210. Maeda H, Dudareva N. 2012. The Shikimate pathway and aromatic amino acid biosynthesis in plants. Annual Review of Plant Biology 63: 73–105. Mariani V, Biasini M, Barbato A, Schwede T. 2013. lDDT: a local superposition-free score for comparing protein structures and models using distance difference tests. Bioinformatics 29: 2722. Marquardt DW. 2006. An algorithm for least-squares estimation of nonlinear parameters. Journal of the Society for Industrial and Applied Mathematics 11: 431–441. Marshall J, Johnsen S. 2017. Fluorescence as a means of colour signal enhancement. Philosophical Transactions of the Royal Society B: Biological Sciences 372, 1724. Martin C. 2013. The interface between plant metabolic engineering and human health. Current Opinion in Biotechnology 24: 344–353. Matile P. 1978. Biochemistry and function of vacuoles. Annual Review of Plant Physiology 29: 193–213. Merin U, Gagel S, Popel G, Bernstein S, Rosenthal I. 1987. Thermal degradation kinetics of prickly- pear-gruit red pigment. Journal of Food Science 52: 485–486. Miller G, Suzuki N, Ciftci-Yilmaz S, Mittler R. 2010. Reactive oxygen species homeostasis and signalling during drought and salinity stresses. Plant, Cell & Environment 33: 453–467. Mirdita M, Schütze K, Moriwaki Y, Heo L, Ovchinnikov S, Steinegger M. 2021. ColabFold - Making protein folding accessible to all. bioRxiv: 2021.08.15.456425. Mosco A. 2012. Tissue localization of betacyanins in cactus stems. Revista Mexicana de Biodiversidad 83: 413–420. Mueller LA, Hinz U, Uzé M, Sautter C, Zryd JP. 1997. Biochemical complementation of the betalain biosynthetic pathway in Portulaca grandiflora by a fungal 3,4-dihydroxyphenylalanine dioxygenase. Planta 203: 260–263. Musso H. 1979. The pigments of fly agaric, Amanita muscaria. Tetrahedron 35: 2843–2853. Nakashima T, Araki T, Ueno O. 2011. Photoprotective function of betacyanin in leaves of Amaranthus cruentus L. under water stress. Photosynthetica 49: 497–506. Nakashima KK, Bastos EL. 2019. Rationale on the high radical scavenging capacity of betalains. Antioxidants 8: 222. Netzel M, Stintzing FC, Quaas D, Straß G, Carle R, Bitsch R, Bitsch I, Frank T. 2005. Renal excretion of antioxidative constituents from red beet in humans. Food Research International 38: 1051–1058. Nguyen LT, Schmidt HA, Von Haeseler A, Minh BQ. 2015. IQ-TREE: A fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Molecular Biology and Evolution 32: 268–274. de Nicola MG, Amico V, Sciuto S, Piattelli M. 1975. Light control of amaranthin synthesis in isolated Amaranthus cotyledons. Phytochemistry 14: 479–481. A. Timoneda, PhD thesis, 2022 148 Nicolson TH. 1959. Mycorrhiza in the Gramineae. Transactions of the British Mycological Society 42: 421- IN3. Osbourn A. 2010. Gene clusters for secondary metabolic pathways: An emerging theme in plant biology. Plant Physiology 154: 531–535. Ostroverkhova O, Galindo G, Lande C, Kirby J, Scherr M, Hoffman G, Rao S. 2018. Understanding innate preferences of wild bee species: responses to wavelength-dependent selective excitation of blue and green photoreceptor types. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology 204: 667–675. Owen C, Patron NJ, Huang A, Osbourn A. 2017. Harnessing plant metabolic diversity. Current Opinion in Chemical Biology 40: 24–30. Pak H, Guo Y, Chen M, Chen K, Li Y, Hua S, Shamsi I, Meng H, Shi C, Jiang L. 2009. The effect of exogenous methyl jasmonate on the flowering time, floral organ morphology, and transcript levels of a group of genes implicated in the development of oilseed rape flowers (Brassica napus L.). Planta 231: 79– 91. Parida AK, Das AB. 2005. Salt tolerance and salinity effects on plants: A review. Ecotoxicology and Environmental Safety 60: 324–349. Parniske M. 2008. Arbuscular mycorrhiza: The mother of plant root endosymbioses. Nature Reviews Microbiology 6: 763–775. Pasch JH, von Elbe JH. 1979. Betanine stability in buffered solutions containing organic acids, metal cations, antioxidants or sequestrants. Journal of Food Science 44: 72–75. Paszkowski U, Jakovleva L, Boller T. 2006. Maize mutants affected at distinct stages of the arbuscular mycorrhizal symbiosis. The Plant Journal 47: 165–173. Pavlov A, Georgiev V, Kovatcheva P. 2003. Relationship between type and age of the inoculum cultures and betalains biosynthesis by Beta vulgaris hairy root culture. Biotechnology Letters 25: 307–309. Pearce G, Marchand PA, Griswold J, Lewis NG, Ryan CA. 1998. Accumulation of feruloyltyramine and p-coumaroyltyramine in tomato leaves in response to wounding. Phytochemistry 47: 659–664. Petrus-Vancea A, Cachiţă-Cosma D, Pop L, Purcărea C. 2010. Histo-anatomical specific features of red beet and sugar beet vitroplantlets. Analele Universităţii din Oradea, Fascicula: Protecţia Mediului 15: 740- 749. Phillips JM, Hayman DS. 1970. Improved procedures for clearing roots and staining parasitic and vesicular-arbuscular mycorrhizal fungi for rapid assessment of infection. Transactions of the British Mycological Society 55: 158-IN18. Phytolon. 2021. Phytolon, Natural Food Colors. Date accessed: 14-12-2021. https://www.phytolon.com/ Piatelli M. 1981. The betalains: Structure, biosynthesis, and chemical taxonomy. Secondary Plant Products: 557–575. Polturak G, Aharoni A. 2017. ‘La vie en rose’: Biosynthesis, sources and applications of betalain pigments. Molecular plant 11: 7-22. Polturak G, Breitel D, Grossman N, Sarrion-Perdigones A, Weithorn E, Pliner M, Orzaez D, Granell A, Rogachev I, Aharoni A. 2016. Elucidation of the first committed step in betalain biosynthesis enables the heterologous engineering of betalain pigments in plants. The New phytologist 210: 269–283. Polturak G, Grossman N, Vela-Corcia D, Dong Y, Nudel A, Pliner M, Levy M, Rogachev I, Aharoni A. 2017. Engineered gray mold resistance, antioxidant capacity, and pigmentation in betalain-producing crops and ornamentals. Proceedings of the National Academy of Sciences of the United States of America 114: 9062–9067. Bibliography 149 Polturak G, Heinig U, Grossman N, Battat M, Leshkowitz D, Malitsky S, Rogachev I, Aharoni A. 2018. Transcriptome and metabolic profiling provides insights into betalain biosynthesis and evolution in Mirabilis jalapa. Molecular Plant 11: 189–204. Price MN, Dehal PS, Arkin AP. 2010. FastTree 2 – Approximately maximum-likelihood trees for large alignments. PLOS ONE 5: e9490. Pupko T, Pe’er I, Hasegawa M, Graur D, Friedman N. 2002. A branch-and-bound algorithm for the inference of ancestral amino-acid sequences when the replacement rate varies among sites: Application to the evolution of five gene families. Bioinformatics 18: 1116–1123. Pupko T, Pe’er I, Shamir R, Graur D. 2000. A fast algorithm for joint reconstruction of ancestral amino acid sequences. Molecular Biology and Evolution 17: 890–896. Purohit A, England JK, Douma LG, Tondnevis F, Bloom LB, Levitus M. 2017. Electrostatic interactions at the dimer interface stabilize the E. coli β sliding clamp. Biophysical Journal 113: 794–804. Qingzhu H, Chengjie C, Zhe C, Pengkun C, Yuewen M, Jingyu W, Jian Z, Guibing H, Jietang Z, Yonghua Q. 2016. Transcriptomic analysis reveals key genes related to betalain biosynthesis in pulp coloration of Hylocereus polyrhizus. Frontiers in Plant Science 6: 1179. Quan C. 2018. A family of small secreted proteins from Rhizophagus irregularis contribute to arbuscular mycorrhiza symbiosis. PhD thesis. University of Cambridge, Cambridge, UK. Ramsay NA, Glover BJ. 2005. MYB-bHLH-WD40 protein complex and the evolution of cellular diversity. Trends in plant science 10: 63–70. Rao S, Ostroverkhova O. 2015. Visual outdoor response of multiple wild bee species: highly selective stimulation of a single photoreceptor type by sunlight-induced fluorescence. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology 201: 705–716. Rey T, Chatterjee A, Buttay M, Toulotte J, Schornack S. 2015. Medicago truncatula symbiosis mutants affected in the interaction with a biotrophic root pathogen. New Phytologist 206: 497–500. Rippert P, Matringe M. 2002a. Molecular and biochemical characterization of an Arabidopsis thaliana arogenate dehydrogenase with two highly similar and active protein domains. Plant Molecular Biology 48: 361–368. Rippert P, Matringe M. 2002b. Purification and kinetic analysis of the two recombinant arogenate dehydrogenase isoforms of Arabidopsis thaliana. European journal of biochemistry / FEBS 269: 4753–61. Rippert P, Puyaubert J, Grisollet D, Derrier L, Matringe M. 2009. Tyrosine and phenylalanine are synthesized within the plastids in Arabidopsis. Plant physiology 149: 1251–60. Rodríguez-Monroy M, Galindo E. 1999. Broth rheology, growth and metabolite production of Beta vulgaris suspension culture: A comparative study between cultures grown in shake flasks and in a stirred tank. Enzyme and Microbial Technology 24: 687–693. Rognes T. 2011. Faster Smith-Waterman database searches with inter-sequence SIMD parallelisation. BMC Bioinformatics 12: 221. Saguy I, Kopelman1 IJ, Mizrahi S. 1978. Thermal kinetic degradation of betanin and betalamic acid. Journal of Agricultural and Food Chemistry 26: 360–362. Sakurai K, Oobatake M, Goto Y. 2001. Salt-dependent monomer–dimer equilibrium of bovine β- lactoglobulin at pH 3. Protein Science : A Publication of the Protein Society 10: 2325. Santos-Díaz S, Potos SL, Gonz MM, Potos SL. 2005. Pigment production by callus of Mammillaria candida Scheidweiler (Cactaceae). Agrociencia: 619–626. A. Timoneda, PhD thesis, 2022 150 Sapers GM, Hornstein JS. 1979. Varietal differences in colorant properties and stability of red beet pigments. Journal of Food Science 44: 1245–1248. Sasaki N, Abe Y, Goda Y, Adachi T, Kasahara K, Ozeki Y. 2009. Detection of DOPA 4,5-Dioxygenase (DOD) activity using recombinant protein prepared from Escherichia coli cells harboring cDNA encoding DOD from Mirabilis jalapa. Plant and Cell Physiology 50: 1012–1016. Sasaki N, Adachi T, Koda T, Ozeki Y. 2004. Detection of UDP-glucose:cyclo-DOPA 5-O- glucosyltransferase activity in four o’clocks (Mirabilis jalapa L.). FEBS letters 568: 159–62. Sasaki N, Wada K, Koda T, Kasahara K, Adachi T, Ozeki Y. 2005. Isolation and characterization of cDNAs encoding an enzyme with glucosyltransferase activity for cyclo-DOPA from four o’clocks and feather cockscombs. Plant & cell physiology 46: 666–70. Savitha BC, Thimmaraju R, Bhagyalakshmi N, Ravishankar GA. 2006. Different biotic and abiotic elicitors influence betalain production in hairy root cultures of Beta vulgaris in shake-flask and bioreactor. Process Biochemistry 41: 50–60. Schenck CA, Chen S, Siehl DL, Maeda HA. 2015. Non-plastidic, tyrosine-insensitive prephenate dehydrogenases from legumes. Nature Chemical Biology 11: 52–57. Schenck CA, Holland CK, Schneider MR, Men Y, Lee SG, Jez JM, Maeda HA. 2017. Molecular basis of the evolution of alternative tyrosine biosynthetic routes in plants. Nature Chemical Biology 2017. 13: 1029– 1035. Schenck CA, Maeda HA. 2018. Tyrosine biosynthesis, metabolism, and catabolism in plants. Phytochemistry 149: 82–102. Schliemann W, Joy IV RW, Komamine A, Metzger JW, Nimtz M, Wray V, Strack D. 1996. Betacyanins from plants and cell cultures of Phytolacca americana. Phytochemistry 42: 1039–1046. Schwartz SJ, von Elbe JH. 1983. Identification of betanin degradation products. Zeitschrift für Lebensmittel-Untersuchung und Forschung 176: 448–453. Sepúlveda-Jiménez G, Rueda-Benítez P, Porta H, Rocha-Sosa M. 2005. A red beet (Beta vulgaris) UDP-glucosyltransferase gene induced by wounding, bacterial infiltration and oxidative stress. Journal of Experimental Botany 56: 605–611. Sheehan H, Feng T, Walker‐Hale N, Lopez‐Nieves S, Pucker B, Guo R, Yim WC, Badgami R, Timoneda A, Zhao L, et al. 2020. Evolution of L-DOPA 4,5‐dioxygenase activity allows for recurrent specialisation to betalain pigmentation in Caryophyllales. New Phytologist 227: 914–929. Shima S, Tziatzios C, Schubert D, Fukada H, Takahashi K, Ermler U, Thauer RK. 1998. Lyotropic-salt- induced changes in monomer/dimer/tetramer association equilibrium of formyltransferase from the hyperthermophilic Methanopyrus kandleri in relation to the activity and thermostability of the enzyme. European Journal of Biochemistry 258: 85–92. Shimada S, Inoue YT, Sakuta M. 2005. Anthocyanidin synthase in non-anthocyanin-producing Caryophyllales species. The Plant Journal 44: 950–959. Shimada S, Takahashi K, Sato Y, Sakuta M. 2004. Dihydroflavonol 4-reductasecDNA from non- anthocyanin-producing species in the Caryophyllales. Plant and Cell Physiology 45: 1290–1298. Shin KS, Murthy HN, Ko JY, Paek KY. 2002. Growth and betacyanin production by hairy roots of Beta vulgaris in airlift bioreactors. Biotechnology Letters 2002 24:24 24: 2067–2069. Shu Z, Shao L, Huang HY, Zeng XQ, Lin ZF, Chen GY, Peng CL. 2009. Comparison of thermostability of PSII between the chromatic and green leaf cultivars of Amaranthus tricolor L. Photosynthetica 47: 548– 558. Bibliography 151 Singer JW, von Elbe JH. 1980. Degradation rates of vulgaxanthine I. Journal of Food Science 45: 489– 491. Sivakumar V, Anna JL, Vijayeeswarri J, Swaminathan G. 2009. Ultrasound assisted enhancement in natural dye extraction from beetroot for industrial applications and natural dyeing of leather. Ultrasonics Sonochemistry 16: 782–789. Smith SE, Smith FA, Jakobsen I. 2003. Mycorrhizal fungi can dominate phosphate supply to plants irrespective of growth responses. Plant Physiology 133: 16–20. Sparkes IA, Runions J, Kearns A, Hawes C. 2006b. Rapid, transient expression of fluorescent fusion proteins in tobacco plants and generation of stably transformed plants. Nature Protocols 1: 2019–2025. Spatafora JW, Chang Y, Benny GL, Lazarus K, Smith ME, Berbee ML, Bonito G, Corradi N, Grigoriev I, Gryganskyi A, et al. 2016. A phylum-level phylogenetic classification of zygomycete fungi based on genome-scale data. Mycologia 108: 1028–1046. Sreekanth D, Arunasree MK, Roy KR, Chandramohan Reddy T, Reddy G V, Reddanna P. 2007. Betanin a betacyanin pigment purified from fruits of Opuntia ficus-indica induces apoptosis in human chronic myeloid leukemia Cell line-K562. Phytomedicine: International Journal of Phytotherapy and Phytopharmacology 14: 739–46. Stafford HA. 1994. Anthocyanins and betalains: evolution of the mutually exclusive pathways. Plant Science 101: 91–98. Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30: 1312–1313. Stevens P. 2021. Angiosperm Phylogeny Website. Version 14. Date accessed: 01-12-2021. http://www.mobot.org/MOBOT/research/APweb/ Stintzing FC, Carle R. 2007. Betalains - Emerging prospects for food scientists. Trends in Food Science and Technology 18: 514–525. Stintzing FC, Herbach KM, Mosshammer MR, Carle R, Yi W, Sellappan S, Akoh CC, Bunch R, Felker P. 2005. Color, betalain pattern, and antioxidant properties of cactus pear (Opuntia spp.) clones. Journal of Agricultural and Food Chemistry 53: 442–451. Strack D, Vogt T, Schliemann W. 2003. Recent advances in betalain research. Phytochemistry 62: 247– 269. Stücheli P, Sieber S, Fuchs DW, Scheller L, Strittmatter T, Saxena P, Gademann K, Fussenegger M. 2020. Genetically encoded betaxanthin-based small-molecular fluorescent reporter for mammalian cells. Nucleic acids research 48: e67. Sugimoto K, Senda T, Aoshima H, Masai E, Fukuda M, Mitsui Y. 1999. Crystal structure of an aromatic ring opening dioxygenase LigAB, a protocatechuate 4,5-dioxygenase, under aerobic conditions. Structure 7: 953–965. Sunnadeniya R, Bean A, Brown M, Akhavan N, Hatlestad G, Gonzalez A, Symonds VV, Lloyd A. 2016. Tyrosine hydroxylation in betalain pigment biosynthesis is performed by cytochrome P450 enzymes in beets (Beta vulgaris). PLOS ONE 11: e0149417. Suresh B, Thimmaraju R, Bhagyalakshmi N, Ravishankar GA. 2004. Polyamine and methyl jasmonate- influenced enhancement of betalaine production in hairy root cultures of Beta vulgaris grown in a bubble column reactor and studies on efflux of pigments. Process biochemistry 39: 2091-2096. Suzuki M, Miyahara T, Tokumoto H, Hakamatsuka T, Goda Y, Ozeki Y, Sasaki N. 2014. Transposon- mediated mutation of CYP76AD3 affects betalain synthesis and produces variegated flowers in four o’clock (Mirabilis jalapa). Journal of plant physiology 171: 1586–90. A. Timoneda, PhD thesis, 2022 152 Takahashi S, Badger MR. 2011. Photoprotection in plants: A new light on photosystem II damage. Trends in Plant Science 16: 53–60. Takahashi K, Takamura E, Sakuta M. 2009. Isolation and expression analysis of two DOPA dioxygenases in Phytolacca americana. Zeitschrift fur Naturforschung - Section C Journal of Biosciences 64: 564–573. Tanaka Y, Sasaki N, Ohmiya A. 2008. Biosynthesis of plant pigments: Anthocyanins, betalains and carotenoids. Plant Journal 54: 733–749. Tenore GC, Novellino E, Basile A. 2012. Nutraceutical potential and antioxidant benefits of red pitaya (Hylocereus polyrhizus) extracts. Journal of Functional Foods 4: 129–136. Terradas F, Wyler H. 1991. 2,3- and 4,5-Secodopa, the biosynthetic intermediates generated from L-DOPA by an enzyme system extracted from the fly agaric, Amanita muscaria L., and their spontaneous conversion to muscaflavin and betalamic acid, respectively, and betalains. Helvetica Chimica Acta 74: 124–140. Tesoriere L, Allegra M, Butera D, Livrea MA. 2004. Absorption, excretion, and distribution of dietary antioxidant betalains in LDLs: potential health effects of betalains in humans. The American journal of clinical nutrition 80: 941–5. Thulin M, Moore AJ, El-Seedi H, Larsson A, Christin PA, Edwards EJ. 2016. Phylogeny and generic delimitation in Molluginaceae, new pigment data in Caryophyllales, and the new family Corbichoniaceae. TAXON 65: 775–793. Tian Y, Fu X, Yang Z, Wang B, Gao J, Wang M, Xu J, Han H, Li Z, Yao Q, et al. 2019. Metabolic engineering of rice endosperm for betanin biosynthesis. New Phytologist 225: 1915–1922. Timoneda A. 2017. Re-directing primary metabolism to increase betalain production. MPhil thesis. University of Cambridge, Cambridge, UK. Timoneda A, Feng T, Sheehan H, Walker‐Hale N, Pucker B, Lopez‐Nieves S, Guo R, Brockington S. 2019. The evolution of betalain biosynthesis in Caryophyllales. New Phytologist 224: 71–85. Timoneda A, Sheehan H, Feng T, Lopez-Nieves S, Maeda HA, Brockington S. 2018. Redirecting primary metabolism to boost production of tyrosine-derived specialised metabolites in planta. Scientific Reports 8: 17256. Timoneda A, Yunusov T, Quan C, Gavrin A, Brockington SF, Schornack S. 2021. MycoRed: Betalain pigments enable in vivo real-time visualisation of arbuscular mycorrhizal colonisation. PLOS Biology 19: e3001326. Trezzini GF, Zrÿd JP. 1991. Characterization of some natural and semi-synthetic betaxanthins. Phytochemistry 30: 1901–1903. Velicanski A, Cvetkovic D, Markov S, Vulic J, Djilas S. 2011. Antibacterial activity of Beta vulgaris L. pomace extract. Acta periodica technologica 42: 263–269. Vierheilig H, Coughlan AP, Wyss U, Piché Y. 1998. Ink and vinegar, a simple staining technique for arbuscular-mycorrhizal fungi. Applied and Environmental Microbiology 64: 5004–5007. Vogt T. 2002. Substrate specificity and sequence analysis define a polyphyletic origin of betanidin 5- and 6-O-glucosyltransferase from Dorotheanthus bellidiformis. Planta 214: 492–5. Vogt T, Grimm R, Strack D. 1999a. Cloning and expression of a cDNA encoding betanidin 5-O- glucosyltransferase, a betanidin- and flavonoid-specific enzyme with high homology to inducible glucosyltransferases from the Solanaceae. The Plant Journal 19: 509–519. Vogt T, Ibdah M, Schmidt J, Wray V, Nimtz M, Strack D. 1999b. Light-induced betacyanin and flavonol accumulation in bladder cells of Mesembryanthemum crystallinum. Phytochemistry 52: 583–592. Bibliography 153 Voigt CC, Roeleke M, Marggraf L, Pētersons G, Voigt-Heucke SL. 2017. Migratory bats respond to artificial green light with positive phototaxis. PLOS ONE 12: e0177748. Walker JF, Yang Y, Feng T, Timoneda A, Mikenas J, Hutchison V, Edwards C, Wang N, Ahluwalia S, Olivieri J, et al. 2018. From cacti to carnivores: Improved phylotranscriptomic sampling and hierarchical homology inference provide further insight into the evolution of Caryophyllales. American Journal of Botany 105: 446–462. Wang C-Q, Chen M, Wang B-S. 2007a. Betacyanin accumulation in the leaves of C3 halophyte Suaeda salsa L. is induced by watering roots with H2O2. Plant Science 172: 1–7. Wang CQ, Liu T. 2007. Involvement of betacyanin in chilling-induced photoinhibition in leaves of Suaeda salsa. Photosynthetica 45: 182–188. Wang M, Lopez-Nieves S, Goldman IL, Maeda HA. 2017. Limited tyrosine utilization explains lower betalain contents in yellow than in red table beet genotypes. Journal of Agricultural and Food Chemistry 65: 4305–4313. Wang M, Schäfer M, Li D, Halitschke R, Dong C, McGale E, Paetz C, Song Y, Li S, Dong J, et al. 2018. Blumenols as shoot markers of root symbiosis with arbuscular mycorrhizal fungi. eLife 7: e37093. Wang C-Q, Xu C, Wei J-G, Wang H-B, Wang S-H. 2007b. Enhanced tonoplast H+-ATPase activity and superoxide dismutase activity in the halophyte Suaeda salsa containing high level of betacyanin. Journal of Plant Growth Regulation 27: 58–67. Wasserman BP, Eiberger LL, Guilfoy MP. 1984. Effect of hydrogen peroxide and phenolic compounds on horseradish peroxidase-catalyzed decolorization of betalain pigments. Journal of Food Science 49: 536–538. Weber E, Engler C, Gruetzner R, Werner S, Marillonnet S. 2011. A modular cloning system for standardized assembly of multigene constructs. PLoS ONE 6: e16765. Wendel M, Nizinski S, Tuwalska D, Starzak K, Szot D, Prukala D, Sikorski M, Wybraniec S, Burdzinski G. 2015. Time-resolved spectroscopy of the singlet excited state of betanin in aqueous and alcoholic solutions. Physical Chemistry Chemical Physics 17: 18152–18158. Werner S, Engler C, Weber E, Gruetzner R, Marillonnet S. 2012. Fast track assembly of multigene constructs using Golden Gate cloning and the MoClo system. Bioengineered Bugs 3: 38–43. Winter Y, López J, Von Helversen O. 2003. Ultraviolet vision in a bat. Nature 425: 612–614. Wroblewska M, Juskiewicz J, Wiczkowski W. 2011. Physiological properties of beetroot crisps applied in standard and dyslipidaemic diets of rats. Lipids in Health and Disease 10, 178. Wruss J, Waldenberger G, Huemer S, Uygun P, Lanzerstorfer P, Müller U, Höglinger O, Weghuber J. 2015. Compositional characteristics of commercial beetroot products and beetroot juice prepared from seven beetroot varieties grown in Upper Austria. Journal of Food Composition and Analysis 42: 46–55. Wyler H, Meuer U, Bauer J, Stravs-Mombelli L. 1984. Cyclodopa glucoside ( = (2S)-5-(β-D- glucopyranosyloxy)-6-hydroxyindoline-2-carboxylic acid) and its occurrence in red beet (Beta vulgaris var. rubra L.). Helvetica Chimica Acta 67: 1348–1355. Xie F, Hua Q, Chen C, Zhang L, Zhang Z, Chen J, Zhang R, Zhao J, Hu G, Zhao J, et al. 2020. Transcriptomics-based identification and characterization of glucosyltransferases involved in betalain biosynthesis in Hylocereus megalanthus. Plant Physiology and Biochemistry 152: 112–124. Xu W, Dubos C, Lepiniec L. 2015. Transcriptional control of flavonoid biosynthesis by MYB-bHLH-WDR complexes. Trends in plant science 20: 176–85. Yang RYK, Bayraktar O, Pu HT. 2003. Plant–cell bioreactors with simultaneous electropermeabilization and electrophoresis. Journal of Biotechnology 100: 13–22. A. Timoneda, PhD thesis, 2022 154 Yang Y, Moore MJ, Brockington SF, Soltis DE, Wong GK-S, Carpenter EJ, Zhang Y, Chen L, Yan Z, Xie Y, et al. 2015. Dissecting molecular evolution in the highly diverse plant clade Caryophyllales using transcriptome sequencing. Molecular Biology and Evolution 32: 2001–2014. Zhang D, Lanier SM, Downing JA, Avent JL, Lum J, McHale JL. 2008. Betalain pigments for dye- sensitized solar cells. Journal of Photochemistry and Photobiology A: Chemistry 195: 72–80. Zhao S-Z, Sun H-Z, Gao Y, Sui N, Wang B-S. 2011. Growth regulator-induced betacyanin accumulation and dopa-4,5-dioxygenase (DODA) gene expression in euhalophyte Suaeda salsa calli. In Vitro Cellular & Developmental Biology - Plant 47: 391–398. Zhou Y, Karl T, Lewis DH, McGhie TK, Arathoon S, Davies KM, Ryan KG, Gould KS, Schwinn KE. 2021. Production of betacyanins in transgenic Nicotiana tabacum increases tolerance to salinity. Frontiers in Plant Science 12: 653147. Zielińska-Przyjemska M, Olejnik A, Dobrowolska-Zachwieja A, Grajek W. 2009. In vitro effects of beetroot juice and chips on oxidative metabolism and apoptosis in neutrophils from obese individuals. Phytotherapy Research 23: 49–55. Zimmermann IM, Heim MA, Weisshaar B, Uhrig JF. 2004. Comprehensive identification of Arabidopsis thaliana MYB transcription factors interacting with R/B-like BHLH proteins. Plant Journal 40: 22–34. Zou D, Brewer M, Garcia F, Feugang JM, Wang J, Zang R, Liu H, Zou C. 2005. Cactus pear: a natural product in cancer chemoprevention. Nutrition journal 4, 25. Bibliography 155 A. Timoneda, PhD thesis, 2022 156 Appendix 1. List of transcript sequences used in this thesis. Chapter 2. Betalain pigments as in vivo visual markers for arbuscular mycorrhizal colonisation of root systems AM specific promoter sequences used for vector construction. All sequences were selected immediately upstream the start codon of the gene. >MtPT4 promoter (836 bp) CTCGATCCACAACAAAGATTAATTTTTGTTCAAACAACTAATACTATCTGAAAATACACTTCCAAAAATATGTTAAATTTTTGTTTTT TTTAACATGGAGTGAAAAAAAGTTGCTCCCTTATTATTTAATTGGCACACATTTTTGTATGTTTAGTAAATGACACGCATTCTATTCG GTGGTAACGTTTGTCCTCACATACACCTCATTTGAATTGCATTTGTATCCTACTACCTCCCTTTCCAATTATATAAGATATTTTTGAA TTTTTTTTGTTCCTTTTTATAAGACCTCCTTGTTATTTCCAATTAAATTAATTATTTCTTTACATATATGTCCCTATTTATTATACAT TTTTTTCTTCAATCAGCAATAAATAGCATGTTGAAAATATAAACTATCTCTCTTTCTAAAGGATAAAATTGTAAAATCAATAGTGATT ACAAATAACTTTAATACAAATACTCACTTTCTTAATTCCTGTGATTTTGGCAAAAAGGTCTTGTATATAGGGACGGAGGTAGTATTAT GTTAGCAAGCAGCTAAAAAACAAAACAATAATCCTCTTTTAACAGCTAACTGTAACCAAAACAAAAGCACTTAGCATTTTTGAACATT CCCTATAAGAATATCCTAATTAATTGATCAACACTAGGTGAATATCCATAACATTTAGCTAGCAATCTCAATTTTTCTTGTTCTCATC TCCTAAGCTAACTTAGGACTAAACGTCACATTGTTACATTATCAAACACATATCAATGCATAGCCATAATAATATATATAAGAGCATA CACATAAACACCTAAACCAAATTAACTCCAAAAACCAACTTGAG >MtBCP1 promoter (1108 bp) AATGCGTTGCTTAAGAGTTGGTTAATCTATTTGAAATGTTGCGATTATAGTAGATCTTGTGCTCCCATCTTTAATATATCTTTTACAC TTACTCCTTAGTTATACAAACATTAAACATCATAACTTCTAAGGTAATTGATACGTGCCAAATTTTCTAAACTCCATTAAATTTGGTT TGAAAGAAATTTCATGAAATTCCTCTAGTTCGTTATAAATATTAGATAATTAGGTTTTGAGTAGCATTGTAGCTACCATGACCTATTT CCATTACATTTGTAGGTTTAAAAGATCAAAGATGACTTGGGGTATCAATTAACGAGTGAGCATATATGCGAGCAAGGAAAGAGATAAG TCAGAGTGTTTAAGGAAAGCAAGGCAAGAAAATTCACTTAGGGAGCCAAAACAAGGTCAATACTCGATTACATCTCTAATTTAAGCTT GAGAACTTGATATATTCGAGACTCCTCAAGTATTTGTACTCCAAGTGTCAATACTCGATTACATCTCTAATTTTCCAAATAAAGAAAC TCGTAAGAAACAAACATTATTTATGTGTAAACTTTCTGCATGGAGACAACTCAATTATAACGAAGCCCGGAAGAGGAACTATATATAC TGTTTCTATTTCCCTTTGACACGGTCCACACTATATTGCATACTTGTTTACTGCCAAATAAAAACAAGTATGTCCACGTGGTTAAATA AAAACAAAAGAAAACTTTCTTTTTGTTCAAGTGATAGAAAGTTTCTTTACATTAATATAAAATAATATAAGTGTATTCCAAAAAGGTA AAAAAAAAAATAATAAGTTGTGGAAATTCAACCTCGTCTTAGTAAAAAGCCGACCAAATCCGGTAGCCGTTTATAATAGCAAACAAAT ATTCTATTATTGTTGTGACGATAAAATCAATACACCCACACATGACAACTATATATAGCCTAAGTTGATATGCACAATTCATCAAATC AAATCAAAACAATTACATTTGCTAACTTGCTATTCCTTTGATTGAGATTTTATTCTAGGAAATAAAATTTTGACACTCATTTTTTTAA AGTTAATTTCTACATTTAGAGAGAAATATTTTCCAAAATCAAATTTTGATCA >NbPT5b promoter (1068 bp) GTGAGAAATTTACCAAATACCTATAAATAATGTATCAGCCGTATATAAATAACGTATAAGAGATGTTTATACACTATTATATATTACT TATATAATGTTATACATTATTATACACTATTATACAAATGAATCATATCAATGGAGCTTCACTGATTCTTCTTCTTCTTTGGGCATCT TCTGAAATTCAACTCAAATCTACCTCAAATCTGCTCCAAATTATTTTAAATTTGAGTTTTGAACTTCTCTTGACGATCCCAATCAATT GGGCCAACACCTAATCCAAACAACTAACAAAATCGAAAAATCTAATTTTAAAATTGAAGCTTCAAATGACCTTTAATGGTGACTCCAA AAATAATAATTCTCTTACGTGGTATTTTTTCTTCTCATCTTAATAGTGATTATCAATTTTGAATATGGAAGGAGAATTAAAAAATATA TATAATGGTAGGGAAACTTTGGGGTGCAAAGTCGTGAAAGAAGAGGAAGAAGAAGAACGAAGAAAAGAAAAAAATAGGGTGGGCTAAC TGGTGAAATTTTTTTCCAAATTATATATAACCTAGGCTATATCATGTAAGAACTTCAAATGTGGGCCATTTTTTGATGGATTGTTGGA CCAAATGACAAGGAGCGTAATTCCCCCTTCATTTTATATGGCACACTAGGATATTATGAGTTTAATAAATAAAGAATTTTTTTTATGA TTTTAAACATGTCGTGACTATAAAATTGTGTTATTAAAGTAAACTAAGAAGTTTGAAGTTAGATTATATCAATTATATAGATTTGTTT TTTTTTTGACACCAACTAACAAGAAAAAGTTGTCATATAAAGCATAAGCATGCGCGCTTTTTGTCCTAGGTTATTTTTGATTGAAGCA TATGCTTGATTTAAGCAAAGAAGAAGGATATCCTGTGGATACTTATCCTCAATTTCTTGTTGTAGCCACTGAGCGCTTAGGATTAAAC GTCATTTAATTCACATTGTGCGCTGAGTCTATATATGGGAGTTACATACACCAAAACAGTACCATGAAAACACAACCAAAATTTGCAG TAAAAAGCAAAC >NbBCP1b promoter (1231 bp) ACTGGTATTAGAATAATAGCCTGTTTAGCCAAAAATTGAAGTGTTTGGTTTTGGAAGGAAAAAAAAGTATTTTGAGGAGAGGCAGAAG CAGAAAAAAGTAGCTTCTCTCCAAAAGTACTTTTTTGAGAAGCATTTTTGAGAAAAATACACTTAGAAGCAGTTTTTTAAAGCTTGGC CAATCACTAATTGCTGCTCAGAAGTGCTTTTCAAACTAATTAGCCAAACACAAACTACTTCTCATCAAAAGTACTTTCGAAAAAAACA CTTATTAAAGTAAGCTAATTTTTGCAGCTTGGCTAAATGGATTATAAATCACGTATCTGTAAAACTACAAATTATATTTTATTGTTCC TCACATTTTTCGAGGTAAACATAAATATAAGTAAAATATAATGGTAATGTACTCCACTTGCTGAATAACTTAGTGTTTCATAGCAAGC ATATGCTCCACAAATTAAGAGTATGGAGCCATAATATTTCCAATATGAGTGACCCTATAGTTACAATAACAACAACAATCCAGTATAA TCTCACATAGTGGGGTCTGGTGAGGGTAGTATGTACGTAGACCTTATCTCTATCCTGGGGTAGAGAGGCTGTTTCCAAATAGACTCCC GACATCATTCTCTCCAAGAACTTCTCACCTTGCTCTTGGGGAAACTCGAACTCACAACCTCTATAGTTAAAACATGATAAATTATGGT CATATTTGTATTCTCTTTTTTTGCTTTAATTTTGGGTATGATAAAATTCCTTATAAATGCTAACAACCTACTCCGTCCTGTTCCAAAT AGTAATCGTTTTAGCATGCTGAAATTGTTTCAAAATATTTGGCATTTTAGAAAAATAAAATATATTTGTTATTTTTTTATTAACCTTA CTACTCCAATTAGTCGATTGTTAATAAGATGAGACATTTAAGAGAAAAAAACAACGATAAGTTAGACAAATAATCTTATAATTAAATA TGTCAAAAATCATAAAAGACAAATAATATGGAATAGAGTAAATATCTACCTACAATACTAAGGTTTTCTTCTCCTTTGAACTTTACGA GAATAATTAGTGGAATATTTGCCAGCCTAACTCATTGGAATTCACGTACTAGAATTCAAATTAATGAATCACACTTCGTTATCCTCGG ATCTCGCTAAGCTCTCTTGTTAAGTTAAGGTATAAATAGCCATACCAATTTTTCTTTTTTCAACTAACCCTTAAAGAAGAATTTTCC Appendices 157 Chapter 3. Search for ADH variants with putative relaxed feedback inhibition outside of the order Caryophyllales ADH sequences used for vector construction. S. lycopersicum sequences were amplified from cDNA libraries. M. esculenta sequences were codon optimized and synthesized by Genewiz (Bishop’s Stortford, UK). >SlADH1 (1134 bp) ATGTTGTCTTTCACCCCACTTCAATCCAAGCCAACCCCAACTTCCAATTCTAATCGTTTCTCTAATCTCACCAACCCCACTAGCAGTA CCAGCCGCCGCCGGCATTTCTCTGTCTCTTCTCCGTCACAAGTCAGTCACCACCATGGACGCCGTCGTCTGAGTATCAAAGCTATTGA TGCTGCACAGCCTTATGATTATGAAGCATTGGTGTCAAATCAATATGCTCAGTCAGGTAGGCTAAAAATTGCTATAGTGGGGTTTGGC AATTTTGGTCAGTTTCTTGCAAAATCCTTTGTTAGTAAAGGTCATTTTGTGTTGGCTCATTCAAGAACAGATTATTCCCAAATTGCTA TTTCTTTAGGGGTTTCCTTTTTTCAAGATCCACATGATCTTTGTGAGCAACACCCTGATGTTATTGTACTTTGTACTTCGATTATTTC AACTGAGACTGTTCTTAGGTCATTACCAATTCAAAGGCTGAAGAGGAACACATTGTTTGTTGATGTTTTATCAGTCAAAGAATTTCCA AAAAATATTTTTCTTCAAGTTTTGCCAACCCATTTTGATATTCTTTGTACTCATCCCATGTTTGGACCTGAAAGTGGTAAGGATAGTT GGAAAGATTTGATCTTTATGTTTGACAAAGTTCGAATTGGTGAAGGGAGATCAAGAACAGCAAGGGTTGACAAGTTTCTTGATATATT TGAGAAAGAAGGGTGTAGGATGGTGCCAATGACATGTGCTGAGCATGATAAGCATGCTGCAGGTTCACAGTTCATTACACATACAATG GGGAGGGTGTTGGAGAAGTTGGGTTTGGAGTCAACTCCTATTAATACTAAAGGGTATGAGACTTTGTTGAATCTGGTGGATAATACTG CTAGTGATAGCTTTGACTTGTACTATGGGCTGTTTATGTACAATAAAAATGCTATGGAGGAGTTGGAAAGACTTGACTTGGCTTTTGA GGCTTTGAAGAAGGAGTTATTCGGGCATTTACATGATTTGTTGAGGAAGCAATTGTTTGGGAAGGCGGAGGAAGCAGGACAGAGGCGT GTCTTATCCAAGTTGCCCAGGAATGGCTATGCGCTGCCAGCCCCTTCATCGGATGCTGTTAAACCTGAGAACAACTGA >SlADH2 (1173 bp) ATGTTTTCCCTTTCATCTATACAATCTAACAATATTCAATCTCAATCATCTTCGTCGCTACTCTTCAATCATCATCACCAGCATTCAA CTATTTCAACTCGGTTTCATCACCACCGCCTACTCTTCCCTCTCCGTGCCCAAAATAGCGACTTAACTACAGCCACCACCAATAACAA CTATGTCGATCTTGATGACAATCTAACCAGACTTGATAAATTTTCAAAATCATTAAGTATTTCGAATATCGAAGAAAATACATCATTA AATCCCCTCTTATGTTCCAATAACAAGCTCAAAATAGCTATCATAGGCTTTGGAAACTTTGGACAATTTATTGCCAAATCCTTTATCA AACAAGGCCATGTTGTATTAGCTCATTCACGTAGTGATTATTCCCTCATAGCACAATCCCTTAATGTCCACTTCTTTCAAGATCCTAA TGACTTATGTGAACAACATCCTGACGTTATTTTACTTTGCACATCCATCAATTCACTCGAAAACGTCATTCGTTCCCTTCCCATCCAA AAGCTTAAACGTAACACACTTTTCGTAGACGTATTATCAGTCAAAGAATTCCCGAAAAACATTTTTCTTCAATCACTACCAAAAGAAT TTGATATTTTGTGTACTCATCCTATGTTTGGTCCAACAAGTGGTAAAGACAATTGGAAAGGACTACCATTTATGTATGACAAAGTTAG AATTGGACAAGAAGAGTCAAGAATTAAAAGAGTCAACAATTTTATCAACATTTTTGTAAAAGAAGGTTGTAGAATGGTTGAAATGAGT TGTAGTGAACATGACAAGTATGCTGCTGGATCACAATTTATTACACATACTATTGGAAGAATGTTACAAAGACTTGGGACACAAACAA CTCCTATAAACACAAAAGGATATGAAAGTTTGTTGAATTTGATGGAGAATACAACTAGTGATAGTTTTGATTTGTATTGTGGTTTGTT TATGTATAACAATAATTCAATGGAGGTGTTAGAGAAACTAGATGCAGCATTGGATAGTTTGAAAAGGGAATTATTTGGACAAGTTCTT CAAAAGTTGGAGAAAAGAGTGGAAAAGGGAAGTAGGTTAGCTTTACCTACTCCTGATTTTAGTAAGAAAATTGAAAAATTGAAAGTTG AGAGGAAGGAACTAGAGGCACTTTCTTGA >MeADH1 (1116 bp) ATGTTACCCTTTAGTTCTACCAAACCATCTACATCTTTCTTTCACTTTCCACCTTTTCTCCCTTCCTTCTTTCACAGCGTATCTTTAC CACTGCTGCTTAACCCACCTAAGTCTAGCCACCTTAGATTTCTACCGCTACCGCTTCGGATTCGTTCTATTGATGCAGCGCAACCCTA TGATTATGAGTCACAACTATTATCCCAACATTTAAAATCTCAGAGTCTGAAAATTGCAATCATAGGTTTCGGTAACTTCGCCCAATTT CTCGCAAAGACACTTTCCAGACAGGGTCACACTCTACTCGCTTATTCTAGGACAAATCATGGTGGAGCCGCAAAGGAATTAGGAGTTA CCTTTTATACTAATCCCCACGATTTGTGTGAATCTCACCCAGAGGTGTTAATCCTTTGTACAAGCATTCTATCCAGCGAGAAAGTTCT TAAAAGTTTTCCCTTCCAACGACTAAAGAGATCTACTCTATTCGTAGACGTTCTCAGTGTAAAGGAGTTCGCTAAGAACATACTGCTT AAATACTTACCTCTCGAATTCGACATCTTGTGTACGCACCCGATGTTCGGACCAGAGTCTGGTAAGAACTCATGGGTGGGCCTTCCAT TCGTATACGACAAGGTCCGGATCGGAAACGAGGAAGAGAGAATTAACCGCTGCGATAAGTTCTTGGACATCTTCGCAAAAGAGGGGTG CAGAATGGTGGAGATGTCATGCGCAGAACACGACCGATATGCAGCTGGTAGTCAATTCGTAACACATACTATGGGCCGCGTCCTCGAA AAGTTCGGACTTGAAAGTTCTCCGATCAACACTAAGGGGTACGAAACATTACTCGACCTCGTTGAAAACACGGCAGGAGACTCTTTCG AACTCTACTACGGCCTTTTCATGTACAACCAAAACGCAATGGAACAACTTGAAAGGCTAGACATGGCTTTCGAAGCGATAAAGAAAGA ACTCTTCGGCAAATTACACCAGGTCTACCGTCGCCAACTATTCGGTACCGTGGAAGGCTCCGAAGAACGCCCAAAGATCCAAAAGCTC TTGCACAACGGAGCCCCACTCGAGCCAAGTAAGGACACGATGGAGCAGGAGAGGTCATAG >MeADH2 (1122 bp) ATGATTTCACTCTCTTCTTTTCATCCTTCCTCCACCACCGCCACCGCCACCGCCGCCGCCGCCACCACCCACCCACCTCAACAATGTC CCGCTTTTTCCTCTCCTCCGTCGCATCTCTCGCTTCCTTTACGCCACCCTCGCCAACACCTTGTAGTTCGGTGCGGTGGAGGTGGTTC GGCCTCCGAATCGGTATTTAACCGTGATGCACAGGACTTCAACTTTAAGAGTTTAAAGATTGCCATTATTGGATTTGGTAATTTCGGA CAATTCTTGGCGAAAACTATCAGTAGGCAGGGTCACACATTGCTGGCCTATAGTAGAACGAACTATTCAGACGCGGCAAAGAACCTGG GCGTGACCTTTTATTCAAATCCTCACGACCTATGTGAGACACACCCAGAGGTACTTATTCTTTGTACGAGCATAATAAGCTCCGAGAA CATGTTGAAATTCTTCCCTTTCCATCGGCTTAAGAGAAACATTCTATTCGTCGACGTCTTATCAGTGAAGGAGTTCGCAAAGAACATC CTTTTAAAGCACCTTCCAGCAGAATTCGACATTTTGTGCACACACCCTATGTTCGGTCCAGACTCTGGAAAGTACTCTCTGTTGGGGC TACCTTTCGTATACGACAAGGTGCGAATCGGTAAGCAAGAGAATAGAATTCATCGTTGCGAAAACTTCCTCGACATATTCGCAAAGGA AGGCTGTCGGATGGTAGAGATGAGTTGTGCGGACCACGATAGATATACTGCTGAAAGCCTTTTCGTTACCCATACGATGGGAAGAGTT CTTGAAAAGTTCGGCCTGGAAAGTAGTCCCATCAATACAAAGGGGTATGAAACCTTACTAAACTTGGTTGATAACACAGCGCGTGACT CATTCGAATTGTACTACGGCCTCTTCATGTACAACAAGAACGCCATGGAACAATTAGAAAGGCTTGATATGGCCTTCGAGGCGGTAAA GAAGGAACTATTCTGGAAACTTCACCAGGACTTCCGTCGACAACCTTTCGCCACTGCTGAAGGGTTGGAAGAGAGGCCCAAGAAACAA TTGCTCGAGAACGGCGCGCCAGTTGAGCCACCCTTAATGGAGAAGAAGGAGTTGGAGCACCAGTAA A. Timoneda, PhD thesis, 2022 158 Chapter 4. Bioproduction of betalains in Escherichia coli bioreactors and characterization of a highly active DODA enzyme in the Cactaceae. DODAα1 sequences used for vector construction from Beta vulgaris, Carnegiea gigantea, Mesembryanthemum crystallinum, and Stegnosperma halimifolium. All sequences were codon optimized for E. coli, domesticated for NdeI and NotI and synthesized by Twist Biosciences (San Francisco, CA, USA). >BvDODAα1 (828 bp) ATGAAGATGATGAACGGCGAGGATGCGAACGACCAGATGATTAAGGAATCTTTCTTTATTACCCATGGTAATCCTATCCTGACCGTCG AAGATACCCACCCTCTCCGCCCATTCTTCGAAACCTGGCGCGAAAAGATTTTCAGCAAGAAACCGAAAGCCATTCTGATCATCAGCGG ACACTGGGAAACCGTCAAGCCGACCGTGAACGCGGTTCACATTAACGACACGATTCACGACTTCGACGATTATCCAGCCGCCATGTAT CAATTTAAATACCCGGCCCCAGGAGCCCCGGAGCTGGCTCGCAAGGTGGAAGAAATTTTAAAGAAGAGCGGGTTTGAGACAGCCGAGA CAGATGAGAAGCGCGGCTTAGACCACGGGGCGTGGGTGCCGCTGATGTTGATGTACCCCGAAGCGGACATCCCTGTGTGCCAACTGAG CGTGCAACCCCACCTCGACGGTACTTATCACTATAATCTGGGTCGCGCGCTGGCGCCGTTAAAGAATGACGGAGTGCTGATTATCGGC AGCGGTTCCGCGACGCATCCGTTAGACGAGACACCGCACTACTTCGATGGCGTCGCCCCGTGGGCCGCCGCGTTTGACTCTTGGCTTC GTAAGGCACTTATCAATGGCCGTTTCGAGGAAGTCAACATTTACGAGACAAAGGCCCCGAACTGGAAGCTGGCCCACCCATTTCCCGA GCACTTCTACCCGCTGCACGTGGTCTTGGGTGCAGCGGGCGAGAAGTGGAAAGCGGAACTCATCCACTCTAGCTGGGATCACGGGACA CTGTGCCATGGTTCATATAAATTTACCAGCGCATAA >CgDODAα1 (816 bp) ATGGGAGTCGGTAACGAGGTCAGCTTTAAAGAAACCTTTTACGTTTCACACGGAAACCCTGCAATGCTGGCAGACGAATCATTTATCG CCCGTAATTTTCTGTTGGGATGGAAGAAGAACGTGTTTCCGATTAAGCCGAAAAGCATTTTAGTCGTGTCCGCACATTGGGAAACCGA CGTACCATGCGTTTCGGCAGGGGAGCACCCGGACGTAATCTATGACTTTTCGGACGTGCCAGATTGTATGTTTCAAATGAAGTATCCG GCGCCGGGTTCTTCGAAGCTTGCGAAGCGTGTACAAGAACTGCTCATTGCGGGTGGCTTTAAAACCGCATCGTTGGATGAAAATCGCG GATTTGATCATTCGAGCTGGGTACCCTTGTCGCTGATGTATCCGGAAGCGGATATTCCCGTCTGTCAACTGTCCGTGCAACCACATTT GTCTGCATCCCATCATTTTGATGTGGGACGCGCCCTGGCCCCACTTAAAGAAGAGGGTGTGTTGTTTATCGGCAGCGGCGGCGCGGTT CATCCGTCCGACGATACGCCTCATTGGTTCGACGGTGTAGCGCCGTGGGCGGCGGAATTCGATCAGTGGTTAGAGGATGCGCTGATCA ACGGTCGTTATGACGACGTAAACAACTACCAGACTAAGGCCCCGTCCGGCTGGAAGATTGCGCACCCCATCCCTGAGCATTTCCTGCC GTTACACGTGGCTATGGGCGCTGCGGGCGAGAAGAGCAAAGCCGAATTAATCTACCGCACCTGGGACCACGGGACCTTGGGTTACGCG TCTTATAAATTTACATCAATTTAA >McDODAα1 (831 bp) ATGGTGGCCGTCGCGACTGGTGAAAAGATTCGTGAAACGTATTTCCTGACTCACGGCGACCCGATCATGTACGTAGACAAAACAATTA AGCTTCGTCACTTTCTTGAAGGCTGGAAGAAGAACGTCTATATGAAGAAAGCACCGAAGAGCATTTTAATGATTAGCGCCCATTGGGG CACCGACAACCCGACCGTGACAACAGTAGACCAGCCCGATACGATTCACGACTACGACAATTATCCCGACCCGTTATATCAAATTAAA TATCCGGTATTCGGTGCGCCGGAGCTTGCGAAGCGTGTTCAGGAACTGCTCGGGAGCAGTGGCCTGAAGTGCGAAGAGGATAATAAAC GCGGCTTCGACCACGGGGTTTGGTTTCCCCTGCAATTTATGTATCCCGAGGCTGATATTCCTGTGTGTCAGCTGTCAGTGCAACCCTC TATGGATGGCGCCCATCACTTCAACATGGGTAAGGCATTAGCACCGCTGATGGATGAGGGTATTTTGATTATTGGTTCGGGCGGTGCC GTACACCCGAGCGACGATACACCCCACTGCCCAAATGCCGTCGCTCCGTGGGCTGCCGAATTTGACGATTGGCTGTGCGATGCGGTGA TTAAGGGGCGTTACGAGGACGTAAATAACTACAATAAACTGGCTCCGAATTGGGAAATTGCGCACCCTGGCCCGGAGCACCTGTATCC GCTGCACGTGGCGTTAGGCGCCGCGGGAGAAAAGAGTATCGCGGAGACAATCCACCATTCATGGGCACGTAACGGCGTATTCGGGTAC GCGAGCTTTAAATTCACTAGTACGTCCTCTACGTTGTGA >ShDODAα1 (792 bp) ATGAAAATGACCGAAACCTATTTTATTTCTCACGGAAGCCCTGAGATGCCTAGCAGCGACTTTCTCGAGGGCTGGGAAGAAAAGCTTT GTCTGAAGCGTCCTAAATCGATCTTAGTTATTGGCGCACACTGGGTGACCGACGAACCGACGGTGAATAGCCTGTTAGACTCGTCCCT GGATATTACGATTGACCCGAGTACAAATACGAAACAAGTGAAATATCCCGCGAAGGGCGCCCCCGAGCTTGCCAAGCGCGTGCAGGAA CTGTTAACTAAGAGCGGCCTCGTTAAACAGGTGAACGTAGACGAAAAGCGTGGCTTGGATCAAAGCGCCTGGGATCCCCTGTTCTTTA TGTACCCGATGGCCAACGTGCCAGTGTGCCAACTGTCCGTGCAGGCGCATCTGGACGGCGCTTACCATTACAATATCGGTCGCGCCCT GAGCCCCCTGTTGGAAGAGGGCGTTCTGATCATCGGAAGCGGGAGCGCGACCCACAACATGGACGCTCTGACCACCACTACAAGCGGC CACGGACAACTCCATAGTTGGGCGAAGGATTTCGATACATGGCTCGAGGAAGCGCTGACTTCCGGGCGCTTCGAGGACGTTAACAATT ACGAGAAGAAGGCTCCTAACGCGAAGATGGCCCACCCGACGCCGGAACATTTCTATCCCCTGCACGTGGCAATCGGTGCAGCCGGCGA GCATGCCAAAGCGGAATTATTCCATCGTAACTGGTCAAAGGGTATCTTTAGTAACGCGAGCTATAAATTTACCATTCCGACCAACTGA Appendices 159 Chapter 5. Identification of residues responsible for high DODA activity in CgDODAα1. Extant and reconstructed ancestral DODA sequences from the Globular Inclusion clade used for expression in yeast were codon optimized for yeast, domesticated for BsmBI, BsaI and NotI and synthesized by Twist Biosciences (San Francisco, CA, USA). >CgDODA_α1_Sc (816 bp) ATGGGAGTGGGAAACGAAGTATCTTTTAAAGAGACTTTCTATGTGTCCCACGGGAATCCTGCAATGTTAGCAGACGAGTCTTTTATCG CTAGGAATTTCCTACTGGGGTGGAAGAAGAACGTATTCCCAATCAAGCCTAAGTCCATTCTTGTTGTATCGGCCCACTGGGAAACTGA CGTTCCGTGCGTTTCCGCCGGAGAGCACCCAGATGTAATATATGACTTCAGTGACGTACCTGACTGTATGTTTCAGATGAAGTATCCA GCTCCCGGCTCCTCTAAGCTGGCAAAGAGAGTCCAGGAATTGCTAATAGCCGGTGGATTTAAAACTGCTTCTTTGGACGAGAATCGTG GCTTTGACCACTCCTCATGGGTGCCACTTTCGTTAATGTATCCCGAGGCAGACATCCCCGTATGCCAGTTATCGGTTCAGCCACACTT ATCAGCCAGCCACCACTTCGACGTCGGTAGGGCCCTAGCCCCTTTAAAAGAAGAGGGCGTTCTATTTATAGGTTCAGGTGGAGCAGTT CATCCGAGTGACGACACACCACATTGGTTTGACGGAGTAGCACCTTGGGCAGCCGAATTCGACCAGTGGTTAGAAGATGCTTTGATTA ATGGAAGATACGACGACGTAAACAACTATCAGACAAAAGCGCCGTCAGGTTGGAAAATTGCACATCCGATTCCCGAGCACTTTCTACC TTTGCATGTCGCCATGGGAGCAGCGGGAGAGAAAAGTAAAGCTGAGCTGATTTACAGAACATGGGACCACGGCACGCTAGGTTACGCC TCATATAAATTCACGTCTATTTAA >CgDODA_α2_Sc (804 bp) ATGGGCGAACAGGAAGCGATCAGAGAAACATTCTATATCTCGCACGGAACTCCTAAGATGTCGATTGATGACTCTATTCCAGCGCGTA AATTCTTTCAGGAGTGGAAGGAGAAGGTATACTCGAAGCGTCCTAAAAGCATGTTGGTTATTTCTGCCCATTGGGAAACAGACGTTCC GGCTGTCAACGCTGTTAACCATTCGGATCTTATTTACGATTTCAGAGGATTCCCCGCCATCATGTACCAGCTTAAATATCCTGCTCCG GGTGCCCCAGACTTAGCTAGGCGTGTGGAAGAGTTGCTTACTGCATCGGGCTTTTCTTGCGTGGTCGACAAGAAGAGAGGTCTAGATC ACGGATCATGGGTCCCACTGATGTTAATGTACCCAGAGGCCGATATTCCGGTGTGTCAATTGTCAGTCCAGTCACATCTTGACGGAAC TCACCACTATAACTTGGGGAAAGCATTGGCACCTTTGAAAGAGGAAGGGGTCCTGATAATAGGATCAGGCTCTGCTGTGCACCCGTCA AACTCTACTCCTATTTCACCGGACGGCGTCGCTCCTTGGGCAGCAGCATTTGACAGTTGGCTTGAGGAAGCTTTGAAATCGGGCCGTT ATGAGGATGTTAACAACTATCAGGCTAAAGCACCTAACTGGAAGTTAGCCCATCCATGGCCTGAGCATTTCTACCCGTTGCACGTTGC AATGGGCGCGGCTGGCGAGAATAGTAAGGCCGAGTTAATACATAGGAGCTGGGATCACGGCACCTTAGGATATGCTTCCTATAAGTTC ACATCCTGTTGA >McDODA_α1_Sc (831 bp) ATGGTGGCTGTCGCAACTGGTGAGAAAATCCGTGAAACTTATTTTCTTACTCACGGAGATCCTATAATGTACGTGGACAAAACTATTA AGTTAAGACACTTCCTAGAAGGCTGGAAGAAGAACGTCTATATGAAGAAAGCGCCAAAGTCCATTTTGATGATATCCGCGCATTGGGG CACGGACAACCCTACGGTTACTACGGTAGACCAGCCGGATACGATTCACGACTACGACAATTATCCAGACCCTTTATATCAAATAAAA TATCCGGTATTCGGGGCCCCTGAGTTGGCTAAGCGTGTGCAGGAACTGCTTGGTTCTAGTGGCTTGAAGTGCGAAGAGGATAATAAAA GAGGCTTCGACCACGGGGTTTGGTTTCCCTTGCAATTTATGTATCCCGAGGCAGATATTCCAGTATGTCAGTTATCGGTACAACCATC AATGGATGGGGCACATCACTTCAACATGGGTAAGGCTTTAGCTCCATTGATGGATGAGGGTATATTGATTATAGGCAGTGGTGGGGCG GTACACCCATCGGACGATACACCACACTGCCCAAATGCTGTCGCCCCATGGGCTGCCGAATTTGACGATTGGTTGTGCGATGCAGTTA TTAAGGGTCGTTACGAGGACGTCAATAACTACAATAAATTGGCTCCCAATTGGGAAATTGCCCACCCCGGCCCCGAGCACCTTTATCC TTTGCACGTTGCTCTTGGAGCGGCCGGCGAAAAGAGTATAGCCGAGACAATCCACCATTCATGGGCAAGGAACGGAGTTTTCGGTTAC GCATCTTTTAAATTCACATCGACCTCTTCGACTTTGTGA >CgDODA_α1/a2_anc_MAP (804 bp) ATGGGTGGCGAGGAGATGATTAGGGAGACTTTCTACATTAGCCATGGCACTCCAATTATGAGCATCGACGAATCAATCCCCGCCCGTC ACTTCCTACAGGGCTGGAAAGAGAAAGTCTACTCTAAGAAGCCAAAATCAATATTGGTCATATCTGCGCACTGGGAGACTGACGTACC AACGGTCAACGCTGTAGATCACTCCGATACTATATACGATTTTAGAGGGTTCCCTGCTCCAATGTACCAGTTGAAATATCCAGCGCCA GGAGCACCTGATTTAGCCAAAAGAGTTCAGGAGCTGCTTACCGCCTCAGGTTTCAAGTGCGCTGTGGACAAGAAACGTGGACTTGACC ATGGCTCATGGGTGCCTCTGATGTTCATGTACCCTGAGGCTGACATACCGGTTTGCCAGCTGAGCGTTCAGAGCCACTTGGATGGGAC ACACCATTACAATATGGGAAAGGCTCTTGCCCCTTTGAAGGAAGAGGGAGTCTTGATAATCGGTTCTGGCAGCGCCGTTCATCCTTCA AACGATACCCCACACTCATTCGACGGGGTTGCCCCATGGGCAGCTGAGTTTGACGACTGGTTAGAGGAAGCGTTAACTAGTGGCCGTT ATGAAGATGTTAACAATTATCAAACTAAAGCGCCTAACTGGAAGATCGCTCACCCCTGGCCCGAGCACTTCTATCCATTACATGTTGC TATGGGCGCTGCTGGTGAAAACTCCAAAGCAGAGCTTATTCATAGGTCGTGGGATCATGGAACGTTGGGATATGCTTCCTACAAATTC ACCTCTACGTAG >CgDODA_α1/a2_anc_AltAll (804 bp) ATGGGTGGCGAGGACATGATTCGTGAAACTTTCTATATATCACATGGGACACCAATCATGTCTATAGACGAGTCCATACCTGCTAGAC ATTTCTTACAGGGGTGGAAGGAAAAGGTCTACAGCAAGAAACCAAAATCCATATTAGTTATAAGCGCTCACTGGGAAACTGATGTACC GACAGTCAATGCAGTTGACCATAGCGATACCATTTATGATTTCAGAGGATTCCCAGCGCCGATGTACCAGCTAAAGTACCCAGCTCCC GGGGCCCCTGATCTGGCCAAAAGGGTCCAAGAGTTGCTGACTGCCAGTGGGTTCAAGTGCGTAGTCGACAAGAAGAGGGGCCTAGACC ATGGCTCCTGGGTTCCATTAATGTTAATGTACCCTGAGGCAGACATACCGGTATGCCAGTTGTCAGTCCAATCACACCTTGATGGGAC CCATCACTACAATTTGGGCAAGGCACTAGCACCGTTGAAGGAAGAGGGAGTTTTAATCATTGGCTCAGGATCTGCTGTACACCCGTCT AACGACACGCCCCACAGCCTTAATGGCGTTGCGCCGTGGGCTGCAGAGTTCGACAATTGGCTTGAGGAAGCGTTGACAAGTGGACGTT ACGAAGATGTGAATAATTATCAGACGAAAGCACCTAACTGGAAGATCGCCCACCCATGGCCTGAGCATTTCTATCCCCTACACGTCGC A. Timoneda, PhD thesis, 2022 160 TATGGGCGCTGCAGGCGAGAACTCGAAAGCAGAACTGATTCACAGATCATGGGACCACGGAACATTGGGCTACGCTTCGTACAAATTC ACCAGTTCATGA >CgDODA_a2_anc_MAP (804 bp) ATGGGTGGTGAAGAGATGATTCGTGAGACTTTCTATATATCGCATGGTACGCCTATGATGTCAATTGACGAAAGCATCCCAGCAAGAC ACTTCTTGCAAGGTTGGAAGGAAAAGGTGTATAGTAAGAAACCAAAGTCCATTTTAGTTATATCTGCTCATTGGGAGACTGATGTACC TACTGTCAATGCGGTTGACCACTCGGATTTGATCTACGATTTCAGAGGATTCCCAGCGCCTATGTATCAGCTGAAATACCCCGCCCCA GGCGCCCCAGATTTGGCTAAGAGAGTGCAGGAGCTTCTAACCGCATCTGGATTCAAGTGTGCCGTGGACAAGAAAAGAGGGCTTGACC ACGGTTCATGGGTCCCACTTATGTTCATGTACCCTGAAGCGGACATACCAGTGTGCCAGTTATCTGTGCAGTCCCACTTAGATGGCAC TCACCACTATAACTTAGGAAAGGCCTTGGCACCGCTTAAAGAGGAAGGTGTCTTAATTATCGGTTCAGGTTCAGCAGTACACCCAAGC AACGACACACCACATTCGGTCGATGGGGTTGCACCCTGGGCCGCCGAGTTCGACAACTGGTTGGAAGAGGCATTGACCTCGGGAAGGT ACGAGGACGTTAATAACTATCAGACGAAAGCGCCGAATTGGAAAATCGCTCACCCTTGGCCAGAACACTTCTACCCGCTTCACGTTGC TATGGGAGCAGCGGGCGAGAACTCCAAAGCAGAGCTTATTCACCGTTCCTGGGATCACGGGACATTAGGTTATGCGTCATACAAATTC ACCTCTACTTGA >CgDODA_a2_anc_AltAll (804 bp) ATGGGTGGCGAGGACATGATACGTGAGACATTTTATATTAGTCACGGCACTCCTATGATGAGTATTGACGAGTCCATCCCAGCTAGAC ACTTTCTACAGGGATGGAAGGAAAAGGTCTATTCCAAGAAGCCTAAATCGATCTTGGTAATTAGCGCCCATTGGGAGACAGACGTTCC TACCGTTAATGCTGTAGACCACTCCGACCTTATATATGACTTTCGTGGATTCCCGGCTCCAATGTACCAACTAAAATACCCGGCCCCG GGTGCACCAGATTTGGCCAGAAGGGTGCAGGAGCTATTGACCGCATCAGGTTTCAAGTGTGCAGTCGATAAGAAAAGGGGCTTAGACC ACGGCAGCTGGGTCCCATTGATGTTAATGTATCCCGAGGCTGATATACCGGTCTGTCAGTTGAGCGTTCAATCGCATCTAGACGGCAC CCACCACTACAATTTAGGAAAGGCCTTAGCACCCCTAAAAGAAGAGGGAGTCCTTATAATCGGATCCGGCTCAGCCGTACACCCGTCA AATAACACCCCTCATTCATTGAACGGAGTTGCACCCTGGGCGGCCGAATTCGACAACTGGTTGGAAGAAGCCTTAACCAGCGGGCGTT ACGAGGACGTCAACAACTACCAAACCAAGGCCCCAAACTGGAAGATTGCCCACCCATGGCCCGAACACTTTTACCCATTACACGTTGC TATGGGCGCCGCCGGTGAAAATTCCAAGGCAGAGCTGATTCATCGTTCATGGGACCACGGAACATTAGGTTATGCTTCATACAAGTTC ACTAGTTCATGA >CgDODA_a1_anc_MAP (804 bp) ATGGGTGGGGAAGAGATGATCAGGGAAACATTCTACATTAGTCACGGAAACCCAATCATGTCGATTGATGAGAGCATTCCAGCTCGTC ATTTCCTGGAAGGATGGAAAGAGAAAGTGTACAGCAAGAAACCGAAATCTATACTGGTGATTTCAGCTCACTGGGAAACCGACGTGCC AACTGTTAACGCAGTTGACCATCCGGATACTATCTACGACTTTTCCGATTTCCCGGCACCAATGTATCAATTAAAGTACCCAGCCCCT GGCGCTCCAGACTTAGCCAAGAGAGTGCAGGAATTGCTTACAGCCAGCGGCTTTAAATGCGAAGTGGACAAGAAGAGGGGCTTTGACC ACGGTTCATGGGTCCCACTTATGTTCATGTACCCAGAGGCTGACATACCAGTCTGTCAGTTGTCTGTTCAGCCCCACTTGGACGGCAC ACACCACTACAACATGGGAAAGGCACTTGCACCACTGAAAGAGGAAGGTGTACTTATTATAGGCTCTGGCGGTGCAGTCCACCCTTCC GACGACACACCCCACTGCTTCGACGGCGTTGCCCCTTGGGCAGCCGAGTTTGATGACTGGTTGGAAGAGGCCTTAATTTCTGGTCGTT ACGAGGACGTAAACAATTACCAGACCAAGGCACCTAACTGGAAGATAGCGCACCCAGGTCCCGAGCACTTCTACCCTTTACATGTAGC CATGGGAGCAGCAGGCGAAAACTCTAAAGCGGAGTTAATACACAGGTCCTGGGACCACGGTACACTAGGCTACGCTTCCTACAAATTC ACAAGCACCTAG >CgDODA_a1_anc_AltAll (804 bp) ATGGGCGGCGAGGACATGATTAGAGAGACATTCTACATCTCCCACGGGAATCCAATAATGCTTATCGACGAGTCGATTCCAGCTAGGC ATTTTCTGCAAGGCTGGAAAGAGAAAGTTTACAGCAAGAAGCCTAAGTCGATTTTAGTCATCTCCGCACACTGGGAAACGGACGTCCC CACGGTCAATGCCGTAGACCACCCTGATACAATACACGACTTTAGGGACTACCCTGCCCCAATGTACCAGTTGAAATATCCTGCTCCT GGAGCTCCAGAACTAGCAAAACGTGTACAAGAGCTATTGACCGCTTCAGGTTTCAAGTGCGAGGTTGACAAGAAGAGAGGCCTAGATC ACGGTTCTTGGGTCCCGCTAATGTTAATGTACCCCGAGGCTGATATACCAGTCTGTCAGTTATCCGTTCAGCCACATCTTGACGGCAC ACACCACTACAACTTAGGAAAAGCATTGGCACCTCTGAAAGAGGAAGGGGTCCTAATTATAGGTTCAGGTGGAGCTGTACACCCTAGT GACGACACACCGCACTCTTTAAATGGTGTCGCACCATGGGCCGCAGAGTTCGACGATTGGTTAGAGGACGCTCTAATATCTGGTAGGT ACGAGGATGTGAACAACTATCAGACTAAGGCTCCTAACTGGAAGATTGCACACCCATGGCCAGAGCACTTTTATCCTTTGCATGTGGC TATGGGTGCTGCTGGCGAAAACTCCAAGGCTGAGCTGATCCACCGTTCATGGGACCACGGAACACTGGGCTATGCTTCATACAAGTTC ACGTCAACATGA >PortDODA_a1_anc_MAP (804 bp) ATGGGAGTGGGGAAAGAAGTATTTAGGGAGACTTTCTACGTGTCGCACGGTAACCCGGCCATGTTGGCAGACGAGTCGTTTATAGCTA GGAACTTCCTGTTAGGGTGGAAAGAAAATGTGTTCCCAATTAAGCCCAAATCAATCTTGGTAGTATCTGCGCATTGGGAAACAGATGT ACCATCGGTATCTGCCGGAGAGCATCCTGATGTTATCTACGACTTCTCCGACGTTCCTGACTGCATGTTTCAAATGAAGTACCCAGCT CCTGGGAGTCCTGAGTTGGCCAAGAGAGTCCAGGAGCTATTAACTGCCTCTGGCTTCAAGACGGTCGACCTGGACAAATCTAGAGGCT TCGATCACTCATCGTGGGTTCCTCTAAGTCTTATGTACCCAGAAGCTGATATACCTGTTTGCCAACTAAGTGTGCAGCCACATCTATC TGGGACTCACCACTACAATGTTGGAAGGGCTCTTGCACCATTGAAAGAAGAGGGCGTTTTATTTATAGGGTCAGGCGGCGCCGTACAT CCTTCTGACGATACACCACACTGGTTTGACGGTGTAGCACCGTGGGCCGCTGAGTTCGACCAGTGGTTAGAGGATGCGCTGATATCTG GCAGGTACGAGGACGTTAATAACTACCAGACAAAGGCTCCTTCCGGATGGAAGATCGCACATCCTATACCGGAGCACTTTCTGCCCCT GCATGTGGCCATGGGAGCAGCTGGCGAGAAATCAAAGGCTGAACTAATCTATAGAACCTGGGATCACGGTACTCTAGGTTACGCCTCT TATAAGTTCACCTCGACGTAG Appendices 161 >PortDODA_a1_anc_AltAll (804 bp) ATGGGTGTGGGGAAGGAAGTATTCAGAGAAACATTTTATGTATCCCATGGTAACCCGGCAATGCTAGCCGATGAATCTTTCATTGCGC GTAACTTTCTGCTAGGATGGAAAGAAAATGTTTTCCCAATTAAGCCGAAATCAATTCTAGTAGTCTCGGCTCATTGGGAAACGGACGT TCCGTCTGTTTCCGCAGGCGAACATCCAGATGTCATCTATGATTTCTCTGACGTTCCCGATTGCATGTTCCAAATGAAATACCCAGCG CCCGGTAGCCCTGAATTAGCTAAGAGGGTTCAAGAATTGCTTACAGCCTCAGGCTTTAAAACCGTTGATTTAGATAAGTCCAGGGGTT TTGATCACAGCAGCTGGGTGCCATTATCTCTTATGTACCCCGAAGCTGATATACCAGTCTGTCAACTTAGTGTGCAACCTCACCTGTC GGCTACGCACCACTATAATGTCGGCAGGGCGCTAGCACCCCTGAAGGAAGAAGGCGTACTATTTATAGGTTCGGGTGGTGCTGTCCAC CCTTCAGATGACACCCCGCACTGGTTTGATGGTGTCGCACCTTGGGCAGCTGAATTTGATCAATGGTTGGAAGATGCCCTTATTTCTG GGAGATACGAAGATGTGAATAACTATCAAACAAAAGCACCTTCTGGTTGGAAGATTGCGCACCCAATCCCTGAACATTTCTTACCTTT GCACGTAGCAATGGGAGCTGCTGGTGAGAAATCGAAGGCCGAGTTGATTTACAGGACGTGGGATCATGGAACATTAGGGTATGCGTCG TATAAATTCACATCCACCTGA >RaphDODA_a1_anc_MAP (804 bp) ATGGGTGAAGAAGAGATGATTCGTGAAACCTTTTACCTAACCCACGGGGACCCCATAATGTACATTGACAAGACCATTAAGCTAAGAC ACTTTCTAGAGGGCTGGAAAGAGAAGGTTTACTCAAAGAAGCCCAAATCCATACTTGTTATTAGCGCTCACTGGGAGACTGACAACCC AACCGTAAACACCGTGGATCACCCAGACACTATTCATGACTTTGACGACTATCCTGCACCACTATACCAGATCAAATATCCCGCACCG GGAGCTCCTGAGTTGGCTAAACGTGTTCAGGAGTTACTTACGGGTTCTGGCTTCAAATGCGAGGTCGATAAGAAACGTGGCTTCGACC ATGGTGCTTGGTTCCCATTGATGTTCATGTACCCTGAAGCCGACATCCCTGTATGCCAGCTATCAGTCCAGCCTCACTTGGATGGCAC ACACCACTATAACATGGGCAAGGCTCTAGCTCCCTTAATGGAAGAGGGAGTTCTAATTATTGGTAGTGGTGGCGCAGTTCACCCAAGT GACGACACACCACATTGCCCGAATGGAGTGGCCCCTTGGGCGGCAGAATTCGACGACTGGTTAGAGGACGCACTTATCTCTGGAAGGT ATGAAGATGTTAATAATTACAAGACTAAGGCGCCCAATTGGGAGATCGCTCACCCGGGTCCTGAGCACCTTTATCCGTTGCACGTCGC CTTAGGAGCCGCTGGAGAAAATAGTAAGGCGGAGCTTATACACCACTCGTGGGCTGCAAACGGTGTTTTCGGATACGCCTCATATAAG TTTACCTCTACGTAA >RaphDODA_a1_anc_AltAll (804 bp) ATGGGCGAAGAGGAGATGATTCGTGAAACCTTCTATATTACTCACGGCGATCCAATTATGTACATCGATAAGACCATAAAATTGAGAC ACTTCTTAGAAGGGTGGAAGGAAAACGTTTATAGCAAGAAGCCAAAGAGTATATTGGTGATCAGCGCTCACTGGGATACCGATGTTCC GACAGTCAATGCTGTTGACCACCCAGACACCATACACGACTTCGACGACTACCCTGCTCCATTATACCAAATTAAATACCCTGCACCA GGTGCCCCAGAGTTAGCAAAGAGAGTTCAAGAGTTGCTTACTAGTTCAGGTTTTAAATGCGAAGTTGACAACAAGAGGGGATTAGATC ACGGCGCCTGGTTTCCTCTGATGTTTATGTATCCTGAGGCTGACATTCCAGTCTGCCAACTTTCAGTGCAACCTCACCTAGACGGAAC TCACCACTACAACATGGGAAAAGCATTAGCTCCACTACTGGAAGAAGGAGTTTTAATTATAGGAAGTGGTGGTGCAGTACACCCATCT GACGATACACCGCACTGCCCTAACGGCGTTGCCCCATGGGCAGCCGAGTTTGACGACTGGTTAGAGGACGCACTGATTTCAGGCAGAT ATGAGGACGTAAATAACTACAAGACGAAGGCTCCGAACTGGGAGATCGCACACCCAGGGCCAGAACACCTGTACCCGTTGCATGTCGC CCTAGGAGCTGCTGGAGAAAATTCCAAAGCCGAGCTGATACATCGTTCTTGGGCCGCAAACGGCGTTTTCGGGTATGCTTCTTATAAG TTCACTTCCACATAG >GiDODA_a1_anc_N1 (N17T) (804 bp) ATGGGCGGTGAGGAAATGATCCGTGAAACGTTTTACATTAGTCATGGAACGCCAATTATGAGTATTGATGAGTCTATCCCAGCAAGAC ATTTCTTAGAGGGGTGGAAGGAGAAGGTCTATAGTAAGAAACCGAAATCAATTCTGGTTATATCAGCCCACTGGGAAACCGACGTGCC AACTGTTAACGCTGTCGATCACCCTGACACCATCTACGACTTTTCCGACTTCCCTGCTCCAATGTATCAACTGAAATACCCTGCACCC GGAGCTCCCGACCTAGCAAAGAGGGTCCAGGAATTACTGACTGCTTCGGGTTTCAAGTGCGAAGTTGACAAGAAGCGTGGCTTTGACC ACGGGTCTTGGGTTCCCTTGATGTTTATGTATCCAGAGGCTGATATACCTGTCTGCCAGTTGTCAGTCCAACCTCACTTGGACGGCAC ACACCACTACAACATGGGAAAGGCACTAGCGCCATTAAAAGAGGAAGGTGTCTTAATAATTGGATCAGGCGGTGCCGTCCATCCTAGT GACGACACTCCACACTGCTTCGATGGCGTGGCACCTTGGGCCGCCGAATTCGATGACTGGCTGGAAGAAGCTCTAATATCTGGTCGTT ACGAGGACGTCAATAATTATCAAACTAAGGCGCCTAATTGGAAGATCGCCCATCCGGGGCCAGAGCACTTCTATCCACTACATGTAGC GATGGGAGCAGCTGGTGAGAACTCTAAGGCCGAGCTAATCCACAGGAGTTGGGACCACGGCACCTTGGGCTATGCGTCATACAAATTT ACTAGTACATAA >GiDODA_a1_anc_N2 (E33Q) (804 bp) ATGGGTGGCGAGGAAATGATTAGAGAGACTTTCTACATAAGCCATGGCAATCCTATTATGTCTATAGATGAGTCAATCCCGGCTAGGC ACTTCTTGCAAGGATGGAAAGAGAAAGTCTATTCCAAGAAGCCGAAGAGCATATTGGTGATTTCAGCGCATTGGGAAACCGACGTACC CACTGTTAATGCCGTGGACCACCCAGATACGATATACGATTTCTCTGACTTCCCGGCACCGATGTACCAGTTAAAGTATCCCGCTCCA GGAGCTCCTGACCTTGCGAAAAGGGTACAAGAGCTTTTAACAGCAAGTGGTTTCAAGTGCGAAGTTGATAAGAAAAGAGGATTCGACC ACGGCAGCTGGGTTCCCCTAATGTTCATGTATCCAGAGGCCGACATTCCAGTCTGCCAGTTGTCTGTCCAGCCCCACCTGGACGGAAC CCACCACTATAACATGGGAAAAGCATTGGCTCCATTAAAAGAAGAGGGTGTCCTAATTATAGGATCGGGCGGAGCAGTGCATCCCTCG GATGATACTCCTCATTGCTTTGATGGTGTAGCACCATGGGCTGCTGAGTTTGACGATTGGTTAGAGGAAGCTCTGATCTCCGGCAGAT ATGAGGACGTGAACAATTATCAGACGAAGGCCCCTAATTGGAAGATAGCACACCCGGGCCCTGAGCATTTCTACCCTCTTCACGTAGC AATGGGAGCAGCAGGAGAGAATTCGAAGGCAGAGTTGATCCACAGAAGTTGGGATCACGGTACTTTGGGATATGCTTCTTACAAATTC ACTTCTACTTAG A. Timoneda, PhD thesis, 2022 162 >GiDODA_a1_anc_N3 (P67S) (804 bp) ATGGGCGGCGAGGAGATGATTAGGGAAACTTTCTACATTAGTCACGGCAACCCCATTATGAGCATCGATGAATCCATCCCAGCGAGGC ACTTTCTTGAAGGTTGGAAGGAGAAAGTGTATAGCAAGAAGCCGAAATCGATACTGGTCATATCAGCTCATTGGGAGACTGATGTGCC TACAGTAAACGCGGTCGATCATTCAGACACCATATACGATTTCAGTGATTTCCCTGCACCCATGTACCAGTTGAAATATCCGGCTCCT GGCGCTCCTGACCTGGCTAAGCGTGTCCAGGAGCTGTTAACGGCTTCAGGATTTAAATGCGAGGTAGATAAGAAGCGTGGCTTCGATC ACGGTTCCTGGGTGCCTTTGATGTTTATGTACCCAGAGGCTGACATACCCGTTTGCCAGCTATCAGTGCAGCCACACTTGGATGGCAC TCATCACTACAATATGGGGAAGGCTTTGGCACCCTTAAAGGAAGAGGGCGTACTTATCATTGGTAGCGGTGGGGCCGTACACCCCTCA GACGACACCCCGCACTGTTTCGACGGTGTGGCACCCTGGGCCGCCGAATTCGACGACTGGTTAGAGGAAGCCTTAATCAGTGGCAGGT ACGAAGATGTTAACAACTACCAAACAAAAGCTCCGAACTGGAAGATTGCACATCCTGGCCCAGAGCACTTTTACCCTCTTCATGTGGC TATGGGCGCCGCTGGTGAGAATTCAAAGGCTGAGTTAATACACCGTTCTTGGGACCATGGTACACTTGGTTACGCGAGTTACAAATTC ACATCAACCTGA >GiDODA_a1_anc_N4 (S74R) (804 bp) ATGGGTGGGGAAGAGATGATCAGGGAAACATTCTATATATCGCACGGCAACCCCATTATGAGTATTGATGAGTCCATACCAGCAAGAC ATTTCCTTGAGGGATGGAAGGAGAAAGTTTACTCTAAGAAGCCTAAGTCCATTCTAGTCATATCAGCCCACTGGGAGACTGACGTACC CACTGTTAATGCTGTTGACCATCCGGACACGATATATGACTTTAGAGACTTCCCAGCACCTATGTATCAACTTAAGTACCCTGCGCCA GGAGCCCCTGACTTGGCCAAAAGAGTTCAAGAGTTATTAACTGCGTCGGGATTTAAATGCGAGGTCGATAAGAAGAGGGGTTTTGACC ACGGTTCGTGGGTCCCGCTAATGTTTATGTACCCCGAGGCAGACATACCTGTTTGTCAGCTGTCAGTTCAACCTCACCTGGATGGGAC TCATCACTATAACATGGGTAAAGCTTTGGCTCCATTAAAGGAAGAGGGTGTTTTAATAATCGGTTCCGGCGGAGCAGTCCACCCTAGC GATGACACGCCCCACTGTTTTGATGGCGTTGCACCCTGGGCCGCGGAGTTCGACGACTGGTTAGAGGAAGCACTAATCTCCGGAAGGT ACGAGGACGTCAACAATTACCAGACCAAGGCCCCCAACTGGAAAATTGCCCACCCAGGTCCAGAGCACTTTTATCCATTGCACGTAGC TATGGGAGCTGCTGGGGAGAACTCCAAAGCTGAATTAATTCACCGTTCCTGGGACCACGGCACTTTAGGTTACGCTTCCTACAAGTTC ACCAGTACGTAG >GiDODA_a1_anc_N5 (D75G) (804 bp) ATGGGTGGCGAGGAAATGATACGTGAAACCTTCTATATTTCCCACGGCAACCCCATAATGTCTATAGATGAGTCAATTCCTGCACGTC ACTTCCTAGAGGGATGGAAAGAGAAAGTCTATTCTAAGAAACCAAAGTCTATTCTGGTTATAAGTGCCCATTGGGAAACCGACGTTCC CACAGTGAACGCTGTAGACCACCCGGACACAATATACGACTTTTCGGGATTCCCGGCGCCAATGTACCAGTTGAAATACCCAGCACCT GGCGCTCCCGATTTAGCCAAGAGAGTCCAGGAGCTATTAACAGCCTCTGGATTCAAATGCGAGGTGGACAAGAAGAGGGGATTTGACC ATGGGTCTTGGGTACCACTGATGTTCATGTACCCCGAGGCGGACATTCCCGTGTGTCAGCTGTCTGTGCAGCCCCACCTAGATGGCAC TCACCACTATAACATGGGGAAAGCACTTGCGCCATTAAAGGAAGAGGGTGTGCTAATCATAGGCTCTGGTGGAGCTGTACACCCTTCA GACGACACGCCTCATTGCTTTGACGGTGTTGCGCCTTGGGCAGCCGAGTTCGATGATTGGCTAGAGGAAGCGTTAATATCAGGAAGAT ATGAGGACGTGAATAACTATCAGACAAAGGCTCCTAATTGGAAGATAGCCCACCCCGGACCTGAACACTTCTATCCATTACACGTAGC AATGGGAGCTGCCGGCGAAAATTCGAAGGCCGAATTGATTCATCGTTCTTGGGATCACGGAACGTTGGGGTATGCATCTTATAAATTT ACATCCACTTAA >GiDODA_a1_anc_N6 (E109A) (804 bp) ATGGGTGGAGAGGAGATGATCAGAGAAACGTTCTACATTTCACACGGAAACCCCATCATGTCCATCGATGAAAGCATTCCAGCAAGAC ACTTCCTAGAGGGCTGGAAGGAGAAAGTTTACAGTAAGAAGCCTAAGTCAATTTTAGTTATATCAGCACATTGGGAAACAGACGTACC CACTGTTAATGCAGTCGACCACCCAGACACCATATACGACTTCTCTGATTTTCCTGCGCCTATGTACCAGCTGAAGTATCCTGCGCCG GGCGCCCCAGATCTAGCCAAACGTGTCCAGGAGTTACTAACTGCTTCGGGATTCAAGTGCGCGGTCGATAAGAAAAGAGGCTTCGATC ATGGTAGTTGGGTACCTTTAATGTTCATGTATCCAGAGGCTGACATTCCTGTTTGTCAACTGTCAGTGCAGCCACATTTAGATGGAAC CCACCATTACAACATGGGAAAGGCATTGGCACCGTTAAAGGAAGAGGGAGTCCTGATAATTGGAAGCGGTGGTGCAGTACACCCCAGT GACGACACCCCACACTGCTTCGACGGCGTGGCTCCGTGGGCGGCTGAGTTCGACGATTGGCTTGAGGAAGCTTTAATATCAGGACGTT ACGAAGATGTGAACAACTACCAGACCAAGGCTCCCAATTGGAAGATTGCGCACCCAGGTCCCGAGCACTTTTACCCCTTACACGTTGC CATGGGTGCCGCAGGCGAGAACAGTAAAGCTGAGTTAATCCACCGTTCTTGGGACCACGGTACTTTAGGCTATGCTAGCTACAAGTTT ACGAGTACGTGA >GiDODA_a1_anc_N7 (F116L) (804 bp) ATGGGTGGCGAGGAGATGATCAGAGAAACCTTTTATATTTCTCACGGTAACCCTATCATGTCTATCGACGAGTCTATACCCGCACGTC ATTTCCTAGAGGGTTGGAAAGAGAAGGTATACTCAAAGAAGCCCAAGAGCATTTTAGTTATATCAGCTCACTGGGAGACAGACGTCCC GACTGTGAACGCTGTCGACCACCCCGATACAATATATGACTTCAGTGATTTCCCGGCTCCAATGTACCAGTTAAAGTACCCAGCACCT GGCGCACCCGACTTGGCCAAGAGAGTGCAGGAGCTTTTGACGGCATCGGGCTTTAAATGCGAGGTGGACAAGAAGAGGGGTTTAGACC ATGGTAGTTGGGTACCACTAATGTTCATGTATCCAGAGGCAGACATCCCTGTGTGCCAGCTTTCCGTACAGCCTCACCTTGATGGCAC CCACCATTACAATATGGGTAAGGCCTTAGCACCATTGAAAGAGGAAGGAGTGCTGATTATAGGATCAGGTGGTGCCGTGCACCCCTCA GACGACACCCCGCATTGCTTCGACGGGGTAGCTCCCTGGGCTGCAGAGTTCGACGACTGGCTGGAAGAGGCATTAATCAGCGGAAGAT ATGAGGACGTTAATAACTACCAGACTAAAGCTCCTAACTGGAAGATAGCACATCCAGGCCCAGAGCATTTCTATCCTCTACACGTTGC AATGGGTGCGGCAGGAGAAAACTCTAAAGCTGAGCTGATTCACCGTAGTTGGGACCATGGCACTCTTGGTTACGCGTCATACAAATTC ACATCCACATGA Appendices 163 >GiDODA_a1_anc_N8 (P142S) (804 bp) ATGGGTGGTGAAGAGATGATAAGAGAAACATTTTATATTTCACACGGAAACCCTATAATGTCGATAGACGAGTCTATTCCCGCCCGTC ATTTCTTAGAGGGTTGGAAAGAAAAGGTATACTCTAAGAAACCTAAATCAATCCTAGTGATCTCTGCTCACTGGGAAACGGACGTTCC TACCGTTAACGCAGTTGACCACCCAGACACCATTTACGACTTCAGTGACTTCCCTGCTCCCATGTACCAGTTAAAGTACCCGGCTCCA GGGGCACCTGATCTAGCCAAAAGGGTACAAGAGCTGTTGACTGCCAGTGGATTCAAATGCGAGGTCGACAAGAAGAGAGGATTTGACC ATGGTTCTTGGGTCCCTTTAATGTTTATGTATCCAGAGGCAGACATACCCGTTTGTCAGTTGTCTGTACAGAGCCACCTGGACGGTAC GCACCATTACAATATGGGTAAAGCACTAGCGCCACTAAAGGAAGAGGGTGTGCTTATAATCGGTTCAGGTGGAGCCGTGCATCCAAGT GACGACACACCACACTGCTTCGATGGTGTGGCTCCTTGGGCAGCTGAGTTCGACGACTGGTTAGAGGAAGCCCTTATAAGCGGTAGGT ACGAGGATGTAAACAACTACCAGACCAAAGCTCCCAACTGGAAGATCGCTCACCCTGGTCCTGAGCATTTCTACCCTTTGCACGTTGC TATGGGTGCGGCCGGTGAAAACAGTAAGGCCGAGCTGATACACAGAAGTTGGGATCACGGGACTTTGGGCTATGCCTCTTACAAGTTT ACGTCAACATAG >GiDODA_a1_anc_N9 (G171S) (804 bp) ATGGGCGGCGAGGAGATGATCAGGGAGACTTTCTACATATCCCATGGGAACCCTATTATGTCTATCGACGAATCCATTCCAGCACGTC ATTTCCTGGAAGGCTGGAAAGAAAAGGTATACAGTAAGAAGCCAAAGTCAATTCTAGTTATTTCAGCACATTGGGAGACAGACGTACC TACTGTTAATGCCGTAGACCACCCTGACACGATATATGACTTTTCGGATTTCCCGGCTCCCATGTACCAATTGAAATACCCAGCCCCT GGAGCCCCAGACCTAGCAAAGAGAGTCCAGGAATTGTTAACCGCTTCCGGATTCAAGTGTGAGGTCGACAAGAAGAGGGGTTTCGATC ATGGCAGTTGGGTCCCATTAATGTTCATGTATCCTGAGGCAGACATCCCGGTTTGCCAGCTTTCCGTGCAACCACATCTTGACGGTAC ACACCATTATAACATGGGCAAGGCATTGGCACCGTTGAAGGAAGAGGGTGTACTAATTATTGGCTCGGGATCTGCAGTTCACCCTAGC GATGATACCCCTCATTGCTTCGACGGTGTTGCTCCTTGGGCCGCAGAATTCGATGACTGGCTGGAAGAGGCCTTAATTTCGGGAAGAT ATGAGGACGTCAATAACTATCAAACGAAGGCCCCAAACTGGAAAATCGCACACCCCGGGCCTGAACATTTCTATCCATTACACGTCGC CATGGGAGCAGCCGGCGAGAATTCAAAAGCCGAGCTAATTCATAGATCTTGGGATCACGGGACTTTAGGATATGCCTCTTACAAATTT ACAAGCACGTAG >GiDODA_a1_anc_N10 (D177N) (804 bp) ATGGGCGGAGAGGAGATGATTCGTGAGACATTCTATATAAGTCACGGAAACCCGATAATGTCTATCGACGAATCAATCCCTGCACGTC ATTTTCTAGAAGGCTGGAAGGAGAAAGTTTATTCTAAGAAGCCGAAAAGTATACTAGTTATTTCGGCTCACTGGGAAACAGATGTACC TACGGTTAACGCAGTCGACCACCCGGACACTATTTATGACTTTTCTGACTTCCCCGCACCTATGTATCAATTAAAGTACCCAGCCCCG GGAGCACCAGACCTAGCCAAGAGAGTCCAAGAGCTATTGACGGCATCAGGTTTCAAGTGTGAGGTTGATAAGAAGCGTGGCTTTGATC ATGGCTCTTGGGTTCCTTTGATGTTCATGTACCCTGAAGCGGATATACCAGTGTGCCAGTTATCCGTTCAGCCGCATTTGGATGGCAC ACACCATTACAACATGGGTAAGGCCCTAGCGCCTCTGAAAGAGGAAGGCGTGCTAATCATAGGTTCCGGTGGTGCCGTACACCCCTCG AATGACACTCCGCACTGTTTTGACGGAGTTGCACCGTGGGCAGCAGAGTTCGACGACTGGCTGGAAGAGGCACTAATTTCGGGTAGAT ATGAGGACGTCAACAACTACCAGACTAAGGCTCCTAACTGGAAAATCGCCCACCCAGGTCCGGAGCACTTTTATCCTTTACATGTAGC GATGGGCGCAGCGGGTGAGAACTCAAAAGCAGAATTAATCCACAGAAGCTGGGATCACGGTACGTTGGGCTACGCCAGTTACAAATTT ACGTCAACATAG >GiDODA_a1_anc_N11 (C182S) (804 bp) ATGGGCGGTGAGGAAATGATCAGAGAAACCTTCTATATCAGTCATGGTAATCCCATCATGTCCATCGACGAATCCATTCCAGCCCGTC ACTTCCTTGAGGGCTGGAAGGAGAAAGTTTACAGCAAGAAACCTAAGTCTATATTGGTAATTAGTGCACATTGGGAGACAGATGTACC TACTGTTAACGCTGTCGATCATCCAGACACTATATACGATTTCTCTGACTTTCCTGCACCGATGTACCAGTTAAAATATCCGGCTCCA GGAGCACCCGATCTAGCTAAGAGGGTCCAGGAGCTGCTGACAGCATCAGGATTTAAATGCGAAGTTGACAAGAAGAGAGGTTTTGATC ATGGATCCTGGGTCCCTTTAATGTTCATGTACCCAGAGGCAGATATACCAGTCTGTCAGCTATCTGTGCAACCACACTTAGACGGGAC ACACCACTACAATATGGGAAAAGCACTAGCACCACTGAAAGAGGAAGGAGTGCTTATAATCGGTAGTGGTGGTGCAGTCCACCCCAGT GATGACACACCGCATTCATTCGATGGAGTAGCTCCTTGGGCTGCCGAGTTCGACGACTGGCTTGAAGAGGCGTTGATAAGCGGGAGAT ACGAGGACGTTAACAACTACCAGACAAAGGCTCCTAACTGGAAAATCGCCCATCCAGGTCCAGAGCACTTTTACCCATTACACGTAGC GATGGGAGCTGCCGGAGAGAACTCTAAGGCTGAGCTAATACATAGGTCCTGGGACCACGGCACTTTGGGTTACGCTAGTTATAAGTTC ACGTCGACTTGA >GiDODA_a1_anc_N12 (I202T G225W) (804 bp) ATGGGTGGAGAGGAAATGATAAGGGAGACTTTCTATATTTCTCACGGAAACCCAATCATGTCAATCGACGAGTCCATACCGGCTAGAC ACTTTCTTGAGGGCTGGAAAGAAAAGGTATATTCTAAGAAACCGAAGAGTATTTTGGTAATTAGTGCGCACTGGGAAACAGACGTACC AACAGTTAATGCTGTGGATCATCCTGACACCATCTATGACTTCAGTGACTTCCCTGCGCCCATGTACCAGTTGAAGTATCCCGCCCCT GGGGCACCCGACTTAGCAAAACGTGTGCAGGAGCTGCTTACCGCGTCTGGTTTCAAATGCGAGGTTGACAAGAAAAGGGGATTCGACC ATGGTAGTTGGGTCCCCTTGATGTTTATGTACCCGGAAGCTGACATACCTGTTTGCCAACTATCGGTGCAGCCACACTTAGACGGAAC GCATCACTACAATATGGGAAAAGCATTAGCCCCTCTAAAGGAAGAGGGAGTATTGATCATTGGTTCTGGTGGGGCTGTGCACCCTTCA GACGACACTCCACACTGCTTCGACGGTGTCGCACCTTGGGCTGCAGAGTTTGACGACTGGTTAGAGGAAGCCCTTACCTCAGGAAGAT ACGAGGACGTAAACAATTACCAAACTAAAGCACCTAACTGGAAGATCGCACACCCAGGTCCCGAGCACTTTTACCCATTGCACGTCGC TATGGGAGCGGCGGGCGAGAACTCAAAAGCCGAGTTGATACATCGTTCATGGGACCACGGTACCCTAGGTTACGCTTCGTACAAATTT ACATCTACCTAA A. Timoneda, PhD thesis, 2022 164 >GiDODA_a1_anc_N13 (G225W) (804 bp) ATGGGTGGGGAAGAAATGATAAGAGAGACATTTTATATTAGTCACGGCAATCCGATAATGAGTATCGACGAGTCCATTCCTGCTAGAC ACTTCCTTGAGGGATGGAAAGAAAAGGTATACAGTAAGAAGCCGAAATCCATACTGGTAATCAGCGCTCATTGGGAGACAGACGTCCC AACTGTCAACGCAGTAGATCATCCTGATACTATATACGACTTCTCCGATTTTCCTGCCCCAATGTACCAGTTGAAATACCCAGCGCCA GGAGCCCCCGACTTGGCTAAACGTGTCCAAGAGTTATTAACTGCCAGTGGGTTCAAATGCGAGGTGGATAAGAAACGTGGGTTCGATC ACGGCAGCTGGGTGCCTCTGATGTTCATGTACCCAGAGGCGGACATACCCGTTTGTCAGCTGTCTGTGCAGCCACACTTAGACGGTAC GCACCACTACAATATGGGTAAGGCATTAGCACCCTTGAAGGAAGAGGGAGTTCTAATAATTGGCAGCGGCGGTGCCGTGCACCCTAGC GATGACACACCTCACTGCTTTGATGGCGTGGCCCCCTGGGCTGCCGAATTTGACGATTGGTTAGAGGAAGCGCTAATCTCTGGGAGAT ACGAGGATGTGAACAACTACCAGACTAAAGCTCCGAACTGGAAGATCGCTCACCCCTGGCCTGAGCACTTTTACCCTCTACATGTCGC GATGGGTGCGGCAGGAGAGAATTCAAAGGCTGAGTTGATACACAGGAGTTGGGACCATGGTACCCTGGGTTACGCATCATATAAGTTC ACATCTACCTAG >GiDODA_a1_anc_N8+N9 (804 bp) ATGGGTGGCGAAGAGATGATCCGTGAGACATTCTACATTTCACATGGGAACCCAATAATGTCTATAGACGAATCTATACCTGCTCGTC ATTTCTTAGAAGGATGGAAAGAGAAAGTATACAGCAAGAAGCCTAAATCCATATTAGTTATCAGCGCCCACTGGGAAACTGACGTCCC GACAGTCAACGCAGTTGATCACCCAGACACCATTTACGACTTCAGTGACTTTCCAGCGCCGATGTATCAACTTAAGTATCCAGCTCCA GGGGCACCTGATCTGGCTAAAAGAGTCCAGGAGCTTCTGACAGCATCAGGGTTTAAGTGTGAGGTCGACAAGAAGAGGGGCTTTGACC ACGGTTCTTGGGTTCCATTGATGTTCATGTACCCTGAAGCAGACATACCGGTCTGTCAATTGTCAGTTCAATCTCACCTAGATGGAAC TCACCACTATAATATGGGAAAGGCCCTAGCACCACTGAAGGAAGAAGGCGTATTGATCATTGGTAGTGGTAGTGCTGTTCACCCTAGC GATGATACCCCGCATTGCTTTGACGGTGTTGCTCCGTGGGCAGCAGAGTTCGATGACTGGTTAGAGGAAGCTCTAATAAGCGGAAGGT ACGAAGATGTGAATAACTATCAAACAAAAGCACCTAATTGGAAGATCGCTCACCCTGGACCAGAGCACTTCTACCCCCTACACGTAGC AATGGGCGCTGCAGGAGAGAACTCGAAAGCTGAGCTGATTCATCGTTCGTGGGATCATGGAACATTGGGTTATGCTTCGTATAAATTC ACATCGACATAA >GiDODA_a1_anc_N8+N10 (804 bp) ATGGGTGGCGAAGAGATGATACGTGAAACCTTCTACATCTCGCACGGAAATCCAATCATGTCCATCGACGAGAGTATCCCTGCTAGGC ACTTCCTAGAGGGATGGAAAGAGAAAGTATACTCTAAGAAGCCTAAAAGCATACTTGTAATTTCTGCGCACTGGGAAACGGACGTCCC TACCGTCAACGCAGTAGACCATCCTGACACCATCTACGACTTTAGCGATTTCCCAGCCCCAATGTACCAGCTAAAATACCCCGCTCCT GGAGCACCTGATTTGGCTAAGAGGGTCCAAGAATTGTTAACGGCATCAGGCTTTAAATGTGAGGTTGACAAGAAAAGGGGTTTCGACC ATGGTTCGTGGGTGCCATTAATGTTTATGTACCCTGAGGCGGACATTCCCGTGTGTCAATTGTCAGTACAGTCTCATTTGGATGGGAC CCATCACTATAATATGGGTAAAGCTTTAGCACCTTTAAAAGAAGAGGGAGTCCTTATTATTGGGTCCGGTGGTGCAGTTCACCCGTCA AATGATACTCCTCACTGTTTTGACGGCGTCGCCCCCTGGGCAGCAGAGTTCGACGATTGGTTGGAAGAAGCATTGATATCTGGACGTT ATGAGGATGTCAATAATTACCAGACCAAGGCCCCTAATTGGAAAATTGCACACCCAGGTCCCGAGCACTTCTACCCCCTACATGTAGC AATGGGCGCTGCTGGCGAAAATAGCAAGGCTGAGCTGATCCATAGGTCTTGGGATCACGGGACACTGGGTTATGCAAGCTACAAATTC ACTAGCACATAG >GiDODA_a1_anc_N9+N10 (804 bp) ATGGGCGGTGAAGAGATGATTCGTGAAACGTTCTATATCTCACATGGAAACCCTATTATGTCCATAGACGAATCAATTCCCGCACGTC ACTTTCTAGAAGGTTGGAAGGAGAAAGTTTATTCGAAGAAGCCAAAGAGCATTTTGGTCATTTCTGCTCATTGGGAGACAGATGTTCC GACGGTGAATGCCGTCGATCACCCAGACACAATATACGATTTTAGTGACTTTCCTGCACCGATGTATCAGCTTAAATACCCTGCTCCT GGCGCGCCCGACCTGGCTAAGAGAGTACAAGAGTTGTTAACAGCCTCTGGTTTTAAGTGTGAGGTGGACAAGAAAAGAGGTTTCGACC ATGGTTCTTGGGTGCCACTGATGTTTATGTACCCTGAGGCCGATATACCCGTATGCCAATTATCTGTACAGCCTCATCTGGATGGCAC TCACCACTACAATATGGGAAAAGCATTGGCTCCTTTGAAGGAAGAGGGTGTACTAATCATCGGGTCTGGTTCTGCAGTCCACCCGTCA AACGACACACCGCACTGCTTCGACGGAGTTGCACCTTGGGCAGCTGAATTCGACGACTGGCTTGAGGAAGCGTTAATCTCCGGTAGAT ACGAGGATGTGAATAACTATCAAACTAAAGCACCCAACTGGAAGATCGCTCATCCTGGTCCAGAACACTTCTACCCTTTACACGTCGC AATGGGTGCGGCGGGAGAAAACTCAAAAGCCGAATTAATTCATAGATCATGGGACCACGGAACATTGGGCTACGCTTCCTATAAGTTC ACATCCACTTGA >GiDODA_a1_anc_N8+N9+N10 (804 bp) ATGGGTGGAGAGGAGATGATTCGTGAAACCTTTTATATCAGTCATGGTAACCCGATCATGAGTATTGACGAGTCAATACCGGCTAGGC ATTTCTTAGAGGGCTGGAAGGAGAAAGTTTACTCTAAGAAGCCCAAGTCGATTCTTGTTATATCAGCTCACTGGGAGACTGATGTACC AACAGTAAACGCGGTAGACCATCCAGATACTATCTACGACTTTAGTGACTTTCCTGCCCCAATGTACCAGCTTAAGTATCCCGCTCCA GGGGCGCCTGATTTGGCAAAGAGGGTTCAGGAGTTGCTTACTGCGTCTGGTTTCAAATGCGAAGTCGACAAGAAACGTGGGTTCGATC ACGGATCTTGGGTTCCACTGATGTTCATGTACCCAGAGGCTGACATCCCTGTATGCCAGCTTTCGGTGCAGTCACATCTTGACGGCAC TCATCACTACAATATGGGTAAAGCCCTTGCTCCTTTAAAGGAAGAGGGCGTCCTTATCATTGGCTCCGGAAGTGCCGTGCATCCCTCA AACGATACCCCACATTGCTTCGACGGTGTCGCACCCTGGGCTGCGGAATTTGATGATTGGTTAGAAGAGGCATTGATTTCGGGAAGAT ATGAAGATGTCAACAACTACCAAACTAAAGCCCCAAACTGGAAAATCGCGCACCCCGGTCCGGAACACTTCTACCCATTACATGTTGC GATGGGAGCTGCCGGTGAGAACTCCAAAGCCGAGTTAATTCATAGGTCATGGGACCATGGTACGTTAGGGTACGCTTCGTATAAGTTT ACCAGTACTTAG Appendices 165 >GiDODA_a1_a1_anc_S1 (804 bp) ATGGGTGGAGAAGAGATGATAAGGGAAACTTTCTATATCTCACATGGTACGCCAATCATGAGCATCGACGAGAGTATCCCAGCTAGGC ACTTCTTACAGGGGTGGAAAGAGAAAGTCTACTCCAAGAAACCAAAATCTATTCTAGTCATATCTGCACACTGGGAGACTGACGTTCC TACGGTCAACGCAGTAGACCACTCCGACACGATTTATGACTTCAGAGGATTTCCCGCACCAATGTACCAGCTTAAATATCCAGCCCCA GGTGCGCCGGATTTGGCAAAGCGTGTACAGGAGTTGTTGACGGCAAGTGGATTTAAGTGCGCTGTAGATAAGAAGAGGGGCTTCGATC ACGGTAGTTGGGTACCATTAATGTTCATGTATCCAGAGGCAGACATTCCTGTATGTCAGTTGTCTGTGCAACCACACTTAGACGGGAC ACACCATTACAATATGGGAAAGGCTCTAGCGCCTTTGAAGGAAGAGGGTGTTCTTATCATTGGCTCTGGTGGTGCGGTGCATCCGTCC GATGACACTCCACATTCCTTTGACGGTGTGGCTCCTTGGGCTGCTGAATTCGATGACTGGTTGGAAGAAGCCCTTACTAGTGGGCGTT ACGAGGATGTGAATAATTACCAGACCAAGGCTCCTAACTGGAAGATTGCTCACCCATGGCCTGAGCACTTCTATCCCCTGCACGTAGC TATGGGCGCAGCCGGAGAGAATTCAAAAGCAGAGTTGATACACCGTTCTTGGGATCACGGAACACTGGGTTACGCCAGCTACAAGTTT ACTAGCACATGA >GiDODA_a1_a1_anc_S2 (804 bp) ATGGGTGGGGAAGAGATGATCAGAGAAACCTTTTACATTTCTCATGGTACACCTATAATGTCTATCGATGAGTCCATACCCGCCAGAC ACTTTCTTCAAGGGTGGAAAGAGAAGGTTTACAGTAAGAAGCCAAAATCGATTCTAGTTATTAGCGCACACTGGGAAACGGACGTCCC AACCGTTAACGCTGTCGATCATTCCGACACGATTTATGATTTCAGAGGCTTCCCAGCTCCAATGTATCAGCTAAAGTATCCCGCTCCT GGGGCTCCAGACTTGGCGAAGCGTGTCCAAGAACTACTTACAGCTTCTGGATTCAAGTGTGCAGTGGATAAGAAGCGTGGCTTGGACC ATGGATCATGGGTACCTTTAATGTTTATGTACCCTGAAGCAGATATACCTGTTTGCCAATTGTCTGTCCAGCCTCATCTGGACGGCAC TCATCACTATAACATGGGAAAAGCGTTAGCTCCTTTAAAAGAGGAAGGAGTGTTAATAATCGGCTCCGGCGGTGCGGTGCATCCTAGT GATGACACGCCTCATTCATTTGACGGAGTTGCTCCATGGGCAGCTGAATTCGACGACTGGCTTGAAGAGGCGTTAACTTCCGGTAGGT ACGAGGATGTTAACAATTACCAGACAAAGGCACCTAACTGGAAAATCGCACATCCATGGCCTGAACACTTCTATCCTTTACACGTCGC CATGGGTGCTGCTGGCGAGAACAGTAAGGCTGAGCTTATCCATAGATCCTGGGACCACGGAACGTTGGGCTATGCGTCATATAAATTC ACATCCACATGA A. Timoneda, PhD thesis, 2022 166 Appendix 2. List of oligonucleotide primers used in this thesis. Primer ID Sequence 5’-3’ Purpose Chapter 2. Betalain pigments as in vivo visual markers for arbuscular mycorrhizal colonisation of root systems M13F GTAAAACGACGGCCAGT Check and sequencing of pBlueScript inserts. M13R CAGGAAACAGCTATGAC MoClo_lvl0_F0015 CGTTATCCCCTGATTCTGTGGATAAC Check and sequencing of level 0 vectors MoClo_lvl0_R0016 GTCTCATGAGCGGATACATATTTGAATG MoClo_lvl1_F0229 GAACCCTGTGGTTGGCATGCACATAC Check and sequencing of level 1 vectors MoClo_lvl1_R0230 CTGGTGGCAGGATATATTGTGGTG MoClo_lvl2_F0231 GTGGTGTAAACAAATTGACGC Check and sequencing of level 2 vectors MoClo_lvl2_R0232 GGATAAACCTTTTCACGCCC MtPT4pro_F CTCGATCCACAACAAAGATT Cloning of MtPT4 promoter from gDNA MtPT4pro_R CTCAAGTTGGTTTTTGGAGT MtBCP1pro_F CACATCTAGAGAGAGGGAGATGTGTT Cloning of MtBCP1 promoter from gDNA MtBCP1pro_R TCTCGGATCCTGCAATTGCAACTGATGAAAG NbPT5bpro_F GTGAGAAATTTACCAAATACCTATAAATAATGTATCAGC Cloning of NbPT5b promoter from gDNA NbPT5bpro_R GTTTGCTTTTTACTGCAAATTTTGGTTG NbBCP1bpro_F ACTGGTATTAGAATAATAGCCTGTTTAGCC Cloning of NbBCP1b promoter from gDNA NbBCP1bpro_R GGAAAATTCTTCTTTAAGGGTTAGTTG BvCYP76AD1_qF AACTGCAACAACAACGACGA RT-PCR of T1 N. benthamiana plants BvCYP76AD1_qR AAAATGTCGACGAGCAAATGG BvDODAa1_qF TGCTGCTATGTACCAGTTCAAGT RT-PCR of T1 N. benthamiana plants BvDODAa1_qR CAGTTTCCGCCGTTTCGAAA MjcDOPA5GT_qF CAACATTCATCATCTGATAA RT-PCR of T1 N. benthamiana plants MjcDOPA5GT_qR GTGATCCAAATGAGATGTAT MtPT4_lv0_F TGAAGACATGGAGCTCGATCCACAACAAAGATT MoClo Golden gate cloning of promoters MtPT4_lv0_R TGAAGACATCATTCTCAAGTTGGTTTTTGGAGT MtBCP1_lv0_F TGAAGACATGGAGAATGCGTTGCTTAAGAGTTG MoClo Golden gate cloning of promoters MtBCP1_lv0_R TGAAGACATCATTTGATCAAAATTTGATTTTGG NbPT5b_lv0_F TGAAGACATGGAGGTGAGAAATTTACCAAATACCTATAAATAATGTATCAGC MoClo Golden gate cloning of promoters NbPT5b_lv0_R TGAAGACATCATTGTTTGCTTTTTACTGCAAATTTTGGTTG NbBCP1b_lv0_F TGAAGACATGGAGACTGGTATTAGAATAATAGCCTGTTTAGCC MoClo Golden gate cloning of promoters NbBCP1b_lv0_R TGAAGACATCATTGGAAAATTCTTCTTTAAGGGTTAGTTG Appendices 167 Primer ID Sequence 5’-3’ Purpose Chapter 3. Search for ADH variants with putative relaxed feedback inhibition outside of the order Caryophyllales GeneRacer_oligoT GCTGTCAACGATACGCTACGTAACGGCATGACAGTG(T)20 cDNA production SlADH1_F ATGTTGTCTTTCACCCCACT Amplifying from cDNA libraries SlADH1_R TCAGTTGTTCTCAGGTTTAAC SlADH2_F ATGTTTTCCCTTTCATCTATACAAT Amplifying from cDNA libraries SlADH2_R TCAAGAAAGTGCCTCTAGTT SlADH1_lv0_F TGAAGACATAATGTTGTCTTTCACCCCACT Cloning into level 0 vector SlADH1_lv0_R TGAAGACATAAGCTCAGTTGTTCTCAGGTTTAAC SlADH2_lv0_F TGAAGACATAATGTTTTCCCTTTCATCTATACAAT Cloning into level 0 vector SlADH2_lv0_R TGAAGACATAAGCTCAAGAAAGTGCCTCTAGTT MeADH1_lv0_F TGAAGACATAATGTTACCCTTTAGTTCTACC Cloning into level 0 vector MeADH1_lv0_R TGAAGACATAAGCCTATGACCTCTCCTGCTC MeADH2_lv0_F (BvADHa_cTP_lv0_F) TGAAGACATAATGATTTCACTCTCTTCTTTTCA Cloning into level 0 vector MeADH2_lv0_R TGAAGACATAAGCTTACTGGTGCTCCAACTCC SlADH1_E200T_F TTTGGACCTACAAGTGGTAAGGATAGTTGGAAAG PCR site-directed mutagenesis SlADH1_E200T_R ACCACTTGTAGGTCCAAACATGGGATGAGTAC SlADH1_E200D_F TTTGGACCTGATAGTGGTAAGGATAGTTGGAAAG PCR site-directed mutagenesis SlADH1_E200D_R ACCACTATCAGGTCCAAACATGGGATGAGTAC MeADH1_E195D_F CGGACCAGATTCTGGTAAGAACTCATGGGTGG PCR site-directed mutagenesis MeADH1_E195D_R CCAGAATCTGGTCCGAACATCGGGTGCGTAC SlADH2_T218E_inF TGAAGACATAGAAAGTGGTAAAGACAATTGGAA MoClo site-directed mutagenesis SlADH2_T218E_inR TGAAGACATTTCTGGACCAAACATAGGATGAG MeADH2_D197E_inF TGAAGACATGAGTCTGGAAAGTACTCTCTGTTG MoClo site-directed mutagenesis MeADH2_D197E_inR TGAAGACATACTCTGGACCGAACATAGGG Chapter 4. Bioproduction of betalains in Escherichia coli bioreactors and characterization of a highly active DODA enzyme in the Cactaceae pET_F TAATACGACTCACTATAGGG Check and sequencing of pET28a inserts pET_R CTAGTTATTGCTCAGCGGT Chapter 5. Identification of residues responsible for high DODA activity in CgDODAα1 ConLS_F ACAAGCAACGATCTCCAGGA Check and sequencing of pTMP137 inserts ConLS_R GCTTAGTTGTGAGTCGCCAG