1 of 19 F-actin coordinates spindle morphology and function in Drosophila meiosis 1 Benjamin W. Wood, Sera Shi and Timothy T. Weil‡ 2 Department of Zoology, University of Cambridge, Downing Street, Cambridge, CB2 3EJ, UK 3 ‡Author for correspondence (e-mail: tw419@cam.ac.uk) 4 ABSTRACT 5 Meiosis is a highly conserved feature of sexual reproduction that ensures germ cells have the correct 6 number of chromosomes prior to fertilization. A subset of microtubules, known as the spindle, are 7 essential for accurate chromosome segregation during meiosis. Building evidence in mammalian 8 systems has recently highlighted the unexpected requirement of the actin cytoskeleton in 9 chromosome segregation; a network of spindle actin filaments appear to regulate many aspects of 10 this process. Here we show that Drosophila oocytes also have a spindle population of actin that 11 appears to regulate the formation of the microtubule spindle and chromosomal movements 12 throughout meiosis. We demonstrate that genetic and pharmacological disruption of the actin 13 cytoskeleton has a significant impact on spindle morphology, dynamics, and chromosome alignment 14 and segregation during maturation and the metaphase-anaphase transition. We further reveal a role 15 for calcium in maintaining the microtubule spindle and spindle actin. Together, our data highlights 16 potential conservation of morphology and mechanism of the spindle actin during meiosis. 17 AUTHOR SUMMARY 18 The actin cytoskeleton is a fundamental component of eukaryotic cells. During meiosis, cytoplasmic 19 actin has been shown to aid in spindle positioning in order to enable asymmetric division. One largely 20 unexplored role of actin is its contribution during the meiotic segregation of chromosomes. Previously 21 thought to be primarily regulated by the microtubule cytoskeleton, the discovery of a population of 22 actin within the meiotic spindle of mouse and human oocytes has opened an entirely new avenue of 23 investigation within the field of meiosis. Here, we reveal in Drosophila melanogaster oocytes a novel 24 population of meiotic actin. Utilising a combination of genetic and pharmacological approaches, we 25 demonstrate the importance of this spindle actin in the regulation of microtubule spindle morphology 26 and chromosomal segregation during the first metaphase-anaphase transition following egg 27 activation. We further explore the potential input from calcium signalling at the spindle. We thus 28 demonstrate great similarity of spindle actin morphologies and mechanisms when compared with 29 starfish, Xenopus and mammals. This suggests a level of conservation that reflects the fundamental 30 nature of meiosis itself. 31 mailto:tw419@cam.ac.uk) 2 of 19 INTRODUCTION 32 Meiosis is a well studied and documented process that is essential to produce haploid gametes 33 during sexual reproduction. In mammals, fetal oogonia initiate meiosis synchronously, arresting at 34 prophase I and pausing in this state until sexual maturity, whereupon an oocyte or small subsets of 35 oocytes are released periodically, and meiosis is resumed [1]. Nuclear envelope breakdown initiates 36 this resumption, resulting in the formation of a bipolar spindle network as microtubules polymerize, 37 capture chromosomes and then align them on the metaphase plate [2]. The spindle interacts closely 38 with the cytoplasmic actin, which aids in the asymmetric positioning of the spindle adjacent to the 39 cortex, enabling asymmetric cell division to leave a single oocyte containing the necessary maternal 40 components [2-7]. Oocytes are arrested in metaphase II until fertilization and subsequent egg 41 activation. An intracellular rise in calcium then triggers the completion of meiosis, resulting in the 42 formation of the maternal pronucleus which undergoes fusion with the paternal pro-nucleus to form 43 a diploid zygote [8,9]. 44 In Drosophila melanogaster, the process is similar but with a few differences. Oocytes are generated 45 continuously when environmental conditions are favorable. Connected to supporting cells via 46 cytoplasmic bridges and surrounded by a mono-layer of epithelial cells, the oocyte passes through 47 14 morphologically distinct stages during oogenesis. Meiosis is held in late prophase I for the majority 48 of oogenesis, until the germinal vesicle breakdown (GVBD) (often used interchangeably with nuclear 49 envelope breakdown) and the prophase to metaphase transition is completed [10-12]. Key features 50 of this step can be observed within the Drosophila oocyte, such as GVBD and formation of a bipolar 51 spindle as microtubule spindles capture the meiosis I chromosomes [13,14]. The spindle is a 52 comparatively small structure in relation to the rest of the oocyte and lies parallel to the cortex at the 53 dorsal-anterior tip of the oocyte, just below the dorsal appendages. 54 Unlike mammals and most vertebrates, the final meiotic arrest in Drosophila oocytes is at metaphase 55 I, and the bipolar spindle structure that forms is more defined and has more focused spindle poles 56 when compared to the “barrel” shaped spindles of mammals. Activation in Drosophila occurs prior 57 to fertilization as the oocyte passes into the oviduct, which results in a calcium influx through transient 58 receptor potential melastatin (TrpM) ion channels in the plasma membrane of the oocyte [15,16]. 59 This calcium event enables the resumption of meiosis from its arrested state and can be observed 60 using a variety of microtubule labelling tools [17]. 61 62 Detailed cytological studies of Drosophila oocytes revealed that at the metaphase I arrest, the 63 chromosomes exist as a central mass [18], including the non-exchange chromosomes, which can 64 be visible as a separate entity during pro-metaphase or anaphase [17-19]. The metaphase to 65 anaphase transition during egg activation is described as a stereotypical series of events that leaves 66 spindle elongated and perpendicular to the cortex [20]. 67 3 of 19 The mature Drosophila oocyte itself is approximately 500 μm in length, compared to the spindle 68 which is approximately 10 μm (~50:1 ratio of oocyte to spindle). In contrast, the mouse oocyte has 69 a ratio of approximately 5:1 oocyte to spindle length [21]. The Drosophila oocyte itself is vitellogenic 70 and surrounded by protective outer casings, reflecting the need for the future embryo to be as robust 71 as possible as they ultimately develop externally to the organism. These features can initially make 72 distinguishing the components of the Drosophila spindle challenging. However, with a variety of 73 genetic tools available, visualization and manipulation of spindle components can now be achieved, 74 highlighting Drosophila as an important model system for understanding and future research of this 75 field. 76 It has been observed that a population of actin exists within the mammalian oocyte that forms a 77 spindle-like structure and has been shown to regulate chromosome alignment and segregation [21]. 78 Treatment of these oocytes with cytochalasin D (cytoD) and knockdown of Formin-2, a key nucleator 79 of spindle-like actin in mice, results in the misalignment of chromosomes during metaphase I and 80 chromosome segregation errors during anaphase, often resulting in aneuploidy. Actin was shown to 81 regulate chromosomal movements in part due to control of the kinetochore microtubules (K-fibres), 82 indicating a likely role of actin in microtubule organization more generally [21]. Multi-color 3D-83 fluorescence microscopy revealed that human oocytes display a population of spindle actin similar 84 to the population in mice, and additionally demonstrated that there is co-localization of γ-tubulin rich 85 minus ends with filamentous actin clusters at the spindle poles [22]. Pharmacological manipulations 86 revealed a co-operation of actin and microtubules at the meiotic spindle, as disruption of the 87 microtubule spindle morphology is directly mirrored by changes to the spindle actin [22]. Taken 88 together, this data suggests that the spatiotemporal organization of actin during oocyte maturation 89 is similar to microtubule dynamics. 90 In this study, we utilize advanced imaging in conjunction with pharmacological and genetic 91 manipulation to demonstrate that a population of spindle actin exists and appears to surround the 92 microtubule spindle and chromosomes in the metaphase I arrested mature Drosophila oocyte, we 93 show that the mammalian Formin-2 homologue, Cappuccino (Capu), is required for the formation of 94 the spindle actin network, which undergoes strikingly similar morphological changes to the 95 microtubule spindle during egg activation. Disruption of the actin cytoskeleton reveals a 96 chromosomal phenotype that parallels that of a mammal: chromosome alignment and segregation 97 appear disrupted. Whilst the actin cytoskeleton has been shown to be required for spindle positioning 98 [23], such chromosomal segregation errors have not yet been explored in Drosophila, and these 99 phenotypes may be attributed to the novel population of spindle actin demonstrated herein. 100 Moreover, visualization and manipulation of calcium ions at this transition reveals the importance of 101 calcium signaling for maintaining the morphology of the metaphase spindle and chromosome 102 segregation. Taken together, our data suggests that actin is required upstream of the microtubules 103 to regulate formation of the spindle. 104 4 of 19 RESULTS 105 Actin is present at the spindle 106 To test if the recently established novel population of actin within the spindle in mouse [21] and 107 human [22] oocytes is conserved in Drosophila, we first sought to visualize actin and microtubules 108 simultaneously in late-stage egg chambers. Using an endogenous F-actin stain (SiR-actin) and the 109 microtubule binding protein Jupiter (Jup) fused to GFP (Jup-GFP), we show that prior to GVBD, the 110 oocyte nucleus remains large and without clear microtubule filaments. An enrichment of actin is 111 detected around the nuclear envelope and the DNA is condensed in a small area of the nucleoplasm 112 (Fig. 1A, A¢). As the oocyte undergoes maturation and transitions from prophase I arrest to 113 prometaphase I, we observe actin filaments forming around and throughout the nucleus as the 114 nuclear membrane breaks down (Fig. 1B, B¢). As this stage progresses, we observe that a ring of 115 actin forms around the DNA, as it continues to condense, while the microtubules appear to be mostly 116 diffuse (Fig. 1B¢¢). At the end of oogenesis, stage 14, when the mature oocyte is arrested in 117 metaphase I, the microtubule spindles form an elliptical structure around the centrally lying 118 chromosomes and what appears to be an outer ring of actin (Fig. 1C, C¢). Higher resolution imaging 119 of the nucleus in a mature oocyte transitioning to the metaphase I shows a population of actin 120 surrounding the DNA and microtubules as well as associating with the microtubule spindles (Fig. 121 1D). These observations are reminiscent of an actin fishnet structure identified in starfish after GVBD 122 [24,25]. 123 To verify these findings, we used two well-established genetically encoded actin markers: Act5C-124 GFP [26], the actin 5C monomer conjugated to a GFP; and Lifeact-GFP [27], the actin binding protein 125 Lifeact conjugated to a GFP. We again observe a distinct spindle actin population with the 126 metaphase chromosomal mass located in the center (Fig. 1E, F). 127 To further characterize this structure in the mature oocyte, we sought to observe calcium at this 128 region. Previous work has shown a clear link between calcium and actin at Drosophila egg activation 129 [28] and recent evidence in mature Xenopus oocytes highlights the presence of enriched calcium 130 and necessity of localized calcium signaling at the spindle for regulation of microtubules [29]. Using 131 GCaMP3 [30], a genetically encoded calcium sensor to visualize calcium in vivo, we observe an 132 increased fluorescence intensity at the spindle indicative a calcium enrichment (Fig. 1G). 133 Finally, we used UtrCH-GFP, the calponin-homology domain of Utrophin (Utr), required for the actin 134 binding capacity of Utr, conjugated to GFP, to visualize actin in live samples. This tool has been 135 used previously in other systems to successfully label the spindle actin for live analysis [21]. We 136 once again show a highly distinct population of actin (Fig. 1H) that resembles the microtubule spindle 137 (Fig. 1I), when visualized in live samples using Jup-GFP. The actin appears to mirror the microtubule 138 5 of 19 elliptical shape with focused poles, a uque feature of the Drosophila meiotic spindle, which will 139 henceforth be referred to as spindle actin. 140 Spindle actin regulates metaphase microtubule spindle and chromosomes 141 To test the function of spindle actin, we first examined the relationship between the spindle actin and 142 the microtubule spindle. Oocytes expressing markers for microtubules (Jup-GFP) or actin (UtrCH-143 GFP) were incubated in the microtubule depolymerizing agent colchicine (Fig. 2A, B). As expected, 144 we observe a complete loss of the microtubule spindle (Fig. 2A), however, the spindle actin remained 145 (Fig. 2B). Next, depolymerization of the actin cytoskeleton using cytoD resulted in an elongated 146 morphology of the microtubule spindle (Fig. 2C), as well as the expected loss of spindle actin (Fig. 147 2D). These data suggest that the spindle actin population is not dependent on the presence of the 148 microtubule spindle filaments and that spindle actin could be important for organizing the microtubule 149 spindle. 150 To test the role of calcium at the spindle, mature oocytes were incubated in the membrane permeable 151 calcium chelating agent BAPTA-AM. The removal of calcium results in the loss of both the 152 microtubule spindle (Fig. 2E) and the spindle actin (Fig. 2F), suggesting that calcium signalling is 153 required to maintain the spindle apparatus but not showing that this is a direct effect. This is, 154 however, similar to results in Xenopus in which BAPTA incubation caused microtubule 155 depolymerization [29]. 156 Considering the function of the microtubule spindle in chromosome segregation, we next asked if 157 the spindle actin is involved in this process. First, we observed the difference in chromosome 158 alignment between the metaphase-arrested spindle in control media (Fig. 2G, H) compared to after 159 disruption of actin with cytoD (Fig. 2I, J). Separately using DAPI and a centromere marker, CENP-A 160 homolog centromere identifier (CID) fused to an enhanced GFP (Cid-EGFP) [31], we show that there 161 is alteration to the alignment as we detect multiple chromosomal masses spreading to either pole of 162 the spindle axis (Fig. 2G-J). When analyzed, both DAPI and Cid show similar data for the spread of 163 chromosomes and make us confident that DAPI is dependable to measure phenotypes. Moreover, 164 BAPTA-AM incubations result in compaction of the chromosomes that closely associates with the 165 oocyte cortex (Fig. 2K). This phenotype is likely explained by the previous results in which BAPTA 166 causes complete loss of the microtubules and actin within the spindle [29]. 167 To further examine the function of spindle actin, we used genetics to disrupt the Drosophila 168 homologue of mammalian Formin-2, Capu, which functions as part of the Capu-Spire actin 169 nucleating complex (Fig. 3A-H) [32,33]. Homozygous capu mutants resulted in lethality, therefore 170 we used several alleles to generate different heterozygous and trans-heterozygous backgrounds for 171 analysis. Live visualization of actin showed that both a capu heterozygous and a capu / spire trans-172 heterozygous oocytes background was sufficient to disrupt the formation of this spindle actin (Fig. 173 6 of 19 3A). We then observed microtubule spindle in various capu and spire mutant backgrounds and show 174 that the microtubule spindle are significantly elongated or no longer form (Fig. 3B, E, F). 175 176 Similarly, in capu and spire mutant combinations a clear disruption to the centrally congressed 177 metaphase chromosomes can be observed (Fig. 3C). Measuring this spread of the metaphase I 178 chromosomes as the maximum chromosomal distance reveals a significant increase in those 179 oocytes with a disrupted spindle actin (Fig. 3C, F). In addition, analysis of the angle between the 180 spindle and the nearest cortex reveals a significant change when actin is disrupted (Fig. 3C, G). We 181 also find that disruption of the spindle actin with a capu mutant results in a loss of enrichment of 182 calcium at the spindle and multiple chromosomal masses (Fig. 3D). 183 184 Finally, we used fluorescence recovery after photobleaching (FRAP) to test the dynamics of 185 microtubule recruitment in wild-type and capu / spire trans-heterozygous oocytes (Fig. 3H). When 186 the spindle actin is disrupted, we observed a small but ultimately significant change in the recovery 187 dynamics of the microtubule spindle. The failure of the spindle to recover fluorescence to a wild-type 188 level, suggests the spindle actin population may play a role in the recruitment of microtubules to the 189 spindle itself. 190 Taken together, our analyses reveal an important relationship between the spindle actin and 191 microtubule spindle, as actin is required for accurate formation of the microtubule spindle and 192 regulation of its morphology. This appears conserved with studies in mice and humans, in which the 193 spindle actin has been shown to be required for formation of K-fibres and recovery of the spindle 194 structure [21,22]. 195 Functional importance of the spindle actin during anaphase I 196 In order to test if the functional importance of this actin population extends throughout meiosis, we 197 observed the spindle actin during egg activation. In Drosophila, egg activation occurs as the oocyte 198 passes into the oviduct but can consistently be recapitulated ex vivo through incubation in a 199 hypotonic buffer (activation buffer (AB)) [15,16,34]. This method was first established in the 1980s 200 and has since been extensively utilized to understand the details of Drosophila egg activation. The 201 first event of egg activation is swelling, which results in TrpM calcium ion channels opening and 202 initiates a calcium transient to pass through the cell and trigger the metaphase-anaphase transition 203 [15,16]. Microtubule spindle undergo a classical rearrangement at egg activation as the spindle 204 initially elongates, then contracts and rotates in relation to the cortex, ultimately becoming 205 perpendicular to the cortex [17]. Co-visualization with DAPI and centromeres, show the 206 chromosomes associated in the center of the actin and expectedly perpendicular with the cortex 207 (Fig. 4A, A¢). Our observation of microtubes and actin at egg activation reveals a similar 208 morphological change for the spindle actin as it appears to contract and remain around the 209 microtubule spindle (Fig. 4A, A¢). In addition, we observe a more even distribution of calcium at the 210 7 of 19 spindle, no longer concentrated at the poles, after egg activation (Fig. 4B). Importantly, we see a 211 similar result with observing actin live during egg action (Fig. 4C), thus reinforcing our conclusion 212 that the interplay between spindle actin and the microtubule spindle continues in anaphase I. 213 We next observed actin at anaphase I and show that spindle actin remains around the separating 214 chromosomes, perhaps less enriched than before egg activation (Fig. 5A). Disruption of the 215 anaphase spindle actin was achieved through addition of cytoD following AB treatment for 10 216 minutes to ensure oocytes had entered anaphase. This resulted in significant disruption to the 217 segregation of the chromosomes, with frequent occurrence of aberrant chromosomal masses, which 218 we define as chromosomes separate from the main axis of segregating chromosomes or a mass 219 causing obvious non-uniformity (Fig. 5B, E). In wild-type and non-disrupted anaphase oocytes the 220 chromosomes separate as what appear visually as singular connected masses, except chromosome 221 4 which appear as small polar units. However, when the actin at the spindle is disrupted, such 222 separating masses, which usually appear as a singular unit are disrupted to the extent that there 223 appear to be several separating units, often not aligned with the rest of the separating chromosomes. 224 This is a drastic phenotype, and one that can be visualised and quantified easily as such. 225 Similarly, visualization of the chromosomes in capu heterozygous and capu / spire trans-226 heterozygous backgrounds revealed microtubule abnormalities and aberrations in chromosome 227 segregation (Fig. 5C, D). Individual aberrant chromosomal masses were clearly identifiable and all 228 test oocytes demonstrating a significant increase in the number of these masses as compared to 229 wild-type oocytes (Fig. 5A). Furthermore, in the case of capu / spire trans-heterozygous oocytes, it 230 was common to observe the 4th non-exchange chromosomes being closely located, suggesting a 231 loss of spindle polarity (Fig. 5D). This perhaps suggests a disruption to the bipolar spindle axis, as 232 each of the non-exchange chromosomes would be expected to be at one pole of the spindle with 233 the remaining chromosomes in-between. 234 Despite clear loss of accurate segregations, measurements of the maximum chromosomal distances 235 and spindle length did not reveal any significant differences from control oocytes (Fig. 5F, G). This 236 may suggest that the chromosomes are still able to separate during anaphase, but with a loss of 237 accuracy. In addition, we observed no significant difference in the angle of the spindle-cortex 238 between cytoD treated oocytes in anaphase versus anaphase controls and cytoD treated metaphase 239 oocytes (Fig. 5H). We do, however, still observe a significant increase in the angle as compared to 240 metaphase I controls. Together, this suggests that many anaphase events may still be capable of 241 occurring when the spindle actin is disrupted. However, in the absence of spindle actin this appears 242 to be brought about prematurely and results in a potential loss of organization, as we see dramatic 243 aberrations in the quality of chromosome segregation. 244 8 of 19 DISCUSSION 245 This study establishes the existence of a population of spindle actin in close proximity with the 246 microtubule spindle and chromosomes in Drosophila mature oocytes that appears to be required for 247 the regulation of meiosis. Alteration of the actin through pharmacological or genetic disruption results 248 in chromosome segregation errors, which may be attributable to the role of the spindle actin. We are 249 aware that these approaches target all actin in the oocyte and are not specifically acting on the 250 spindle population. For example, while we demonstrate the requirement of Capu in the formation of 251 this spindle actin, one should consider that capu and spire have previously been shown to act 252 together at stage 9 of oogenesis to organize the cytoskeleton which is required for axis formation 253 [35,36]. 254 At metaphase, the spindle lacks accurate chromosome alignment and congression, and at anaphase 255 aberrant chromosomal masses are frequent. We also identify calcium as important in maintaining 256 the spindle throughout metaphase. Together, we suggest that the spindle actin is required to 257 mediate the accurate segregation of chromosomes through regulation of the microtubule spindle. 258 Our data, together with recent work that reveals a population of spindle actin in mammals [21,22] 259 and calcium enrichment in Xenopus [37], suggests there is a high level of evolutionary conservation 260 at the spindle. 261 Prior to the formation of the spindle, we observed actin surrounding the nucleus reminiscent of the 262 actin shell in mice oocytes [38], which aids the tearing and depolymerization of nuclear lamina during 263 GVBD. In addition, our observation of the filamentous actin projecting through the nuclear envelope 264 are similar to the spike-like structures in starfish oocytes [24]. To see whether this population of actin 265 later forms transient fishnet structures that gather all chromosomes and coordinate their capture by 266 microtubule spindles as in starfish oocytes [25], further exploration with higher resolution microscopy 267 and better labelling tools are required. 268 The conservation, from mammals to Drosophila, we see in both functional and morphological 269 similarities between spindle actin populations. Whilst variations to be expected in comparison of 270 meiotic mechanisms between species, such as final meiotic arrest occurring at metaphase I in 271 Drosophila, in comparison to metaphase II in mammals, it appears that distinct populations of spindle 272 actin may be another fundamental feature of meiosis. The potential conservation of this population 273 of actin extends from its nucleation by the Formin-2 homolog Capu, to its functional role in the 274 regulation of spindle microtubules and chromosomal movements. 275 However, there does appear to be some variation in the morphology and function of this spindle 276 population. Much like the microtubule spindle, the spindle actin forms an elliptical shape with highly 277 focused poles, unlike mammalian meiotic spindle structures which are more ‘barrel like' in shape [2]. 278 This population also seems to be resilient to disruption of the microtubule spindle, appearing to be 279 9 of 19 important for recruitment of microtubules to the spindle and regulating overall spindle morphology, 280 therefore suggesting there may be an upstream requirement of the actin cytoskeleton. When 281 disrupted, either through disruption of capu or depolymerization by cytoD, significant defects in 282 chromosome alignment can be observed. Observation of metaphase I oocytes indicated a spreading 283 of the chromosomes along the metaphase spindle, with separation of chromosomes appearing 284 reminiscent of pro-metaphase oocytes [19]. This could suggest a requirement of the spindle actin in 285 the prophase-metaphase transition, a much earlier stage than has been observed in mammals to 286 date. Perhaps this population of actin plays a more significant role in Drosophila meiosis I as mature 287 oocytes arrest at metaphase I, which may indicate that the actin is important in meiotic arrest and 288 release from this arrest during egg activation and onset of the metaphase-anaphase transition. 289 During egg activation, a global transient of calcium triggers a multitude of events, including global 290 rearrangements of the actin cytoskeleton and resumption of meiosis [28]. There is building evidence 291 that ties calcium and actin as two interlinked molecules in many signaling pathways; in Drosophila 292 egg activation, dispersion of cortical actin enables entry of calcium in the form of a wave, which, 293 downstream, effects a wave of reorganizing F-actin [28]. With many actin binding proteins (ABPs) 294 being calcium sensitive, such as α-actinin and the villin family, and many calcium-sensitive proteins 295 having downstream effects on the actin cytoskeleton, such as Calmodulin and calcineurin [39], it is 296 possible that the calcium wave has a direct effect on the population of spindle actin, potentiating the 297 release from meiotic arrest. 298 We have further observed what appears to be an enrichment of the calcium indicator GCaMP3 at 299 the metaphase arrested spindle, suggesting an increased local concentration of calcium. 300 Introduction of a calcium chelator results in depolymerisation of both the actin and microtubule 301 spindle networks, indicating the likely requirement of calcium in maintenance of these populations, 302 consistent with observations in Xenopus [29]. 303 As demonstrated, disruption of Capu-Spire actin nucleating complex often results in the loss of 304 formation of the microtubule spindle, indicating that the spindle actin may be required for recruitment 305 of microtubules. This recruitment could be direct, however, there may be a contribution from the 306 localized action of calcium. For example, microtubule associated proteins (MAPs), many of which 307 are calcium sensitive and associate with both the actin and microtubule cytoskeleton [37], may 308 mediate an actin-dependent recruitment of microtubules. Without the spindle actin, it is likely that the 309 localization and coordination of these proteins are lost, resulting in a loss of calcium signaling at the 310 spindle and concomitant loss of the microtubule spindle itself. Such mechanisms should be explored 311 further in both Drosophila and mammals, as contributions from the spindle actin, microtubule spindle 312 and calcium signaling are likely to demonstrate a degree of conservation. 313 10 of 19 ACKNOWLEDGEMENTS 314 We are grateful to Jens Januschke, Torsten Krude, Jose Casal, and Paul Conduit for discussions 315 and advice; Elise Wilby for detailed feedback on the manuscript; the Zoology Imaging Facility and 316 Matt Wayland for assistance with microscopy; the Bloomington Drosophila Stock Center and 317 Drosophila community for fly lines. 318 FUNDING 319 The Wellcome Trust Institutional Strategic Support Fund, University of Cambridge grant number 320 097814 (to TTW); Sir Isaac Newton Trust Research Grant (Ref 18.07ii(c)) (to TTW); Biotechnology 321 and Biological Sciences Research Council Doctoral Training Partnerships studentship (to BWW); 322 The funders had no role in study design, data collection and analysis, decision to publish, or 323 preparation of the manuscript. 324 AUTHOR CONTRIBUTIONS 325 B.W.W. was responsible for conceptualization, data curation, formal analysis, investigation, 326 methodology, writing original draft, and writing review and editing; S.S. was responsible for data 327 curation, formal analysis, investigation, methodology, and writing review and editing; T.T.W was 328 responsible for conceptualization, funding acquisition, investigation, methodology, supervision, and 329 writing review and editing. 330 COMPETING INTERESTS 331 The authors have declared that no competing interests exist. 332 11 of 19 MATERIALS AND METHODS 333 334 Fly Maintenance 335 Fly stocks were raised on Iberian recipe fly food at 18°C, 21°C and 25°C. For dissection of mature 336 oocytes, approximately 30 female flies with 5 male flies were transferred into a vial with Iberian recipe 337 fly food and wet yeast for 48 hours at 25°C. 338 339 Fly lines 340 matα-GAL4::VP16, UASp-GCaMP3 [40] ; tub-GAL4VP16 (S. Roth); UASp-Utrophin-CH-GFP/Tm3 341 [41]; UASp-LifeactGFP/Tm3 (BL58717); UASp-Act5CGFP/Tm3 (BL7309); P{PTTGA}JupiterG00147 342 (BL6836); P{PTT-un}CamP00695/CyO (BL50843); cid-EGFP, His2Av-mRFP (BL91708); capuEY12344 343 [41]; spire2F/CyO (BL8723); spire1 (gift from I. Palacios). 344 345 Preparing oocytes for live imaging 346 Ovaries from flies fattened for 48 hours on yeast were dissected onto a 22 by 40 mm cover slip, into 347 series 95 halocarbon oil using forceps (11251-30 Dumont #5 forceps, Fine Science Tools) and a 348 dissecting probe (0.25 mm straight 10140-01, Fine Science Tools) as described previously [42]. 349 Mature oocytes were gently teased out of the ovaries and left for 10 minutes prior to imaging to allow 350 them to settle onto the coverslip. 351 352 Preparing in vivo activated eggs 353 Ovaries from flies fattened for 48 hours on yeast were carefully dissected onto a 22 by 40mm cover 354 slip, such that the oviduct was still intact, into series 95 halocarbon oil. Oocytes were then gently 355 teased out of the oviduct, carefully removing the surrounding oviduct tissues. 356 357 Preparation of fixed samples 358 Ten to twenty ovaries were dissected from flies fattened for 48 hours on yeast into Schneider’s Insect 359 Medium (Sch) (GibCo). Ovaries were splayed open, and oocytes gently teased out using fine forceps 360 (11251-30 Dumont #5 forceps, Fine Science Tools) and a dissecting probe (0.25 mm straight 10140-361 01, Fine Science Tools). Mature oocytes were transferred into a 0.5mL eppendorf tube using a glass 362 pipette. Sch was removed and 500 μL 4% paraformaldehyde (PFA) stabilized with phosphate buffer 363 (Thermofisher Scientific) was added for 10-15 minutes on a rotary machine (PTR-35 360 vertical 364 multi-function rotator, Thermofisher Scientific). Oocytes were then washed for 10 minutes, three 365 times in 0.1% PBST (0.1% Triton X-100 (ThermoFisher Scientific) in PBS. Oocytes were then 366 incubated for 2 hours in 1% PBST at room temperature with the following labelling probes: Alexa-367 Fluor Phalloidin 568, 1:500 (Molecular Probes); Alexa-Fluor Phalloidin 637, 1:500 (Molecular 368 Probes); ChromoTek GFP-booster, 1:500 (Proteintech). When SiR-Actin is utilized, fixed oocytes 369 were incubation for 40 minutes in 1% PBST with 1 μM SiR-Actin Probe 652 (Spirochrome) at 37°C. 370 http://flybase.org/reports/FBti0100219 12 of 19 Oocytes were then washed, stained in glycerol and DAPI and mounted on a glass slide in 371 Vectashield with DAPI (Vector Laboratories). 372 373 Ex vivo egg activation 374 Mature oocytes were activated ex vivo through addition of the hypotonic, 260 mOsm, activation 375 buffer (AB): 3.3 mM NaH2PO4, 16.6 mM KH2PO4, 10mM NaCl, 50 mM KCl, 5% polyethylene glycol 376 (PEG) 8000, 2mM CaCl2, brought to pH 6.4 with a 1:5 ratio of NaOH:KOH [34]. Oocytes typically 377 activate within two minutes of addition of AB. For fixation or drugs treatments, mature oocytes were 378 activated with AB for 10 minutes before further processing. 379 380 Pharmacological treatments 381 All pharmacological incubations were validated with either a Schneider’s Insect Medium (Sch) or 382 PBS control and a DMSO control made to the appropriate dilution. Oocytes were incubated in Sch 383 or PBS for 10 minutes whereupon samples were flooded with AB and visualized. 384 CytoD (Sigma Aldrich) was made to a final concentration of 2-20 μM in Sch or PBS. Oocytes were 385 dissected into Sch or PBS a glass bottom dish. The Sch or PBS was carefully removed and replaced 386 with the cytoD. Oocytes were incubated for 10 to 30 minutes in this solution prior to fixation or live 387 imaging. Colchicine (Sigma Aldrich) was made to a final concentration of 50 μM in Sch or PBS. 388 Oocytes were incubated for at least 30 minutes in this solution prior to fixation or live imaging, as 389 above. 390 391 Imaging with the Inverted Olympus FV3000 system 392 A 1.05 NA 30X silicone objective was used for whole oocyte imaging and a 1.35 NA 60X silicone 393 objective for visualizing intracellular components. For high resolution imaging of the spindle, oocytes 394 were oriented on the coverslip such that the dorsal appendages were in contact with the surface of 395 the cover slip, therefore the dorsal side of the oocyte becomes the shallowest plane of visualization. 396 Parameters for image collection were: 1.35 NA 60x silicon immersion objective, 10 μm Z-stack, 397 0.5 μm between each Z-slice, 1024×1024 pixels, approximately 15 seconds per stack. 398 399 FRAP analysis 400 For FRAP of spindle components Jup-GFP was bleached for 10 seconds. Time lapse series of 401 recovery was recorded every 5 seconds in single plane imaging of the cortex or every 30 seconds 402 in Z-stack imaging of the spindle, both using the 488nm laser channel, 2 Airy unit pinhole, 1024x1024 403 pixels. For all FRAP series, background correction was performed by subtracting the fluorescence 404 intensities of the unbleached cytoplasmic area from fluorescent intensities of bleached regions, with 405 percentage fluorescence of the maximum plotted in graphs. 406 13 of 19 FIGURE 1 407 408 Fig. 1 A spindle actin population is present in the metaphase-arrested Drosophila mature oocyte. 409 410 A-D. Schematic and confocal Z-projection (10 μm) labeled for microtubules (Jup-GFP, green), actin 411 (SiR-Actin, white), and DNA (DAPI, cyan). In stage 12 oocytes prior to egg maturation (A,A¢) the 412 DNA is condensed (white arrowhead, pointing right) and actin appears to enrich around the nucleus 413 (white arrowhead, pointing left). (A¢) n = 24 stage 12 oocytes and the depicted features were 414 observed in 24 out of 24 samples (24/24). In stage 13 at maturation (B-B¢¢) actin spikes penetrate 415 the nucleus (B¢, white arrowhead) and then surround the DNA (B¢¢, white arrowhead actin inset) with 416 the microtubule material appearing diffuse (B¢¢). (B¢) n = 12/15 early stage 13 oocytes (B¢¢) n = 3/3 417 late stage 13 oocytes. In stage 14 metaphase I arrested oocytes (mature oocytes) (C,C¢) the 418 microtubule spindle forms and a population of actin, spindle actin, appears to associate with 419 microtubules. Insets show induvial channels. (C¢) n = 125/130 mature oocytes. Higher resolution 420 image of early metaphase I arrest nucleus (D), actin appears to form a cage around the microtubules 421 and DNA. In some cases, the 4th non-exchange chromosomes are visible as a smaller mass at each 422 tip of the main body of chromosomes. (D) n = 56/60 mature oocytes. 423 424 E-F. Confocal Z-projections (10 μm) labeled for actin (Act5C-GFP, green) (E), (Lifeact-GFP, green) 425 (F), and DNA (DAPI, cyan) of a mature oocytes. The MI stage is marked by the spindle becoming 426 parallel in orientation to the cortex. The cortical actin is visualized here as vertical green band of 427 fluorescence between the oocyte nucleus and the follicle cell nuclei (far left) (E). The spindle actin is 428 visible surrounding a central chromosomal mass using both markers (E, F). Insets show induvial 429 channels, n > 15 mature oocytes per genetic actin marker. 430 431 G. Confocal Z-projection (10 μm) labeled for Ca2+ (GCaMP3, green) and DNA (DAPI, cyan) of a MI 432 oocyte. Higher calcium signal detected at the tip of the spindle pole, n = 15 mature oocytes. 433 434 H-I. Confocal Z-projections (10 μm) from live time series labeled for actin (UtrCH-GFP) (H) or 435 microtubules (Jup-GFP) (I) of MI oocytes. Similar structures and orientations are observed for actin 436 and microtubule as compared to fixed samples. n > 37 mature oocytes per genetically encoded 437 marker. 438 439 Scale bar: 10 μm (A-C, H-I), 5 μm (D-G). 440 14 of 19 FIGURE 2 441 442 Fig. 2 The actin cytoskeleton promotes maintenance of spindle microtubules and regulation of 443 chromosomal alignment. 444 445 A-F. Confocal Z-projections (10 μm) from live time series labeled for actin (UtrCH-GFP) or 446 microtubules (Jup-GFP) of MI oocytes treated with colchicine (A-B), cytoD (C-D) and BAPTA-AM 447 (E-F). Microtubules (A) first appear as a typical spindle structure (t = 0¢) and depolymerize post 448 colchicine treatment (t = 30¢), whereas spindle actin (B, white arrowhead) is not affected. (A-B) n = 449 10 mature oocytes per genetically encoded marker treated. Before addition of cytoD (t = 0¢), the 450 microtubule spindle appears as a typical elliptical structure of approximately 10 μM (C). Post-cytoD 451 treatment (t = 10¢) the spindle has undergone a distinct morphological change as it elongates (C). 452 Spindle actin (D, white arrowhead) appears parallel to the cortex before the addition of cytoD (t = 0¢) 453 and becomes depolymerized post treatment (t = 10¢). (C-D) n = 9 mature oocytes per genetically 454 encoded marker treated. Both microtubule spindle (E) and spindle actin (F, white arrowhead) 455 depolymerize post BAPTA-AM treatment (t = 15¢), suggesting calcium is required to maintain the 456 metaphase spindles. (E,F) n = 8 mature oocytes per genetically encoded marker treated. 457 458 G-J. Confocal Z-projections (10 μm) labeled for microtubules (Jup-GFP, green), actin (SiR-Actin, 459 white), and DNA (DAPI, cyan) (G,I) or centromere (Cid-EGFP, green) and DNA (DAPI, cyan) (H,J) 460 of MI oocytes incubated in control media (G-H) or cytoD (I-J). Spindle actin and the microtubule 461 spindle appear to surround the chromosomes in MI oocytes incubated in control media (G). 462 Centromeres are detected on each aligned chromosome (H, white arrowhead). (G-H) n = 40 mature 463 oocytes per genetically encoded marker. Spindle actin depolymerizes post-cytoD treatment and the 464 microtubule spindle is significantly elongated (I). Distance between centromeres also increases and 465 the number of centromeres is unequally distributed between the two poles (J). (I,J) n = 50 mature 466 oocytes per genetically encoded marker treated, (I) unpaired t-test, P value < 0.005. Insets show 467 induvial channels (G,H,K). 468 469 K. Confocal Z-projections (10 μm) labeled for microtubules (Jup-GFP, green), actin (SiR-Actin, 470 white), and DNA (DAPI, cyan) of MI oocytes incubated in BAPTA-AM. A loss of the spindle and a 471 round chromosomal mass in close proximity of the cortex are detected post-treatment. n = 10 mature 472 oocytes. Insets show induvial channels. 473 474 Scale bar: 10 μm (A-K). 475 15 of 19 FIGURE 3 476 477 Fig. 3 The spindle actin is required for recruitment and the position of the metaphase I spindle and 478 to maintain chromosomal integrity. 479 480 A-B. Confocal Z-projections (10 μm) from live time series of WT, capuEY12344/+, and capuEY12344/spire1 481 or capuEY12344/spire2F in MI oocytes, labelled for actin (UtrCH-GFP) (A) or microtubules (Jup-GFP) 482 (B). While the cortical actin (A, right panel, top right corner) could still be observed, spindle actin is 483 not detected in the heterozygous or trans-heterozygous mutants. (A) n = 30 mature oocytes per 484 genotype. Microtubules are elongated (B) and no longer parallel to the cortex in both the 485 heterozygous and the trans-heterozygous mutants. (B) n = 30 mature oocytes per genotype. (Note: 486 capuEY12344 referred to as capuEY in panels in all figures). 487 488 C. Confocal Z-projections (10 μm) labeled for microtubules (Jup-GFP, green) and DNA (DAPI, cyan) 489 of capuEY12344/+ (left) and capuEY12344/spire2F (right) in MI oocytes. DNA stains in both mutants reveal 490 a separation of the chromosomal mass into two units that begin to migrate along the spindle axis. 491 Insets show induvial channels. n = 20 mature oocytes per genotype. 492 493 D. Confocal Z-projection (10 μm) labeled for calcium (GCaMP3, green) and DNA (DAPI, cyan) of a 494 capuEY12344/+ MI oocyte. A loss of the calcium signal in a spindle shape and separation of the 495 chromosomes into two separate masses are detected. Insets show induvial channels. n = 13 mature 496 oocytes. 497 498 E-G. Graphs summarising the proportion of microtubule spindle phenotype (E), maximum 499 chromosomal distance and microtubule spindle length (F), and the angle of spindles from cortex (G), 500 for WT, post-colchicine, post- cytoD, capuEY12344/+ and capuEY12344/spire2F MI oocytes. The proportion 501 of oocytes with a WT microtubule spindle phenotype are significantly decreased after drug treatment 502 and in mutant backgrounds (E). Most microtubule spindles found in the experimental groups are 503 either disrupted (elongated in cytoD and mutants or shortened in colchicine) or not detected. We do 504 see a proportion of WT oocytes with no spindle detected, which is likely a technical issue due to the 505 orientation of the oocyte or labelling efficiency. (E) n > 16 mature oocytes per treatment or genotype, 506 Fishers Exact Test, P value <0.05. Comparison of the maximum chromosomal distance and the 507 microtubule spindle length indicate a significant increase in cytoD treated and mutant backgrounds 508 compared to WT (F). In the few mature oocytes where microtubule spindle where still detected 15 509 minutes after colchicine addition (n = 11 mature oocytes), the spindle length is reduced without any 510 significant impact on the maximum chromosomal distance. (F) n >16 mature oocytes per treatment 511 or genotype, student’s t-test, ***< 0.005. Comparison of the spindle-cortex angle (degrees) indicates 512 a significant increase in cytoD, colchicine treated, and mutant background compared to WT. (G) n 513 >16 mature oocytes per treatment or genotype, student’s t-test, ***< 0.005. 514 515 H. Recovery of fluorescence intensity following photobleaching of microtubules in WT and 516 capuEY12344/spire2F mutant. WT and mutant oocytes initially show similar recovery dynamics, 517 however, a significant difference between the mutant and WT is observed over time (after 270 s). n 518 = 6 mature oocytes per genotype, student’s t-test, *< 0.05. 519 520 Scale bar: 10 μm (A-D). 521 16 of 19 FIGURE 4 522 523 Fig. 4 The spindle actin population is dynamic at egg activation. 524 525 A. Schematic and confocal Z-projection (10 μm) labeled for centromere (Cid-EGFP, green) and DNA 526 (DAPI, cyan) (A¢) or microtubules (Jup-GFP, green), actin (SiR-Actin, white), and DNA (DAPI, cyan) 527 (A¢¢) in early (A¢) and later (A¢¢) anaphase I (AI) oocytes. Post egg activation, the chromosomes start 528 to separate to opposite poles while the spindle first increases in width (A¢) and then in length (A¢¢). 529 The completion of AI is marked by the spindle becoming perpendicular in orientation to the cortex. 530 Insets show induvial channels. (A¢) n = 34/34 early activated mature oocytes. (A¢¢) n = 20/20 late 531 activated mature oocytes. 532 533 B. Confocal Z-projection (10 μm) labeled for calcium (GCaMP3, green) and DNA (DAPI, cyan) of an 534 AI oocyte. Calcium signal spread evenly across the elongated spindle with separating chromosomal 535 mass. n = 10 activated mature oocytes. 536 537 C. Confocal Z-projections (10 μm) labeled for actin (UtrCH-GFP) from live time series of MI oocytes 538 (left, t = 0¢) incubated in activation buffer (AB) (right, t = 5¢). Similar changes in shape and rotation 539 observed as with fixed samples. n = 25 activated mature oocytes. 540 541 Scale bar: 5 μm (A-C). 542 17 of 19 FIGURE 5 543 544 Fig. 5 The spindle actin is required for accurate segregation of chromosomes during anaphase I. 545 546 A-B. Confocal Z-projections (10 μm) labeled for actin (Lifeact-GFP, green) and DNA (DAPI, cyan) 547 of AI oocytes incubated in control media (A) and cytoD (B). Spindle actin surrounds the segregating 548 AI chromosomes in control media (A), but appears less defined than in MI oocytes, and is not 549 detected after cytoD treatment (B). Chromosome segregation phenotype is disrupted in cytoD 550 treated oocytes, many showing chromosomal units separated from the main mass (B¢-B¢¢) compared 551 to untreated controls (A¢-A¢¢). Insets show induvial channels and additional chromosomal examples. 552 n = 30 activated mature oocytes per treatment. 553 554 C-D. Confocal Z-projections (10 μm) labeled for microtubules (Jup-GFP, green) and DNA (DAPI, 555 cyan) of capuEY12344/+ (C) and capuEY12344/spire2F (D) AI oocytes. Microtubule spindle are detected in 556 both backgrounds, but chromosome segregation is misregulated. Examples of the variety of 557 disrupted chromosome segregation phenotypes indicate chromosomal units separated from the 558 main mass (C¢-C¢¢, D¢-D¢¢). Insets show induvial channels and additional chromosomal examples. n 559 = 17 activated mature oocytes per genotype. 560 561 E-G. Graphs summarising the number of aberrant chromosomal masses (E), maximum 562 chromosomal distance (F), and microtubule spindle length (G) for WT, cytoD treated, capuEY12344/+, 563 and capuEY12344/spire2F AI oocytes. To quantify the aberrant chromosomal masses, we first defined 564 the main chromosomal mass as the aligned materials in close association (< 1 μm), not including 565 chromosome 4. Aberrant chromosomal masses were counted when material was disconnected from 566 main mass by > 1 μm or when present > 1 μm outside of the spindle shape observed in wild type 567 segregation. We also completed a 3D analysis to check the Z distance of > 1 μm from the main 568 mass. Comparison of the number of aberrant chromosomal masses shows a significant increase in 569 cytoD treated and mutant backgrounds compared to the WT. (E) n > 20 mature oocytes per treatment 570 or genotype, student’s t-test, * < 0.05. Comparison of the maximum chromosomal distance and the 571 microtubule spindle length shows no significant difference in cytoD treated and mutant backgrounds 572 compared to the WT. (F-G) n > 30 mature oocytes per treatment or genotype, student’s t-test, not 573 significant (ns). 574 575 H. Graph summarising the angle of spindles from cortex before (left) and after (right) egg activation 576 in WT and cytoD treated oocytes. Comparison of the angle between the spindle and cortex shows 577 cytoD only has a significant effect on MI oocytes. There are no significant differences between the 578 WT AI oocyte, and the cytoD treated MI and AI oocytes. n > 20 mature oocytes per treatment and 579 stage, student’s t-test, not significant (ns). 580 581 Scale bar: 5 μm (A-D). 582 18 of 19 REFERENCES 583 1. Handel, M.A. & Schimenti, J.C. Genetics of mammalian meiosis: regulation, dynamics and 584 impact on fertility. Nat. Rev. Genet. 11, 124-136 (2010). 585 2. Bennabi, I., Terret, M.E. & Verlhac, M.H. Meiotic spindle assembly and chromosome segregation 586 in oocytes. J. Cell. Biol. 215, 611-619 (2016). 587 3. Yi, K. et al. Sequential actin-based pushing forces drive meiosis I chromosome migration and 588 symmetry breaking in oocytes. J. Cell. Biol. 200, 567-576 (2013). 589 4. Field, C.M. & Lenart, P. Bulk cytoplasmic actin and its functions in meiosis and mitosis. Curr. 590 Biol. 21, 825-830 (2011). 591 5. Mori, M. et al. Intracellular transport by an anchored homogeneously contracting F-actin 592 meshwork. Curr. Biol. 21, 606-611 (2011). 593 6. Azoury, J., et al. Spindle positioning in mouse oocytes relies on a dynamic meshwork of actin 594 filaments. Curr. Biol. 18, 1514-1519 (2008). 595 7. Azoury, J. et al. Symmetry breaking in mouse oocytes requires transient F-actin meshwork 596 destabilization. Development 138, 2903-2908 (2011). 597 8. Nixon, V.L. et al. Ca(2+) oscillations promote APC/C-dependent cyclin B1 degradation during 598 metaphase arrest and completion of meiosis in fertilizing mouse eggs. Curr. Biol. 12, 746-750 599 (2002). 600 9. Bolcun-Filas, E. & Handel, M.A. Meiosis: the chromosomal foundation of reproduction. Biol. 601 Reprod. 99,112-126 (2018). 602 10. Avilés-Pagán EE, Orr-Weaver TL. Activating embryonic development in drosophila. Seminars 603 in Cell & Developmental Biology. 2018;84:100–10. doi:10.1016/j.semcdb.2018.02.019 604 11. McLaughlin, J.M. & Bratu, D.P. Drosophila melanogaster Oogenesis: An Overview. Methods 605 Mol. Biol. 1328, 1-20 (2015). 606 12. Hughes, S.E. et al. Female Meiosis: Synapsis, Recombination, and Segregation in Drosophila 607 melanogaster. Genetics 208, 875-908 (2018). 608 13. Giovanni Bosco, T.L.O.-W. The cell cycle during oogenesis and early embryogenesis in 609 Drosophila. Advances in Developmental Biology and Biochemistry 12,107-154 (2002). 610 14. Theurkauf, W.E. & Hawley R.S. Meiotic spindle assembly in Drosophila females: behavior of 611 nonexchange chromosomes and the effects of mutations in the nod kinesin-like protein. J. Cell. 612 Biol. 116,1167-1180 (1992). 613 15. York-Andersen, A.H. et al. Osmolarity-regulated swelling initiates egg activation in Drosophila. 614 Open. Biol. 11, p210067 (2021). 615 16. Hu, Q. & Wolfner M.F. The Drosophila Trpm channel mediates calcium influx during egg 616 activation. Proc. Natl. Acad. Sci. 116, 18994-19000 (2019). 617 17. Endow, S.A. & Komma D.J. Spindle dynamics during meiosis in Drosophila oocytes. J. Cell. Biol. 618 137,1321-1336 (1997). 619 18. Gilliland WD, Hughes SF, Vietti DR, Hawley RS. Congression of achiasmate chromosomes to 620 the metaphase plate in drosophila melanogaster oocytes. Developmental Biology. 621 2009;325(1):122–8. doi:10.1016/j.ydbio.2008.10.003 622 19. Majekodunmi, A., Bowen A.O. & Gilliland W.D., Comparative Cytology of Female Meiosis I 623 Among Drosophila Species. G3 (Bethesda) 10, 1765-1774 (2020). 624 20. Riparbelli MG, Callaini G. The meiotic spindle of thedrosophilaoocyte: The role of Centrosomin 625 and the central aster. Journal of Cell Science. 2005;118(13):2827–36. doi:10.1242/jcs.02413 626 21. Mogessie, B. & Schuh M., Actin protects mammalian eggs against chromosome segregation 627 errors. Science 357, (2017). 628 22. Roeles, J. & Tsiavaliaris G. Actin-microtubule interplay coordinates spindle assembly in human 629 19 of 19 oocytes. Nat. Commun. 10, p4651 (2019). 630 23. Laband K, Le Borgne R, Edwards F, Stefanutti M, Canman JC, Verbavatz J-M, et al. 631 Chromosome segregation occurs by microtubule pushing in oocytes. Nature Communications. 632 2017;8(1). doi:10.1038/s41467-017-01539-8 633 24. Mori M, Somogyi K, Kondo H, Monnier N, Falk HJ, Machado P, et al. An arp2/3 nucleated F-634 actin shell fragments nuclear membranes at nuclear envelope breakdown in starfish oocytes. 635 Current Biology. 2014;24(12):1421–8. doi:10.1016/j.cub.2014.05.019 636 25. Burdyniuk M, Callegari A, Mori M, Nédélec F, Lénárt P. F-actin nucleated on chromosomes 637 coordinates their capture by microtubules in oocyte meiosis. Journal of Cell Biology. 638 2018;217(8):2661–74. doi:10.1083/jcb.201802080 639 26. Kelso, R.J., Hudson, A.M. & Cooley L. Drosophila Kelch regulates actin organization via Src64-640 dependent tyrosine phosphorylation. J. Cell. Biol. 156, 703-713 (2002). 641 27. Riedl, J. et al., Lifeact: a versatile marker to visualize F-actin. Nat. Methods 5, 605-607 (2008). 642 28. York-Andersen, A.H. et al. A calcium-mediated actin redistribution at egg activation in 643 Drosophila. Mol. Reprod. Dev. 87, 293-304 (2020). 644 29. Li, R. et al., Spindle function in Xenopus oocytes involves possible nanodomain calcium 645 signaling. Mol. Biol. Cell 27, 3273-3283 (2016). 646 30. Nakai, J., Ohkura, M. & Imoto, K. A high signal-to-noise Ca(2+) probe composed of a single 647 green fluorescent protein. Nat. Biotechnol. 19, 137-141 (2001). 648 31. Schuh M, Lehner CF, Heidmann S. Incorporation of drosophila CID/CENP-A and CENP-C into 649 centromeres during early embryonic anaphase. Current Biology. 2007;17(3):237–43. 650 doi:10.1016/j.cub.2006.11.051 651 32. Chang, C.W. et al., Anterior-posterior axis specification in Drosophila oocytes: identification of 652 novel bicoid and oskar mRNA localization factors. Genetics 188, 883-896 (2011). 653 33. Drechsler, M. et al. Optical flow analysis reveals that Kinesin-mediated advection impacts the 654 orientation of microtubules in the Drosophila oocyte. Mol. Biol. Cell 31, 1246-1258 (2020). 655 34. Mahowald AP, Goralski TJ, Caulton JH. In vitro activation of drosophila eggs. Developmental 656 Biology. 1983;98(2):437–45. doi:10.1016/0012-1606(83)90373-1 657 35. Dahlgaard K, Raposo AASF, Niccoli T, St Johnston D. Capu and spire assemble a cytoplasmic 658 actin mesh that maintains microtubule organization in the drosophila oocyte. Developmental 659 Cell. 2007;13(4):539–53. doi:10.1016/j.devcel.2007.09.003 660 36. Quinlan ME. Direct interaction between two actin nucleators is required in drosophila oogenesis. 661 Development. 2013;140(21):4417–25. doi:10.1242/dev.097337 662 37. Aaboud, M. et al., Measurement of the Inelastic Proton-Proton Cross Section at sqrt[s]=13 TeV 663 with the ATLAS Detector at the LHC. Phys. Rev. Lett. 117, p182002 (2016). 664 38. Grossman H, Har-Paz E, Gindi N, Levi M, Miller I, Nevo N, et al. Regulation of GVBD in mouse 665 oocytes by Mir-125a-3p and Fyn kinase through modulation of actin filaments. Scientific Reports. 666 2017;7(1). doi:10.1038/s41598-017-02071-x 667 39. Descazeaud, V. et al., Calcineurin regulation of cytoskeleton organization: a new paradigm to 668 analyse the effects of calcineurin inhibitors on the kidney. J. Cell. Mol. Med. 16, 218-227 (2012). 669 40. Kaneuchi, T. et al. Calcium waves occur as Drosophila oocytes activate. Proc. Natl. Acad. Sci. 670 112, 791-796 (2015). 671 41. Rauzi, M.,Lenne, P.F. & Lecuit, T. Planar polarized actomyosin contractile flows control epithelial 672 junction remodelling. Nature 468,1110-1114 (2010). 673 42. Weil, T.T., Parton, R.M. & Davis I. Preparing individual Drosophila egg chambers for live imaging. 674 J Vis Exp. 60, (2012). 675 Final Manu Fig1 Fig2 Fig3 Fig4 Fig5