Male meiotic spindle poles are stabilized by TACC3 and cKAP5/chTOG differently from female meiotic or somatic mitotic spindles in mice

Transforming acidic acid coiled-coil protein 3 (TACC3) and cytoskeleton associated protein 5 (cKAP5; or colonic hepatic tumor overexpressed gene, chTOG) are vital for spindle assembly and stabilization initiated through TACC3 Aurora-A kinase interaction. Here, TACC3 and cKAP5/chTOG localization with monospecific antibodies is investigated in eGFP-centrin-2- expressing mouse meiotic spermatocytes. Both proteins bind spermatocyte spindle poles but neither kinetochore nor interpolar microtubules, unlike in mitotic mouse fibroblasts or female meiotic oocyte spindles. Spermatocytes do not display a liquid-like spindle domain (LISD), although fusing them into maturing oocytes generates LISD-like TACC3 condensates around sperm chromatin but sparse microtubule assembly. Microtubule inhibitors do not reduce TACC3 and cKAP5/chTOG spindle pole binding. MLN 8237 Aurora-A kinase inhibitor removes TACC3, not cKAP5/chTOG, disrupting spindle organization, chromosome alignment, and impacting spindle pole γ-tubulin intensity. The LISD disruptor 1,6-hexanediol abolished TACC3 in spermatocytes, impacting spindle bipolarity and chromosome organization. Cold microtubule disassembly and rescue experiments in the presence of 1,6-hexanediol reinforce the concept that spermatocyte TACC3 spindle pole presence is not required for spindle pole microtubule assembly. Collectively, meiotic spermatocytes without a LISD localize TACC3 and cKAP5/chTOG exclusively at spindle poles to support meiotic spindle pole stabilization during male meiosis, different from either female meiosis or mitosis.

Understanding the mechanism of spindle assembly and maintenance between male or female meiosis and mitosis, including commonalities and distinctions, will permit a greater understanding of how we can avoid chromosomal separation errors that lead to infertility, genetic and developmental defects, miscarriages, and cancers.Such studies remain exciting avenues for understanding the fundamental differences in meiotic spindle assembly and stabilization between male and female meiosis and somatic cell mitosis.Accurate bipolar spindle assembly in meiosis and mitosis is vital for faithful chromosome alignment and segregation to avoid generating chromosomal aneuploidy errors that can form compromised cells or gametes.

Results
Two monospecific TACC3 and one polyclonal rabbit cKAP5/chTOG antibody was used to track these spindle proteins in mouse spermatogenic cells (Suppl Fig. 1S).We relied primarily on the rabbit monoclonal TACC3 antibody, owing to its greater specificity in immunolabeling experiments using somatic mouse CF-1 fibroblasts and meiotic oocytes in mice.The distinctions in the staining patterns observed in these different cell types investigated along with antibody validation criteria (Suppl Fig. S1) provided confidence in these antibodies to track TACC3 and cKAP5/chTOG in spermatocytes.
The distinct TACC3 staining patterns observed between male and female meiotic and mitotic fibroblasts spindles is shown in Fig. 2. In meiotic spermatocytes, TACC3 often interfaces between the spindle pole γ-tubulin and the lateral surface of a centriole doublet but does not bind to intrapolar or kinetochore spindle microtubules nor the plus ends of microtubules near kinetochores (Fig. 2, panel 1).In meiotic oocytes, TACC3 is concentrated in the LISD extending beyond the spindle lattice and spindle microtubules, but without specific binding to spindle pole MTOC's or maternal GFP CETN2-expressing foci (Fig. 2, panel 2).KTMTs strongly labeled with TACC3 in female meiotic spindles, but not microtubule plus-ends near kinetochores.In mitotic fibroblasts, TACC3 is bound to the spindle pole centrosomes and intrapolar as well as KTMT in the spindle proper, but typical not microtubule plus-ends near kinetochores.
The divergent patterns of TACC3 and cKAP5/chTOG detection observed in meiotic or mitotic spindles suggest that these proteins may have distinctly different roles in supporting spindle organization, stabilization, and functions.To explore this further, spermatocytes were fused into a mouse GV-stage meiotic oocyte using HVJ-E Sendai extract fusion protein to investigate how TACC3 protein would organize at the paternal centrosome within the maternal cytoplasmic environment.After confirming fusion by differential interference microscopy (DIC; Fig. 3A,D), oocytes resumed meiosis by washout of dbcAMP.Round spermatids (RS) mostly silence GFP CETN2 expression and never immunolabel with TACC3 antibody (Fig. 3A, upper inset; see Suppl Fig. S2, panel 3).By 2 h post fusion, the incorporated RS showed GFP CETN2-expressing centrioles in the oocyte's cytoplasm surrounded by large maternal TACC3 condensates, but no microtubule organization as observed around the condensing female chromosomes and assembling TACC3 LISD (Fig. 3A-C).By 4-h post-oocyte fusion, weak disorganized microtubules surrounded the paternal GFP CETN2-expressing centrioles and slightly decondensed sperm nucleus as extensive maternal TACC3 condensates surrounded the paternal DNA and co-immunostained the sparse microtubules, similar in appearance to TACC3 staining of the female spindle microtubules and LISD (Fig. 3D-F).Thus, spermatocytes that typically restrict TACC3 binding to their centrosomes assemble TACC3 LISD condensates around the paternal DNA when introduced into the maternal cytoplasmic environment, despite poorly organized microtubules and incomplete DNA decondensation.
To further explore spindle pole TACC3 role in spindle microtubule assembly in spermatocytes, microtubule depolymerization and recovery experiments in the presence of 1,6-hexanediol were performed.Spermatocytes were placed in ice-cold media for 15 min to depolymerize spindle microtubules, then recovered in warm culture media (34C) with or without 3.5% 1,6-hexanediol before examining spindle pole TACC3 and spindle microtubules in fixed samples (Fig. 7).Compared to controls, 15 min in ice-cold media completely depolymerized spindle microtubules, significantly reduced spindle pole TACC3 detection, and induced chromosome misalignment (Fig. 7, panel 1 D-F; panel 2; second bar; **; p < 0.011470).Rescue experiments in warm culture media for 15 min restored spindle pole TACC3 even greater than in control spermatocytes (Fig. 7, panel 1 G-I; panel 2, third bar; ***; p < 0.006156) and assembled robust intrapolar and kinetochore microtubules with aligned meiotic chromosomes, although the spindle microtubules did not label with TACC3 (Fig. 7, panel 1H).When recovery experiments were performed for 15 min in the presence of 3.5% 1,6-hexanediol, strong centrosomal astral microtubule assembled from both poles, despite no spindle pole TACC3 detection or chromosome realignment (Fig. 7, panel 1 J-L).3-dimensionnal image rotations of rescued spermatocytes in the presence of 1,6-hexanediol showed abundant cortical and cytoplasm TACC3 foci after 15 min in warm culture conditions without specific binding to spindle pole centrioles or reformed spindle microtubules (Fig. 7, panel 1 K-L; panel 2, 4th bar; ***; p < 0.001419).Thus, spindle pole TACC3 may not be required for initial microtubule polymerization from spindle pole centrosomes, but bipolar organization with proper chromosome alignment may require TACC3.
Further experiments explored the role of microtubule dynamics on spindle pole TACC3 and cKAP5/chTOG in male meiotic spindles (Suppl Fig. S5).Meiotic spermatocytes exposed to 10 µM nocodazole microtubule depolymerization agent for 30 min showed cells without assembled microtubules but retained spindle pole TACC3 and cKAP5/chTOG detection in close association with GFP CETN2-exrpessing centrioles (Suppl Fig. S5; panel 1 D-F and panel 2 D-F).Likewise, 30 min in 10 µM paclitaxel, a microtubule stabilizing agent that significantly increases spindle microtubules, did not remove spindle pole TACC3 or cKAP5/chTOG detection, but appeared to reduce tight centriole association (Suppl Fig. S5, panel 1 G-I and panel 2 G-I).Analysis of spindle pole TACC3 and cKAP5/chTOG after nocodazole or paclitaxel application did not show significant reduction in these spindle pole proteins compared to untreated control spermatocytes (Suppl Fig. S5, panel 3).These results imply that microtubule dynamics may not play a significant role in spindle pole TACC3 or cKAP5/chTOG retention in male meiotic spermatocytes.
Figure 8 summarizes our major findings on TACC3 and cKAP5/chTOG detection in spermatocytes and how disrupting or recovery of spindle pole TACC3 impacts spindle organization and integrity in male meiotic spindles.

Discussion
In a recent important study of male meiosis, spermatocytes from AURKA conditional knockout mouse lacked centriole colocalized TACC3 despite maintaining normal cellular TACC3 protein levels, underscoring the importance of Aurora A kinase in TACC3 centrosome recruitment to male meiotic spindle poles.TACC3 disruption appeared to induce misoriented chromosomes in meiosis-I spermatocytes 18 .But no information has yet been reported on the role(s) of TACC3 and cKAP5/chTOG in spermatocyte meiotic spindle assembly and stabilization during critical stages of chromosomal divisions.
No meiotic male spindle LISD was observed in mouse spermatocytes as observed in female oocytes (Fig. 1) 47 .We were curious if a male spermatocyte centrosome introduced into the female oocyte cytoplasm could attract TACC3 LISD condensates to the paternal centrioles to assemble a bipolar spindle with aligned paternal chromosomes (Fig. 3).From fusions produced and culture in vitro, we first found extensive maternal TACC3 LISD aggregation surrounding the male centriole doublets, remarkably like TACC3 detected in male meiotic spindle poles (Fig. 1, panel 2).However, by 4-h post-fusion, the TACC3 LISD aggregates co-immunolabel with sparse, disorganized microtubules, not the unduplicated paternal centrioles, near the incomplete decondensed male chromosomes, a pattern like the LISD around female condensing chromosomes.The paternal centrosome is most likely inactive in spindle organization within the oocyte's cytoplasm, consistent with observations showing that mouse oocytes repress centrioles and their activities during meiotic stages 14 .Thus, mouse spermatocytes introduced into maturing mouse oocytes assemble a TACC3 LISD condensate like female spindles, but the paternal LISD is insufficient to support a fully functioning paternal bipolar spindle, perhaps suggesting other significant spindle protein recruitment is required to complete proper spindle assembly.
In mitotic cells, siRNA TACC3 depletion studies severely impacted spindle pole astral microtubule assembly by interfering with γ-TuRC recruitment to the centrosomes.However, no major impact on spindle bipolarity was reported despite some reduced spindle microtubule density and misaligned chromosomes (Gergely et al. 29 ; Singh et al. 25 ; Rajeev et al. 26 ).Meiotic spermatocytes exposed to the AURKA inhibitor MLN 8237 lost spindle pole TACC3 detection (Fig. 4 and 5; Wellard et al. 18 ).Spermatocytes, which do not assemble spindle pole astral microtubules, showed significant reductions of meiotic spindle microtubule density, bipolar organization, poleto-pole spindle length, and misaligned meiotic chromosomes after MLN 8237 depletion of TACC3.Aberrant spindle phenotypes observed in MLN 8237 exposed spermatocytes appeared more severe in comparison to human mitotic cells after TACC3 depletion 25,26,29 but perhaps like oocyte meiotic spindles from AURKA or pericentrin knockout mice lacking the ability to assemble the spindle TACC3 LISD 45,46 .
Spermatocyte spindle pole TACC3 depletion by MLN 8237 did not fragment the PCM or produce supernumerary spindle poles, as reported in Aurora A RNAi depletion experiments in mitotic cells 53 .However, we observed a measurable increase in spindle pole γ-tubulin fluorescent intensity compared to spermatocytes recovered from MLN 8237 exposure for 1-h, although not from untreated control spermatocytes (Fig. 4, panel 4).This observation is surprising given the critical role phosphorylated TACC3 plays in regulating assembly of γ-TuRC for controlling microtubule nucleation rev,28 .Additionally, spindle microtubules in spermatocytes treated with MLN 8237 were largely disassembled despite this increase in spindle pole γ-tubulin intensity.The opposite occurred upon MLN 8237 washout-partial spindle microtubule recovery with a significant decrease in spindle pole γ-tubulin intensity (Fig. 4, panel 1).In siRNA-TACC3 depleted human cells, centrosomal γ-tubulin detection was either not impacted or significantly reduced at spindle pole 25,26,28,54 .More recently, human mitotic cells expressing TACC3 shRNA-resistant mutant proteins engineered to be incapable of binding ch-TOG at the spindle poles, thus uncoupling TACC3 from cKAP5/chTOG activity at the centrosomes, also demonstrated an increase in spindle pole γ-tubulin because of elevated spindle pole TACC3 phosphorylation increasing γ-TuRC protein recruitment to the poles 27 .Here, it is not clear why spindle pole γ-tubulin intensity increased after MLN 8237 disruption of TACC3.Perhaps experimental variability is a factor, given our spermatocyte cells are treated for an hour with MLN 8237 inhibitor as opposed to transfection protocols requiring hours-to-days to initiate impacts on spindle pole TACC3.Alternatively, perhaps other members of the TACC family of proteins (i.e.TACC1 or TACC2) could be active and compensating for the spindle pole TACC3 disruption in spermatocytes [55][56][57] .
Spermatocyte spindle pole TACC3 disruption by MLN 8237 also showed slightly reduced spindle pole cKAP5/ chTOG with a significant increase of cKAP5/chTOG fluorescent intensity detected within the central spindle apparatus, but without specific binding to assembled spindle microtubules (Fig. 5).In human cells, siRNA-TACC3 spindle pole disruption did not impact centrosomal chTOG but did significantly reduced spindle microtubule ch-TOG binding 29 .More recently, uncoupling spindle pole TACC3 from ch-TOG significantly reduce spindle pole ch-TOG detection with a concomitant increase in TACC3-free ch-TOG binding to intra-spindle microtubules 27 .Taken together, the TACC3: cKAP5/chTOG dyad are critical regulators in male meiotic spindle pole organization with direct impacts on spindle microtubule organization, density, spindle pole length, and meiotic chromosomal alignment, observations unlike reports in human mitotic cells 29,58,59 but perhaps like reports in mouse meiotic spindles [45][46][47] .
In mouse oocytes, TACC3 is concentrated in a liquid-like condensate (LISD) disassembled by exposure to the disrupting agent 1,6-hexanediol 51,52 .Mouse oocyte spindles treated with 1,6-hexanediol depleted a majority of TACC3 LISD but remained bipolar with TACC3 detected on spindle pole microtubules 47 .In spermatocytes, application of 3.5% 1,6-hexanediol for 30 min removed all spindle pole TACC3 detection, disrupting normal bipolar spindle organization, spindle pole centriole GFP CETN2 expression and/or locality, and chromosome alignment (Fig. 6).Unlike MLN 8237 AURKA inhibitor treatment, spermatocytes significantly recovered centrosomal TACC3 after hexanediol washout, with good spindle bipolarity, tightly focused spindle pole microtubules but overall reduced microtubule density and unaligned chromosomes (Fig. 6, panel 1).Analysis after recovery from 1,6-hexanediol showed significantly higher spindle pole microtubule intensity measurements when TACC3 could reform at the centrosomes as opposed to recovered spermatocytes from MLN 8237 treatment that did not restore spindle pole TACC3, although both treatments were significantly reduced in spindle pole microtubules compared to control spermatocytes (Fig. 6, panel 3).Finally, we investigated male meiotic spindle assembly in cold depolymerization-rescue experiments in the presence of 1,6-hexanediol that depletes spindle pole TACC3 (Fig. 7).We observed robust microtubule aster assembly from the sperm centrosomes at both spindle poles, despite severely reduced centrosomal TACC3 detection but abundant cytoplasmic or cortical TACC3 aggregates.Yet, we observed incomplete bipolar spindle assembly and no meiotic chromosome realignment.Taken together, these findings support observations that spindle pole TACC3 in spermatocytes does not influence initial microtubule assembly from male spermatocyte centrosomes but plays a critical role in organizing spindle poles to support and stabilize male meiotic bipolar spindles.
Spermatocyte spindle pole TACC3 and cKAP5/chTOG appear independent of spindle microtubule dynamics based on analysis of spermatocytes treated with the microtubule depolymerization drug nocodazole or the microtubule stabilizing drug paclitaxel.Neither drug treatment significantly impacted centrosomal TACC3 or cKAP5/chTOG detection in spermatocytes (Suppl Fig. S5), supporting observations in human mitotic cells 29,62 .But, in mouse female oocytes, nocodazole exposure removed meiotic spindle microtubules to expose TACC3 protein, but not cKAP5/chTOG, in the unique LISD 43,47 .Taxol enhanced meiotic oocyte spindles, including astral microtubules at the poles, were all immunolabeled with cKAP5/chTOG 43 .We conclude that microtubule dynamics probably are not significant in maintaining spindle pole TACC3 or cKAP5/chTOG localization in male meiosis, as reported in mitotic cells, but distinctly different from mouse female meiotic oocytes.
Figure 8 summaries our preliminary findings for TACC3 and cKAP5/chTOG in male meiotic spindles.We provide evidence for similarities and differences on spindle assembly and maintenance mechanisms for preserving accurate chromosome segregation in male and female meiosis versus somatic mitotic cells.TACC3 and cKAP5/chTOG are remarkable divergence in spindle localization with overlapping but also unique functions in supporting spindle integrity.With nearly 200 essential microtubule associated proteins identified mediating spindle assembly and disassembly activities, the foundations for understanding the molecular underpinnings of meiotic and mitotic spindle assembly in directing proper chromosome segregation lay ahead.Understanding their commonalities and distinctions will help avoid chromosomal separation errors that lead to infertility, genetic and developmental defects, as well as cancers.(G-I) A metaphase-II spermatocyte exposed for 15 min to cold conditions plus 15 min rescue in warm conditions shows bipolar spindle reassembly (G, H: microtubules, red) with kinetochore microtubules (H: red, arrows), realigning chromosomes (G: DNA, blue) and spindle pole TACC3 (H, I: blue, arrowheads) at the GFP CETN2 centrioles (G, I: green, arrowheads).Assembling spindle kinetochore microtubules do not label with TACC3 (H: blue, arrows).(I) insets: details, GFP CETN2 centrioles (green) and TACC3 (blue).(J-L) A metaphase-I spermatocyte exposed for 15 min to cold conditions plus 15 min rescue in warm conditions in the presence of 3.5% 1,6-hexanediol shows extensive cytoplasmic TACC3 without binding at the GFP CETN2 centrioles (J, L: green, arrowheads), significant microtubule assembly from both spindle poles (J, K: red) with kinetochore microtubules (K: red, arrows) and still misaligned chromosomes (J: DNA, blue).The reassembling kinetochore microtubules do not label with TACC3 (K: blue, arrows).(L) Insets: details, GFP CETN2 centrioles (green) and TACC3 (blue).Scale bars = 5 µm.Panel 2: graphic analysis of TACC3 spindle pole detection shows significant deviation from non-treated controls following cold exposure (**; p < 0.01147), cold with 15 min rescue (***; p < 0.006156), and cold with 15 min rescue in presence of 3.5% 1,6-hexanediol (***; p < 0.01419).approved methods by the American Veterinary Medical Association (AVMA) and our University-approved IACUC protocols.Tissues were harvested within min post-euthanization.We investigated isolated spermatogenic cells from 54 testes from 27 male mice and produced 983 spermatogenic cells for analysis.
GFP expression determination by PCR.Genomic DNA was isolated for GFP detection in GFP CETN2expressing mice, using tail tip tissues (< 5 mm) and PCR with MyTaq Extract-PCR Kit (Bioline, Taunton, MA) as previously described 17 .
Figure 8. Male meiotic spindle organization is lost after MLN 8237 Aurora A kinase inhibitor and 1,6-hexanediol disruption of spindle pole TACC3.Meiotic spermatocytes have anastral bipolar metaphase spindles with aligned chromosomes and canonical centrosomes at both spindle poles.TACC3 and cKAP5/ chTOG function as a dyad at the poles, strongly interfacing between the γ-tubulin pericentriolar material and centriole doublets.TACC3 phosphorylation is almost certainly necessary to bind the dyad to spindle poles 34 .Spindle pole TACC3 appears required for male meiotic spindle bipolarity, as exposure to MLN 8237 Aurora A kinase inhibitor (top panel) or hexanediol (middle panels) shows reduce, shorter, and highly disarrayed spindle microtubules with misaligned chromosomes.Spindle pole organization is also impacted by TACC3 disruption, with MLN 8237 exposure slightly enhancing spindle pole γ-tubulin intensity but reducing spindle pole cKAP5/ chTOG and shifting the protein to the spindle lattice, although without significant binding on the remaining intact microtubules.1,6-hexanediol exposure either displaces centrioles from the spindle pole or removes centriole GFP CETN2-expression at the disorganized bipolar spindles.Interestingly, TACC3 does not appear required for spindle pole microtubule assembly, as recovery from MLN 8237 AURKA inhibitor did not block partial microtubule recovery, including assembly of kinetochore microtubules, despite no detection of spindle pole TACC3 after washout.Conversely, rescue from 1,6-hexanediol did restore centriole-associated spindle pole TACC3 with better spindle pole microtubule organization.However, neither recovery from MLN 8237 inhibitor nor 1,6-hexanediol instituted chromosome realignment to the spindle equator.Cold microtubule disassembly and regrowth experiments in the presence of 1,6-hexanediol (lower panels) reinforces that spindle pole TACC3 is not required for microtubule assembly as spindle pole astral microtubules reassemble without TACC3 detection at the centrioles despite abundance of TACC3 protein detection in the cytoplasm.Here, TACC3 does not bind to centrioles perhaps because it cannot be phosphorylated to target the centrosomes in the presence of 1,6-hexanediol.AURKA: Aurora A kinase.
For transduction with lentiviral CETN2-GFP (a generous gift from Dr. Jeffrey Salisbury; Mayo Clinic, Rochester, MN), viral particles were produced using the ViraPower Lentiviral Packaging Mix (Thermo Fisher) according to the manufacturer's recommendations.CF-1 primary mouse embryonic fibroblast cell lines were transduced after plating cells into a T-25 cell culture-treated flask at 37 °C overnight with viral particles and 6 mg/mL of polybrene (Millipore).After 24 h, transduced cells were washed once with PBS, and transduction medium was replaced with Feeder Cell Complete cell culture medium.Confirmation of centriole and centrosome expression was confirmed by indirect immunofluorescence on a Nikon inverted microscope equipped with appropriate filters for visualizing GFP.
Fixation of GFP CETN2 CF-1 primary fibroblasts was accomplished in 1% paraformaldehyde in DMEM without protein for 40 min at 37 °C after overnight seeding on sterile 22-mm 2 coverslips.After PBS washes, coverslips were permeabilized, blocked, and stained as described below.

Mature germinal vesicle (GV)-stage oocyte collection and zona-free in vitro fusion with spermatogenic cells, using Sendai extract
Mature GV-stage oocytes were collected from GFP CETN2-expressing CB6F1 ovaries in EmbryoMax M2 (EMD Millipore, Billerica, MA) in the presence of 100 µg/ml dibutyryl adenosine 3′,5′-cyclic monophosphate (dbcAMP; Sigma-Aldrich) to prevent in vitro maturation as previously described 15 .Briefly, sterile excised ovaries were minced mechanically using sterile needles to collect GVs.Cumulus cells were removed with a 75-µm tip (Stripper pipet; Origio Mid-Atlantic Diagnostics, Trumbull, CT).Mature GV oocytes ≥ 65 µm in diameter were segregated from the population of GV's after determining diameters using Elements software (Nikon USA, Melville, NY) and after image capture on a Nikon Digital Sight camera (DS Fi1).GV oocyte maturation was initiated after 3× washes in M-2 and placing mature GVs in advanced KSOM (Sigma-Aldrich) at 37 °C in a humified 5% CO 2 incubator until fixation.
To accomplish zona-free GV oocyte in vitro fusion with isolated spermatogenic cells, we utilized an inactivated Sendai virus cell fusion kit (HVJ Envelope: HVJ-E; Cosmo Bio, Tokyo, Japan) as previously described 17 .Briefly, zona-free GV oocytes were prepared by a 35-to 45-s treatment with warm EmbryoMax acidic Tyrode's culture medium (EMD Millipore), washed 3 × in M-2, and then placed in individual 15-μl droplets of ice-cold 1× fusion buffer containing a 1:25 dilution of Sendai HVJ-E extract in a sterile 35-mm petri dish coated with parafilm to prevent zona-free oocyte adhesion.Isolated GFP-CETN2-expressing spermatocytes were laid over the zona-free oocytes in Sendai extract and pipetted repeatedly to adhere spermatocytes to the GV oocytes.www.nature.com/scientificreports/Fusion of the spermatocyte with zona-free GV oocyte couplet was accomplished by gently flooding with 500-μl of warm fusion buffer at 37 °C for 15 min.After fusion incubation, oocytes were transferred individually to 15-μl droplets of advanced KSOM medium under mineral oil until fixation at 2-4 h post-fusion and processed for immunocytochemistry analysis as described below.
Cytoskeletal, Aurora A kinase inhibitor, cold and hexanediol treatment 10-µM stock solutions of nocodazole and paclitaxel (Sigma-Aldrich), along with 5 mM stock of MLN 8237 Aurora A Kinase inhibitor (Cayman Chemical, Ann Arbor, MI) were prepared in DMSO and stored at − 80 °C.All drug stocks were diluted to final concentrations in mEKRB.Rescue experiments from inhibitors for isolated spermatogenic cells were performed by centrifuging cells at 125×g for 3 min, discarding supernatant (inhibitor), and adding back 2 ml of inhibitor-free mEKRB to cell pellets for 15-60 min recovery at 34 °C.Cold treatment was performed in ice-cold water at 4 °C in mEKRB for 15 min, with rescue experiments performed by a 15-min recovery at 34 °C.Hexanediol (Sigma-Aldrich) was suspended in mEKRB at a final concentration of 3.5% just before use 47 .

Equipment, imaging, analysis, and settings
Imaging of fixed slides was accomplished as described 17 .Briefly, images were collected with a Nikon A1 fourlaser line confocal microscope equipped with Elements A1 Plus compact GUI acquisition software (version 4.20; Nikon USA) at 1024 × 1024 size at ¼ frame per second, using a pinhole size of 79.2 mm and a z-depth of 0.25 mm through the entire spermatogenic cell, with a differential interference contrast (DIC) Plan Fluor ×100 (1.4 NA) objective.We collected 5-× 12-bit depth images (nd2 files), using the same laser photomultiplier tube settings for each channel across specimens (5% laser power, except UV, for DNA imaging, at 10.24%) to facilitate comparison between meiotic spermatogenic control versus drug-treated cells.Fluorescent intensity ratios, surface intensity plots, area, or volume measurements were performed on binarized images, using the threshold tool and region-of-interest statistical menu in the Elements software, and downloaded to Microsoft Excel for statistical analysis.For image panel presentation, generated confocal nd2 files were first denoised in the A1 software before performing a subtracted background image, collected from outside of the spermatogenic cells.All channels were then subjected to the deconvolution software module in Elements (Landweber algorithm) using the point-scan confocal command and same filter (noisy) at twenty iterations for all images.Final panels from deconvolved images were prepared in Photoshop (Adobe Systems, San Jose, CA).

Statistics
Means ± standard deviations were determined using calculator.net(Maple Tech International, LLC; The Woodlands, TX).We used Excel (Microsoft; Redmond, WA) to prepare graphs and box plots, which show median (horizontal lines), 25th and 75th percentiles (small boxes), and min and max (whiskers)
14:4808 | https://doi.org/10.1038/s41598-024-55376-z Statistical significance was determined by Student's T-test (GIGA calculator; Web Focus, LLC; Sofia, Bulgaria).Significance was determined at p < 0.05.Graphical analyses shown are indicative of average values ± standard deviation.For most experiments, more than three trials were performed, and data are representative of all trials.