Time-lapse in utero imaging of the meiotic spindle and an endoplasmic reticulum marker (ER) revealed that the ER is distributed throughout the zygote but appears as undulating lines during metaphase I (n=8), changes to a dispersed appearance during anaphase I (n=9), transitions back to undulating lines interspersed with large blobs during metaphase II (n=6), then changes to a dispersed pattern during anaphase II (n=5) (Figure 1A; Figure 1—video 1). Previous electron microscopy studies have indicated that the undulating lines correspond to sheet-like ER and the dispersed appearance corresponds to tubular ER (Poteryaev et al., 2005; Gong et al., 2024). A similar transition to sheet-like ER during mitotic M phase has been reported in HeLa cells (Lu et al., 2009) and Xenopus egg extracts (Wang et al., 2013), however, this may not be universal in all cell types (Puhka et al., 2012). Pertinent to this study, ER morphology was used to determine cell-cycle stages during live imaging reported below in Figure 5.

In striking contrast with the ER filling the entire zygote, yolk granules are packed inward, away from the cortex (n=9 metaphase I; Figure 1B) as previously described (McNally et al., 2010). When Mitotracker-treated males were mated to hermaphrodites expressing GFP-labeled yolk granules, the paternal mitochondria were found in a discrete cloud around the paternal DNA (n=34; Figure 1B–C) as previously described (Sato and Sato, 2011). Both the cloud of paternal mitochondria (n=19) and the meiotic spindle (n=9 metaphase I) were observed in cortical regions that are free of yolk granules (Figure 1B). Maternal mitochondria labeled with COX-4::GFP were also packed inward, away from the cortex, and were excluded from the cloud of paternal mitochondria and excluded from the spindle (n=5 metaphase I; Figure 1C). Time-lapse imaging (n=21) revealed that the cloud of paternal mitochondria remained together and separate from maternal yolk granules even when moving long distances with cytoplasmic streaming (Figure 1D).

Because yolk granules and maternal mitochondria are packed inward during meiosis (Figure 1B and C) and sperm must enter from the outside, it is possible that the exclusion of maternal yolk granules from the volume of paternal mitochondria might simply be a consequence of inward packing. In kca-1(RNAi) meiotic embryos, yolk granules (McNally et al., 2010) and mitochondria (Figure 2A and B) do not pack and instead extend to the plasma membrane. Maternal mitochondria were still excluded from the volume of paternal mitochondria in kca-1(RNAi) embryos (n=11 control, 8 RNAi; Figure 2B). This result indicated that the volume of paternal mitochondria excludes maternal mitochondria. In addition, paternal mitochondria also formed a discrete cluster in meiotic embryos of a nematode species with giant sperm (Vielle et al., 2016) (n=4; Figure 2C) indicating that the unique properties of the ball of paternal organelles in the meiotic zygote are conserved.

In contrast with maternal yolk granules and maternal mitochondria, maternal ER was observed penetrating into the cloud of paternal mitochondria and enveloping the paternal DNA in a shell that appeared as a ring in confocal sections of both live and fixed meiotic embryos (n=10; Figure 3A). To elucidate how the maternal ER enters the ball of paternal organelles, we monitored seven instances of sperm-egg fusion by time-lapse microscopy. Previous studies have demonstrated that C. elegans sperm fuse with the egg as opposed to being phagocytosed (Takayama and Onami, 2016). Mitotracker-labeled males were mated with hermaphrodites with a maternally-expressed ER marker and a maternally-expressed plasma membrane marker (mCherry::PH). 20 s after the apparent entry of paternal mitochondria into the egg (0:20 in Figure 3B), a sperm-sized volume devoid of maternal ER was observed within the maternal plasma membrane. One minute after apparent sperm-egg fusion, the maternal ER had moved into the plasma membrane behind the sperm (1:25 in Figure 3B; 1:00 in Figure 3C). The maternal ER ring enveloping the sperm DNA was not discernible until 7 min after apparent sperm-egg fusion (7:24 in Figure 3C), although it might form earlier because the zygote undergoes dramatic movement through two sequential sphincters into the spermatheca then into the uterus during this time. The envelopment of the sperm DNA by maternal ER may be an initial step in nuclear envelope assembly (Penfield et al., 2020; Barger et al., 2023), which has been halted until after the completion of meiosis. These results show that maternal ER is initially excluded from the sperm at fusion. Since maternal mitochondria and yolk granules are excluded later, this suggests that all maternal membranes are initially excluded from the sperm at fusion. The plasma membrane marker diffuses in behind the sperm first, followed by plasma-membrane-associated ER, followed by envelopment of the sperm DNA by maternal ER by 7 min after fusion. Maternal yolk granules and maternal mitochondria, however, are excluded from the sperm volume for much longer (Figure 1).

If the maternal ER envelope around sperm DNA was sealed and impermeable during meiosis, this could both prevent the sperm DNA from inducing ectopic spindle assembly and prevent the sperm DNA from interacting with meiotic spindle microtubules. To test whether the ER envelope around the sperm DNA is permeable to proteins, we analyze BAF-1, a chromatin-binding component of the inner nuclear envelope (Gotzmann and Foisner, 1999). Maternally provided GFP::BAF-1 was not detected on maternal or paternal chromatin during metaphase I or metaphase II but labeled the surface of both maternal and paternal chromatin during anaphase I and anaphase II (Figure 4A, B and C). Because BAF-1 is a chromatin-binding protein, this result indicates that BAF-1 can pass freely through holes in the ER envelope surrounding the sperm DNA between metaphase I and anaphase I.

C. elegans meiotic cytoplasmic streaming (McNally et al., 2010; Panzica et al., 2017; Yang et al., 2003; Kimura et al., 2017) has the potential to bring the sperm contents into close proximity with the meiotic spindle, however, a previous study found that the male pronucleus very rarely forms at the same end of the embryo as the female pronucleus (Kimura and Kimura, 2020). This indicates that limitations on cytoplasmic streaming might be involved in maintaining distance between the sperm contents and the meiotic spindle. We, therefore, asked whether increasing cytoplasmic streaming would cause collisions between the meiotic spindle and sperm contents. Meiotic cytoplasmic streaming requires microtubules (Yang et al., 2003) and kinesin-1 (McNally et al., 2010). Depletion of katanin by mei-1(RNAi) increases the mass of cortical microtubules and the extent of yolk granule streaming (Kimura et al., 2017) and depletion of kinesin-13 by klp-7(RNAi) also increases the mass of cortical microtubules (Gigant et al., 2017). We found that mei-1(RNAi) or klp-7(RNAi) increased the maximum displacement of the sperm contents in both the long and short axes of the ellipsoid embryo relative to control L4440(RNAi) embryos (Figure 5A, B, C and D; Figure 5—videos 1–3). The only demonstrably significant difference between the excessive cytoplasmic streaming in these two depletions was that excessive streaming in klp-7(RNAi) persists for a longer period of time, well into metaphase II (Figure 5D).

Among 15 control L4440(RNAi) time-lapse sequences, the closest center-to-center distance between spindle and sperm contents was 18 µm. In contrast, among 51 time-lapse sequences of mei-1(RNAi) meiotic embryos, the sperm came within 5.5 µm center to center distance of the spindle in 12 cases, and in 12/12 of these cases, the sperm stopped moving relative to the spindle indicating a capture event (Figure 5B; Figure 5—video 2). Among a subset that could be tracked through the pronuclear stage, a single male pronucleus formed adjacent to multiple small female pronuclei. Among 25 time-lapse sequences of klp-7(RNAi) embryos, the sperm DNA became stuck against the meiotic spindle (4.6 µm center to center distance) in only one case. In all other cases, the sperm streamed past a stationary spindle. However, the closest sperm to spindle distance was 7.7 µm in one case. These results indicate that limiting cytoplasmic streaming is important for maintaining a distance between spindle and sperm and thus preventing the capture of the sperm by the meiotic spindle. However, mei-1(RNAi) spindles are apolar and do not undergo normal polar body extrusion, and the single capture event in a klp-7(RNAi) embryo resulted in cell cycle arrest. These results, therefore, do not reveal what would happen to paternal DNA captured by a normal meiotic spindle during polar body extrusion.

An important question is why the sperm contents move more than the meiotic spindle. During metaphase I, when the ER is sheet-like, the movement of the sperm contents is limited. When the ER disperses during anaphase I, the spindle rotates and one pole is moved closer to the cortex by cytoplasmic dynein (Figure 1—video 1; Ellefson and McNally, 2011). In contrast, the distance of the sperm contents from the cortex (Figure 5—figure supplement 1) and the movement of the sperm contents both increase when the ER disperses (Figure 5C; Figure 5—video 4).

Because maternal ER penetrates the sperm contents and envelops the sperm DNA, and because the C. elegans ortholog of ataxin-2, ATX-2, affects ER organization (Del Castillo et al., 2019), we analyzed the contribution of ATX-2 to the cohesion of the sperm contents in meiotic embryos. We first introduced an auxin-induced degron and GFP tag to the 3’end of the endogenous atx-2 gene. Endogenously tagged ATX-2 was observed throughout oocytes and meiotic embryos (Figure 6B and C; Figure 6—figure supplement 1). ATX-2 did not uniquely co-localize with ER (Figure 6—figure supplement 1). Dark holes were observed suggesting exclusion from the lumens of larger membranous organelles (Figure 6C; Figure 6—figure supplement 1). We then compared the intensity of the GFP signal after three different depletion treatments, 1- hr auxin, 24 hr GFP(RNAi), or 27 hr atx-2(RNAi). All three methods resulted in a significant reduction of the GFP signal (Figure 6B, C and D).

The extent of scattering of paternal mitochondria was analyzed in Z-stacks of fixed meiotic embryos depleted of ATX-2 by each of the three methods. This analysis was restricted to embryos from anaphase I through anaphase II because our streaming data (Figure 5C) and that of Kimura (Kimura and Kimura, 2020) indicate that the sperm contents had not moved significantly before anaphase I. Example Z-projections are shown in Figure 7. Mitochondria can exist as a tubular network (Okamoto and Shaw, 2005) and tubules in close proximity cannot always be resolved by light microscopy. To account for the apparent heterogeneous sizes of foci of paternal mitochondria, the fluorescence intensity of larger foci was divided by the fluorescence intensity of the smallest foci such that a focus with three times the fluorescence intensity was counted as three ‘mitochondria.’ ATX-2 depletion by each of the three methods resulted in a significantly increased mean distance of paternal mitochondria from the sperm DNA (Figure 8A) as well as significantly increased standard deviation of individual distances (Figure 8B). Standard deviation is an important measure because an increase in the number of mitochondria both closer and further from the sperm DNA would result in no change in the mean. It remains possible that paternal mitochondria scatter is caused by pressure applied during fixation, pressure resulting from ovulation through the spermatheca valves, or cytoplasmic streaming. We were not able to unambiguously track scattering by live imaging because significant movement occurs during the acquisition of a complete z stack and because the embryos cannot withstand the additional photodamage. These results still support the hypothesis that ATX-2 is required to maintain the integrity or cohesiveness of the ball of paternal mitochondria that surrounds the sperm DNA to resist external forces.

Because we observed increased cytoplasmic streaming in klp-7(RNAi) and disruption of the integrity of the sperm contents after ATX-2 depletion, we hypothesized that double depletion of KLP-7 and ATX-2 would result in an increased frequency of spindle microtubules capturing the sperm DNA. Among 24 time-lapse sequences of atx-2(AID +auxin) klp-7(RNAi) meiotic embryos, the ER envelope around the sperm DNA moved extensively in all cases but came within a threshold distance of 5.5 µm (center to center) of the meiotic spindle in only five cases. In 5/5 of these cases, a bundle of microtubules extended from the meiotic spindle into the ER envelope around the sperm DNA, and the sperm DNA became stuck in this position (Figure 9A and B). In one case, the ER envelope around the sperm DNA was transiently captured by the meiosis I spindle and stretched in the direction of streaming (Figure 9C). The sperm DNA then released, continued streaming, and was captured by the meiosis II spindle. All cases of stable capture caused a cell-cycle arrest so that the consequences of polar body extrusion could not be determined.

Because ATX-2 depletion alters ER morphology, we were not able to score cytoplasmic streaming with the cell-cycle accuracy shown in Figure 5. However, the maximum long-axis and short-axis displacement at any cell-cycle time was increased in atx-2(AID) with auxin vs without auxin (Figure 9—figure supplement 1). No spindle capture events were observed, however, among 15 ATX-2 single depletion time-lapse sequences and the closest distance between sperm and spindle was 8.2 µm with 1 hr auxin and 19.3 µm without auxin.

The frequency of sperm capture by the meiotic spindle (Figure 9D) was significantly higher than wild-type controls in klp-7(RNAi) atx-2(AID) double-depleted embryos (p=0.011 Fisher’s exact test). Although the number of single mutant embryos analyzed was too low to demonstrate a significant difference between single and double mutant embryos, these results qualitatively support the hypothesis that limiting cytoplasmic streaming and maintaining the integrity of the ball of paternal mitochondria are both important for preventing capture events between the meiotic spindle and sperm DNA.

Genotypes of strains used in this study are listed in Supplementary file 1. atx-2(syb5389; ATX-2::AID::GFP) was generated by SunyBiotech using CRISPR/Cas9.

Adult hermaphrodites were anaesthetized with tricaine/tetramisole in PBS as described (McCarter et al., 1999; Kirby et al., 1990) and then placed on 2% agarose pads on slides. Extra anesthetic was gently pipetted onto the agarose pad and a coverslip was placed on top. The slide was inverted and placed on the stage of an inverted microscope. Meiotic embryos were identified by bright-field microscopy before initiating time-lapse fluorescence. For all live imaging, the stage and immersion oil temperature were 21°C–24°C. For all time-lapse data, single-focal plane images were acquired with a Solamere spinning disk confocal microscope equipped with an Olympus IX-70 stand, Yokogawa CSU10, either Hamamatsu ORCA FLASH 4.0 CMOS (complementary metal oxide semiconductor) detector or Hamamatsu ORCA-Quest qCMOS (quantitative complementary metal oxide semiconductor) detector, Olympus 100 x UPlanApo1.35 oil objective, 100 mW Coherent Obis lasers (405, 640, 488, 561 nm) set at 30% power, and MicroManager software control. Pixel size was 65 nm for the ORCA FLASH 4.0 CMOS detector and 46 nm for the ORCA-Quest qCMOS detector. Exposures were 200ms for the ORCA FLASH 4.0 qCMOS detector and 100ms for the ORCA-Quest qCMOS detector. Time interval between image pairs or trios was 5 s. Focus was adjusted manually during time-lapse imaging.

For the ATX-2 images in Figure 6, z-stacks of –1 oocytes of anesthetized live worms were captured with a Zeiss LSM 980 confocal microscope with Airyscan 2 and a Zeiss Objective LD LCI Plan-Apochromat 40 x/1.2 Imm Corr DIC M27 for water, silicon oil or glycerine.

C. elegans meiotic embryos were extruded from hermaphrodites in 0.8x egg buffer by gently compressing worms between a coverslip and a slide, flash frozen in liquid N2, permeabilized by removing the coverslip, and then fixed in ice-cold methanol before staining with antibodies and DAPI. The primary antibodies used in this work were mouse monoclonal anti-tubulin (DM1α; Thermo Fisher Scientific; 1:200) and rabbit anti-GFP (NB600-308; Novus Biologicals; 1:600). The secondary antibodies used were Alexa Fluor 488 anti-rabbit (A-21206; Thermo Fisher Scientific; 1:200), Alexa Fluor 488 anti-mouse (A-21202; Thermo Fisher Scientific; 1:200), and Alexa Fluor 594 anti-mouse (A-21203; Thermo Fisher Scientific; 1:200). Z-stacks were captured at 1 μm steps for meiotic embryos using the same microscope described above for live imaging.

For sperm ER ring filming, the ring was observed throughout meiosis I and II and manually kept in the focal plane by adjusting the stage (z-axis). Filming typically began at spermatheca exit or meiosis I and ended at pronuclear formation or cell arrest. Due to the rigors of filming an embryo with optimal orientation and positioning in the uterus, videos that started and ended mid-phases were still filmed and included in measurements so long as a complete phase was included between starting and ending filming (e.g. AI and MII were measured in videos starting mid-MI and ending mid-AII). In these cases when filming started mid-phase and/or ended prematurely mid-phase, the incompletely filmed phases were not included in quantifications. The phases at which cell arrest occurred were not included in quantifications, but the phases prior were still used.

Measurements of the ER rings’ dynamics in videos of embryos were tracked manually using the Fiji plugin MTrackJ. All embryos quantified were rotated such that at the beginning of each video the pole containing the meiotic spindle was at the left. Only embryos measured at >35 µm long were used for ER ring tracking. Embryos measured <35 µm were tilted and not ideal for tracking. The starting and ending frames of each cell phase were then determined based on ER morphology. Individual tracks for each phase were manually made by clicking on the center of the ER ring, or in cases when it was briefly out of focus, the mitochondria that most closely followed where the ER ring was previously seen. The following criteria was used for determining phases to be quantified by tracks:

Displacement was measured as the distance between the first point of a phase’s track and the last point.

Distance Traveled was measured as the total length of the track that the sperm ring traveled.

Maximum Average Velocity was measured by taking the maximum of 3-point moving averages of velocities in each phase. MTrackJ was used to measure the velocity of the ER ring between each video frame.

Duration of Phases were calculated by subtracting the first frame number of a phase from the last one and multiplying by 5 due to the 5- s intervals.

Paternal mitochondrial scattering was calculated by using a circular ROI to measure the mean value fluorescence of Deep Red MitoTracker FM or Red MitoTracker CMXros labeling paternal mitochondria in meiotic embryos. Mean pixel values were taken using a circle ROI with an 18-pixel diameter to cover the entire area of a region of fluorescence that appeared as a punctum (unless specified otherwise due to a different resolution). This was repeated until all puncta fluorescence in an embryo were measured. For every embryo, a region of cytoplasm without mitochondria had its fluorescence measured. This value was subtracted from every mitochondrial measurement to correct for noise. The distances between the center of each punctum and the center of the paternal DNA were then recorded with a line tool. The Pythagorean theorem was used to determine the distance when the mitochondria were in a different z-stack than the paternal DNA. These distances were recorded in column scatter graphs to observe the distribution of mitochondrial distance from sperm DNA quantitatively. To measure large amorphous masses of mitochondria, the same sized ROI measuring fluorescence in distinct puncta was used to cover a portion of the mass, make measurements, and then moved to another portion to make more measurements. This was repeated until the entire area of the mass had its fluorescence measured. The corresponding distances to the paternal DNA were recorded for every ROI used. The mean value fluorescence of the amorphous masses were typically much greater than the distinct puncta and as such presumably had a greater density of mitochondria per ROI than the puncta. In order to account for this density, the mean values measured from the ROI’s over the masses were divided by the average of the mean values of all of the puncta in an embryo. This resulting number was then used to determine how many times the distance between the paternal DNA and the ROI was recorded in the column scatter graphs (e.g. if the rounded value was 2 then the distance was recorded twice). The averages and standard deviations of each distribution of mitochondrial distances from paternal DNA in the embryos were then measured in order to compare the scattering of mitochondria in different experimental treatments. Quantifications were done in meiotic embryos of all phases except metaphase I. Since this phase is right after fertilization, we believe embryos do not have the time to exhibit scattering as a phenotype.

L4 fog-2(q71) males were picked onto an OP50 plate and treated with 200 µL 0.05 mM working stock of MitoTracker Deep Red FM (Invitrogen) or MitoTracker Red CMXros (Invitrogen) in M9 overnight. Subsequently, all the males were moved to a fresh OP50 plate and incubated for 15 minu to ‘wash’ away excess MitoTracker. This washing step was conducted three times to fully remove excess MitoTracker. After the third wash, the males were then moved to plates with hermaphrodites 24 hr before they were to be filmed or used for immunofluorescence. A minimum ratio of 1:1 males to hermaphrodites was used for matings.

L4 sdhc-1::mCherry; him-5(e1490) males were added to plates with hermaphrodites 24 hr before they were to be filmed. A minimum ratio of 1:1 males to hermaphrodites was used for matings.

At least 20 hr before imaging, hermaphrodites expressing HaloTag were treated with 100 µl of 2.5 µM Janelia Fluor HaloTag Ligand 646 or 549 in M9 added to the bacterial lawn of a 60 mm MYOB agar plate.

For RNA interference, L4 hermaphrodites were placed on RNAi plates with an RNAi bacterial lawn for a set period of time before being used for live imaging or fixed slides. In 48 hr treatment, the worms were moved to a fresh RNAi plate after 24 hr. RNAi plates were always seeded the day before adding worms. For mei-1, atx-2, gfp, and L4440 (RNAi), worms fed on RNAi bacterial lawns for 24–28 hr. For klp-7 (RNAi) and its corresponding control L4440 (RNAi), worms fed on RNAi bacterial lawns for 48–52 hr.

For auxin-induced degradation, L4’s of strains endogenously tagged with auxin-inducible degrons and a TIR1 transgene were placed on a fresh plate of OP50. After 24 hr, the hermaphrodites were moved to auxin plates with lawns of OP50 for 1–3 hr before use in live or fixed experiments. 4 mM auxin plates were made by adding 400 mM auxin (indole acetic acid) in ethanol to molten agar which was then poured and seeded with OP50 bacteria. Depletion of ATX-2::AID::GFP was confirmed by the reduction of ATX-2::AID::GFP signal in –1 oocytes.

In AID klp-7 RNAi experiments, L4’s were placed on klp-7 (RNAi) and then transferred to a fresh klp-7 (RNAi) lawn after 24 hr. After a total of 47 hrs of treatment, the worms were moved to auxin/RNAi plates seeded with klp-7 (RNAi). 1 hr later the worms were then used in live or fixed experiments. Auxin/RNAi plates consisted of 4 mM auxin, 1 mM IPTG, and 200 ug/mL ampicillin in agar. Stocks of 400 mM auxin, 1 M IPTG, and 200 mg/mL ampicillin were added to molten agar and mixed to create the plates.

The depletion of atx-2 was measured by taking the mean value of an ROI over the cytoplasm of –1 oocytes, taking caution to not include the nucleus. The z-stack at which the nucleus appeared most in focus was used for measurements in each embryo. The mean value of the cytoplasm was then subtracted by the mean value of an area not containing any part of the worm to correct for noise.

GFP::BAF-1 fluorescence was measured by manually tracing the outline of chromosomes during metaphase I and II and then taking the mean value fluorescence. The same ROI was also used to measure the mean value fluorescence of the middle of the embryo’s cytoplasm. The mean value of the fluorescence over the chromosomes was then divided by the cytoplasmic mean value in order to measure and compare BAF-1 fluorescence in metaphase I vs anaphase I.

Shapiro-Wilks tests through GraphPad Prism were used to test for normality in all data in which statistical tests were used to compare means. If the test determined the data was normal, p-values were calculated in GraphPad Prism using Welch’s T-tests for comparing means of only two groups and ANOVA tests for comparing means of three or more groups. If the data was not normal, p-values were calculated in GraphPad Prism using Mann-Whitney tests for comparing means of only two groups and Kruskal-Wallis tests for comparing means of three or more groups.