Programmed mitophagy at the oocyte-to-zygote transition promotes lineage endurance

Siddharthan B Thendral; Sasha Bacot; Ian T Ryde; Katherine S Morton; Qiuyi Chi; Isabel W Kenny-Ganzert; Joel N Meyer; David R Sherwood

doi:10.1038/s41556-025-01854-z

. Author manuscript; available in PMC: 2026 Feb 20.

Published in final edited form as: Nat Cell Biol. 2026 Jan 12;28(2):268–284. doi: 10.1038/s41556-025-01854-z

Programmed mitophagy at the oocyte-to-zygote transition promotes lineage endurance

Siddharthan B Thendral ¹, Sasha Bacot ², Ian T Ryde ², Katherine S Morton ², Qiuyi Chi ¹, Isabel W Kenny-Ganzert ¹, Joel N Meyer ², David R Sherwood ^1,^*

PMCID: PMC12918445 NIHMSID: NIHMS2137744 PMID: 41526527

Abstract

The quality of mitochondria inherited from the oocyte determines embryonic viability, lifelong metabolic health of the progeny, and lineage endurance. High levels of endogenous reactive oxygen species and exogenous toxicants pose threats to mitochondrial DNA (mtDNA) in fully developed oocytes. Deleterious mtDNA is commonly detected in mature oocytes, but is absent in embryos, suggesting the existence of a cryptic purifying selection mechanism. Here, we discover that in C. elegans, the onset of oocyte-to-zygote transition (OZT) developmentally triggers a rapid mitophagy event. We show that mitophagy at OZT (MOZT) requires mitochondrial fragmentation, the macroautophagy pathway, and the mitophagy receptor FUNDC1, but not the prevalent mitophagy factors PINK1 and BNIP3. MOZT reduces the transmission of deleterious mtDNA and as a result, protects embryonic survival. Impaired MOZT drives the increased accumulation of mtDNA mutations across generations, leading to the extinction of descendant populations. Thus, MOZT represents a strategy that preserves mitochondrial health during the mother-to-offspring transmission and safeguards lineage continuity.

Introduction

Mitochondria fuel eukaryotic life. Originating from ancient endosymbiosis, they retain a genome—mitochondrial DNA (mtDNA), which encodes essential proteins involved in respiration¹. In most eukaryotes mitochondria are maternally inherited from the oocyte, and the integrity of this mitochondrial genome is vital for embryonic development².

Mitochondrial quality control is fundamental to organismal and population health. Deleterious mtDNA and resulting mitochondrial dysfunction cause embryonic lethality, metabolic disorders, neurodegeneration, and infertility^{3, 4}. Oocyte mtDNA is susceptible to disruption by maternal diet, toxicants, radiation, ROS, and replication errors^5–7. Limited mtDNA repair machinery causes animal mtDNA to accumulate damage and mutate more rapidly than nuclear DNA over evolutionary time^{8, 9}. mtDNA mutations can accumulate irreversibly over generations, risking species extinction^{10, 11}. However, few deleterious variants are transmitted across generations¹², implying mechanisms exist to eliminate deleterious mtDNA—a process called purifying selection.

In the germline, purifying selection has been observed during primordial germ cell development, germ cell entry into oogenesis, and mid-oogenesis. Primordial germ cells in C. elegans employ PINK1-dependent pathways to reduce mutant mtDNA (uaDf5, a 3054-bp mtDNA deletion), promoting the quality of the germline’s founder mtDNA pool¹³. However, when these primordial germ cells proliferate to form germline stem cells (GSCs), uaDf5 mutant mtDNA replicates preferentially, resulting in a rapid increase in the mutation load¹³. In Drosophila, GSCs counteract high mtDNA heteroplasmy when they enter meiosis. During this transition, mitochondrial fragmentation¹⁴ and programmed mitophagy selectively removes mutant mtDNA¹⁵. These processes ensure that oogenesis begins with a low mtDNA mutant load. During mid-oogenesis, when mitochondria undergo substantial mtDNA replication^{15, 16}, PINK1 accumulates on mitochondria harboring defective mtDNA and inhibits local translation to favor healthy mtDNA expansion^{16, 17}. Additionally, the programmed cell death (PCD) pathway functions throughout C. elegans germline development and restricts the expansion of uaDf5 mutant mtDNA¹⁸. This provides a secondary, cell-level checkpoint to protect mtDNA integrity. Collectively, these mechanisms preserve mtDNA quality throughout the earliest stages of germline development and oogenesis.

The last step in oogenesis, oocyte maturation, is an energy-intensive process that requires a critical threshold of mitochondrial abundance^{19, 20}. Insufficient mtDNA copy number and ATP production in oocytes compromise fertilization and embryonic development^{19, 21, 22}. During maturation, oocytes generate high levels of reactive oxygen species (ROS) within both cytoplasmic and mitochondrial compartments²³. Moreover, as mature oocytes await fertilization, their mitochondria, already burdened by ROS, remain highly vulnerable to exogenous insults. This poses a challenge for the mature oocyte—maintaining abundant, active mitochondria while ensuring the transfer of high-quality mitochondria to the zygote. Several quality control mechanisms have been proposed to act in mature oocytes, including the selective expulsion of mutant mtDNA during the first polar body extrusion in human oocytes²⁴ and, to a lesser extent, programmed cell death^{18, 25, 26}. While these mtDNA quality control mechanisms are not fully elucidated, compelling evidence from cross-generational studies demonstrates that mtDNA is subject to purifying selection during maternal transmission.

In many species, mutant mtDNA variants are common in mature oocytes^{27, 28}, but are lost in the resulting progeny²⁹, suggestive of organelle-level purifying selection. Notably, the “common deletion” (a 4977-bp deletion) has been detected in 30-50% of human oocytes, along with several mtDNA rearrangements. Intriguingly, the frequency of these deleterious mtDNA variants is substantially diminished in embryos^30–32. Similarly, mouse oocytes harboring abundant deleterious mtDNA give rise to progeny with reduced mtDNA mutant load²⁹. The mitophagy receptor BCL2L13 and the macroautophagy pathway have recently been implicated in mediating this intergenerational purifying selection in mice³³. In C. elegans, mutant mtDNA variants are also reduced during transmission from mother to progeny, although the underlying mechanism is unknown^{18, 34}. These investigations reveal a recurring pattern—the mtDNA mutation load is reduced during maternal transmission from the oocyte to the offspring. However, the precise site of action and molecular mechanism underlying this cross-generational purifying selection remain elusive, as do the consequences of losing this selection.

Here, we uncover a rapid mitochondrial fragmentation and mitophagy event at the oocyte-to-zygote transition (OZT). Designated as mitophagy at OZT (MOZT), this phenomenon involves FUNDC1, a conserved mitochondrial membrane protein³⁵, as the molecular link coordinating fragmentation and mitophagy. MOZT drives mtDNA purifying selection during the mother-to-offspring transmission. Notably, the loss of MOZT sensitizes embryonic survival to maternally inherited deleterious mtDNA, leading to fertility decline and eventual population collapse.

Results

C. elegans oocytes reduce mitochondrial content during transition to zygote

In the syncytial C. elegans germline, stem cells (GSCs) proliferate in the distal mitotic zones³⁶. During this time, mutant mtDNA (uaDf5) rapidly expands, reaching ~55% of total mtDNA¹³. GSCs then exit the mitotic zone, differentiate and enter meiotic prophase I. As they progress along the germline bend, they cellularize and grow before arresting in diakinesis as oocytes (Fig. 1a). At the proximal gonad, oocytes stack by age. The oldest (−1 oocyte) is positioned most proximally, with successively younger oocytes (−2, −3, etc.) arranged distally (Fig. 1a). The sperm releases major sperm protein (MSP) and triggers the −1 oocyte to undergo the oocyte-to-zygote transition (OZT), a ~20 min process encompassing nuclear envelope breakdown, cortex rearrangement, ovulation, and fertilization³⁶ (Fig. 1a). Because GSCs seed the oocyte with their highly heteroplasmic mtDNA, we investigated whether mtDNA mutational load is reduced before transmission to the zygote.

Fig. 1: — a, A schematic of the adult *C. elegans* hermaphrodite germline (left) and a timeline of the oocyte-to-zygote transition (OZT) (right).

b, Confocal fluorescence images of mitochondria during *C. elegans* OZT. Endogenous outer mitochondrial membrane protein TOMM-20::mNG (top) and inner mitochondrial membrane protein NDUV-2::mNG (bottom) are reduced in zygotes compared to −2 oocytes. Blue boxes indicate −2 oocytes and pink boxes zygotes. The −2 oocytes and zygotes enclosed in boxed regions are magnified. Scale bars, 10μm.

c, Total fluorescence intensity of endogenous TOMM-20::mNG (top) and NDUV-2::mNG (bottom) in −2 oocytes and zygotes. The data represent the mean ± s.d. (n = 23, 30 animals per strain) from three biological replicates. P values were calculated using two-tailed paired Student’s t-tests.

d, Time-lapse images showing changes in mitochondrial abundance over 40 min in a −2 and −1 oocyte expressing endogenous TOMM-20::mNG (*tomm-20::mNG*). Fluorescence images show unchanged mitochondrial abundance in the −2 oocyte throughout its progression to the −1 position (top), compared to mitochondrial reduction in the −1 oocyte during transition to the zygote (bottom). The time-lapse is representative of six independent experiments.

e, Mitochondrial volume (μm³) determined from endogenous NDUV-2 (*nduv-2::mNG*) signal distribution in −2 oocytes, −1 oocytes, and zygotes. Mitochondrial volume is actively reduced in the −1 oocytes (as shown in d and Extended Data Fig. 2). When measured at any time point, the −1 oocyte, on average, contains reduced mitochondrial volume compared to the −2 oocyte. The data represent the mean ± s.d. (n = 30 animals) from three biological replicates. P values were calculated using Kruskal-Wallis followed by Dunn’s multiple comparisons test.

Refer to Extended Data Fig. 2c,d for dynamic measurements of oocyte mitochondrial volume over time.

We visualized mitochondria throughout oogenesis and fertilization using endogenous mNeonGreen knock-ins of TOMM-20 (outer mitochondrial membrane, OMM) and NDUV-2 (inner mitochondrial membrane, IMM). We observed an unexpected ~2-fold reduction in mitochondrial protein levels and volume from the −2 oocyte to the zygote (Fig. 1b,c,e) and confirmed this mitochondrial reduction using a general mitochondrial dye NAO (Extended Data Fig. 1a,b).

Using HMG-5::GFP₁₁ (TFAM), a marker for mtDNA nucleoids¹³, we found that both overall TFAM abundance and mtDNA nucleoid number/μm³ decreased ~2-fold in zygotes relative to −2 oocytes (Extended Data Fig. 1c–e). We directly visualized mtDNA using the dsDNA-specific dye PicoGreen and confirmed this reduction (Extended Data Fig. 1f, g). A published proteomic database profiling the C. elegans oocyte-to-zygote transition (http://elegans.mdc-berlin.de)³⁷ further corroborates the decrease in mitochondrial (TOMM-40, TOMM-70, ATAD-3, LONP-1, SDHB-1) and mtDNA nucleoid protein (TFAM) abundance during OZT. Together, these distinct observations indicate a rapid decrease in mitochondrial content during OZT.

Mitochondrial reduction at OZT is developmentally programmed

To determine the timing of mitochondrial reduction, we performed time-lapse imaging of arrested −2 oocytes and −1 oocytes undergoing OZT. We found that mitochondrial reduction was specific to the −1 oocyte (Fig. 1d and Extended Data Fig. 2a,c,d), beginning with the onset of OZT and ending upon completion of the transition to zygote (Extended Data Fig. 2d).

This temporal coupling led us to hypothesize that oocyte mitochondrial reduction was developmentally programmed to be triggered by sperm signaling. To test this, we inhibited sperm signaling using RNAi knockdown of fem-3, a sex-determination gene required for spermatogenesis³⁸. In sperm-containing hermaphrodites, all −1 oocytes underwent OZT and reduced their mitochondrial volume, such that at any given time point the −1 oocyte consistently contained lower mitochondrial abundance than the −2 oocyte (Fig. 1e, Extended Data Fig. 2b). In the absence of sperm (fem-3 RNAi), the mitochondrial volume difference between −1 and −2 oocytes was eliminated (Extended Data Fig. 2b). Time-lapse imaging confirmed that when OZT was blocked, mitochondrial reduction failed to occur in the −1 oocyte (Extended Data Fig. 2e). We conclude that mitochondrial reduction is developmentally programmed to occur at OZT and is induced by sperm signaling.

Without sperm, oocyte mitochondria shifted from interconnected networks to donut-shaped structures, suggesting an energy-conserving, arrested state of the cell (Extended Data Fig. 2f and Supplementary Fig. 1a–e,g). When oocytes mature, sperm signaling reorganizes oocyte mitochondria into interconnected networks to support high ATP production^{20, 39}. Our findings reveal that sperm signaling regulates two key processes in the oocyte: the organization of mitochondria into tubular networks and the subsequent reduction of mitochondria during OZT.

Macroautophagy is required for programmed mitochondrial degradation during OZT

We hypothesized that oocytes activate macroautophagy⁴⁰ during OZT to degrade mitochondria. We tested this by using RNAi knockdown to disrupt macroautophagy at multiple steps: initiation of autophagosome membrane formation (vps-34⁴⁰), membrane elongation (atg-16.1 and atg-16.2⁴⁰), autophagosome-lysosome fusion (vps-41⁴⁰), and lysosomal acidification (V-ATPase components vha-2, vha-9, and vha-12³⁹). We quantified mitochondrial reduction at OZT (zygote/−2 oocyte mitochondrial signal = ~0.5 in controls, Fig. 1c) in control and knockdown conditions. Knockdown of each of these macroautophagy genes eliminated mitochondrial reduction at OZT (Fig. 2a,b).

Fig. 2: — a, Confocal fluorescence images of mitochondria (NDUV-2::mNG) in the −2 oocyte and zygote of empty vector control (L4440) compared to animals treated with RNAi targeting genes involved in macroautophagy. Scale bar, 10μm.

b, Quantification of mitochondrial reduction during OZT in controls vs RNAi-treated animals. Data represent the ratio of total NDUV-2::mNG fluorescence intensity in the zygote to that in the −2 oocyte for each animal compared between control and RNAi knockdown conditions represented in (a). The data represent the mean ± s.d. (n = 30 animals each condition) from three biological replicates. ***P<0.0001 from one-way ANOVA followed by Tukey’s multiple comparison test.

c, Merged confocal fluorescence images of mitochondria (NUO-1::mKate2) and lysosomes (SCAV-3::GFP) in the −2 oocyte (top) and −1 oocyte (bottom). Scale bar, 10μm. Inset: Boxed regions are magnified. White arrows indicate regions of mitochondria-lysosome colocalization. Scale bar, 2μm.

d, Pearson’s correlation coefficient (r) measuring overall colocalization of mitochondria and lysosomes compared between −2 oocytes and −1 oocytes. Data represent the mean ± s.d. (n = 17 animals) from three biological replicates. P value using a two-tailed paired Student’s t-test.

e, Confocal fluorescence images of mitochondria (NUO-1::mKate2), lysosomes (SCAV-3::GFP), and merged, from an example region in the −1 oocyte. Scale bar, 5μm. N, nucleus. Boxed region in the merged image is magnified to the right, showing the lysosome signal clustered around mitochondrial region. Scale bar, 2μm. An additional example of mitochondrial lysosome engulfment from a different z-slice from the same cell is shown in Supplementary Fig. 6c (Example 2).

f, Graph represents line scans (2.0μm) drawn across the mitochondrial region engulfed by the lysosome shown in e. Data shows the normalized fluorescence intensity of the mitochondrial region (magenta) and lysosome (green) along the dotted yellow line drawn on the magnified image from e. Data represents similar results from 7 proximal oocytes (from n=7 animals) across three biological replicates. An additional line scan showing lysosome engulfment of mitochondria is shown in Supplementary Fig. 6c.

Autophagosome biogenesis protein ATG-9::GFP, autophagosomes (germline-expressed mCherry::LC3) and lysosomes (CTNS-1::wrmScarlet and SCAV-3::GFP) were upregulated in −1 oocytes compared to −2 oocytes, temporally coinciding with mitochondrial reduction (Extended Data Fig. 3a–h). Oocytes also increase lysosomal acidity through the assembly of V-ATPase components as they approach OZT³⁹. Spatiotemporal transcriptomics of the C. elegans germline reveal marked upregulation of autophagy, lysosomal membrane, and lysosomal acidification gene expression as germ cells transition into oocytes⁴¹ (Supplementary Fig. 2a–c). Together, these results suggest that C. elegans oocyte development involves assembly of the autophagosome-lysosome system, and its specific activation during OZT for mitochondrial degradation.

To directly test mitophagy during OZT, we employed two approaches: assessing mitochondria-lysosome co-localization and visualizing mitochondrial acidification. While green fluorophores such as eGFP are quenched in acidic pH, red fluorophores like mKate2 are acidic pH-resistant and allow imaging in acidic autophagosomes/lysosomes^{42, 43}. We therefore examined co-localization between mKate2-labeled mitochondria and eGFP-labeled lysosomes. In −1 oocytes, mitochondria exhibited increased lysosomal co-localization (Fig. 2c,d), with clustering of lysosomal signal (SCAV-3::GFP) around mitochondrial regions (Example 1: Fig. 2e,f, and Example 2: Supplementary Fig. 6c), consistent with lysosome-mediated mitochondrial degradation.

We visualized mitophagy using mitochondria endogenously tagged with pH-sensitive eGFP and pH-insensitive mKate2, allowing distinction between neutral and acidified mitochondria. In −2 oocytes, both fluorophores robustly labelled mitochondria, whereas in zygotes, eGFP fluorescence decreased relative to mKate2, resulting in a reduced eGFP/mKate2 ratio (Example 1: Extended Data Fig. 4a–c, and Example 2: Supplementary Fig. 6d) and an increase in acidified mitochondria—mKate2-enriched, eGFP-diminished puncta (Extended Data Fig. 4d,e). These observations reveal a mitophagy event during OZT.

Collectively, these findings demonstrate that oocytes assemble the autophagy-lysosome system and specifically activate it during OZT, which drives mitochondrial degradation. We define this event as mitophagy at OZT (MOZT).

Mitochondrial fragmentation is necessary but not sufficient for MOZT

Mitochondrial networks often undergo fragmentation into small, discrete mitochondria more conducive to autophagic degradation⁴⁴. We examined whether oocytes undergo mitochondrial fragmentation at OZT by comparing mitochondrial network parameters between −2 oocytes and zygotes. Zygote mitochondria were smaller, more spherical, and discrete compared to the tubular interconnected oocyte mitochondria (Fig. 3a,b). Time-lapse imaging confirmed rapid fragmentation in the −1 oocyte (Fig. 3c, n=15/15 animals). Average mitochondrial volume decreased from ~2.0 μm³ (−2 oocyte) to ~0.3 μm³ (zygote) (Fig. 3b), smaller than the −1 oocyte autophagosomes (~0.6 μm³) (Extended Data Fig. 6d). Static measurements of mitochondrial morphology are sensitive to fluorescence intensity of mitochondrial protein visualized⁴⁵. To account for the mitochondrial fluorescence decrease in zygotes, we used a photoconversion approach (TOMM-20::Dendra2) to further test for mitochondrial fragmentation (Extended Data Fig. 5a). Tracking photoconverted regions in −1 oocyte mitochondria over 20s, before mitochondrial signal drop, revealed rapid fragmentation (Extended Data Fig. 5b). Quantification of mitochondrial network parameters immediately upon photoconversion (t=0s) and after 20s revealed a trend towards fragmented mitochondria (Extended Data Fig. 5c–e). These results demonstrate rapid mitochondrial fragmentation during OZT.

Fig. 3: — a, Top: Confocal fluorescence images showing the change in mitochondrial morphology (visualized with NDUV-2::mNG) between the −2 oocyte and zygote in empty vector control (L4440) compared to a *drp-1* RNAi-treated animal. Scale bar, 10μm. Bottom: Skeletonized representation of mitochondria shown in the upper panels. Boxed regions are magnified. Scale bars, 5 μm.

b, Quantification of mitochondrial network parameters compared between −2 oocytes and zygotes in control and *drp-1* RNAi-treated animals. Data represent mean ± s.d. (n = 30 animals per condition) from three biological replicates. P values using two-tailed Student’s t-tests (mitochondrial count and sphericity) and a two-tailed Mann-Whitney test (mean mitochondrial volume).

c, Time-lapse images of mitochondria (NDUV-2::mNG) undergoing fragmentation in the −1 oocyte. White arrows indicate the site of fragmentation on mitochondria. Example represents mitochondrial fragmentation events recorded at a single focal plane in the −1 oocyte and was observed in 15/15 animals over 4 min of observation in each. Scale bar, 2μm.

d, Quantification of mitochondrial reduction during OZT in control vs *drp-1* RNAi-treated animals. Data represent the ratio of total NDUV-2::mNG fluorescence intensity in the zygote to that in the −2 oocyte for each animal. The data represent the mean ± s.d. (n = 30 animals) from three biological replicates. P value using a two-tailed Mann-Whitney test.

e, Endogenous GFP::DRP-1 puncta distribution in a −2 oocyte and −1 oocyte. Scale bar, 10μm.

f, Increase in DRP-1 puncta number (left) and mean puncta volume (right) from −2 oocytes to −1 oocytes. Data represent mean ± s.d. (n = 30 animals) from three biological replicates. P values using two-tailed Wilcoxon tests.

To identify the molecular drivers of mitochondrial fragmentation during OZT, we knocked down mitochondrial morphology effectors and assessed fragmentation in each condition (Table 1). Only drp-1 knockdown blocked mitochondrial fragmentation and resulted in interconnected, elongated mitochondria in zygotes (Fig. 3a,b). DRP-1 is a conserved large GTPase that regulates mitochondrial fission⁴⁴. Notably, when mitochondrial fragmentation was blocked during OZT, mitochondrial reduction was also inhibited (Fig. 3a,d). To test whether mitochondrial fragmentation is essential for MOZT, we measured mitochondrial acidification after drp-1 knockdown and found strongly inhibited MOZT (Extended Data Fig. 4a,f–i). We next examined endogenous GFP::DRP-1 localization during OZT. GFP::DRP-1 formed puncta throughout the oocytes, with a ~2-fold increase in number and ~1.5-fold increase in size in −1 oocytes versus −2 oocytes (Fig. 3e,f). drp-1 transcripts also increase ~4-fold in oocytes compared to the distal germline⁴¹ (Supplementary Fig. 2d). These findings indicate that oocytes upregulate DRP-1 during OZT to promote mitochondrial fragmentation necessary for mitophagy.

Table 1. RNAi screen to identify effectors of change in mitochondrial morphology during the C. elegans oocyte-to-zygote transition.

The genes targeted by RNAi and the effects on mitochondrial morphology observed in oocytes and zygotes of empty vector controls (L4440) and RNAi treated animals across three replicates. Rows in bold represent genes that when knocked-down, led to significant disruption of oocyte and zygote morphology.

Mitochondrial fission-fusion RNAi screen		Mitochondrial fragmentation during OZT
Gene targeted by RNAi	Number of animals	Oocyte mitochondrial morphology	Zygote mitochondrial morphology
L4440 (empty vector control)	60	Elongated (n=60/60)	Punctate (n=60/60)
pink-1	30	Elongated (n=27/30) Donut shaped (n=3/30); P=0.1037, ns	Punctate (n=30/30)
*drp-1*	42	Elongated (n=42/42)	Elongated (n=39/42); P=0.0001
dyn-1	30	Elongated (n=30/30)	Punctate (n=30/30)
mff-1	30	Elongated (n=30/30)	Punctate (n=30/30)
slc25A46	30	Elongated (n=30/30)	Punctate (n=30/30)
mtp-16	30	Elongated (n=30/30)	Punctate (n=30/30)
fis-1	30	Elongated (n=30/30)	Punctate (n=28/30) Elongated (n=2/30); P=0.1086, ns
fis-2	30	Elongated (n=30/30)	Punctate (n=30/30)
chch-3	30	Elongated (n=30/30)	Punctate (n=30/30)
eat-3	30	Elongated (n=30/30)	Punctate (n=30/30)
mfn-1	30	Elongated (n=30/30)	Punctate (n=30/30)
*fzo-1*	38	Punctate (n=24/38) Donut + punctate shaped (n=14/38); P=0.0001	Punctate (n=38/38)
miga-1	30	Elongated (n=30/30)	Punctate (n=30/30)

Open in a new tab

Statistical analysis was conducted using 2x2 contingency table- Fisher’s exact test. P values denoted are for comparisons between L4440 control and respective RNAi treated animals.

Sustained mitochondrial fragmentation can trigger mitophagy in some cells^{14, 15}. To test this in oocytes, we knocked down the mitochondrial fusion protein Mitofusin (encoded by fzo-1)¹⁴. Upon fzo-1 knockdown, the −2 oocyte mitochondria were fragmented and small compared to controls (Extended Data Fig. 6a,b). However, the mitochondria in the −2 oocyte did not undergo reduction—mitochondrial reduction remained restricted to OZT (Extended Data Fig. 6c). Thus, mitochondrial fragmentation is necessary but not sufficient for MOZT.

FUNDC1 regulates mitophagy and mitochondrial fragmentation during OZT

During programmed mitochondrial degradation events, mitophagy receptors assemble on the mitochondrial surface and activate downstream pathways to ensure selectivity in degradation⁴⁶. The most studied pathways are PINK1/Parkin-mediated and BNIP3-mediated mitophagy⁴⁶; FUNDC1-mediated mitophagy has been examined in response to hypoxia³⁵ but is understudied in developmental contexts.

Hence, we investigated the role of mitophagy receptors PINK-1 (C. elegans pink-1), BNIP3 (dct-1), and FUNDC1 (fndc-1) in MOZT. pink-1 and dct-1 mutants (alleles: tm1779 and luc194) showed normal mitochondrial reduction at OZT. In contrast, loss of fndc-1 (rny14) completely inhibited mitochondrial reduction (Fig. 4a–c) and mtDNA nucleoid reduction during OZT (Supplementary Fig. 3a–c). We also noted that fndc-1 transcripts are ~4-fold more abundant than PINK1 and BNIP3 in oocytes, and exhibit localized enrichment near OZT⁴¹ (Supplementary Fig. 2d). These results establish FUNDC1 as the primary mediator of mitochondrial and mtDNA reduction during OZT.

Fig. 4: — a, Confocal fluorescence images of mitochondria (visualized with NDUV-2::mNG) in the −2 oocyte and zygote of a control animal (wild-type) compared to null mutants of *pink-1*(*tm1779*), *dct-1*(*luc194*), and *fndc-1*(*rny14*). Scale bar, 10μm.

b, Total NDUV-2::mNG fluorescence intensity compared between −2 oocytes and zygotes in control animals and null mutants of *pink-1*, *dct-1*, and *fndc-1*. Data represent mean ± s.d. (n = 30 animals per strain) from three biological replicates. P values using two-tailed Student’s t-tests (control, *dct-1*, and *fndc-1*) and a two-tailed Mann-Whitney test (*pink-1*).

c, Quantification of reduction in mitochondrial volume during OZT in wild-type controls, *pink-1*, *dct-1*, and *fndc-1* mutant animals. Data represents the ratio of mitochondrial volume in the zygote to that in the −2 oocyte for each animal. Data represent mean ± s.d. (n = 30 animals per strain) from three biological replicates. P values using a Kruskal Wallis followed by Dunn’s multiple comparisons test.

d, Top: Confocal fluorescence images showing the change in mitochondrial morphology (NDUV-2::mNG) between the −2 oocyte and zygote in a wild-type control compared to a *fndc-1* mutant animal. Bottom: Skeletonized representation of mitochondria shown in upper panels. Scale bars, 5μm.

e, Quantification of mitochondrial network parameters compared between −2 oocytes and zygotes in a wild-type control and *fndc-1* mutant animals. Data represent mean ± s.d. (n = 30 animals per strain) from three biological replicates. P values using two-tailed Student’s t-tests and two-tailed Mann-Whitney tests, depending on the normality of distribution.

To test FUNDC1’s requirement for mitochondrial acidification during OZT, we examined mitochondria tagged with eGFP and mKate2 and found that fndc-1 mutant zygotes showed markedly fewer acidified mitochondria versus wild-type controls (Extended Data Fig. 4j–m). Mitochondrial morphology analysis further revealed that mitochondrial fragmentation at OZT was significantly inhibited in fndc-1 mutants (Fig. 4d,e). We conclude that FUNDC1 plays a crucial role in mediating MOZT and mitochondrial fragmentation.

Stable FUNDC1 puncta localize at sites of mitochondrial fragmentation during OZT

FUNDC1 (OMM protein) mediates mitochondria-autophagosome interaction through its LIR (LC3-interacting region)⁴⁷. It also accumulates at mitochondrial-ER contact sites (MERCs) in response to hypoxia and directly recruits DRP-1 for mitochondrial fragmentation⁴⁷. We examined FUNDC1 localization during OZT using an endogenous mRb3::FNDC-1 knock-in⁴⁸ and a mitochondria marker (NDUV-2::mNG). A previous study did not detect any mRb3::FNDC-1 signal in oocytes⁴⁸. However, using our sensitive live-imaging system, we observed diffuse mRb3::FNDC-1 signal across the mitochondrial surface in −2 and −1 oocytes (Fig. 5a), consistent with OMM localization.

We also detected FUNDC1 puncta (Fig. 5a), which increased in number and signal intensity in the −1 oocytes (Fig. 5b,c). Since FUNDC1 accumulates at MERCs⁴⁷, we visualized mRb3::FNDC-1 puncta together with ER and mitochondrial markers. The puncta were usually localized on or adjacent to the ER and mitochondria, consistent with MERC localization (n=20/20, Extended Data Fig. 7a,b). As the mRb3::FNDC-1 appeared punctate, and we noted mitophagy during OZT, we also tested if the mRb3::FNDC-1 puncta were in autophagic structures and lysosomes. However, we observed minimal signal overlap between mRb3::FNDC-1 puncta and ATG-9::GFP or lysosomes (SCAV-3::GFP)(Extended Data Fig. 7c,d). mRb3::FNDC-1 puncta co-localized significantly more with the ER and mitochondria (Extended Data Fig. 7e,f) further suggesting MERC association. Since mitochondrial fission-fusion is required for MERC formation^{47, 49, 50}, we induced mitochondrial hyperfusion (drp-1 RNAi) and found that mRb3::FNDC-1 puncta were reduced in −1 oocytes (Extended Data Fig. 7g,h) and FUNDC1 shifted to diffuse OMM localization (Extended Data Fig. 7i,j). These data support the notion that FUNDC1 puncta are localized at MERCs in the oocyte.

In −2 oocytes, ~35% of mRb3::FNDC-1 puncta co-localized at specific loci on the mitochondrial surface, which increased to ~60% in −1 oocytes (Fig. 5d). Time-lapse imaging showed that FUNDC1 puncta dwell time on −1 oocyte mitochondria ranged from 10 s to 8.6 min (Fig. 5e). We observed two distinct FUNDC1 puncta behaviors (Fig. 5e and Extended Data Fig. 7k; Supplementary Video 2): ~70% of FUNDC1 puncta (n=20 puncta from 12 animals) co-localized for short time frames ranging from 10 s to 2 min without affecting mitochondrial morphology (n = 14/14; Fig. 5e and Extended Data Fig. 7k,l—Yellow graph). In contrast, ~30% of FUNDC1 puncta (n=20) localized on mitochondria for longer timeframes ranging from 4 min to 8.6 min (Fig. 5e and Extended Data Fig. 7k,l—Blue graph). This consistently led to a mitochondrial fragmentation event at the site (Supplementary Video 1; Fig. 5f,g [n=10/10]). These results suggest that stable FUNDC1 puncta promote mitochondrial fragmentation, which is necessary for MOZT (Extended Data Fig.7m).

Oocytes upregulate FUNDC1 in response to diverse mitochondrial damage

Cancer cells, mouse cardiomyocytes, and C. elegans body wall muscles increase FUNDC1 expression and MERC accumulation under hypoxic stress^{47, 51}. Whether this occurs in response to other mitochondria-damaging stimuli is unknown.

We tested oocyte FUNDC1 response to oxidative damage (paraquat⁵²), mitochondrial uncoupling (FCCP⁵³), and 6PPD-Q treatment (a rubber tire oxidation product that disrupts mitochondrial complexes I and II⁵⁴). Oocyte FUNDC1 levels increased upon FCCP and 6PPD-Q exposure (Extended Data Fig. 8a–c and Supplementary Fig. 3d,f). 6PPD-Q treatment led to the highest FUNDC1 upregulation and coincided with the appearance of donut-shaped mitochondria (Supplementary Fig. 1f and Supplementary Fig. 3d), indicating mitochondrial dysfunction. These results reveal that oocytes sense diverse mitochondrial damage and respond by increasing FUNDC1 levels.

We next tested whether oocyte FUNDC1 is increased in response to mtDNA damage and mutations. We treated animals with UVC radiation, which creates irreparable DNA lesions⁵⁵. Nuclear DNA lesions are repaired by NER, while mtDNA lesions are not⁵⁵. UV treatment dramatically upregulated FUNDC1 on oocyte mitochondrial membranes and increased MERC-associated puncta (Extended Data Fig. 9a,b and Supplementary Fig. 3e). This coincided with increased frequency of donut-shaped mitochondria in oocytes (Supplementary Fig. 1f). To test how mtDNA-specific disruptions alter oocyte FUNDC1 levels, we examined oocytes carrying the uaDf5 deletion (Fig. 6c). uaDf5 mtDNA elevated FUNDC1 levels both on oocyte mitochondrial membranes and in MERC-associated puncta (Fig. 6a,b).

Fig. 6: — a, Endogenous mRb3::FNDC-1 distribution in a −2 oocyte of a wild-type versus a *uaDf5/+* animal. Blue and yellow arrows indicate OMM and MERC-associated signals, respectively. Scale bar, 10μm.

b, Upregulation of mRb3::FNDC-1 fluorescence intensity in −2 oocytes harboring *uaDf5*, relative to wild-type controls. n = 30 animals per strain across three replicates. P values using two-tailed Mann-Whitney tests (left and right graphs) and two-tailed Student’s t-test (middle graph).

c, A schematic of *C. elegans* mtDNA, specifying the *uaDf5* deletion region.

**d,e,** Percent *uaDf5* heteroplasmy in oocytes and zygotes, from wild-type; *uaDf5* (d) and *fndc-1; uaDf5* (e) animals. n = 12, 11 preparations of oocyte and zygote samples, across three experiments for (d) and (e), respectively. P values using two-tailed paired Student’s t-tests.

f, Data from (d) and (e) presented as percent change in *uaDf5* heteroplasmy from oocytes to zygotes. P value using a two-tailed Student’s t-test.

g, DIC images of representative F1 progeny of wild-type; *uaDf5* and *fndc-1;uaDf5* parents (P0). Scale bars, 10μm. Data represents F1 progeny observed from n=10 parents per strain across three biological replicates. Grey boxes were added at the corners of the DIC images covering black background outside the field of view of the imaging plane.

h, Percent embryonic lethality among F1 progeny of control and *uaDf5/+* animals in wild-type and *fndc-1* mutant backgrounds. n = 10 (wild-type), 12 (*fndc-1*) animals were used as P0 parents across three replicates. P values using two-tailed Mann-Whitney tests.

i, Overview of population bottlenecking experimental design to accelerate mtDNA mutation accumulation.

**j,k,** The introduction of *uaDf5*, followed by five generations of mutation accumulation via population bottlenecking (G0 to G5) has differential effects on fertility (j) and embryo survival (k) in wild-type and *fndc-1* backgrounds. Datapoints represent the normalized total live progeny count (j) and proportion of embryonic progeny that survive (%) (k). Circles denote mean with s.e.m. error bars. Green denotes wild-type lines; magenta denotes *fndc-1* lines. n = 10 animals per strain from three independent experiments. For (j), P values using two-tailed Student’s t test (wild-type control vs G0 *uaDf5*) and all other P values using two-tailed Mann-Whitney tests. For (k) P values using two-tailed Mann-Whitney tests (Control vs G0) and two-tailed Student’s t-tests (G0 vs G5).

l, Extinction rates of wild-type;*uaDf5* lines (green) and *fndc-1*;*uaDf5* lines (magenta), over 5 generations of serial bottlenecking.

m,n, Shift in zygotic *uaDf5* heteroplasmy over 5 generations of population bottlenecking in wild-type(m) and *fndc-1*(n) backgrounds, compared to pre-bottleneck conditions. n = 13 (G0) and 9 (G5) (m), 17 each (n) individual preparations of zygote samples (3-4 zygotes each) for wild-type and *fndc-1* mutant lines, respectively, across three biological replicates. P value using a two-tailed Welch’s t test (m) and a two-tailed Mann-Whitney test (n).

o, *uaDf5* heteroplasmy in zygotes from bottlenecked *fndc-1;uaDf5* lines with corresponding embryonic lethality and fertility quantifications. Line 1 had the lowest, and line 2 had the highest zygote heteroplasmy (out of 5 surviving lines at G5). *uaDf5* heteroplasmy of arrested embryos from *fndc-1;uaDf5* lines was consistently ~40%, suggesting reduced tolerance for mutant load in *fndc-1* mutants. Data from two biological replicates.

p, Proposed model of MOZT’s role in cross-generational mtDNA purifying selection.

Data from **b-h**, **m-o** represent mean ± s.d, and data from **j, k** represent mean± s.e.m. For **d-f** and **m-o** larger circles represent the mean of each color-coded experiment; each smaller dot represents data calculated from a single real-time qPCR run (biological replicate). Overall means are represented by central horizontal lines with s.d. as error bars.

Given FUNDC1’s roles in mitochondrial fragmentation and MOZT, we asked whether UV-induced FUNDC1 upregulation correspondingly led to increased mitochondrial fragmentation and MOZT. Upon UV exposure, both fragmentation and mitochondrial reduction remained unchanged (Extended Data Fig. 8d–f). ATG-9::GFP vesicles, an indicator of autophagosome biogenesis, were also not elevated upon UV treatment (Extended Data Fig. 8g,h). These findings suggest that the oocyte sets mitophagy levels developmentally, independent of damage levels. Increased FUNDC1 levels upon mitochondrial damage might instead promote tolerance to mtDNA-damaging agents like UV radiation, or uaDf5 heteroplasmy. Alternatively, increased FUNDC1 expression may enhance selectivity during MOZT.

Oocyte-expressed FUNDC1 enhances reproductive tolerance to UV-induced damage

mtDNA damage can affect oocyte fertility, and when inherited by the embryo, could severely compromise progeny survival⁵⁶. We tested whether oocyte-expressed FUNDC1 promotes tolerance to mtDNA-damaging factors, such as UV radiation. We treated wild-type and fndc-1 mutant parents (P0) with UV-radiation and assessed F1 embryonic survival (Extended Data Fig. 9c,d). UV-treated wild-type parents always produced viable F1 progeny. Untreated fndc-1 parents exhibited 17-25% F1 embryonic lethality, while UV-treated fndc-1 parents (P0) gave rise to ~50% arrested F1 embryos (Extended Data Fig. 9d,e). These findings indicate that loss of FUNDC1 increases embryonic vulnerability to maternal UV exposure.

fndc-1 (UV) embryos arrested primarily at the 160-180 cell gastrula stage (n=30, Extended Data Fig. 9f). We thus tested whether maternal UV exposure caused mitochondrial dysfunction in the progeny at the gastrula stage, which could lead to arrest. Arrested embryos from UV-treated fndc-1 mothers showed loss of interconnected tubular mitochondrial networks, and instead harbored donut-shaped mitochondria (Extended Data Fig. 9g–j). TMRE staining revealed that the donut-shaped mitochondria had diminished mitochondrial membrane potential (MMP) (Extended Data Fig. 9k,l), and only ~50% of the mitochondria in these arrested embryos had detectable MMP (Extended Data Fig. 9k,m), indicating mitochondrial dysfunction. Although we cannot rule out the possibility that embryonic arrest could indirectly impair mitochondria, these data support the notion that maternal UV-treatment in fndc-1 mutants induces mitochondrial dysfunction in embryos, which leads to arrest.

To assess effects of accumulating UV-induced damage over multiple generations in wild-type and fndc-1 mutant animals, we performed a population bottlenecking (mutation accumulation) experiment⁵⁷. We UV-irradiated a single L4 animal each generation and used it as progenitor for the next generation to exacerbate damage accumulation. We bottlenecked 10 sublines each from wild-type and fndc-1 mutants for 10 generations (Extended Data Fig. 9n). Over generations, reproductive fitness declined at a rapid rate in the fndc-1 (UV) lines compared to wild-type (UV) lines (Extended Data Fig. 9o). Upon just one generation of UV exposure (generation 2, G2), fndc-1 lines exhibited ~40% decline in fertility, while wild-type lines showed no fertility change (Extended Data Fig. 9o). After 10 generations, wild-type (UV) lines exhibited an insignificant fertility decline compared to untreated controls. In contrast, fndc-1 (UV) lines showed severely reduced fecundity (~75% decline) (Extended Data Fig. 9o). Moreover, by G10, only 19% of wild-type (UV) lines were extinct, in contrast to 82% of fndc-1 (UV) lines (Extended Data Fig. 9p).

To test whether accumulated UV-induced damage impairs overall mitochondrial function in descendant populations, we measured mitochondrial respiration in wild-type (UV) and fndc-1 (UV) lines. However, fndc-1 mutants on the whole-body level showed unaltered mitochondrial respiration, even in UV-damage accumulated lineages (Supplementary Fig. 4). This result suggests that the heightened sensitivity of fertility and embryo viability in fndc-1 mutants is not from global mitochondrial dysfunction, but likely from specific impairment of embryo mitochondrial function upon maternal UV exposure (Extended Data Fig. 9k–m). This also suggests that other mechanisms counter UV-induced mitochondrial dysfunction in somatic tissues.

To test whether deteriorating sperm quality over the 10 generations contributes to the phenotypes, we crossed fourth generation (G4) fndc-1 (UV) hermaphrodites with untreated wild-type males and assessed fertility (methods). Notably, reproductive fitness of the crossed fndc-1 (UV) lines did not improve (Extended Data Fig. 10a–c), ruling out paternal effects.

Our results suggest that oocytes increase FUNDC1 expression to counter UV-induced damage and promote embryo survival. Alternately, FUNDC1 may function embryonically to enhance survival and tolerance to maternally inherited UV-damage. To test when FUNDC1 functions, we rescued FUNDC-1 expression specifically after fertilization in fndc-1 (UV) embryos (Extended Data Fig. 10d). FUNDC1 expression was robust by the 90-100 cell stage of embryogenesis (Extended Data Fig. 10f), which is before arrest occurs in fndc-1 (UV) embryos. Importantly, these embryos were derived from oocytes that lacked FUNDC1 (methods, Extended Data Fig. 10f). However, embryonic FUNDC1 overexpression did not improve survival (Extended Data Fig. 10d–f). These results offer strong evidence that FUNDC1 functions at OZT to protect embryonic survival and cross-generational reproductive success from UV damage.

MOZT is a cross-generational mtDNA purifying selection event

Although the UV irradiation protocol induces persistent mtDNA damage and nDNA damage is known to be repaired⁵⁵, the reduced embryonic viability, fertility, and population collapse observed in fndc-1 (UV) lines might have resulted from UV-induced nuclear DNA damage or mutations. To directly assess whether impaired mtDNA integrity underlies the phenotypes observed in MOZT-deficient animals, and to investigate if MOZT performs purifying selection, we investigated oocyte-to-zygote transmission of the uaDf5 mtDNA deletion (Fig. 6c) and how loss of FUNDC1 affected uaDf5 dynamics at OZT.

uaDf5 mutant mtDNA exists in heteroplasmy with wild-type mtDNA. We introduced uaDf5 mtDNA into wild-type and fndc-1 mutant animals and measured mtDNA heteroplasmy in oocytes and zygotes. In wild-type animals, uaDf5 heteroplasmy decreased by ~12.5% during OZT (Fig. 6d). However, in fndc-1 mutants, uaDf5 heteroplasmy remained the same across OZT (Fig. 6e). This indicates that mtDNA purifying selection at OZT is mediated by FUNDC1 (Fig. 6f) and suggests that MOZT reduces the mtDNA mutant load during zygotic transmission.

MOZT promotes lineage continuity

Since MOZT operates across generations, we hypothesized that disrupting it would impair both parental fertility (P0) and progeny (F1) survival in the presence of mutant mtDNA. To test this, we evaluated fertility and embryonic viability in wild-type and fndc-1 mutant animals carrying uaDf5 mtDNA. All wild-type; uaDf5 lines were fertile, while ~28% of fndc-1; uaDf5 lines generated were infertile (2 out of 7 lines generated). Introducing uaDf5 mtDNA increased F1 embryonic lethality modestly in wild-type (~1.80%), but dramatically in fndc-1 mutants (~34.61%) (Fig. 6g,h). uaDf5 mtDNA reduced wild-type fertility by ~20%, but in fndc-1 mutants fertility dropped by ~40% (Fig. 6j). Thus, loss of MOZT severely impairs fertility and embryonic survival in the presence of uaDf5 mtDNA.

When uaDf5 heteroplasmy compromises fitness in nuclear mutant backgrounds, organism-level selection acts against animals carrying high uaDf5%⁵⁸. Interestingly, we found that fndc-1; uaDf5 hermaphrodites displayed a lower uaDf5 heteroplasmy (~32%) than wild-type; uaDf5 hermaphrodites (~53%; Supplementary Fig. 5a). This aligns with prior work showing reduced heteroplasmy on a whole-body level in fndc-1 mutants vs wild-type (~30% vs ~60%)⁴⁸. Together with embryonic lethality and reproductive decline, these data indicate that fndc-1 mutants have reduced tolerance for uaDf5 heteroplasmy than wild-type.

To minimize organism-level selection against uaDf5 and examine mtDNA heteroplasmy drift across generations, we performed population bottlenecking experiments on wild-type; uaDf5 and fndc-1; uaDf5 populations (Fig. 6i). Each independent bottlenecked lineage was established by isolating a single L4 animal per generation for five generations. If MOZT mediates cross-generational purifying selection, we hypothesized that diminished organism-level selection would cause fndc-1; uaDf5 lines to accumulate higher heteroplasmy, while wild-type; uaDf5 lines would maintain or reduce heteroplasmy levels. We measured fertility and embryonic viability after 5 generations (G5) of population bottlenecking (N=1 animal isolated per generation). At G5, wild-type; uaDf5 lines maintained fertility and embryonic survival comparable to G0, whereas fndc-1; uaDf5 animals exhibited ~40% fertility decline (Fig. 6j) and embryonic survival dropped from ~75% to ~50% (Fig. 6k). Extinction rates were much higher in fndc-1; uaDf5 lines (50%) than wild-type; uaDf5 lines (10%) (Fig. 6l). Critically, fndc-1 animals homoplasmic for wild-type mtDNA subjected to the same bottlenecking regime showed no declines in fertility, embryonic survival, or population continuity (Supplementary Fig. 5b–d). Thus, reproductive decline in fndc-1; uaDf5 lines resulted specifically from uaDf5 heteroplasmy, rather than fndc-1 loss alone (Supplementary Fig. 5e–f).

To investigate whether loss of MOZT leads to accumulation of mutant mtDNA across generations, we also measured zygotic uaDf5 heteroplasmy in wild-type and fndc-1 populations at G0 and G5. At G0, wild-type; uaDf5 zygotes contained ~55% heteroplasmy, whereas fndc-1; uaDf5 zygotes contained ~30%, consistent with the overall reduced heteroplasmy levels in fndc-1 adults (Supplementary Fig. 5a). At G5, wild-type zygotic heteroplasmy decreased (Fig. 6m), indicating active mtDNA purifying selection across generations. In contrast, fndc-1; uaDf5 zygotic heteroplasmy increased (Fig. 6n), showing increased accumulation of mutant mtDNA across generations due to loss of purifying selection. Embryonic lethality in fndc-1; uaDf5 lines correlated with heteroplasmy levels: among the 5 surviving G5 lines, the line with ~30% heteroplasmy exhibited ~59% embryonic lethality, whereas the line with ~40% heteroplasmy showed ~87% embryonic lethality (Fig. 6o). ~50% of G5 fndc-1; uaDf5 zygotes had ~40% heteroplasmy (mean±s.d.= 41.46 ± 0.77% uaDf5, n=8/16 zygote preparations; Fig. 6n), correlating with ~50% embryonic lethality in these lines (Fig. 6k). Arrested fndc-1; uaDf5 embryos exhibited heteroplasmy around 40% (Fig. 6o), with no embryos detected above this percentage, whereas wild-type embryos exceed 40% uaDf5 mtDNA and develop normally. These results further support the idea that FUNDC1 enables embryonic tolerance of high mutant mtDNA heteroplasmy (50-60%); without FUNDC1, this tolerance is reduced to ~40%. In summary, FUNDC1 prevents mutant mtDNA accumulation, sustains embryonic tolerance to mutant mtDNA, and ensures lineage continuity (Fig. 6p).

Because uaDf5 reduces sperm fitness⁵⁹, we tested whether paternal effects underlie the fertility and embryonic survival decline in bottlenecked fndc-1; uaDf5 lines. We crossed fndc-1; uaDf5 hermaphrodites (G4) with wild-type males and found fndc-1; uaDf5 mothers (G4) retained diminished fertility, and their offspring maintained low embryonic survival (Fig. 7a–d). These results indicate that the fertility and embryonic survival defects are maternally determined, not paternal effects.

Fig. 7: — a, Experimental design to test if sperm with healthy wild-type mtDNA rescues fertility and progeny embryonic survival in *fndc-1;uaDf5* bottlenecked lines (generation 4, G4). The schematic shows merged confocal images of a *fndc-1;uaDf5* (G4) hermaphrodite mother, wild-type male used for cross-fertilization (tagRFP::LMN-1), and progeny (live and arrested embryos). Red, tagRFP::LMN-1 signal; grey, DIC signal. Red signal in progeny confirms successful cross-fertilization.

b, Experimental design to test if embryonic expression of FUNDC1 after fertilization rescues embryonic lethality in *fndc-1;uaDf5* bottlenecked lines (G4). The schematic shows merged confocal images of a *fndc-1;uaDf5* (G4) hermaphrodite mother, wild-type male used for cross-fertilization (mRb3::FNDC-1), and progeny. Magenta, mRb3::FNDC-1 signal; grey, DIC. Magenta signal in gastrula-stage embryos indicates robust FUNDC-1 expression (sperm-contributed).

c,d, Total live progeny count (c) and embryonic lethality among F1 progeny (d) from hermaphrodites from *fndc-1;uaDf5* bottlenecked lines (G4), self-fertilized vs crossed with wild-type males carrying healthy, wild-type mtDNA. Data represent mean ± s.d.; n = 7 (self), 12 (cross) (c) and n = 6 (self), 12 (cross) (d), from three biological replicates. P values using two-tailed Welch’s t tests.

e, Experimental design for plasmid-mediated embryonic FUNDC1 overexpression in F1 progeny of hermaphrodites from *fndc-1;uaDf5* bottlenecked lines (G3). The schematic shows merged confocal images of a P0 hermaphrodite germline and the resulting embryos. Grey, DIC; magenta, mRb3::FNDC-1; green, co-injection markers. Magenta signal in ~100-cell embryos confirmed successful post-fertilization FUNDC1 overexpression. Lack of FUNDC1 signal in P0 germline confirmed that the plasmid was silenced in the maternal germline.

f, Embryonic lethality among F1 progeny of hermaphrodites from *fndc-1;uaDf5* bottlenecked lines (G3), compared with F1 progeny of P0 hermaphrodites (from the same lines) injected with a FNDC-1 rescue plasmid [*pie-1p::mRb3::FNDC-1::tbb-2* 3’UTR]. Data represent mean ± s.d.; n = 7 (non-injected) and 15 (injected), from three biological replicates. P values using a two-tailed Welch’s t-test.

g, Model for how FUNDC-1 mitophagy at the oocyte-to-zygote transition regulates mtDNA quantity, morphology, and quality, safeguarding fertility and embryonic viability across generations.

Because oocyte-specific FUNDC1 (but not embryonic FUNDC1) preserves progeny survival against UV-damage, we similarly tested site of action for FUNDC1 function against mutant mtDNA heteroplasmy. We first cross-fertilized fndc-1; uaDf5 hermaphrodites (G4) with untreated wild-type males expressing mRb3::FNDC-1. This resulted in sperm-contributed expression of mRb3::FNDC-1 in gastrula stage embryos (Fig. 7b). We independently overexpressed mRb3::FNDC-1 specifically in early embryos (Fig. 7e). In both cases, embryo survival did not improve (Fig. 7f), showing embryonic FUNDC1 is insufficient to tolerate/survive uaDf5 heteroplasmy conditions.

Taken together, these observations indicate that oocyte-expressed FUNDC1, through purifying selection of mtDNA, promotes a cross-generational drift towards lower mtDNA heteroplasmy. This sustains embryonic tolerance to mutant mtDNA, maintains fertility and ensures lineage continuity (Fig. 7g).

Discussion

Early oocytes employ multiple mitochondrial quality control strategies: complex I inactivation reduces ROS⁶⁰, genetic bottlenecking controls the number of mtDNA copies that each oocyte inherits⁶¹, and purifying selection before oogenesis ensures founder mtDNA pool integrity^{13, 15}. As oocytes mature, they upregulate mitochondrial respiration^{20, 60} and mtDNA replication⁶² to support ATP and biosynthetic demands of fertilization and embryogenesis. However, elevated mitochondrial activity generates ROS, leading to oxidative damage of mtDNA and proteins. Compounding this, mtDNA replication errors generate deleterious mtDNA in mature oocytes, leading to deletions³⁰ and rearrangements³². When present at high levels, these diminish the fertilization potential and disrupt embryonic development⁶³. How oocytes balance mitochondrial abundance with selective transmission of healthy mitochondria to progeny remains unresolved. In this study, we identify MOZT as a developmentally programmed process that maintains this balance.

Developmentally programmed mitophagy is not unique to oocytes, as it also occurs in keratinocytes and erythrocytes during differentiation, mediated by mitophagy receptors NIX and BNIP3⁶⁴. MOZT is distinct from most known programmed mitophagy events in its regulation by FUNDC1. Unlike NIX and BNIP3, which have BH3 domains that interact with apoptosis regulators⁶⁴, FUNDC1 contains motifs that bind to mitochondrial fission-fusion proteins, including DRP-1⁴⁷. This specialization suggests a unique requirement for network remodeling over apoptotic signaling in oocytes. In C. elegans mature oocytes, mitochondria form interconnected networks²⁰. During OZT, FUNDC1 accumulates at MERCS and DRP-1 is upregulated, collectively driving rapid fragmentation. Live imaging revealed FUNDC1 puncta trafficking across the mitochondrial network and stabilizing at specific sites to promote DRP-1–dependent fission. When we impaired mitochondrial fragmentation, MOZT was strongly inhibited, establishing the essentiality of fission. FUNDC1 enables macroautophagy-dependent clearance of oocyte mitochondria, likely through its canonical LIR³⁵. Thus, we identify FUNDC1 as a critical regulator of MOZT, influencing each step—mitochondrial fission and mitochondrial degradation. About 2.5 hours post-MOZT, during embryogenesis, FUNDC1 drives another programmed mitophagy event: paternal mitochondria elimination⁴⁸. FUNDC1 regulates both maternal and paternal mitochondria and is therefore a central coordinator of mitochondrial inheritance.

We investigated additional FUNDC1 roles in oocyte mitochondrial damage response. Beyond developmental upregulation during OZT, FUNDC1 levels further increase upon presence of deleterious mtDNA (UV damage and uaDf5 mutation). To test functional importance, we compared wild-type and fndc-1 mutants carrying UV-damaged mtDNA or uaDf5 mtDNA. Specifically in the presence of deleterious mtDNA, fndc-1 mothers exhibited diminished fertility. Their F1 progeny showed two-fold diminished embryonic viability. These embryos exhibited collapsed mitochondrial networks with severely diminished membrane potential—hallmarks of mitochondrial dysfunction. Interestingly, fndc-1 embryos also have reduced tolerance to deleterious mtDNA: fndc-1 embryos carrying ~40% uaDf5 arrest, while wild-type embryos develop normally at 50-60% uaDf5. Loss of Mitofusin (fzo-1 mutants) results in embryonic lethality at a similar uaDf5 heteroplasmy level (~40%)⁵⁸. It has been proposed that proper mitochondrial mixing via fission and fusion distributes mutant mtDNA across mitochondria, limiting the mtDNA mutational load within individual organelles⁶⁵. Disrupted fission (fndc-1) or fusion (fzo-1) could cause mutant mtDNA to concentrate in individual mitochondria in the embryo, causing mitochondrial injury and leading to arrest. Additional functions of FUNDC1 may work in concert with fission-fusion balance to enhance tolerance to deleterious mtDNA. In mice, FUNDC1 promotes mitochondrial Ca²⁺ balance by regulating the Ca²⁺ exchanger IP3R3 at MERCs⁶⁶. C. elegans OZT is accompanied by a surge in cytosolic calcium⁶⁷, and FUNDC1 may similarly regulate Ca²⁺ dynamics to prevent mitochondrial injury. We thus conclude that oocyte FUNDC1 promotes reproductive and embryonic tolerance to deleterious mtDNA.

FUNDC1 upregulation also occurs upon exposure to toxicants (6PPD-Q), and mitochondrial uncouplers unrelated to mtDNA, indicating that MOZT responds to mitochondrial stress broadly. C. elegans oocytes exhibit elevated baseline ROS, which further increases upon exposure to exogenous stressors²³. Mitochondrial ROS levels sharply decline after oocytes transition into zygotes²³. Similar to yeast⁶⁸ and cancer cells⁶⁹ that use mitophagy to reduce oxidative burden, MOZT likely plays a similar function in preventing oxidative injury during embryogenesis. Notably, while oocytes with damaged mitochondria upregulate FUNDC1, they still undergo mitochondrial reduction to the same degree as undamaged oocytes. Autophagy is not upregulated in these oocytes either, indicating that autolysosomal capacity could limit mitochondrial clearance even while quality control demands rise. We propose that a developmental threshold has evolved to ensure embryos inherit an essential number of mtDNA copies^{70, 71}, since mtDNA replication resumes only late, at the third larval stage⁷².

Purifying selection is conserved across species but differs in timing and mechanism (Supplementary Fig. 7). In humans, mtDNA deletions and rearrangements are reduced from mature oocytes to embryos^{30–32, 73}. In mice, selection occurs earlier, during oocyte maturation (between the primordial and antral stages)^{74, 75}. In Drosophila, purifying selection occurs even earlier, with programmed germline mitophagy (PGM) occurring at the beginning of oogenesis¹⁵. In C. elegans, multiple studies have demonstrated mother-to-offspring purifying selection^{18, 34}, prompting us to test whether MOZT performs mtDNA purifying selection. We find that zygotes contain ~12.5% less mutant mtDNA than oocytes, and FUNDC1 is essential for this reduction. The wild-type; uaDf5 lines used here contained around 50-60% uaDf5 (medium heteroplasmy level). It remains unknown whether purifying selection differs at the extremes of heteroplasmy (very low or very high maternal heteroplasmy)—similar to how different selection pressures operate at different heteroplasmy extremes in mice carrying the m.5024C>T mutation⁷⁶. In mice oocytes purifying selection occurs through increased protein translation rather than mitophagy⁷⁴, making it mechanistically and temporally distinct from MOZT. MOZT and Drosophila PGM are similar in that they are the only known programmed mitophagy events dedicated to mtDNA quality control¹⁵. They occur at opposite ends of prophase I: PGM at entry¹⁵ and MOZT at exit. Both require mitochondrial fragmentation, but mechanisms differ—fragmentation is temporally separated from PGM¹⁵ but coincides with MOZT. Moreover, PGM and MOZT employ distinct receptors, BNIP3 and FUNDC1, respectively, suggesting specialization across developmental transitions. Notably, MOZT, PGM, and mouse mtDNA quality control all occur before the oocyte exits prophase I⁷⁴. Given that prophase I exit is an irreversible commitment to oocyte meiotic completion and availability for fertilization, we speculate that evolutionary pressures may have coupled this meiotic stage with purifying selection.

The importance of purifying selection for lineage continuity is evident when considering the consequences of mutant mtDNA accumulation. Maternally inherited mtDNA mutations in mice lead to reduced fertility⁷⁷. In C. elegans mtDNA mutation accumulation driven by dysregulated mtDNA methylation results in sterility within 12 generations⁷⁸. We thus investigated the effects of disrupted MOZT over multiple generations. We reduced organism-level selection against uaDf5 through a population bottlenecking experiment, where a single progenitor was used to establish isogenic lines. At the end of 5 generations, wild-type animals showed reduced zygotic mtDNA mutant load due to purifying selection, whereas fndc-1 mutant zygotes accumulated higher mutant loads. This progressively reduced embryonic survival and fertility in fndc-1; uaDf5 lines, ultimately resulting in the complete extinction of 50% of these lines. Neither embryonic FUNDC1 expression nor introduction of healthy sperm (wild-type mtDNA) could rescue these phenotypes, demonstrating that these defects originate strictly from the oocyte and cannot be compensated post-fertilization. MOZT is therefore an intergenerational safeguard against mutant mtDNA that ensures lineage endurance.

The conservation of MOZT across species remains unclear. While mitochondrial reduction from oocytes to embryos occurs in zebrafish⁷⁹ and pigs⁷⁰, it is absent in mice⁸⁰ and humans³¹. Yet, oocyte expression of FUNDC1—a key component of MOZT—appears conserved across multiple taxa^{60, 81–83}. Proteomic analysis of Drosophila oogenesis reveals significant upregulation of FUNDC1 during oocyte maturation⁸⁴. Despite this conserved expression pattern, FUNDC1’s developmental role in the oocyte remains unexplored. In summary, we demonstrate that FUNDC1 regulates multiple facets of mitochondrial quality control within the oocyte: mitochondrial morphology, acidification, damage tolerance, and purifying selection, all of which occur within a ~20-minute developmental interval.

Methods

C. elegans strains and culture conditions

C. elegans strains were fed Escherichia coli OP50 on plates with nematode growth medium (NGM) and raised under standard conditions at 20°C. N2 Bristol was the wild-type strain⁸⁵. All strains used in this study are listed in Supplementary Table 1.

Genome editing

mNeonGreen was knocked-in at the C-terminus of the endogenous coding sequence of mitochondrial proteins TOMM-20 and NDUV-2, and mKate2 at the C-terminus of NUO-1 protein using CRISPR-Cas9-mediated genome editing as previously described^{86, 87}. The strains generated were outcrossed with the wild-type strain four times before experimental use. sgRNA sequences and primers used for genotyping are included in Supplementary Table 2.

Strain generation through genetic crosses

The following C. elegans strains were created using standard mating techniques: NK3406 (qy157 (nuo-1::mKate2)II; zac389(scav-3::GFP)⁸⁸), NK3407 (zac389(scav-3::GFP)II; rny15(mRuby3::fndc-1)⁴⁸II), NK3408 (qy157(nuo-1::mKate2)II; zu476(cox-4::eGFP::3xFLAG)⁸⁹)I), NK3409 (qy157(nuo-1::mKate2)II; zu476(cox-4::eGFP::3xFLAG); fndc-1(rny14)⁴⁸ II), NK3285 (qy174(nduv-2::mNG)V; rny15(mRuby3::fndc-1⁴⁸)II), NK3286 (ojIs23 [pie-1p::GFP::SP12, unc-119(+)]⁹⁰; rny15(mRuby3::fndc-1)II), NK3410 (ola274[atg-9::gfp]⁹¹; rny15(mRuby3::fndc-1)II ), NK3282 (qy174 (nduv-2::mNG)V; pink-1(tm1779)⁹²II), NK3283 (qy174 (nduv-2::mNG)V; dct-1(luc194)⁹³X), NK3284 (qy174 (nduv-2::mNG)V; fndc-1(rny14⁴⁸)II), NK3287 (rny15(mRuby3::fndc-1)II; qy174 (nduv-2::mNG)V; uaDf5/+ mtDNA⁹⁴), and NK3288 (fndc-1(rny14)II; uaDf5/+ mtDNA). Primers used to verify the presence of uaDf5 mtDNA are listed in Supplementary Table 2.

Microscopy and image analysis

Confocal images were acquired on an upright Carl Zeiss Axio Imager A1 microscope attached to a Yokogawa CSU-W1 spinning disk confocal head, using a Zeiss 63x Plan-Apochromat (NA 1.40) oil-immersion objective, a Hamamatsu ORCA-Fusion or ORCA-Quest camera, with 488nm, 505nm, and 561nm lasers. Image acquisition was performed using the Micro-Manager (v2.0) software. For OZT imaging, day 1 adult worms were anesthetized in 5 mM Levamisole on 5% agar pads. A #1.5 cover slip was placed on top before imaging at room temperature⁹⁵. For time-lapse experiments, the slide was sealed using VALAP to prevent evaporation of the agar pad⁹⁶. Exposure times and laser power were optimized for each experiment based on fluorescence intensity. The collected images were viewed and analyzed using FIJI software (v1.54f). Single confocal z-slices of −2 oocytes, −1 oocytes, and zygotes from the same focal plane were used as representative images in figures. Uncropped confocal images supporting main figures are shown in Supplementary Fig. 6.

Mitochondrial abundance quantification

Regions of interest were manually drawn around the −2 oocytes, −1 oocytes, or zygotes expressing mitochondrial markers TOMM-20 or NDUV-2 or stained with NAO. Sum projections were generated from three central slices after background subtraction in Fiji (Rolling ball 15px). Total fluorescence intensity was calculated for the respective cell using the measure tool (Fiji). Fold reduction in mitochondria across OZT was calculated for each animal by dividing the zygote signal by −2 the oocyte signal, represented as “Zygote/−2 oocyte mitochondrial signal” data in Figs. 2 and 3.

Surface area of −2 oocytes and zygote sections showed no significant difference (P value = 0.1558, from 20 animals using a paired two-tailed Student’s t-test). Prior studies confirm that cell size remains constant during OZT (~21,700μm³ measured for the oocyte⁹⁷, and ~21,000μm³ calculated for the zygote⁹⁸), so normalization for organelle or protein abundance based on cell size was not required. To follow oocyte mitochondrial abundance over time during OZT using time-lapse imaging (Extended Data Fig. 2), five central z-stacks were captured every 5 min for 50 min.

RNAi treatment

RNAi knockdown was performed by using standard feeding of Escherichia coli HT115⁹⁹ expressing the relevant double-stranded RNA. RNAi constructs were obtained from the RNAi libraries constructed by the Ahringer laboratory¹⁰⁰ and the Vidal laboratory¹⁰¹. Bacteria with L4440 or T444T empty vectors were used as relevant controls.

RNAi bacterial strains were cultured in LB media containing 100 mg/mL ampicillin for ~14 h at 37°C. dsRNA expression was induced by the addition of 1 mM IPTG, bacterial cultures were then seeded onto NGM plates treated with 1 mM IPTG and 100 mg/mL ampicillin and dried at room temperature overnight before use. All RNAi knockdown experiments were conditional, with target RNAi fed beginning at the L4 stage to avoid developmental defects in the hermaphrodite germline. In every biological replicate, ~10 synchronized L1 worms were grown on a negative control (L4440 or T444T) for 48 h, followed by transfer onto relevant RNAi plates for 24 h before experimental use.

RNAi knockdown efficiency

Knockdown efficiency of drp-1 and fzo-1 RNAi was measured using animals expressing endogenous gfp::drp-1¹⁰² and fzo-1:gfp¹⁰³, treated from the L4 stage. Net total fluorescence intensity of GFP::DRP-1 and FZO-1::GFP within the −2 oocyte was measured from background-subtracted sum projections (3 z-slices). Refer to Supplementary Fig. 5d–g for knockdown efficiency calculations. The drp-1 and fzo-1 knockdown experiments were performed in parallel, with a common negative control (L4440).

Quantification of organelle/protein abundances at OZT

Confocal z-stack images (0.37μm per section) of −2 oocytes, −1 oocytes, and zygotes expressing HMG-5::GFP₁₁ [mex-5p::mito(matrix)-GFP_1-10::nos-2 3’UTR]¹³ (this strain is referred to as HMG-5::GFP₁₁ in short throughout the manuscript), GFP::DRP-1, ATG-9::GFP⁹¹, mCherry::LGG-2¹⁰⁴, SCAV-3::GFP⁸⁸, CTNS-1::wrmScarlet¹⁰⁵, FZO-1::GFP, mRb3::FNDC-1⁴⁸, COX-4::GFP⁸⁹, or NUO-1::mKate2 were captured as mentioned before, and respective protein abundance was quantified similarly to mitochondrial abundance (see “mitochondrial abundance quantification section”).

For mtDNA nucleoids (PicoGreen staining and HMG-5::GFP₁₁), ATG9 vesicles (ATG-9::GFP), autophagosomes (mCherry::LGG-2), lysosomes (SCAV-3::GFP and CTNS-1::wrmScarlet), and DRP-1 puncta (GFP::DRP-1), the “3D object counter” plugin (Fiji) was used across the central three z-slices after background subtraction, with the average fluorescence intensity of the respective marker in −2 oocyte used as the threshold for object detection. The number of objects, mean volume, and total volume were noted from this analysis.

Data depicting the increase in ATG-9 puncta during OZT (Extended Data Fig. 3a,e) and the changes to the number of ATG-9::GFP vesicles upon UVC treatment (“controls” in Extended Data Fig. 8g,h) were from the same experiment.

Before quantification of the mtDNA nucleoid number (PicoGreen staining and HMG-5::GFP₁₁), co-localization was performed with a mitochondrial marker. We confirmed that the labelled nucleoids were exclusively localized to mitochondria, and extramitochondrial mtDNA was absent.

Germline extrusion for mtDNA nucleoid count quantification

Due to the depth at which oocytes are positioned within the C. elegans body plan, mtDNA nucleoids marked by HMG-5::GFP₁₁ could not be consistently visualized with ~2-fold increased signal over noise. Therefore, we extruded germlines and zygotes from day 1 adults for reliable imaging. To account for changes in germline shape when extruded from the animal, we normalized the mtDNA nucleoid count to the sampled volumes of respective cells and reported “TFAM puncta per μm³ ”.

Staining with mitochondrial, mtDNA dyes

Nonyl Acridine Orange¹⁰⁶ (NAO) was prepared at 125nM final concentration. Picogreen¹⁰⁷ was reconstituted in DMSO and then diluted in M9 to a final concentration of 125nM before adding to medium with OP50. Thirty L4 larval stage animals were picked onto plates containing OP50 with the respective dye and allowed to grow in the dark for 24 h at 20°C. Day 1 adults were isolated, allowed to feed on non-stained OP50 to reduce gut fluorescence for two hours, and then OZT was imaged. Due to photobleaching, single confocal slices were used for the quantification of NAO signal, and 3 z-slices were used for PicoGreen imaging.

Quantification of mitochondrial volume and morphology

The volume of mitochondria was quantified using the Fiji macro, “Mitochondria Analyzer”¹⁰⁸. Three background-subtracted z-slices were converted to 8-bit, and 3D analysis was performed using the settings: Max slope:1.40, gamma 0.90, block size 1.25μm, and C-Value 5. ‘Despeckle’, ‘remove outliers’, and ‘fill 3D holes’ were set to be performed post-processing. The same analysis settings were used for both the −2 oocyte and zygote, as well as across different strains and conditions, to accurately compare changes in mitochondrial volume. This analysis pipeline also yielded mitochondrial network parameters for comparison of morphology. “Total mitochondrial volume”, “mean mitochondrial volume”, “mitochondrial count”, and “sphericity” were noted. For experiments where imaging was restricted to single z-slice imaging (for photoconversion followed by time-lapsing), the “2D analysis” pipeline from the same plugin “Mitochondrial analyzer” was used, and “mitochondrial count”, “mean mito surface area”, and “branches/mito” were noted.

Fold reduction in mitochondrial volume across OZT was calculated for each animal by dividing the zygote signal by −2 the oocyte signal, represented as “Zygote/−2 oocyte mitochondrial volume” data in Fig. 4.

Mitochondrial analysis in female animals

Feminization of C. elegans was achieved through transgenerational RNAi knockdown of fem-3^{38, 109}. L4 stage progeny were serially transferred to fem-3 RNAi plates for three generations. Successful feminization was confirmed by the absence of embryos in the animal’s uterus using confocal microscopy. To avoid confounding factors of cell shape changes in oocytes due to excessively stacking from feminization^110–112, sampling was limited to day 1 females that exhibited only 3-5 oocytes along the proximal germline. Hermaphrodites grown on T444T RNAi plates for three generations were used as controls for mitochondrial analysis.

Mitochondrial acidification

Assessment of mitochondrial acidification during OZT was done by measuring the ratio of green-to-red mitochondrial fluorescence in a strain expressing endogenous cox-4::eGFP and nuo-1::mKate2. ROIs were drawn around the −2 oocyte and the zygote, and overall green and red fluorescence intensities were measured to calculate green-to-red ratios at the cell level.

Using the green-to-red ratio of the overall −2 oocyte mitochondrial network as a reference point, mitochondrial regions in the zygote where the green-to-red ratio dropped by at least 2-fold compared to this reference point were confirmed to be “acidified mitochondria” (in these regions, red fluorescence signal overtook green by 2-fold, relative to oocyte mitochondria). Methodology adapted from Schwartz et al.¹³, which assessed mitochondrial acidification in the C. elegans germline.

To further confirm the presence of acidified mitochondria, a semi-automated quantification method was used (Extended Data Fig. 5e,i,m). The FIJI macro “mito-QC counter” sensitively identifies regions where the red-to-green ratio is significantly higher than the overall sample (pixel-by-pixel), marking acidified mitochondria¹¹³. mitoQC-counter parameters: Smoothing: 0; Ratio threshold: 0.500; Red channel threshold stdDev above mean: 1. Number of mitolysosomes in zygotes quantified using mitoQC counter in L4440 (negative control) treated animals was compared to drp-1 RNAi-treated animals (Extended Data Fig. 5i), and wild-type zygotes was compared to fndc-1 zygotes (Extended Data Fig. 5m).

Quantification of fragmentation events

Time-lapse images of day 1 adults expressing NDUV-2::mNG and mRb3::FNDC-1 were captured by spinning-disk microscopy. Single-slice confocal images were captured at 1.5 s intervals for 8-15 minutes. Line scans spanning 1μm were performed along the thresholded mitochondrial and FUNDC1 signals before and after the fragmentation event to confirm breaks.

For photoconversion experiments, day 1 adults expressing TOMM-20::Dendra2¹¹⁴ in the germline were used. Photoconversion was performed using an iLas2 targeted laser system equipped with an Omicron Lux 60mW 405nm wave laser, controlled by MetaMorph software (v 7.10.3.279). We photoconverted a small region (~10 × 5 μm) in the −1 oocyte mitochondrial network and performed time-lapse imaging to track structural changes in the photoconverted mitochondria. The imaging frame interval was 2s, conducted for 30-60s. The frame captured right after photoconversion was considered t=0s, and the mitochondrial morphology parameters were assessed at t=20s to quantify fission/fusion. Laser intensity was optimized (through serial dosage) to achieve photoconversion without altering mitochondrial structure.

Quantification of mRb3::FNDC-1 abundance at OMM and MERCs

Total mRb3::FNDC-1 fluorescence intensity was measured in 1.83 x 1.83 μm boxes at the mitochondrial surface (NDUV-2::mNG) in −2 and −1 oocytes. Ten regions per cell were sampled, and the mean value was reported as “Average OMM FUNDC1 signal”. We compared the overall −2 oocyte mitochondrial abundance (NDUV-2::mNG) between controls and animals exposed to mitochondrial stressors, and confirmed no significant changes (Supplementary Fig. 3f, g). This confirmed that FNDC-1 increases in oocytes specifically during mitochondrial damage, and the increase is not due to an overall increase in mitochondrial abundance.

The total fluorescence intensity within each mRb3::FNDC-1 puncta was measured after manually drawing regions of interest around each puncta throughout the z-stack, and the mean value per cell was reported as “average FUNDC1 puncta fluorescence intensity”.

Co-localization analysis

Co-localization analyses of mRb3::FNDC-1 signal with the endoplasmic reticulum (pie-1p::GFP::SP12) mitochondria (NDUV-2::mNG), lysosomes (SCAV-3::GFP) and ATG-9::GFP puncta in −2 oocytes were performed using the Fiji plugin, “JACoP”. Regions of interest (10x10μm box) were drawn around mRb3::FNDC-1 puncta and Pearson’s correlation coefficient (r) was calculated for co-localization with the ER, mitochondria and ATG-9 puncta. Since SCAV-3::GFP signal appears as a ring due to its lysosomal membrane localization, instead of Pearson’s correlation coefficient, an object-based colocalization approach was used: distance-based colocalization measurement using the same plugin, “JaCoP”. Co-localization between mitochondria (NUO-1::mKate2) and lysosomes (SCAV-3::GFP) was performed in −2 and −1 oocytes similarly.

The percentage of mRb3::FNDC-1 that localized on mitochondria in −2 oocytes and −1 oocytes was manually quantified with confocal fluorescence stacks with 10 z-stacks (0.37μm per section). Pearson’s correlation coefficient (r) was calculated between mitochondria and each mRb3::FNDC-1 puncta observed in the −2 and −1 oocytes of 10 animals; mRb3::FNDC-1 puncta for which r values were measured to be positive ( >0.40 ) were scored as “localized on mitochondria” and the puncta for which r values were below 0.40 were scored as “not localized on mitochondria”. Most non-mitochondrial localized mRb3::FNDC-1 had r values that were negative or close to zero. This validated our manual quantification of the percentage of mitochondria-localized mRb3::FNDC-1 puncta in −2 oocytes and −1 oocytes.

Quantification of mRb3::FNDC-1 puncta dwell time on mitochondria

Fluorescence images (single z-slice) of −1 oocytes were captured every 1.5 s for 8-15 min in strains expressing NDUV-2::mNG and mRb3::FNDC-1. Dynamic changes in co-localization between mRb3::FNDC-1 and mitochondria in time-lapse experiments conducted in −1 oocytes were validated by measuring changes in Pearson’s correlation coefficient (r) over time. The timepoint at which Pearson’s correlation coefficient (r) values drop from positive values (indicating co-localization) to zero or negative values was used to mark the end of co-localization. The time for which each mRb3::FNDC-1 co-localized on mitochondria was calculated manually, denoted by “dwell time on mitochondria”. Over 30 mRb3::FNDC-1 puncta from 15 animals were tracked and quantified to characterize dynamic behaviors of FUNDC1 puncta in the −1 oocyte.

UVC treatment

10-15 L4 worms per strain were washed once in M9 buffer to remove trace bacteria (to prevent shielding of UVC treatment) and plated onto unseeded NGM plates. The worms were then exposed to 25J/m² radiation of 254nm using a UVC lamp with a built-in sensor (CL-1000 Ultraviolet Crosslinker). Following exposure, worms were grown for 24h on OP50 plates, and oocyte FUNDC1 levels, progeny survival, and mitochondrial health were analyzed.

25J/m² UV was selected as dosage because: (i) causes mtDNA lesions, persistent 24 h after exposure⁵⁵; (ii) nuclear DNA (nDNA) lesions are removed within 24 h after exposure through nucleotide excision repair¹¹⁵, avoiding nDNA mutations⁵⁵; (iii) did not have significant effects on growth and fecundity of wild-type animals. To test this, 80-90 wild-type L4 animals were treated with 25J/m² UVC radiation, and the proportion of gravid adults was counted every 12 h to quantify developmental delays caused by UVC. The number of viable progenies from each animal was counted to quantify reproductive fitness decline. There were no developmental defects or defects in fecundity at a dosage of 25 J/m², suggesting that nuclear DNA damage is alleviated.

TMRE staining

TMRE¹¹⁶ was reconstituted in M9 buffer to a concentration of 2μM, and 500μL was added to plates with L4 animals (P0) on NGM plates with OP50. 24 hrs later, the F1 embryos were isolated for imaging. The animals stained with TMRE expressed endogenous COX-4::eGFP, or NDUV-2::mNG so that membrane potential quantification could be performed at the mitochondrial level (TMRE/NDUV-2::mNG signal ratio). “3D objects counter” plugin was used to quantify the number of TMRE-positive bodies and the total number of mitochondria, from 3 z-slices of gastrula stage embryos. Their ratio gave “percentage of mitochondria with membrane potential”.

Genetic bottlenecking experiments (mutation accumulation line generation)

Using the standardized dosage of UVC treatment (25J/m²) and animals harboring uaDf5 mtDNA, genetic bottlenecking was performed in wild-type and fndc-1 animals, along with corresponding controls. A single random L4 of each strain was isolated (G0) and allowed to lay eggs. Once the offspring reached the L4 stage, one worm was randomly selected and placed onto a new NGM plate with OP50 to create the G1 subline. Ten different G1 sublines were generated per strain and treatment to initiate the bottlenecking experiment. Among their progeny, one L4 animal was randomly selected and transferred to a new plate to serve as the progenitor for the next generation. Every subline was subject to such bottlenecking for 10 generations per strain per treatment/condition. All lines were grown and propagated at 20°C. Two previous generations were stored at 16°C, so that if an individual was arrested before adulthood, dead, or sterile, a back-up individual could be used to continue the line. The line was considered extinct if the backup was also not viable.

In UVC treatment conditions, each L4 progenitor was exposed to 25 J/m² UVC before being allowed to propagate the next generation.

Paraquat, FCCP, and 6PPD-Q treatment

10-15 L4 worms were transferred from plates seeded with OP50, washed three times in complete K medium¹¹⁷, and added to wells containing 1 mL complete K medium. Worms were kept from starvation by feeding 50 mg/ml OP50. Paraquat, FCCP, and 6PPD-Q were diluted in DMSO and added to these wells to achieve final dilutions of 75μM¹¹⁸, 5μM¹¹⁹ , and 33μM, respectively. The equivalent concentration of DMSO was added to the control wells. The plates were gently rocked and incubated at 25°C for 24 hours. Day 1 adults were transferred to slides for confocal imaging.

uaDf5 heteroplasmy quantification

mtDNA heteroplasmy via RT-PCR was conducted by adapting previously described protocols^{13, 120}. The composition of the lysis buffer used was as follows: 60% nuclease-free water, 5% 20mg/μL Proteinase K, and 35% 3.3X lysis buffer. Lysis was set up at ratios of 5-7 oocyte or zygote samples per 20μL buffer. The lysis reaction involved storage at −80°C for 10 min, lysis for 1 h at 65°C, and deactivation of Proteinase K at 95°C for 15 min. RT-PCR was then performed on lysates using 25 μl reaction volumes (2μL lysate, 8.5 μL nuclease-free water, 12.5 μL Power SYBR Green Master Mix, and 1 μL of each 100 μM primer). mtDNA heteroplasmy was determined by using two oligo pairs that specifically detected wild-type mtDNA or total mtDNA. Primer efficiencies were determined by amplifying a series of 7 exponential dilutions of the same sample using the two primer pairs, generating the standard curves: y = −3.3164x + 17.55, R² = 0.9992 (total mtDNA) and y = −3.312x + 18.267, R²= 0.9965 (wild-type mtDNA). Three independent RT-PCR reactions (technical replicates) of the same sample were run simultaneously to determine cycle-threshold (Ct) values for wild-type mtDNA and total mtDNA, and mean Ct values were reported. uaDf5 heteroplasmy percentage was calculated as follows: uaDf5 heteroplasmy (%) = [1-2^{(total mtDNA Ct - WT mtDNA Ct)} ] x 100.

All zygotic uaDf5 measurements obtained from comparison between oocytes and zygotes in Fig. 6d,e were used as datapoints in Fig. 6m and 6n for “pre-bottleneck (G0)” values.

Isolation of oocytes and zygotes

Day 1 adult worms were dissected at the pharynx-gut boundary, in about 5 μL of egg buffer on a glass slide to extrude fully intact germline arms. Additional egg buffer (5 μL) was added gently to promote the separation of the gonad from other tissues. Using two 0.4mm diameter needles, the distal germline and the −1 oocyte were cut off from the stacked oocytes. The remaining oocytes (~3 per animal) were mouth pipetted into a fresh drop of egg buffer for washing. The oocytes were then transferred into a PCR tube containing 20 μL of lysis buffer, and immediately subjected to lysis reaction. From the same dissected worms, embryos were isolated by mouth pipetting into a separate 10 μL drop of egg buffer. The embryos were inspected using a microscope, and 1-cell zygotes (1 or 2 per animal) were segregated into separate drops with egg buffer for washing and finally transferred to a PCR tube containing 20 μL of lysis buffer. The tube containing zygotes was immediately subjected to three cycles of flash freezing in liquid nitrogen and thawing at 37°C for egg-shell permeabilization before the subsequent lysis reaction. Roughly 3-4 independent sample preparations (oocyte and zygote samples) were performed for each independent biological replicate, and three independent biological replicates were performed for mtDNA heteroplasmy quantification through RT-PCR.

Comparison of mtDNA abundance between wild-type and fndc-1 zygotes

For each replicate, five zygotes were isolated from wild-type animals (LB138, 4x outcrossed version) and fndc-1 mutants (NK3288), as described above. After permeabilization of the egg-shell and lysis reaction, RT-PCR was performed with primers specific to mtDNA. Cycle-threshold values (Ct) were measured, and the fold increase in mtDNA abundance in fndc-1 zygotes, relative to wild-type, was calculated using the formula: 2^(Ct [wild-type] – Ct [fndc-1]). Four biological replicates were performed, each with five zygotes per strain.

Quantification of embryonic lethality and fertility

To examine embryonic lethality, ten L4 worms were singled onto individual NGM plates seeded with OP50. The worms were then transferred to fresh plates every 24 h for 6 days. The eggs laid on each plate were counted right after removing the parent worm and added across the 6 days to calculate the total eggs laid per animal. 48 hours after removal of the parent animal, the number of hatched viable progeny was counted in each plate, along with the number of unhatched dead embryos. Embryonic lethality% = [Number of dead embryos]/ [Total number of eggs laid] x 100. Ten parent animals and all embryos laid by each animal were quantified per strain and treatment, with three biological replicates performed. The total number of viable progeny laid in each condition was counted to report “total live progeny laid per animal”, as a quantification of fertility.

Mating experiments

Ten males expressing Itls44 [pie-1p::mCherry::PH(PLC1delta1)]¹²¹, marking germline and embryonic plasma membrane, were crossed with two L4 hermaphrodites from the fndc-1 (UV) lines (generation 4, G4). Ten males expressing bab344 [tagRFP::lmn-1]¹²², marking nuclear lamina, were crossed with two L4 hermaphrodites from the fndc-1;uaDf5 (G4) hermaphrodites. Mating was verified by assessing red fluorescence in the progeny. From each of the 10 fndc-1 uaDf5 (G4) lines, and each of the 10 fndc-1 uaDf5 (G4) lines, two hermaphrodites were isolated and crossed with 8-10 wild-type males. The same was repeated with males expressing rny15 [mRuby3::fndc-1] to rule out the embryonic role of FNDC-1 in uaDf5 experiments.

Embryo-specific expression of fndc-1

The embryonic expression of mRb3::FNDC-1 was achieved by construction of the transgene pie-1p::mRuby3::fndc-1::unc-54 3’UTR. The pie-1 promoter (1.114-kb fragment), full genomic sequence of fndc-1 (535-bp), and the unc-34 3’UTR (868-bp fragment) were fused in the pAP088 plasmid. The construct was injected into the gonads of two young adults from each of the 10 fndc-1 (UV) MA lines (G4), 10 fndc-1 uaDf5 MA lines (G3), along with 2.5ng/mL co-injection markers^{123, 124} (myo-2p::gfp and myo-3::gfp). The construct also contained the coding sequence of a dominant negative sqt-1 mutation that gives rise to a rol phenotype when expressed. Viability assays and fluorescence imaging were performed on F1 offspring from successful injections, identified by the presence of mRb3::FNDC-1 and selectable markers (green pharynx, green body wall muscle, and dominant negative sqt-1 rol phenotype). Multiple copies of extrachromosomal plasmids get silenced by the C. elegans germline and are thus not expressed in the oocyte¹²⁵. The silencing mechanisms persist only until the ~90-100 cell stage of embryos¹²⁶, allowing for an experimental model where FUNDC1 gene expression is absent in oocytes but is activated specifically post-fertilization (Fig. 7e and Extended Data Fig. 10f).

Mitochondrial respiration

Day 1 adult C. elegans from wild-type MA lines, wild-type (UV) MA lines, fndc-1 MA lines, and fndc-1 (UV) MA lines were picked off NGM plates with OP50, and the Seahorse Extracellular Flux Bioanalyzer was used to quantify respiration parameters¹²⁷. Confocal images of the plate were acquired to count the number of individual animals per well, thereby normalizing Oxygen Consumption Rate (OCR) measurements to the individual worm. Five to ten adult animals from each of the 10 MA lines were used per strain per condition across four independent experiments. Means of technical replicates per plate were reported.

Experimental setup

No statistical methods were used to pre-determine sample sizes, but our sample sizes are similar to those reported in previous publications. We conducted a pilot experiment and confirmed sample size was sufficient to consistently assess differences in fluorescence values between oocytes and zygotes, and across treatment conditions. Figure legends include details about sample size and statistical analysis used. C. elegans were randomly chosen from NGM plates for image acquisition, qPCR, and fitness assays. For experiments with different groups (control vs RNAi), all animals (~100-200) were synchronized to L1 larval stage, and 20-30 worms were randomly distributed to control and RNAi plates. Data collection and analysis were not performed blind to conditions of experiment, except for the RNAi knockdown screen reported in Table 1. For which target genes were randomly allocated numbers “1” through “13” before data collection and analysis. No experimental repeat was excluded. For analysis of individual oocytes and zygotes, out of focus confocal images were excluded. In some cases of zygote/embryo imaging, autofluorescence from the eggshell was cropped out before analysis and display to avoid artifacts. No datapoints were excluded after image analysis. Animals that failed to survive throughout time-lapse experiments were eliminated from further data collection or analysis. All animals were included for phenotypic analysis.

Statistical analysis

Statistical analyses were performed in GraphPad Prism 10 and GraphPad 2x2 contingency table. Normality of data distribution was assessed using the Kolmogorov-Smirnov test. Data derived from two different conditions/cells were compared using two-tailed Student’s t-test (Gaussian distribution), Mann-Whitney U test or Wilcoxon test (non-Gaussian distribution). For comparisons of three or more datasets, one-way ANOVA followed by Tukey’s multiple comparison test (Gaussian distribution) or Kruskal-Wallis test followed by Dunn’s multiple comparisons test (non-Gaussian distribution) were performed. Categorical data were compared using two-tailed Fisher’s exact test (RNAi screen in Table 1). Data were considered statistically different at P<0.05. All data are generally expressed as mean ± s.d. except when mentioned to be mean ± s.e.m. in the figure legend.

Illustrations were prepared using Procreate (v5.2.9) and Adobe Illustrator (v.29.2.1). Supplementary Videos were edited using Adobe Premiere Pro (v25.5.0). Confocal images were processed using ImageJ (v.2.1.4.7). Statistical analyses were performed using GraphPad Prism (v.10).

Extended Data

Extended Data Fig. 3 | — a, Endogenous ATG-9::GFP distribution in a −2 oocyte and a −1 oocyte. Scale bar, 10 μm. b, Confocal fluorescence images of autophagosomes visualized by germlinespecific LC3 protein marker (mCherry::LGG-2) in a −2 oocyte and −1 oocyte. Scale bar, 10 μm. c,d, Confocal fluorescence images of lysosomes visualized using SCAV-3::GFP (c) and CTNS-1::wrmScarlet (d) in − 2 oocytes and −1 oocytes. Scale bars, 10 μm. e-h, Number of ATG-9 vesicles (e), autophagosomes (f), and lysosomes (g,h) in −2 oocytes and −1 oocytes. Data represent mean ± s.d. (n = 30 animals per strain) from three biological replicates. P values using two-tailed Wilcoxon tests (e,f) and two-tailed paired Student’s t-tests (g,h).

Extended Data Fig. 4 | — a,b, Top: Confocal fluorescence images of mitochondria endogenously tagged with both eGFP and mKate2 in the oocyte and zygote. Scale bar, 10 μm. Bottom: Magnifications of boxed regions. Scale bar, 1 μm. Blue arrows indicate non-acidified mitochondria with robust green and red fluorescence. Yellow arrows indicate acidified mitochondria, where red fluorescence dominates. Mitochondria marked by arrowheads are magnified in (b). Scale bar, 1 μm. Green-to-red (G/R) ratio of the overall −2 oocyte mitochondria was normalized to be 1, and the relative G/R values of marked regions in (b) are denoted. c, Quantification of overall mitochondrial eGFP/mKate2 ratio in −2 oocytes and zygotes, revealing mitochondrial acidification in zygotes. Average eGFP/mKate2 ratio of −2 oocytes was normalized to be 1. Data represent mean ± s.d. (n = 25 animals) from three biological replicates. P value using a two-tailed Student’s t-test. d, Quantification of eGFP/mKate2 ratio in whole oocyte mitochondrial network and in the eGFPdiminished mitochondrial regions noted in zygotes. Average G/R ratio of the overall −2 oocyte mitochondrial network was normalized to be 1. Relatively decreased G/R ratio of these zygotic mitochondria suggests they are acidified. Data represent mean ± s.d. (n = 24 animals) from three biological replicates. P values using a two-tailed Wilcoxon test. e, Semi-automated quantification of the number of acidified mitochondria (mitolysosomes) in −2 oocytes and zygotes quantified using the ‘mitoQC counter’ plugin. Data represent mean ± s.d. (n = 26 animals per strain) from three biological replicates. P values using a two-tailed Wilcoxon test. f–i, Mitochondrial acidification visualized and quantified in drp-1 RNAi-treated animals, similar to a-e. Scale bars, 10 μm (top) and 2 μm (bottom) (f), 2 μm (g). Data represent mean ± s.d. from three biological replicates. P values using a two-tailed Student’s t-test (n = 22 animals) (h) and a two-tailed Mann– Whitney test (n = 16 animals) (i). j–m, Mitochondrial acidification visualized and quantified in fndc-1 mutant animals. Scale bars, 5 μm (top) and 1 μm (bottom) (j), 2 μm (k). Data represent mean ± s.d. from three biological replicates. P values using a two-tailed Student’s t-test (n = 25 animals) (l) and a two-tailed Mann– Whitney test (n = 16 animals) (m). In all panels, blue arrows indicate non-acidified mitochondria, and yellow arrows indicate acidified mitochondria. ‘G/R ratio’ denotes the eGFP/mKate2 fluorescence ratio.

Extended Data Fig. 5 | — a, A schematic of the experimental setup to assess changes to mitochondrial interconnectedness during OZT— photoconversion of a specific mitochondrial region and tracking changes to its morphology. Mitochondrial fusion would result in the intermixing of the photoconverted region with the greater network, while fragmentation would be reflected as breaks in the photoconverted region. b, Top: Examples of mitochondrial photoconversion experiments performed on the −1 oocyte mitochondria (TOMM-20::Dendra2). Boxed region is photoconverted (t = 0 s), and mitochondrial structure is assessed at t = 20. Scale bar, 10 μm. Bottom: Boxed region is magnified. Scale bar, 5 μm. White arrows indicate fragmentation of the photoconverted mitochondrial region. The second example (animal 2) shows rapid fragmentation occurring between t = 20 s and t = 22 s, as indicated by white arrows. c-e, Network parameters of the photoconverted mitochondrial region compared immediately upon photoconversion (t = 0 s) and after 20 s. Data represent mean ± s.d. (n = 10 animals) from three biological replicates. P values using two-tailed paired Student’s t-tests.

Extended Data Fig. 6 | — a, Top: Confocal fluorescence images showing the change in mitochondrial morphology (visualized with NDUV-2::mNG) between the −2 oocyte and zygote in an empty vector control compared to a fzo-1 RNAi-treated animal. Scale bar, 10 μm. Boxed regions are magnified under original images. Bottom: Skeletonized images of the mitochondria shown in middle panels. Scale bar, 2 μm. b, Quantification of mitochondrial network parameters (NDUV-2::mNG) compared between −2 oocytes and zygotes in control and fzo-1 RNAi-treated animals. Data represent mean ± s.d. (n = 30 animals per condition) from three biological replicates. P values using two-tailed Student’s t-tests and two-tailed Mann–Whitney tests, depending on normality of distribution. c, Quantification of MOZT in control compared to fzo-1 RNAi-treated animals. Data represent ratio of total NDUV2::mNG fluorescence intensity in the zygote to that in the −2 oocyte for each animal. The data represent the mean ± s.d. (n = 30 animals each condition) from three biological replicates. P value using a two-tailed Student’s t-test. d, Quantification of mean volume (μm3 ) of individual fragmented mitochondrion, autophagosome, and lysosome in the −1 oocyte. Data represent mean ± s.d. (n = 30 animals per strain) from three biological replicates. P values using a Kruskal–Wallis test followed by a Dunn’s multiple comparisons test.

Extended Data Fig. 7 | — a-d, Left: mRb3::FNDC-1 puncta visualized together with the endoplasmic reticulum (SP12::GFP) (a), mitochondria (NDUV-2::mNG) (b), ATG-9 puncta (ATG-9::GFP) (c) and lysosomes (SCAV-3::GFP) (d) in the −2 oocyte. Scale bar, 10 μm. Right: Boxed regions are magnified to the right. e, Quantification of Pearson’s correlation coefficient (r) to measure colocalization between mRb3::FNDC-1 puncta and three organelles in the −2 oocyte—the ER, mitochondria, and ATG-9 puncta. Data represent mean ± s.d. (n = 14, 29, 23 animals, respectively) from three biological replicates. P values using one-way ANOVA followed by a Tukey’s multiple comparison test. f, Object-based colocalization analysis measuring proportion of mRb3::FNDC-1 puncta overlapping with ATG-9::GFP positive puncta (n = 15 animals) and lysosomes (SCAV-3::GFP) (n = 19 animals). Data represent mean ± s.d. from three biological replicates. g, Top: Endogenous mRb3::FNDC-1 distribution in the oocyte of a control animal compared to a drp-1 RNAi-treated animal. Yellow and cyan arrows indicate MERC and OMM-localized mRb3::FNDC-1 signal, respectively. Region within dashed lines is segmented below. Bottom: Spectral fluorescence-intensity map of the segmented region, which displays the minimum (L, low) and maximum (H, high) pixel value of the original image. Scale bars, 10 μm. h-j, Number (h), mean fluorescence intensity of mRb3::FNDC-1 puncta (i), and mean fluorescence intensity of OMM-localized mRb3::FNDC-1 signal (j), in oocytes of controls compared to drp-1 RNAi-treated animals. Data represent mean ± s.d. (n = 33 animals per condition) from three biological replicates. P values using two-tailed Mann–Whitney tests. k, Time-lapse images showing dynamic associations of FUNDC1 punctae with mitochondria in the −1 oocyte. Scale bars, 2 μm. Boxes indicate short-dwelling (yellow) and mitochondria-stabilized (blue) FUNDC1 puncta. Note that FUNDC1 puncta observed are not colocalized on mitochondria at t = 0 min, but both gradually localize onto mitochondria. FUNDC1 foci indicated by yellow box starts localizing on mitochondrial surface (NDUV-2::mNG) around t = 2:27 min, fully co-localizes on mitochondria 10 s later, stays associated with mitochondria for 45 s, after which it moves away from mitochondria. FUNDC1 foci indicated by blue box start localizing on mitochondrial surface around t = 1:00 min and stays fully associated with mitochondria for the rest of the time-lapsing experiment (2:37 min). This data represents the two distinct behaviours of FUNDC1 puncta on the −1 oocyte mitochondria shown in Fig. 5e. White arrows indicate the start of FUNDC1 colocalization with mitochondria. Yellow arrow indicates the end of FUNDC1 colocalization with mitochondria. Orange arrow indicates intact mitochondria after short-dwelling FUNDC1 puncta is no longer mitochondrial localized. Supplementary Video 2 shows the time-lapse data from which (k) is derived. l, Pearson’s correlation coefficient (r) plotted over time measuring colocalization of FUNDC1 and mitochondria for both FUNDC1 puncta shown in time-lapse (k). Arrows in representative images from (k) correspond to arrows shown on the line graphs. Data represent FUNDC1 puncta dynamics observed in −1 oocytes of n = 35 animals from 8 independent experiments. m, Proposed model for FUNDC1 localization and function during C. elegans OZT. Panels 1 – 4 show the sequence of molecular events involved in MOZT.

Extended Data Fig. 8 | — a, Endogenous mRb3::FNDC-1 distribution in oocytes of a control animal compared to Paraquat, FCCP, and 6PPD-Q treated animals. Scale bar, 10 μm. Refer to Supplementary Fig. 3d for the same representative images with mitochondrial (NDUV-2::mNG) signal. b,c, Mean fluorescence intensity of OMM-localized (b) and MERC localized (c) mRb3::FNDC-1 in oocytes of control animals compared to animals treated with Paraquat, FCCP, and 6PPD-Q. Data represent mean ± s.d. for control (n = 10 animals), Paraquat (n = 10 animals), FCCP (n = 11 animals), and 6PPD-Q (n = 16 animals), from two independent experiments. P values using a one-way ANOVA followed by a Tukey’s multiple comparison test. d, Confocal fluorescence images showing the change in mitochondrial abundance and morphology (visualized with NDUV-2::mNG) between the −2 oocyte and zygote in a control animal compared to a UV-treated animal. Scale bar, 2 μm. e, Quantification of mitochondrial network parameters compared between −2 oocytes and zygotes in control and UV-treated animals. Data represent mean ± s.d. (n = 30 animals per condition) from three biological replicates. P values using two-tailed Student’s t-tests (mitochondrial count) and a two-tailed Mann–Whitney test (mitochondrial volume). f, Quantification of mitochondrial reduction in control vs UV-treated animals. Data represent the ratio of total NDUV-2::mNG fluorescence intensity in the zygote to that in the −2 oocyte for each animal. The data represent the mean ± s.d. (n = 30 animals each condition) from three biological replicates. P value using a two-tailed Student’s t-test. g, Distribution of endogenous ATG-9::GFP in −1 oocytes of a control and UV-treated animal. Scale bar, 10 μm. The same control oocyte is displayed in Extended Data Fig. 3a to show autophagy activation. h, Quantification of number and size (surface area, μm²) of ATG-9 vesicles in −1 oocytes of control vs UV-treated animals. Data represent mean ± s.d. (n = 30 animals) from three biological replicates. P values were calculated from a two-tailed Mann–Whitney test (number) and a two-tailed Student’s t-tests (size).

Extended Data Fig. 9 | — a, Endogenous mRb3::FNDC-1 distribution in the −2 oocyte of a control animal compared to a UV-treated animal. Blue and yellow arrows indicate OMM and MERC-associated mRb3::FNDC-1 signal, respectively. Scale bar, 10 μm. b, Mean fluorescence intensity of OMM-localized mRb3::FNDC-1 signal (left); number (middle) and mean fluorescence intensity of mRb3::FNDC-1 puncta (right) in −2 oocytes of controls compared to those of UV-treated animals. Data represent mean ± s.d. (n = 30 animals per condition) from three biological replicates. P values using a two-tailed Student’s t-test (left graph) and two-tailed Mann–Whitney tests (middle and right graphs). c, Schematic of experimental design to generate mtDNA lesions in oocytes. Inset: Red dots in mitochondria and nucleus denote DNA lesions. d, DIC images of representative F1 progeny of wild-type and fndc-1 mutant parents (P0) treated with UV radiation at L4 stage. Scale bars, 10 μm. Mean embryonic lethality percentage ± s.d. among F1 progeny is indicated. Data represents 10 progeny observed per strain across three replicates. e, Embryonic lethality (mean ± s.d.) in F1 progeny of untreated controls and UV-treated parents (P0), in wild-type and fndc-1 mutant backgrounds. n = 10 animals were used as P0 parents per strain per condition in three independent experiments. P values using a two-tailed Wilcoxon test (wildtype) and a two-tailed Student’s t-test (fndc-1). f, Quantification of the proportion of fndc-1 embryos arrested at specific stages during embryonic development. Embryos assessed are arrested F1 progeny of UV-treated fndc-1 mothers. g, Merged confocal fluorescence images (top) of mitochondria in F1 embryos from control fndc-1 mutant parents (P0) compared to arrested F1 embryos from UV-treated fndc-1 mutant parents (P0). Green, mitochondrial signal; grey, DIC signal. Scale bars, 10 μm. Boxed regions of mitochondrial signal and their skeletonized representations are magnified below. Scale bar, 2 μm. h-j, Mitochondrial network parameters (NDUV-2::mNG) compared between F1 embryos from untreated (control) and UV-treated fndc-1 mutant parents. Data represent mean ± s.d. (n = 11, 12 embryos per condition) from three biological replicates. P values using a two-tailed Mann–Whitney test (h) and two-tailed Student’s t-test (i,j). k, Top to bottom: Confocal fluorescence images of mitochondria (NDUV-2::mNG), TMRE signal, and merged, in a representative F1 embryo from an untreated fndc-1 mutant parent (P0) compared to two arrested F1 embryos from UV-treated fndc-1 mutant parents (P0). Scale bars, 10 μm. l,m, Mitochondrial membrane potential (TMRE signal per mitochondrial signal) compared between mitochondria of F1 embryos from untreated (control) and UV-treated fndc-1 mutant parents. Data represent mean ± s.d. (n = 18 embryos per condition) from three biological replicates. P values using a two-tailed Welch’s test (l) and a two-tailed Mann–Whitney test (m). n, A Schematic of population bottlenecking experimental design to accelerate mtDNA damage accumulation. Green animals denote untreated control worms and magenta animals denote UV-treated worms. o, Relative brood size as an indicator of reproductive fitness of MA lines across generations. Total brood size was calculated at each generation (G0 to G10) in UV-bottlenecked lines and normalized against that of the respective control bottlenecked lines. Green and magenta lines denote the trajectory of reproductive fitness in wild-type (UV) and fndc-1 (UV) bottlenecked lines, respectively. Graph indicates a decline in reproductive fitness of fndc-1 lines relative to controls across generational time upon UV treatment repeated at each generation. Data points represent mean ± s.e.m (n = 11 animals per strain per treatment) from three independent experiments. P values using Kruskal–Wallis test followed by Dunn’s multiple comparisons test. p, Extinction rates of wildtype (UV) and fndc-1 (UV) bottlenecked lines. Y-axis denotes the percentage of bottlenecked lines present at each generation out of 11 wild-type UV lines (green line) and 11 fndc-1 UV lines (magenta line).

Extended Data Fig. 10 | — a, A schematic of the experimental design to rescue sperm health in fndc-1 UV lines. Green animals denote untreated worms, and magenta animals denote worms with mtDNA damage. b, Quantification of reproductive fitness (brood size) of self-fertilizing hermaphrodites from fndc-1 UV-bottlenecked lines (fourth generation, G4) compared to those crossed with untreated wild-type males. Data represent mean ± s.d. (n = 11 animals [selfed] and 10 animals [crossed]) from three biological replicates. P values using a two-tailed Student’s t-test. c, Verification of successful crossing between wild-type males and hermaphrodites from fndc-1 UV-bottlenecked lines (G4). Merged confocal images showing F1 progeny with paternally inherited plasma membrane signal (mCherry::PH, magenta; DIC, grey). Magenta signal in progeny confirms successful cross-fertilization by wild-type males. Scale bar, 10 μm. d, Schematic of experimental design to rescue embryonic expression of FUNDC1 in fndc-1 UV-bottlenecked lines (G4). Green denotes the untreated condition, and magenta denotes UV-treated animal. e, Embryonic lethality among F1 progeny of parents from the fndc-1 UVbottlenecked lines (G4) compared to F1 progeny of parents (from the same lines) injected with FNDC-1 rescuing embryo expression plasmid [pie-1p::mRb3::fndc1::tbb-2 3’UTR]. Data represent mean ± s.d. (n = 10 animals per condition) from three biological replicates. P values using a two-tailed Mann–Whitney test. f, Merged confocal fluorescence images of mRb3::FNDC-1 (FNDC-1, magenta; DIC, grey) in oocytes of a fndc-1 UV-bottlenecked (G4) parent injected with embryo rescue plasmid and the resulting F1 embryo. Magenta signal in the F1 embryo, but not the parental oocytes confirm embryo-specific rescue of mRb3::FNDC-1 expression. Scale bar, 10 μm.

Supplementary Material

Video 1

Download video file^{(66.8MB, mp4)}

Video 2

Download video file^{(43.4MB, mp4)}

Supl Material

NIHMS2137744-supplement-Supl_Material.pdf^{(43.5MB, pdf)}

Acknowledgements

We thank the members of the laboratories of D.R.S. and J.M. for helpful discussions, Adam W.J. Soh and Lena P. Basta for reading and editing the manuscript, and Litong Yang for help with statistics and maintaining worm strains. FT2296, WEH722, and PHX5270 strains were provided as gifts by Jeremy Nance (UW-Madison), Ann M Wehman (U. Denver), and Patrick Laurent (ULB). We also thank Laura Jameson for technical help with mitochondrial toxicants. This work was supported by R35GM118049 to D.R.S, P42ES010356 and T32ES021432 to J.M.

Footnotes

Competing interests statement

The authors declare no competing interests. Authors received no specific funding for this work.

Data availability

All data supporting the findings of this study are available in the main text, extended data figures and supplementary materials. Uncropped versions of confocal images supporting main figures are displayed in Supplementary Fig. 6. C. elegans strains used in this study are listed in Supplementary Table 1. Oligonucleotides used in the study are listed in Supplementary Table 2. All stable reagents generated in this study are available from the lead contact without restriction upon request. Requests for further information and resources should be directed to and will be fulfilled by the lead contact, David R. Sherwood (david.sherwood@duke.edu). Source data are provided with this article.

Code availability

No custom code was developed or used in this study.

References

1.Johnston IG & Williams BP Evolutionary Inference across Eukaryotes Identifies Specific Pressures Favoring Mitochondrial Gene Retention. Cell Systems 2, 101–111 (2016). [DOI] [PubMed] [Google Scholar]
2.Thouas GA, Trounson AO, Wolvetang EJ & Jones GM Mitochondrial dysfunction in mouse oocytes results in preimplantation embryo arrest in vitro. Biol Reprod 71, 1936–1942 (2004). [DOI] [PubMed] [Google Scholar]
3.Taylor RW & Turnbull DM Mitochondrial DNA mutations in human disease. Nature Reviews Genetics 6, 389–402 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Tesarik J & Mendoza-Tesarik R Mitochondria in Human Fertility and Infertility. Int J Mol Sci 24 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Andreas E, Reid M, Zhang W & Moley KH The effect of maternal high-fat/high-sugar diet on offspring oocytes and early embryo development. Molecular Human Reproduction 25, 717–728 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Malott KF & Luderer U Toxicant effects on mammalian oocyte mitochondria†. Biol Reprod 104, 784–793 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Otten ABC et al. Replication Errors Made During Oogenesis Lead to Detectable De Novo mtDNA Mutations in Zebrafish Oocytes with a Low mtDNA Copy Number. Genetics 204, 1423–1431 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Denver DR, Morris K, Lynch M, Vassilieva LL & Thomas WK High direct estimate of the mutation rate in the mitochondrial genome of Caenorhabditis elegans. Science 289, 2342–2344 (2000). [DOI] [PubMed] [Google Scholar]
9.Fu Y, Tigano M & Sfeir A Safeguarding mitochondrial genomes in higher eukaryotes. Nature Structural & Molecular Biology 27, 687–695 (2020). [DOI] [PubMed] [Google Scholar]
10.Howe DK & Denver DR Muller’s Ratchet and compensatory mutation in Caenorhabditis briggsae mitochondrial genome evolution. BMC Evolutionary Biology 8, 62 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Muller HJ The relation of recombination to mutational advance. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis 1, 2–9 (1964). [DOI] [PubMed] [Google Scholar]
12.Guo Y et al. Very low-level heteroplasmy mtDNA variations are inherited in humans. J Genet Genomics 40, 607–615 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Schwartz AZA, Tsyba N, Abdu Y, Patel MR & Nance J Independent regulation of mitochondrial DNA quantity and quality in Caenorhabditis elegans primordial germ cells. eLife 11, e80396 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Lieber T, Jeedigunta SP, Palozzi JM, Lehmann R & Hurd TR Mitochondrial fragmentation drives selective removal of deleterious mtDNA in the germline. Nature 570, 380–384 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Palozzi JM et al. Mitochondrial DNA quality control in the female germline requires a unique programmed mitophagy. Cell Metabolism 34, 1809–1823.e1806 (2022). [DOI] [PubMed] [Google Scholar]
16.Zhang Y et al. PINK1 Inhibits Local Protein Synthesis to Limit Transmission of Deleterious Mitochondrial DNA Mutations. Mol Cell 73, 1127–1137.e1125 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Hill JH, Chen Z & Xu H Selective propagation of functional mitochondrial DNA during oogenesis restricts the transmission of a deleterious mitochondrial variant. Nature Genetics 46, 389–392 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Flowers S et al. Regulation of defective mitochondrial DNA accumulation and transmission in C. elegans by the programmed cell death and aging pathways. eLife 12, e79725 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Wai T et al. The Role of Mitochondrial DNA Copy Number in Mammalian Fertility1. Biology of Reproduction 83, 52–62 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Charmpilas N & Tavernarakis N Mitochondrial maturation drives germline stem cell differentiation in Caenorhabditis elegans. Cell Death & Differentiation 27, 601–617 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Lee SK et al. The association of mitochondrial potential and copy number with pig oocyte maturation and developmental potential. J Reprod Dev 60, 128–135 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Ge H et al. The importance of mitochondrial metabolic activity and mitochondrial DNA replication during oocyte maturation in vitro on oocyte quality and subsequent embryo developmental competence. Mol Reprod Dev 79, 392–401 (2012). [DOI] [PubMed] [Google Scholar]
23.Braeckman BP, Smolders A, Back P & De Henau S In Vivo Detection of Reactive Oxygen Species and Redox Status in Caenorhabditis elegans. Antioxid Redox Signal 25, 577–592 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.De Fanti S et al. Intra-individual purifying selection on mitochondrial DNA variants during human oogenesis. Human Reproduction 32, 1100–1107 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Krakauer DC & Mira A Mitochondria and germ-cell death. Nature 400, 125–126 (1999). [DOI] [PubMed] [Google Scholar]
26.Perez GI, Trbovich AM, Gosden RG & Tilly JL Mitochondria and the death of oocytes. Nature 403, 500–501 (2000). [DOI] [PubMed] [Google Scholar]
27.Chan CCW et al. Mitochondrial DNA content and 4977 bp deletion in unfertilized oocytes. Molecular Human Reproduction 11, 843–846 (2006). [DOI] [PubMed] [Google Scholar]
28.Gibson TC, Kubisch HM & Brenner CA Mitochondrial DNA deletions in rhesus macaque oocytes and embryos. Molecular Human Reproduction 11, 785–789 (2005). [DOI] [PubMed] [Google Scholar]
29.Ma H et al. Deleterious mtDNA mutations are common in mature oocytes. Biology of Reproduction 102, 607–619 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Brenner CA et al. Mitochondrial DNA deletion in human oocytes and embryos. Mol Hum Reprod 4, 887–892 (1998). [DOI] [PubMed] [Google Scholar]
31.Monnot S et al. Mutation dependance of the mitochondrial DNA copy number in the first stages of human embryogenesis. Human Molecular Genetics 22, 1867–1872 (2013). [DOI] [PubMed] [Google Scholar]
32.Barritt JA, Brenner CA, Cohen J & Matt DW Mitochondrial DNA rearrangements in human oocytes and embryos. Molecular Human Reproduction 5, 927–933 (1999). [DOI] [PubMed] [Google Scholar]
33.Kremer LS et al. A role for BCL2L13 and autophagy in germline purifying selection of mtDNA. PLOS Genetics 19, e1010573 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Gitschlag BL, Tate AT & Patel MR Nutrient status shapes selfish mitochondrial genome dynamics across different levels of selection. eLife 9, e56686 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Liu L et al. Mitochondrial outer-membrane protein FUNDC1 mediates hypoxia-induced mitophagy in mammalian cells. Nature Cell Biology 14, 177–185 (2012). [DOI] [PubMed] [Google Scholar]
36.McCarter J, Bartlett B, Dang T & Schedl T On the Control of Oocyte Meiotic Maturation and Ovulation inCaenorhabditis elegans. Developmental Biology 205, 111–128 (1999). [DOI] [PubMed] [Google Scholar]
37.Stoeckius M et al. Global characterization of the oocyte-to-embryo transition in Caenorhabditis elegans uncovers a novel mRNA clearance mechanism. The EMBO Journal 33, 1751–1766 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Ahringer J, Rosenquist TA, Lawson DN & Kimble J The Caenorhabditis elegans sex determining gene fem-3 is regulated post-transcriptionally. Embo j 11, 2303–2310 (1992). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Bohnert KA & Kenyon C A lysosomal switch triggers proteostasis renewal in the immortal C. elegans germ lineage. Nature 551, 629–633 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Yamamoto H, Zhang S & Mizushima N Autophagy genes in biology and disease. Nature Reviews Genetics 24, 382–400 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Tzur YB et al. Spatiotemporal Gene Expression Analysis of the Caenorhabditis elegans Germline Uncovers a Syncytial Expression Switch. Genetics 210, 587–605 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Shcherbo D et al. Far-red fluorescent tags for protein imaging in living tissues. Biochemical Journal 418, 567–574 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Shang A, Shao S, Zhao L & Liu B Far-Red Fluorescent Proteins: Tools for Advancing In Vivo Imaging. Biosensors (Basel) 14 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Burman JL et al. Mitochondrial fission facilitates the selective mitophagy of protein aggregates. Journal of Cell Biology 216, 3231–3247 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Mitra K & Lippincott-Schwartz J Analysis of mitochondrial dynamics and functions using imaging approaches. Curr Protoc Cell Biol Chapter 4, Unit 4.25.21-21 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Wang S et al. The mitophagy pathway and its implications in human diseases. Signal Transduction and Targeted Therapy 8, 304 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Wu W et al. FUNDC1 regulates mitochondrial dynamics at the ER-mitochondrial contact site under hypoxic conditions. The EMBO Journal 35, 1368–1384-1384 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Lim Y et al. Fndc-1 contributes to paternal mitochondria elimination in C. elegans. Dev Biol 454, 15–20 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Chatel-Chaix L et al. Dengue Virus Perturbs Mitochondrial Morphodynamics to Dampen Innate Immune Responses. Cell Host Microbe 20, 342–356 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Das R & Chakrabarti O Mitochondrial hyperfusion: a friend or a foe. Biochem Soc Trans 48, 631–644 (2020). [DOI] [PubMed] [Google Scholar]
51.Lim Y et al. FNDC-1-mediated mitophagy and ATFS-1 coordinate to protect against hypoxia-reoxygenation. Autophagy 17, 3389–3401 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Dilberger B et al. Mitochondrial Oxidative Stress Impairs Energy Metabolism and Reduces Stress Resistance and Longevity of C. elegans. Oxidative Medicine and Cellular Longevity 2019, 6840540 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Terada H Uncouplers of oxidative phosphorylation. Environmental Health Perspectives 87, 213–218 (1990). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Hua X, Liang G, Chao J & Wang D Exposure to 6-PPD quinone causes damage on mitochondrial complex I/II associated with lifespan reduction in Caenorhabditis elegans. Journal of Hazardous Materials 472, 134598 (2024). [DOI] [PubMed] [Google Scholar]
55.Bess AS, Crocker TL, Ryde IT & Meyer JN Mitochondrial dynamics and autophagy aid in removal of persistent mitochondrial DNA damage in Caenorhabditis elegans. Nucleic Acids Research 40, 7916–7931 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Zhang W & Wu F Effects of adverse fertility-related factors on mitochondrial DNA in the oocyte: a comprehensive review. Reprod Biol Endocrinol 21, 27 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Leuthner Tess C. et al. Resistance of mitochondrial DNA to cadmium and Aflatoxin B1 damage-induced germline mutation accumulation in C. elegans. Nucleic Acids Research 50, 8626–8642 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Meshnik L et al. Mutant C. elegans mitofusin leads to selective removal of mtDNA heteroplasmic deletions across generations to maintain fitness. BMC Biology 20, 40 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Liau WS, Gonzalez-Serricchio AS, Deshommes C, Chin K & LaMunyon CW A persistent mitochondrial deletion reduces fitness and sperm performance in heteroplasmic populations of C. elegans. BMC Genet 8, 8 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Rodríguez-Nuevo A et al. Oocytes maintain ROS-free mitochondrial metabolism by suppressing complex I. Nature 607, 756–761 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Kim K, Kenigsberg S, Jurisicova A & Bentov Y The Role of Mitochondria in Oocyte and Early Embryo Health. OBM Genetics 03, 070 (2019). [Google Scholar]
62.Spikings EC, Alderson J & John JCS Regulated Mitochondrial DNA Replication During Oocyte Maturation Is Essential for Successful Porcine Embryonic Development. Biology of Reproduction 76, 327–335 (2007). [DOI] [PubMed] [Google Scholar]
63.Han L et al. Increased mtDNA mutation frequency in oocytes causes epigenetic alterations and embryonic defects. National Science Review 9 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Field JT & Gordon JW BNIP3 and Nix: Atypical regulators of cell fate. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1869, 119325 (2022). [DOI] [PubMed] [Google Scholar]
65.Lin Y-F et al. Maintenance and propagation of a deleterious mitochondrial genome by the mitochondrial unfolded protein response. Nature 533, 416–419 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
66.Ren J et al. FUNDC1 interacts with FBXL2 to govern mitochondrial integrity and cardiac function through an IP3R3-dependent manner in obesity. Science Advances 6, eabc8561 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Samuel ADT, Murthy VN & Hengartner MO Calcium dynamics during fertilization in C. elegans. BMC Developmental Biology 1, 8 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
68.Kurihara Y et al. Mitophagy plays an essential role in reducing mitochondrial production of reactive oxygen species and mutation of mitochondrial DNA by maintaining mitochondrial quantity and quality in yeast. J Biol Chem 287, 3265–3272 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Mathew R & White E Autophagy, stress, and cancer metabolism: what doesn’t kill you makes you stronger. Cold Spring Harb Symp Quant Biol 76, 389–396 (2011). [DOI] [PubMed] [Google Scholar]
70.Cagnone GL et al. Restoration of normal embryogenesis by mitochondrial supplementation in pig oocytes exhibiting mitochondrial DNA deficiency. Sci Rep 6, 23229 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
71.Chiaratti MR & Meirelles FV Mitochondrial DNA Copy Number, a Marker of Viability for Oocytes1. Biology of Reproduction 83, 1–2 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
72.Tsang WY & Lemire BD Mitochondrial genome content is regulated during nematode development. Biochem Biophys Res Commun 291, 8–16 (2002). [DOI] [PubMed] [Google Scholar]
73.Hsieh R-H et al. Multiple rearrangements of mitochondrial DNA in unfertilized human oocytes. Fertility and Sterility 77, 1012–1017 (2002). [DOI] [PubMed] [Google Scholar]
74.Ru Y et al. Maternal age enhances purifying selection on pathogenic mutations in complex I genes of mammalian mtDNA. Nature Aging 4, 1211–1230 (2024). [DOI] [PubMed] [Google Scholar]
75.Fan W et al. A Mouse Model of Mitochondrial Disease Reveals Germline Selection Against Severe mtDNA Mutations. Science 319, 958–962 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
76.Zhang H et al. Mitochondrial DNA heteroplasmy is modulated during oocyte development propagating mutation transmission. Science Advances 7, eabi5657 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Ross JM et al. Germline mitochondrial DNA mutations aggravate ageing and can impair brain development. Nature 501, 412–415 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
78.Hahn A et al. Misregulation of mitochondrial 6mA promotes the propagation of mutant mtDNA and causes aging in C. elegans. Cell Metabolism 36, 2528–2541.e2511 (2024). [DOI] [PubMed] [Google Scholar]
79.Otten Auke B.C. et al. Differences in Strength and Timing of the mtDNA Bottleneck between Zebrafish Germline and Non-germline Cells. Cell Reports 16, 622–630 (2016). [DOI] [PubMed] [Google Scholar]
80.Cao L et al. The mitochondrial bottleneck occurs without reduction of mtDNA content in female mouse germ cells. Nature Genetics 39, 386–390 (2007). [DOI] [PubMed] [Google Scholar]
81.Zhang Y et al. Polyamine metabolite spermidine rejuvenates oocyte quality by enhancing mitophagy during female reproductive aging. Nature Aging 3, 1372–1386 (2023). [DOI] [PubMed] [Google Scholar]
82.Xu G et al. Expression pattern and functional analysis of fundc1 in rare minnow (Gobiocypris rarus). Gene 626, 149–157 (2017). [DOI] [PubMed] [Google Scholar]
83.Xu G et al. Fundc1 is necessary for proper body axis formation during embryogenesis in zebrafish. Scientific Reports 9, 18910 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
84.Kronja I et al. Quantitative proteomics reveals the dynamics of protein changes during Drosophila oocyte maturation and the oocyte-to-embryo transition. Proceedings of the National Academy of Sciences 111, 16023–16028 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

Methods References

85.Brenner S THE GENETICS OF CAENORHABDITIS ELEGANS. Genetics 77, 71–94 (1974). [DOI] [PMC free article] [PubMed] [Google Scholar]
86.Dickinson DJ, Ward JD, Reiner DJ & Goldstein B Engineering the Caenorhabditis elegans genome using Cas9-triggered homologous recombination. Nat Methods 10, 1028–1034 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
87.Shaner NC et al. A bright monomeric green fluorescent protein derived from Branchiostoma lanceolatum. Nat Methods 10, 407–409 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
88.Fang J et al. Ribo-On and Ribo-Off tools using a self-cleaving ribozyme allow manipulation of endogenous gene expression in C. elegans. Communications Biology 6, 816 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
89.Raiders SA, Eastwood MD, Bacher M & Priess JR Binucleate germ cells in Caenorhabditis elegans are removed by physiological apoptosis. PLOS Genetics 14, e1007417 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
90.Poteryaev D, Squirrell JM, Campbell JM, White JG & Spang A Involvement of the actin cytoskeleton and homotypic membrane fusion in ER dynamics in Caenorhabditis elegans. Mol Biol Cell 16, 2139–2153 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
91.Stavoe Andrea K.H., Kargbo-Hill Sarah E., Hall David H. & Colón-Ramos Daniel A. KIF1A/UNC-104 Transports ATG-9 to Regulate Neurodevelopment and Autophagy at Synapses. Developmental Cell 38, 171–185 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
92.Sämann J et al. Caenorhabditits elegans LRK-1 and PINK-1 act antagonistically in stress response and neurite outgrowth. J Biol Chem 284, 16482–16491 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
93.Mirza Z, Walhout AJM & Ambros V A bacterial pathogen induces developmental slowing by high reactive oxygen species and mitochondrial dysfunction in Caenorhabditis elegans. Cell Rep 42, 113189 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
94.Tsang WY & Lemire BD Stable heteroplasmy but differential inheritance of a large mitochondrial DNA deletion in nematodes. Biochemistry and Cell Biology 80, 645–654 (2002). [DOI] [PubMed] [Google Scholar]
95.Kelley LC et al. Live-cell confocal microscopy and quantitative 4D image analysis of anchor-cell invasion through the basement membrane in Caenorhabditis elegans. Nat Protoc 12, 2081–2096 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
96.Walston T & Hardin J An agar mount for observation of Caenorhabditis elegans embryos. Cold Spring Harb Protoc 2010, pdb.prot5540 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
97.Wolke U, Jezuit EA & Priess JR Actin-dependent cytoplasmic streaming in C. elegansoogenesis. Development 134, 2227–2236 (2007). [DOI] [PubMed] [Google Scholar]
98.Burdet V et al. Ultrastructure Expansion Microscopy applied to C. elegans embryos. MicroPubl Biol 2024 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
99.da Rosa J et al. Optimizing dsRNA engineering strategies and production in E. coli HT115 (DE3). J Ind Microbiol Biotechnol 51 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
100.Kamath RS & Ahringer J Genome-wide RNAi screening in Caenorhabditis elegans. Methods 30, 313–321 (2003). [DOI] [PubMed] [Google Scholar]
101.Rual JF et al. Toward improving Caenorhabditis elegans phenome mapping with an ORFeome-based RNAi library. Genome Res 14, 2162–2168 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
102.Lowry J et al. High-Throughput Cloning of Temperature-Sensitive Caenorhabditis elegans Mutants with Adult Syncytial Germline Membrane Architecture Defects. G3 Genes|Genomes|Genetics 5, 2241–2255 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
103.Fu H et al. Wounding triggers MIRO-1 dependent mitochondrial fragmentation that accelerates epidermal wound closure through oxidative signaling. Nature Communications 11, 1050 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
104.Kolli S, Kline CJ, Rad KM & Wehman AM Phagolysosomes break down the membrane of a non-apoptotic corpse independent of macroautophagy. PLOS ONE 19, e0306435 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
105.Njume FN et al. A lipid transfer protein ensures nematode cuticular impermeability. iScience 25, 105357 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
106.Rodriguez ME et al. Targeting of mitochondria by 10-N-alkyl acridine orange analogues: role of alkyl chain length in determining cellular uptake and localization. Mitochondrion 8, 237–246 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
107.Ashley N, Harris D & Poulton J Detection of mitochondrial DNA depletion in living human cells using PicoGreen staining. Experimental Cell Research 303, 432–446 (2005). [DOI] [PubMed] [Google Scholar]
108.Hemel IMGM, Engelen BPH, Luber N & Gerards M A hitchhiker’s guide to mitochondrial quantification. Mitochondrion 59, 216–224 (2021). [DOI] [PubMed] [Google Scholar]
109.Soh AWJ et al. On-demand delivery of fibulin-1 protects the basement membrane during cyclic stretching in C. elegans. Dev Cell (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
110.Klass MR Aging in the nematode Caenorhabditis elegans: Major biological and environmental factors influencing life span. Mechanisms of Ageing and Development 6, 413–429 (1977). [DOI] [PubMed] [Google Scholar]
111.Hughes SE, Huang C & Kornfeld K Identification of Mutations that Delay Somatic or Reproductive Aging of Caenorhabditis elegans. Genetics 189, 341–356 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
112.Luo S, Kleemann GA, Ashraf JM, Shaw WM & Murphy CT TGF-beta and Insulin Signaling Regulate Reproductive Aging via Oocyte and Germline Quality Maintenance. Cell 143, 299–312 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
113.Montava-Garriga L, Singh F, Ball G & Ganley IG Semi-automated quantitation of mitophagy in cells and tissues. Mech Ageing Dev 185, 111196 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
114.Fan X et al. SapTrap Assembly of Caenorhabditis elegans MosSCI Transgene Vectors. G3 (Bethesda) 10, 635–644 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
115.Schärer OD Nucleotide excision repair in eukaryotes. Cold Spring Harb Perspect Biol 5, a012609 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
116.Crowley LC, Christensen ME & Waterhouse NJ Measuring Mitochondrial Transmembrane Potential by TMRE Staining. Cold Spring Harb Protoc 2016 (2016). [DOI] [PubMed] [Google Scholar]
117.Boyd WA, Smith MV & Freedman JH Caenorhabditis elegans as a Model in Developmental Toxicology, in Developmental Toxicology: Methods and Protocols. (eds. Harris C & Hansen JM) 15–24 (Humana Press, Totowa, NJ; 2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
118.Taylor SKB et al. C. elegans electrotaxis behavior is modulated by heat shock response and unfolded protein response signaling pathways. Scientific Reports 11, 3115 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
119.Srivastava V & Gross E Detection of Mitophagy in Caenorhabditis elegans and Mammalian Cells Using Organelle-Specific Dyes. J Vis Exp (2023). [DOI] [PubMed] [Google Scholar]
120.Leuthner TC, Hartman JH, Ryde IT & Meyer JN PCR-Based Determination of Mitochondrial DNA Copy Number in Multiple Species, in Mitochondrial Regulation: Methods and Protocols. (eds. Palmeira CM & Rolo AP) 91–111 (Springer US, New York, NY; 2021). [DOI] [PubMed] [Google Scholar]
121.Longhini KM & Glotzer M Aurora A and cortical flows promote polarization and cytokinesis by inducing asymmetric ECT-2 accumulation. eLife 11, e83992 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
122.d’Humières J, Wang L, Sherwood DR & Plastino J The actin protrusion deforms the nucleus during invasion through basement membrane. bioRxiv (2025). [Google Scholar]
123.Norris AD, Kim HM, Colaiácovo MP & Calarco JA Efficient Genome Editing in Caenorhabditis elegans with a Toolkit of Dual-Marker Selection Cassettes. Genetics 201, 449–458 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
124.Rieckher M & Tavernarakis N Caenorhabditis elegans Microinjection. Bio Protoc 7 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
125.Kelly WG & Fire A Chromatin silencing and the maintenance of a functional germline in Caenorhabditis elegans. Development 125, 2451–2456 (1998). [DOI] [PMC free article] [PubMed] [Google Scholar]
126.Chen S & Phillips CM Silencing of a NRDE-3 transgene in C. elegans germ cells and early embryos is mediated by the RNAi pathway. MicroPubl Biol 2024 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
127.Luz AL, Smith LL, Rooney JP & Meyer JN Seahorse Xfe 24 Extracellular Flux Analyzer-Based Analysis of Cellular Respiration in Caenorhabditis elegans. Curr Protoc Toxicol 66, 25.27.21–25.27.15 (2015) [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Video 1

Download video file^{(66.8MB, mp4)}

Video 2

Download video file^{(43.4MB, mp4)}

Supl Material

NIHMS2137744-supplement-Supl_Material.pdf^{(43.5MB, pdf)}

Data Availability Statement

No custom code was developed or used in this study.

[R1] 1.Johnston IG & Williams BP Evolutionary Inference across Eukaryotes Identifies Specific Pressures Favoring Mitochondrial Gene Retention. Cell Systems 2, 101–111 (2016). [DOI] [PubMed] [Google Scholar]

[R2] 2.Thouas GA, Trounson AO, Wolvetang EJ & Jones GM Mitochondrial dysfunction in mouse oocytes results in preimplantation embryo arrest in vitro. Biol Reprod 71, 1936–1942 (2004). [DOI] [PubMed] [Google Scholar]

[R3] 3.Taylor RW & Turnbull DM Mitochondrial DNA mutations in human disease. Nature Reviews Genetics 6, 389–402 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Tesarik J & Mendoza-Tesarik R Mitochondria in Human Fertility and Infertility. Int J Mol Sci 24 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] 5.Andreas E, Reid M, Zhang W & Moley KH The effect of maternal high-fat/high-sugar diet on offspring oocytes and early embryo development. Molecular Human Reproduction 25, 717–728 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Malott KF & Luderer U Toxicant effects on mammalian oocyte mitochondria†. Biol Reprod 104, 784–793 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] 7.Otten ABC et al. Replication Errors Made During Oogenesis Lead to Detectable De Novo mtDNA Mutations in Zebrafish Oocytes with a Low mtDNA Copy Number. Genetics 204, 1423–1431 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] 8.Denver DR, Morris K, Lynch M, Vassilieva LL & Thomas WK High direct estimate of the mutation rate in the mitochondrial genome of Caenorhabditis elegans. Science 289, 2342–2344 (2000). [DOI] [PubMed] [Google Scholar]

[R9] 9.Fu Y, Tigano M & Sfeir A Safeguarding mitochondrial genomes in higher eukaryotes. Nature Structural & Molecular Biology 27, 687–695 (2020). [DOI] [PubMed] [Google Scholar]

[R10] 10.Howe DK & Denver DR Muller’s Ratchet and compensatory mutation in Caenorhabditis briggsae mitochondrial genome evolution. BMC Evolutionary Biology 8, 62 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R11] 11.Muller HJ The relation of recombination to mutational advance. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis 1, 2–9 (1964). [DOI] [PubMed] [Google Scholar]

[R12] 12.Guo Y et al. Very low-level heteroplasmy mtDNA variations are inherited in humans. J Genet Genomics 40, 607–615 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] 13.Schwartz AZA, Tsyba N, Abdu Y, Patel MR & Nance J Independent regulation of mitochondrial DNA quantity and quality in Caenorhabditis elegans primordial germ cells. eLife 11, e80396 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Lieber T, Jeedigunta SP, Palozzi JM, Lehmann R & Hurd TR Mitochondrial fragmentation drives selective removal of deleterious mtDNA in the germline. Nature 570, 380–384 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Palozzi JM et al. Mitochondrial DNA quality control in the female germline requires a unique programmed mitophagy. Cell Metabolism 34, 1809–1823.e1806 (2022). [DOI] [PubMed] [Google Scholar]

[R16] 16.Zhang Y et al. PINK1 Inhibits Local Protein Synthesis to Limit Transmission of Deleterious Mitochondrial DNA Mutations. Mol Cell 73, 1127–1137.e1125 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] 17.Hill JH, Chen Z & Xu H Selective propagation of functional mitochondrial DNA during oogenesis restricts the transmission of a deleterious mitochondrial variant. Nature Genetics 46, 389–392 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] 18.Flowers S et al. Regulation of defective mitochondrial DNA accumulation and transmission in C. elegans by the programmed cell death and aging pathways. eLife 12, e79725 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] 19.Wai T et al. The Role of Mitochondrial DNA Copy Number in Mammalian Fertility1. Biology of Reproduction 83, 52–62 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Charmpilas N & Tavernarakis N Mitochondrial maturation drives germline stem cell differentiation in Caenorhabditis elegans. Cell Death & Differentiation 27, 601–617 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Lee SK et al. The association of mitochondrial potential and copy number with pig oocyte maturation and developmental potential. J Reprod Dev 60, 128–135 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] 22.Ge H et al. The importance of mitochondrial metabolic activity and mitochondrial DNA replication during oocyte maturation in vitro on oocyte quality and subsequent embryo developmental competence. Mol Reprod Dev 79, 392–401 (2012). [DOI] [PubMed] [Google Scholar]

[R23] 23.Braeckman BP, Smolders A, Back P & De Henau S In Vivo Detection of Reactive Oxygen Species and Redox Status in Caenorhabditis elegans. Antioxid Redox Signal 25, 577–592 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] 24.De Fanti S et al. Intra-individual purifying selection on mitochondrial DNA variants during human oogenesis. Human Reproduction 32, 1100–1107 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] 25.Krakauer DC & Mira A Mitochondria and germ-cell death. Nature 400, 125–126 (1999). [DOI] [PubMed] [Google Scholar]

[R26] 26.Perez GI, Trbovich AM, Gosden RG & Tilly JL Mitochondria and the death of oocytes. Nature 403, 500–501 (2000). [DOI] [PubMed] [Google Scholar]

[R27] 27.Chan CCW et al. Mitochondrial DNA content and 4977 bp deletion in unfertilized oocytes. Molecular Human Reproduction 11, 843–846 (2006). [DOI] [PubMed] [Google Scholar]

[R28] 28.Gibson TC, Kubisch HM & Brenner CA Mitochondrial DNA deletions in rhesus macaque oocytes and embryos. Molecular Human Reproduction 11, 785–789 (2005). [DOI] [PubMed] [Google Scholar]

[R29] 29.Ma H et al. Deleterious mtDNA mutations are common in mature oocytes. Biology of Reproduction 102, 607–619 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] 30.Brenner CA et al. Mitochondrial DNA deletion in human oocytes and embryos. Mol Hum Reprod 4, 887–892 (1998). [DOI] [PubMed] [Google Scholar]

[R31] 31.Monnot S et al. Mutation dependance of the mitochondrial DNA copy number in the first stages of human embryogenesis. Human Molecular Genetics 22, 1867–1872 (2013). [DOI] [PubMed] [Google Scholar]

[R32] 32.Barritt JA, Brenner CA, Cohen J & Matt DW Mitochondrial DNA rearrangements in human oocytes and embryos. Molecular Human Reproduction 5, 927–933 (1999). [DOI] [PubMed] [Google Scholar]

[R33] 33.Kremer LS et al. A role for BCL2L13 and autophagy in germline purifying selection of mtDNA. PLOS Genetics 19, e1010573 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Gitschlag BL, Tate AT & Patel MR Nutrient status shapes selfish mitochondrial genome dynamics across different levels of selection. eLife 9, e56686 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R35] 35.Liu L et al. Mitochondrial outer-membrane protein FUNDC1 mediates hypoxia-induced mitophagy in mammalian cells. Nature Cell Biology 14, 177–185 (2012). [DOI] [PubMed] [Google Scholar]

[R36] 36.McCarter J, Bartlett B, Dang T & Schedl T On the Control of Oocyte Meiotic Maturation and Ovulation inCaenorhabditis elegans. Developmental Biology 205, 111–128 (1999). [DOI] [PubMed] [Google Scholar]

[R37] 37.Stoeckius M et al. Global characterization of the oocyte-to-embryo transition in Caenorhabditis elegans uncovers a novel mRNA clearance mechanism. The EMBO Journal 33, 1751–1766 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] 38.Ahringer J, Rosenquist TA, Lawson DN & Kimble J The Caenorhabditis elegans sex determining gene fem-3 is regulated post-transcriptionally. Embo j 11, 2303–2310 (1992). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R39] 39.Bohnert KA & Kenyon C A lysosomal switch triggers proteostasis renewal in the immortal C. elegans germ lineage. Nature 551, 629–633 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R40] 40.Yamamoto H, Zhang S & Mizushima N Autophagy genes in biology and disease. Nature Reviews Genetics 24, 382–400 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R41] 41.Tzur YB et al. Spatiotemporal Gene Expression Analysis of the Caenorhabditis elegans Germline Uncovers a Syncytial Expression Switch. Genetics 210, 587–605 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R42] 42.Shcherbo D et al. Far-red fluorescent tags for protein imaging in living tissues. Biochemical Journal 418, 567–574 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R43] 43.Shang A, Shao S, Zhao L & Liu B Far-Red Fluorescent Proteins: Tools for Advancing In Vivo Imaging. Biosensors (Basel) 14 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R44] 44.Burman JL et al. Mitochondrial fission facilitates the selective mitophagy of protein aggregates. Journal of Cell Biology 216, 3231–3247 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R45] 45.Mitra K & Lippincott-Schwartz J Analysis of mitochondrial dynamics and functions using imaging approaches. Curr Protoc Cell Biol Chapter 4, Unit 4.25.21-21 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R46] 46.Wang S et al. The mitophagy pathway and its implications in human diseases. Signal Transduction and Targeted Therapy 8, 304 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R47] 47.Wu W et al. FUNDC1 regulates mitochondrial dynamics at the ER-mitochondrial contact site under hypoxic conditions. The EMBO Journal 35, 1368–1384-1384 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R48] 48.Lim Y et al. Fndc-1 contributes to paternal mitochondria elimination in C. elegans. Dev Biol 454, 15–20 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R49] 49.Chatel-Chaix L et al. Dengue Virus Perturbs Mitochondrial Morphodynamics to Dampen Innate Immune Responses. Cell Host Microbe 20, 342–356 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R50] 50.Das R & Chakrabarti O Mitochondrial hyperfusion: a friend or a foe. Biochem Soc Trans 48, 631–644 (2020). [DOI] [PubMed] [Google Scholar]

[R51] 51.Lim Y et al. FNDC-1-mediated mitophagy and ATFS-1 coordinate to protect against hypoxia-reoxygenation. Autophagy 17, 3389–3401 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] 52.Dilberger B et al. Mitochondrial Oxidative Stress Impairs Energy Metabolism and Reduces Stress Resistance and Longevity of C. elegans. Oxidative Medicine and Cellular Longevity 2019, 6840540 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R53] 53.Terada H Uncouplers of oxidative phosphorylation. Environmental Health Perspectives 87, 213–218 (1990). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R54] 54.Hua X, Liang G, Chao J & Wang D Exposure to 6-PPD quinone causes damage on mitochondrial complex I/II associated with lifespan reduction in Caenorhabditis elegans. Journal of Hazardous Materials 472, 134598 (2024). [DOI] [PubMed] [Google Scholar]

[R55] 55.Bess AS, Crocker TL, Ryde IT & Meyer JN Mitochondrial dynamics and autophagy aid in removal of persistent mitochondrial DNA damage in Caenorhabditis elegans. Nucleic Acids Research 40, 7916–7931 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R56] 56.Zhang W & Wu F Effects of adverse fertility-related factors on mitochondrial DNA in the oocyte: a comprehensive review. Reprod Biol Endocrinol 21, 27 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R57] 57.Leuthner Tess C. et al. Resistance of mitochondrial DNA to cadmium and Aflatoxin B1 damage-induced germline mutation accumulation in C. elegans. Nucleic Acids Research 50, 8626–8642 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R58] 58.Meshnik L et al. Mutant C. elegans mitofusin leads to selective removal of mtDNA heteroplasmic deletions across generations to maintain fitness. BMC Biology 20, 40 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R59] 59.Liau WS, Gonzalez-Serricchio AS, Deshommes C, Chin K & LaMunyon CW A persistent mitochondrial deletion reduces fitness and sperm performance in heteroplasmic populations of C. elegans. BMC Genet 8, 8 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R60] 60.Rodríguez-Nuevo A et al. Oocytes maintain ROS-free mitochondrial metabolism by suppressing complex I. Nature 607, 756–761 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R61] 61.Kim K, Kenigsberg S, Jurisicova A & Bentov Y The Role of Mitochondria in Oocyte and Early Embryo Health. OBM Genetics 03, 070 (2019). [Google Scholar]

[R62] 62.Spikings EC, Alderson J & John JCS Regulated Mitochondrial DNA Replication During Oocyte Maturation Is Essential for Successful Porcine Embryonic Development. Biology of Reproduction 76, 327–335 (2007). [DOI] [PubMed] [Google Scholar]

[R63] 63.Han L et al. Increased mtDNA mutation frequency in oocytes causes epigenetic alterations and embryonic defects. National Science Review 9 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R64] 64.Field JT & Gordon JW BNIP3 and Nix: Atypical regulators of cell fate. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1869, 119325 (2022). [DOI] [PubMed] [Google Scholar]

[R65] 65.Lin Y-F et al. Maintenance and propagation of a deleterious mitochondrial genome by the mitochondrial unfolded protein response. Nature 533, 416–419 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R66] 66.Ren J et al. FUNDC1 interacts with FBXL2 to govern mitochondrial integrity and cardiac function through an IP3R3-dependent manner in obesity. Science Advances 6, eabc8561 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R67] 67.Samuel ADT, Murthy VN & Hengartner MO Calcium dynamics during fertilization in C. elegans. BMC Developmental Biology 1, 8 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R68] 68.Kurihara Y et al. Mitophagy plays an essential role in reducing mitochondrial production of reactive oxygen species and mutation of mitochondrial DNA by maintaining mitochondrial quantity and quality in yeast. J Biol Chem 287, 3265–3272 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R69] 69.Mathew R & White E Autophagy, stress, and cancer metabolism: what doesn’t kill you makes you stronger. Cold Spring Harb Symp Quant Biol 76, 389–396 (2011). [DOI] [PubMed] [Google Scholar]

[R70] 70.Cagnone GL et al. Restoration of normal embryogenesis by mitochondrial supplementation in pig oocytes exhibiting mitochondrial DNA deficiency. Sci Rep 6, 23229 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R71] 71.Chiaratti MR & Meirelles FV Mitochondrial DNA Copy Number, a Marker of Viability for Oocytes1. Biology of Reproduction 83, 1–2 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R72] 72.Tsang WY & Lemire BD Mitochondrial genome content is regulated during nematode development. Biochem Biophys Res Commun 291, 8–16 (2002). [DOI] [PubMed] [Google Scholar]

[R73] 73.Hsieh R-H et al. Multiple rearrangements of mitochondrial DNA in unfertilized human oocytes. Fertility and Sterility 77, 1012–1017 (2002). [DOI] [PubMed] [Google Scholar]

[R74] 74.Ru Y et al. Maternal age enhances purifying selection on pathogenic mutations in complex I genes of mammalian mtDNA. Nature Aging 4, 1211–1230 (2024). [DOI] [PubMed] [Google Scholar]

[R75] 75.Fan W et al. A Mouse Model of Mitochondrial Disease Reveals Germline Selection Against Severe mtDNA Mutations. Science 319, 958–962 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R76] 76.Zhang H et al. Mitochondrial DNA heteroplasmy is modulated during oocyte development propagating mutation transmission. Science Advances 7, eabi5657 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R77] 77.Ross JM et al. Germline mitochondrial DNA mutations aggravate ageing and can impair brain development. Nature 501, 412–415 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R78] 78.Hahn A et al. Misregulation of mitochondrial 6mA promotes the propagation of mutant mtDNA and causes aging in C. elegans. Cell Metabolism 36, 2528–2541.e2511 (2024). [DOI] [PubMed] [Google Scholar]

[R79] 79.Otten Auke B.C. et al. Differences in Strength and Timing of the mtDNA Bottleneck between Zebrafish Germline and Non-germline Cells. Cell Reports 16, 622–630 (2016). [DOI] [PubMed] [Google Scholar]

[R80] 80.Cao L et al. The mitochondrial bottleneck occurs without reduction of mtDNA content in female mouse germ cells. Nature Genetics 39, 386–390 (2007). [DOI] [PubMed] [Google Scholar]

[R81] 81.Zhang Y et al. Polyamine metabolite spermidine rejuvenates oocyte quality by enhancing mitophagy during female reproductive aging. Nature Aging 3, 1372–1386 (2023). [DOI] [PubMed] [Google Scholar]

[R82] 82.Xu G et al. Expression pattern and functional analysis of fundc1 in rare minnow (Gobiocypris rarus). Gene 626, 149–157 (2017). [DOI] [PubMed] [Google Scholar]

[R83] 83.Xu G et al. Fundc1 is necessary for proper body axis formation during embryogenesis in zebrafish. Scientific Reports 9, 18910 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R84] 84.Kronja I et al. Quantitative proteomics reveals the dynamics of protein changes during Drosophila oocyte maturation and the oocyte-to-embryo transition. Proceedings of the National Academy of Sciences 111, 16023–16028 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Programmed mitophagy at the oocyte-to-zygote transition promotes lineage endurance

Siddharthan B Thendral

Sasha Bacot

Ian T Ryde

Katherine S Morton

Qiuyi Chi

Isabel W Kenny-Ganzert

Joel N Meyer

David R Sherwood

Abstract

Introduction

Results

C. elegans oocytes reduce mitochondrial content during transition to zygote

Fig. 1: Mitochondrial reduction during the C. elegans oocyte-to-zygote transition.

Mitochondrial reduction at OZT is developmentally programmed

Macroautophagy is required for programmed mitochondrial degradation during OZT

Fig. 2: Macroautophagy activation is required for mitochondrial degradation during OZT.

Mitochondrial fragmentation is necessary but not sufficient for MOZT

Fig. 3: DRP-1 mediated mitochondrial fragmentation is necessary for MOZT.

Table 1. RNAi screen to identify effectors of change in mitochondrial morphology during the C. elegans oocyte-to-zygote transition.

FUNDC1 regulates mitophagy and mitochondrial fragmentation during OZT

Fig. 4: FUNDC1 is required for mitophagy and mitochondrial fragmentation during OZT.

Stable FUNDC1 puncta localize at sites of mitochondrial fragmentation during OZT

Fig. 5: FUNDC1 puncta localize at sites of mitochondrial fragmentation during OZT.

Oocytes upregulate FUNDC1 in response to diverse mitochondrial damage

Fig. 6: MOZT executes purifying selection against uaDf5 during OZT, preventing accumulation of mutant mtDNA and population collapse over generations.

Oocyte-expressed FUNDC1 enhances reproductive tolerance to UV-induced damage

MOZT is a cross-generational mtDNA purifying selection event

MOZT promotes lineage continuity

Fig. 7: Loss of MOZT renders fertility and progeny survival susceptible to mtDNA heteroplasmy through a maternal effect.

Discussion

Methods

C. elegans strains and culture conditions

Genome editing

Strain generation through genetic crosses

Microscopy and image analysis

Mitochondrial abundance quantification

RNAi treatment

RNAi knockdown efficiency

Quantification of organelle/protein abundances at OZT

Germline extrusion for mtDNA nucleoid count quantification

Staining with mitochondrial, mtDNA dyes

Quantification of mitochondrial volume and morphology

Mitochondrial analysis in female animals

Mitochondrial acidification

Quantification of fragmentation events

Quantification of mRb3::FNDC-1 abundance at OMM and MERCs

Co-localization analysis

Quantification of mRb3::FNDC-1 puncta dwell time on mitochondria

UVC treatment

TMRE staining

Genetic bottlenecking experiments (mutation accumulation line generation)

Paraquat, FCCP, and 6PPD-Q treatment

uaDf5 heteroplasmy quantification

Isolation of oocytes and zygotes

Comparison of mtDNA abundance between wild-type and fndc-1 zygotes

Quantification of embryonic lethality and fertility

Mating experiments

Embryo-specific expression of fndc-1

Mitochondrial respiration

Experimental setup

Statistical analysis

Extended Data

Extended Data Fig. 1 |. Reduction in mtDNA nucleoids during the C. elegans oocyte-to-zygote transition.

Extended Data Fig. 2 |. Mitochondrial reduction during OZT is developmentally programmed and induced by sperm signalling.

Extended Data Fig. 3 |. Upregulation of the autophagy-lysosome system during the C. elegans oocyte-to-zygote transition.

Extended Data Fig. 4 |. Mitochondria are acidified during OZT; DRP-1 and FUNDC1 are necessary for mitochondrial acidification.

Extended Data Fig. 5 |. Rapid mitochondrial fragmentation during the oocyteto-zygote transition.

Extended Data Fig. 6 |. Mitochondrial fragmentation is not sufficient to regulate the timing and extent of mitochondrial reduction at OZT.

Extended Data Fig. 7 |. FUNDC1 puncta associate with ER and mitochondria, and exhibit two distinct behaviours on oocyte mitochondria.

Extended Data Fig. 8 |. Oocytes upregulate FUNDC1 levels upon mitochondrial stress, but OZT mitochondrial reduction and fragmentation are not altered.

Extended Data Fig. 9 |. fndc-1 embryos are vulnerable to maternally inherited UV-induced damage.

Extended Data Fig. 10 |. Reproductive decline and embryonic lethality in fndc-1 (UV) lines are strictly maternal effects.

Supplementary Material

Acknowledgements

Footnotes

Data availability

Code availability

References