Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
1 of 15
LIFE SCI E NCE S
Depletion of oocyte dynamin-related protein 1
shows maternal-effect abnormalities in 
embryonic development
Deepak Adhikari1
*, In-won Lee1
, Usama Al-Zubaidi1,2
, Jun Liu1
, Qing-Hua Zhang1
, Wai Shan Yuen1
,
Likun He1
, Yasmyn Winstanley3
, Hiromi Sesaki4
, Jeffrey R. Mann1
, Rebecca L. Robker1,5
, John Carroll1
*
Eggs contain about 200,000 mitochondria that generate adenosine triphosphate and metabolites essential for
oocyte development. Mitochondria also integrate metabolism and transcription via metabolites that regulate
epigenetic modifiers, but there is no direct evidence linking oocyte mitochondrial function to the maternal
epigenome and subsequent embryo development. Here, we have disrupted oocyte mitochondrial function via
deletion of the mitochondrial fission factor Drp1. Fission-deficient oocytes exhibit a high frequency of failure in
peri- and postimplantation development. This is associated with altered mitochondrial function, changes in the
oocyte transcriptome and proteome, altered subcortical maternal complex, and a decrease in oocyte DNA methylation
and H3K27me3. Transplanting pronuclei of fertilized Drp1 knockout oocytes to normal ooplasm fails to
rescue embryonic lethality. We conclude that mitochondrial function plays a role in establishing the maternal
epigenome, with serious consequences for embryo development.
INTRODUCTION
Mitochondria are essential organelles that generate cellular energy
in the form of adenosine triphosphate (ATP) through oxidative
phosphorylation (OXPHOS). Mitochondria also regulate apoptosis
through control of caspase activity and have roles in Ca2+
signaling
and in maintaining cellular redox balance (1,  2). Mitochondrial
quality and function are dependent on mitochondrial fission and
fusion events, also known as mitochondrial dynamics (2). These
events reestablish mitochondrial function or divert ailing mitochondria
to the mitophagy pathway. Mitochondrial fission is driven
by dynamin-related protein 1 (DRP1) and its adaptor proteins,
while fusion is controlled by mitofusins (MFN1 and MFN2) and
optic atrophy 1 (OPA1) (2–5). Knockout (KO) of Drp1 in mice
results in embryo death, demonstrating the essential role of mitochondrial
fission (3, 5). Furthermore, deletion specifically in neurons
results in neuronal cell death (6, 7) and is associated with marked
effects on cellular signaling and metabolism (8–12).
Female reproductive capacity is dependent on a finite pool of
primordial oocytes (13). The growth of these oocytes from a diameter
of approximately 15 to 80 m is accompanied by a remarkable
mitochondrial biogenesis, resulting in the number of mitochondria
increasing from about 1000 to 300,000 by the time the oocyte is
fully grown (14, 15). This window of mitochondrial biogenesis
coincides with remodeling of the maternal epigenome that is critical
for embryogenesis (16, 17). Throughout oocyte growth, maturation,
and fertilization, mitochondria provide ATP, nucleotide and amino
acid precursors, and tricarboxylic acid (TCA) cycle metabolites
(18, 19). Unlike most other cell types, oocytes and cleavage stage
embryos cannot perform glycolysis and are therefore completely
reliant on mitochondria as their source of ATP. Mitochondria
undergo substantial changes in localization associated with meiotic
progression (20–22) and at fertilization are activated by sperm-­
induced Ca2+
oscillations for increased ATP production required for
fueling the egg-to-embryo transition (23, 24). Thus, mitochondria play
key roles in driving and regulating oocyte and early embryo development,
while, at the same time, there is a need to minimize any
risk to their integrity as they pass from one generation to the next.
Optimal mitochondrial function and quality control in oocytes
is therefore essential to ensure the production of a developmentally
competent oocyte and viable embryo and for the provision of quality
mitochondria to the developing organism. This has been graphically
borne out in mice that have a conditional deletion of oocyte Drp1
(25). Infertility of these mice is attributed to the arrest of ovarian
follicles early in their growth, although some larger oocytes were
recovered. Here, we set out to investigate the question of how inhibition
of mitochondrial fission during oocyte growth, when
mitochondrial numbers are undergoing rapid expansion, affects
subsequent embryo development. We find that Drp1-deleted
oocytes undergo fertilization and preimplantation development but
that peri- and postimplantation embryonic lethality is markedly
increased. This defect is largely assignable to the maternal nuclear
genome and is accompanied by changes to the maternal epigenome.
Our results reveal a previously unknown role for mitochondrial fission
in developmental epigenetic programming in the female germ line.
RESULTS
We generated a conditional Cre/loxP KO of Drp1 (fl) by floxing
exons 3 to 5 (fig. S1A). Oocyte-specific deletion of the floxed segment,
resulting in the KO allele (∆), was achieved by crossing Drp1fl/fl
female mice (3) with transgenic Drp1fl/fl
, Gdf9-Cre males in which
1
Development and Stem Cell Program and Department of Anatomy and Developmental
Biology, Monash Biomedicine Discovery Institute, Monash University,
Melbourne, Victoria 3800, Australia. 2
Applied Embryology Department, High Institute
for Infertility Diagnosis and Assisted Reproductive Technologies, Al-Nahrain
University, Baghdad, Iraq. 3
School of Biomedicine, Discipline of Reproduction and
Development, Robinson Research Institute, The University of Adelaide, South Australia
5005, Australia. 4
Department of Cell Biology, Johns Hopkins University School
of Medicine, 725 N. Wolfe Street, 109 Hunterian, Baltimore, MD 21205, USA. 5
School
of Pediatrics and Reproductive Health, Robinson Research Institute, The University
of Adelaide, Adelaide, South Australia 5005, Australia.
*Corresponding author. Email: deepak.adhikari@monash.edu (D.A.); j.carroll@
monash.edu (J.C.)
Copyright © 2022
The Authors, some
rights reserved;
exclusive licensee
American Association
for the Advancement
of Science. No claim to
originalU.S. Government
Works. Distributed
under a Creative
Commons Attribution
NonCommercial
License 4.0 (CC BY-NC).
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
2 of 15
Cre recombinase is driven by the oocyte-specific Gdf9 (growth
differentiation factor 9) promotor (26). Here, Drp1fl/fl
, Gdf9-Cre
females are referred to as Drp1cKO
mice and their oocytes as Drp1/
oocytes. We confirmed that DRP1 levels were reduced to background
levels in Drp1/
oocytes recovered from 4- to 5-week-old
Drp1cKO
mice (Fig. 1, A to D). These Drp1/
oocytes recovered
from juvenile mice therefore provide an ideal model to investigate
the effect of disrupted mitochondrial fission on oocyte maturation
and subsequent embryo development.
Drp1/
oocytes have altered mitochondrial function
Drp1/
oocytes contained elongated and aggregated mitochondria
(Fig. 1, G to J) compared to the discrete and evenly distributed
mitochondria in the ooplasm of Drp1fl/fl
oocytes (Fig. 1, E, F, I, and J).
In fully grown germinal vesicle (GV)–stage oocytes, we observed
a moderate but significant increase in mitochondrial reactive oxygen
species (ROS) levels in Drp1/
oocytes compared to Drp1fl/fl
controls (Fig. 1K). On measurement of mitochondrial membrane
potential (MMP), Drp1/
oocytes had a reduced MMP compared
to Drp1fl/fl
oocytes (Fig. 1L). These changes were not associated
with any detectable effects on mitochondrial mass (Fig. 1M) or
mitochondrial DNA (mtDNA) copy number between Drp1fl/fl
and
Drp1/
oocytes (Fig. 1N). The differences observed in mitochondrial
activity also did not translate into consistent differences in
oocyte ATP levels (Fig. 1O).
Mitochondrial FADH2 [hydroquinone form of FAD (flavin
adenine dinucleotide)] and NADH (reduced form of nicotinamide
adenine dinucleotide) are generated as a result of reduction in the
TCA cycle and are then oxidized by the mitochondrial OXPHOS
pathway to provide electrons for the electron transport chain (ETC).
FAD++
and NAD(P)H are autofluorescent and can be measured to
provide an indication of the metabolic state of the cell. Drp1/
oocytes were found to have increased levels of FAD++
(Fig. 1P) and
NAD(P)H (Fig. 1Q) compared to Drp1fl/fl
controls. NAD(P)H can
be cytosolic or mitochondrial in origin; the increase in NAD(P)H
fluorescence appears to be primarily in the mitochondria rather
than in the cytosol (Fig. 1Q). These changes provide evidence
that mitochondrial metabolism and/or ETC activity is altered in
Drp1/
oocytes.
Drp1/
oocytes undergo normal meiotic maturation yet
Drp1cKO
females are severely subfertile
To investigate whether DRP1 is necessary to maintain ploidy and
normal spindle function, we monitored meiotic maturation and
assessed the first and second meiotic spindles. The maturation rates,
as indicated by first polar body extrusion (PBE), were comparable
between control (84 ± 10.6%) and KO oocytes (87 ± 15.6%; Fig. 1R).
Consistent with this, mature Drp1/
oocytes formed normal metaphase
II (MII) spindles with aligned chromosomes (Fig. 1, S and U)
similar to control oocytes (Fig. 1, S and T). Histological analysis of
ovaries of 5-week-old Drp1cKO
mice revealed the presence of
follicles at all developmental stages (fig. S1C), similar to Drp1fl/fl
controls (fig. S1B). Furthermore, superovulation of 4- to 5-weekold
Drp1cKO
females and Drp1fl/fl
controls resulted in similar numbers
of ovulated oocytes (fig. S1D). Together, these results indicate
that DRP1 is dispensable for the production of oocytes with intact
spindles and normal chromosome organization.
While Drp1cKO
mice were able to produce oocytes and exhibited
normal mating behavior, six females produced only one litter over
10 weeks of being housed with wild-type (WT) males. By contrast,
control Drp1fl/fl
females regularly produced litters, with an average
size of 6 ± 1.37 pups/litter (fig. S1, E and F). Histological analysis of
ovarian sections showed 10-week-old Drp1cKO
females contained
only primary and secondary stage follicles, with no evidence of
more advanced antral stages (fig. S1, G and H). These findings are
consistent with a previous study that identified the development of
this ovarian phenotype in 10-week-old Drp1cKO
mice (25).
DRP1 is essential for preimplantation development
To investigate the developmental capacity of zygotes derived from
Drp1/
oocytes, we mated 5-week-old control Drp1fl/fl
and Drp1cKO
females with WT (+/+)
males (Fig. 2A). Despite disrupted mitochondrial
organization in Drp1/
oocytes, Drp1mat/+
two-cell stage embryos
displayed equal mitochondrial inheritance between blastomeres
(Fig. 2, B to D), and their rate of preimplantation development was
similar to Drp1fl/+
controls (Fig. 2, E to G). Drp1mat/+
blastocysts
had comparable TMRM (tetramethylrhodamine, methyl ester,
perchlorate) intensity to Drp1fl/+
controls, indicating similar MMP
(fig. S2A). Drp1mat/+
blastocysts had comparable levels of NAD(P)
H (fig. S2B) but slightly lower levels of FAD++
(fig. S2C) compared to
Drp1fl/+
controls. These results indicate that during preimplantation,
development paternal expression of Drp1 normalizes some of the
mitochondrial differences observed in Drp1/
oocytes. Moreover,
preferential immunolabeling of Octamer-binding transcription
factor 4 (OCT-4) and CDX2 revealed no detectable difference in
the number of inner cell mass (ICM) cells in Drp1mat/+
blastocysts
compared to Drp1fl/+
controls, although a 25% reduction in the
number of trophectoderm (TE) cells was observed (Fig. 2, H to J).
Thus, loss of the maternal Drp1 allele appears to preferentially affect
the proliferation or survival of TE cells in blastocysts.
To examine the developmental competence of Drp1/
oocytes
directly, i.e., in the absence of the paternal Drp1 allele, parthenogenetic
activation was used to generate diploid Drp1/
zygotes. The
rate of two-cell formation after parthenogenetic activation of Drp1/
oocytes was significantly reduced compared to Drp1fl/fl
controls
(60% versus 90%, respectively; Fig. 2K) and, remarkably, less than
4% of Drp1/
two-cell parthenotes developed to the blastocyst
stage (Fig.  2,  K  and  M) compared to 68% of Drp1fl/fl
controls
(Fig.  2,  K  and  L). Immunofluorescence confirmed that fertilized
Drp1mat/+
blastocysts recovered DRP1 protein, while it was not
detectable in the few Drp1/
parthenogenetic blastocysts produced
(Fig. 2, N to R). In addition, TMRM labeling showed that fertilized
Drp1mat/+
blastocysts (Fig. 2T) had a mitochondrial distribution
indistinguishable from control fertilized Drp1fl/+
blastocysts (Fig. 2S)
and Drp1fl/fl
control parthenogenetic blastocysts (Fig. 2U), while in
Drp1/
parthenogenetic blastocysts, mitochondria were highly
aggregated into large clumps (Fig. 2V). The catastrophic failure of
Drp1/
embryos to undergo preimplantation development is not
due to parthenogenesis per se but likely mediated by impacts on
mitochondrial function, although we cannot rule out other DRP1-­
dependent factors playing a role. These findings demonstrate an
absolute reliance on oocyte DRP1 for preimplantation development
and that the paternal allele is sufficient to support development
in vitro, albeit with a reduced number of TE cells.
Embryos from Drp1cKO
females exhibit developmental arrest
Previous studies show Drp1 to be haplosufficient because viable
Drp1 heterozygous progeny (Drp1+/−
) are produced at normal
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
3 of 15
Fig. 1. Oocyte-specific deletion of Drp1 and analysis of oocyte mitochondria. (A to D) Oocyte-specific deletion of Drp1. DRP1 (green) and DNA (pink) in Drp1fl/fl
oocytes (A) but Drp1/
oocytes lacked DRP1 (B). Dashed circles represent oocyte margins. (C) DRP1 fluorescence quantification. AU, arbitrary unit. (D) Western blots of
DRP1. Each lane contains 50 GV-stage oocytes from 4- to 5-week-old mice. -Actin was used as a loading control. The experiments were repeated three times, and a representative
image is shown. Mitochondrial morphology in GV-stage Drp1fl/fl
(E and F) and Drp1/
(G and H) oocytes. Branch length mean (I) and network branches mean (J) between
Drp1fl/fl
and Drp1/
oocytes. Oocytes were labeled with mitotracker orange. Number of oocytes analyzed are shown (C, I, and J). (K) Increased mitochondrial ROS(mtROS)
levels in Drp1/
oocytes. Numbers represent total number of oocytes. (L) Reduced MMP in Drp1/
oocytes compared to Drp1fl/fl
oocytes. Carbonyl cyanide-4-phenylhydrazone
(FCCP) was used to demonstrate assay specificity. Numbers represent total number of oocytes. (M) Similar mitochondrial mass between Drp1fl/fl
and Drp1/
oocytes.
MitoTracker Green signals in (L) were analyzed to calculate mitochondrial mass. ns, not significant. (N) Comparable mtDNA copy number between Drp1fl/fl
and
Drp1/
oocytes. Numbers represent total number of oocytes. (O) Similar ATP levels between Drp1/
and Drp1fl/fl
oocytes. Numbers represent experimental replicates.
Drp1/
oocytes have increased levels of FAD++
(P) and NAD(P)H (Q) compared to Drp1fl/fl
controls. Total oocyte numbers shown. Representative autofluorescent images
are also shown. (R) Comparable PBE rates in vitro. Numbers represent experimental replicates. (S to U) MII oocytes showing spindle (green) and DNA (blue). Chromosome
spreads were measured as exemplified in (T) and (U). Total oocyte numbers shown in (S). Results show means ± SD and P values determined by two-tailed
Student’s t test (C and I to S). F.C., fold change.
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
4 of 15
Fig. 2. Preimplantation development of Drp1mat/+
embryos. (A) Schematic representation of mating schemes and the genotype of resultant embryos. (B to D) Comparison
of mitochondrial mass between blastomeres of 2-cell embryos. One-cell zygotes harvested from gonadotropin-stimulated females and mated with WT males
were cultured for 24 hours and labelled with MitoTracker Orange. Numbers represent total number of zygotes. (E to G) After gonadotropin stimulations, 4- to 5-week-old
Drp1fl/fl
and Drp1cKO
mice were mated with WT males to harvest Drp1fl/+
and Drp1matΔ/+
one-cell zygotes. One-cell zygotes were cultured in vitro for 96 hours. Number of
replicates are shown. Bla, blastocyst. (H to J) Blastocysts were labelled with CDX2 and OCT-4 to count the trophectoderm (TE) and inner cell mass (ICM), respectively.
Representative maximum projection images of Drp1fl/+
(I) and Drp1matΔ/+
(J) blastocysts are shown. Number of blastocysts analyzed in (H) are shown. (K to M) Absolute
requirement of DRP1 for preimplantation embryo development. The rate of parthenogenetic activation of Drp1Δ/Δ
oocytes was significantly reduced compared to Drp1fl/fl
controls (K). Numbers of experimental repeats are shown. Bla, blastocyst. Representative images of Drp1fl/fl
control (L) and Drp1Δ/Δ
(M) parthenotes at the end of culture.
(N to V) Recovery of DRP1 and mitochondrial morphology in fertilized Drp1matΔ/+
embryos. DRP1 expression was largely recovered in fertilized Drp1matΔ/+
embryos (N).
Number of blastocysts analyzed are shown. Immunofluorescence of DRP1 (cyan) was evident in Drp1fl/+
(O) and Drp1matΔ/+
(P) blastocysts. Unfertilized Drp1fl/fl
control
parthenote expressed DRP1 (Q) but Drp1Δ/Δ
parthenote blastocysts completely lacked DRP1 immunofluorescence (R). Reflecting the DRP1 expression profiles, TMRM
labelling revealed normal singular mitochondria in control fertilized Drp1fl/+
(S) and Drp1matΔ/+
(T) blastocysts. Unfertilized Drp1fl/fl
control parthenote also contained
dispersed mitochondria (U) but Drp1Δ/Δ
parthenogenetic blastocysts had mitochondria aggregated into large clumps (V). Results show mean ± SD and P values
determined by two-tailed Student’s t test (E, H, K, N).
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
5 of 15
frequency in Drp1+/−
intercrosses (3, 5, 27). However, because of
heterozygous parents, these matings are unable to determine
whether there is a maternal requirement for DRP1. Therefore, to
investigate the role of DRP1 in oocytes, we mated 5-week-old
Drp1cKO
and control Drp1fl/fl
females with WT males (as in Fig. 2A).
We focused on 5-week-old mice as we knew they were capable of
generating oocytes at similar rates to controls (fig. S1D). Plugged
females were assessed between embryonic day 10.5 (E10.5) to E18.5.
A confirmed pregnancy was defined if the uteri of plugged females
contained decidua, resorptions, or live embryos. By this criterion,
pregnancies were significantly reduced in Drp1cKO
females (60%;
n = 25) compared to Drp1fl/fl
females (94%; n = 17; Fig. 3A and table
S1). Of confirmed pregnancies, Drp1cKO
females had a significantly
reduced number of implanted embryos per mating compared to
Drp1fl/fl
females (4.2 ± 2.3 versus 8.0 ± 1.1, respectively; Fig. 3B and
table S1). Notably, in Drp1cKO
females, most of the implanted
embryos had died and were resorbing (76%, compared to 3% in
controls; Fig. 3, C to E, and table S1). In addition, when live fetuses
were present, most of these were severely growth retarded (Fig. 3E,
arrowheads).
With such a marked effect of Drp1 deletion in oocytes on subsequent
embryonic development, we investigated a number of potential
contributory causes. First, we confirmed that the fetuses generated
were of the correct heterozygous Drp1mat/+
genotype (fig. S2D)
and that they were expressing DRP1 from the paternal allele (fig.
S2E). Furthermore, to test whether the heterozygote level of DRP1
was sufficient to support normal mitochondrial morphology, we
derived fetal fibroblasts from E13.5 Drp1mat/+
and control Drp1fl/+
embryos. The fetal fibroblasts from both genotypes exhibited comparable
mitochondrial morphology (fig. S2, F and G), demonstrating
that the DRP1 level in Drp1mat/+
fetuses is sufficient to support
overtly normal mitochondrial organization. An increase in aneuploidy
may be a factor contributing to embryonic failure, but no significant
differences were seen in chromosome counts of blastomeres of
Drp1mat/+
and control Drp1fl/+
four-cell embryos (fig. S2, H to K).
Another source of embryonic failure could potentially be related to
the timing of onset of zygotic genome activation (ZGA) at the
two-cell stage, but no significant difference in 5-ethynyl uridine
(EU) incorporation was found between Drp1fl/+
and Drp1mat/+
embryos (fig. S2, L, N, O, Q, and R). In addition, phosphorylated
RNA polymerase II at serine-2 (pS2, a marker of RNA polymerase
II activation) levels were also similar between Drp1fl/+
and Drp1mat/+
embryos (fig. S2, M, N, P, Q, and S). These results indicate that
there is no marked difference in transcriptional activation between
Drp1fl/+
and Drp1mat/+
embryos. Another common feature of
embryonic lethal mouse models is a defect in placental development
(28), a possibility consistent with the observed decrease in TE cells
in Drp1mat/+
blastocysts (Fig. 2H). However, a detailed histological
analysis of placental structure in surviving E13.5 embryos of
Drp1mat/+
and control embryos uncovered no obvious differences
in placental or embryonic structures (fig. S2, T to W). The developmental
failure of Drp1mat/+
embryos appears not to be caused by a
failure to express sufficient DRP1 in development, or to any
obvious abnormalities in chromosome number, ZGA, or even
placental organization. We therefore returned to a deeper analysis
of Drp1/
oocytes.
Altered transcriptome and proteome in Drp1/
oocytes
To understand the effects of DRP1 deletion on oocytes, we used
single-cell RNA sequencing (RNA-seq) to compare the transcriptome
of Drp1fl/fl
and Drp1/
GV-stage oocytes. This analysis revealed
717 differentially expressed genes (DEGs; adjusted P < 0.05)
between Drp1fl/fl
and Drp1/
oocytes, of which 126 (17.5%) were
up-regulated and 591 (82.5%) were down-regulated in Drp1/
oocytes (Fig.  4,  A and B). Several DEGs between Drp1fl/fl
and
Drp1/
oocytes are known epigenetic regulators, such as Kdm5a,
Kdm5b, Tet2, Dppa4, and Dnmt3a (Fig. 4C and fig. S3, A and B),
indicating that loss of DRP1 during oocyte growth may lead to
alterations in the maternal epigenome. In addition, maternal-effect
genes of the subcortical maternal complex (SCMC), Tle6 and Khdc3,
were significantly down-regulated in Drp1/
oocytes as compared
to Drp1fl/fl
oocytes (fig. S3C). In contrast, there were no significant
changes in the expression of oocyte-specific genes (fig. S3D), consistent
with the minimal impact of Drp1 deletion on oogenesis,
oocyte maturation, and morphology. Perhaps unexpectedly, we also
found no evidence for changes in mtDNA-encoded gene products
(mt-Nd1, mt-Nd2, mt-Nd3, mt-Nd4, mt-Nd5, mt-Nd6, mt-Cytb, and
mt-Co1; fig. S3E), suggesting that the nuclear genome provides the
main mechanism for adapting to mitochondrial dysfunction due to
compromised DRP1 expression.
Proteomic analysis of Drp1/
and Drp1fl/fl
MII oocytes demonstrated
that 69 proteins were differentially accumulated between
Fig. 3. Postimplantation death of Drp1mat/+
embryos derived from Drp1/
oocytes. Five-week-old Drp1cKO
and control Drp1fl/fl
females were mated with WT (+/+)
males and assessed fetal development between E10.5 to E18.5 of pregnancy.
(A) Uteri of 94% of plugged control Drp1fl/fl
females showed embryos as compared
to only 60% of plugged Drp1cKO
females contained any embryo. Number of mice
analyzed are shown. (B) Control females had an average of eight implants per
pregnancy, whereas Drp1cKO
females had only four implants per pregnancy. Number
of mice analyzed are shown. (C) Within confirmed pregnancies, there was a
higher percentage of embryo resorption in Drp1cKO
females (76%) compared to
only 3% in Drp1fl/fl
females. Number of mice analyzed are shown. While the uteri of
E13.5 Drp1fl/fl
females showed normal developing embryos (D), the embryos in
uteri of Drp1cKO
females were either dead (E, arrows) or severely growth retarded
(E, arrowheads). Results show means ± SD (B and C). P value determined by chi-square
test (A), two-tailed Student’s t test (B and C).
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
6 of 15
Drp1fl/fl
and Drp1/
oocytes, of which 24 proteins were increased
and 45 proteins were decreased in Drp1/
oocytes (Fig. 4, D and E,
and table S2). Enrichment analyses indicate that proteins regulating
cellular metabolism (LDHA, CMAS, NDUFA9, SYNJ1, ADH5,
COX5B, PDHB, MTHFD1, TYMS, PGLS, ITPA, SDHC, PAFAH1B2,
and NDUFS2) are significantly affected in Drp1/
oocytes
compared to Drp1fl/fl
oocytes (fig. S3F and table S2). Many of the
affected proteins in Drp1/
oocytes are known regulators of epigenetic
modifications. For example, SAP30 is a regulator of histone
acetylation (29), SDHC is a known regulator of histone modifications
Fig. 4. Altered gene expression profile of Drp1/
oocytes. (A) Volcano plot showing DEGs (down-regulated, blue; up-regulated, red; and not different, black) between
Drp1fl/fl
and Drp1/
oocytes. Highly dysregulated genes are labeled. (B) Heatmap of differential gene expression in Drp1fl/fl
and Drp1/
oocytes. CPM, counts per million.
(C) Heatmap showing differential expression of epigenetic regulators in Drp1fl/fl
and Drp1/
oocytes. Number of single GV-stage oocytes per genotype used for RNA-seq
analysis are shown (A to C). (D) Volcano plot showing differentially abundant proteins (down-regulated, blue; up-regulated, red; and not different, black) between
MII-stage Drp1fl/fl
and Drp1/
oocytes. Highly dysregulated proteins are labeled. (E) Hierarchically clustered heatmap of relative abundancies of proteins in Drp1fl/fl
and
Drp1/
oocytes. Number of replicates per genotype used for proteomic analysis are shown (D and E).
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
7 of 15
in mammalian cells (30) and MTHFD1 plays an essential role in
one-carbon metabolism (31) and maternal Mthfd1 disruption is
known to impair fetal growth (32).
Together, these omics studies show that disrupting mitochondrial
fission in oocytes leads to marked changes to the transcriptomic and
proteomic profile of Drp1/
oocytes and that epigenetic and metabolic
pathways are substantially altered. Given the importance of metabolites
in the regulation of gene expression via modulation of epigenetic
enzyme activity in somatic cells (33–35), we examined the effects of
Drp1 deletion during oocyte growth on the oocyte epigenome.
Altered DNA and histone modifications in Drp1/
oocytes
and Drp1mat/+
zygotes
Given the dysregulation of metabolic and epigenetic gene products
in Drp1/
oocytes, we tested the hypothesis that deletion of Drp1
during oocyte growth is sufficient to disrupt the establishment
of the oocyte epigenome. We examined DNA methylation and
H3K27me3 in oocytes as they are known to regulate embryo development
(36,  37). Immunofluorescence showed that the levels of
5mC and H3K27me3 were significantly decreased in Drp1/
oocyte
nuclei compared to Drp1fl/fl
oocytes (Fig. 5, A to H) while the level
of H3K27ac was significantly higher in the nuclei of Drp1/
than in
Drp1fl/fl
oocytes (Fig. 5, I to O). We found that 5mC and H3K27me3
were also significantly decreased in Drp1mat/+
two-cell embryos
compared to Drp1fl/+
embryos (Fig. 5, P to W). Other epigenetic
marks, H3K4me3 (fig. S4, A to G) and H3K9me3 (fig. S4, H to N),
were not altered. We matched the DEGs between Drp1fl/fl
and
Drp1/
oocytes with previously published genes known to be
marked with H3K27me3 in mouse oocytes (38, 39). About 18% of
DEGs between Drp1fl/fl
and Drp1/
oocytes are among the known
H3K27me3-marked genes (fig. S4O and table S3), suggesting that
altered H3K27me3 level could be a determinant, at least in part, of
differential gene expression in Drp1/
oocytes.
Developmental defects of Drp1mat/+
embryos associated
with the maternal nuclear genome
The epigenetic changes seen in Drp1/
oocytes suggest a potential
explanation for subsequent embryonic failure. To investigate this
hypothesis, we have undertaken two series of embryo transfer
experiments. First, Drp1fl/+
and Drp1mat/+
two-cell embryos were
transferred into identical surrogates for gestation to obtain a baseline
rate of development after embryo transfer (Fig. 6A). Seventy-­
four Drp1fl/+
(8.2  ±  1.6 per transfer, n  =  9) and 65 Drp1mat/+
(8.1 ± 2.3 per transfer, n = 8) two-cell embryos were transferred to
the oviducts of WT pseudo-pregnant recipients (table S4). Recipients
culled between E15.5 to E20.5 of pregnancy revealed that eight of
nine recipients receiving Drp1fl/+
embryos were confirmed pregnant
compared to only four of eight recipients receiving Drp1mat/+
embryos (Fig. 6B and table S4). Of Drp1fl/+
embryos transferred,
40 of 74 (54%) implanted and 39 (53%) formed viable embryos
(Fig. 6, C and D, and table S4). In contrast, 14 of 65 (22%) of
Drp1mat/+
embryos implanted and only 8 (12%) formed morphologically
normal fetuses (P < 0.0001; table S4). These embryo transfer
results confirm that effects on fetal development are independent
of the maternal environment in Drp1cKO
mice and demonstrate that
the absence of DRP1 specifically during oocyte growth markedly
compromises subsequent embryo development.
Next, we performed nuclear transfer at the pronuclear stage
followed by embryo transfer to test the relative contributions of the
nuclear genome (and its epigenetic modifications) and cytoplasm
(with fission defective mitochondria) to the increased rate of embryonic
lethality seen in Drp1mat/+
embryos. Nuclear transfer was
used to swap the pronuclei between Drp1mat/+
and Drp1fl/+
zygotes
to generate Drp1mat/+
pronuclei with WT cytoplasm (KO Pn) and
Drp1fl/+
pronuclei with Drp1mat/+
cytoplasm (KO Cyt; Fig. 6E).
Similar numbers of KO Pn embryos (10.5 ± 2.7 per transfer) and
KO Cyt embryos (10.7 ± 2.2 per transfer) were transferred to individual
oviducts of recipients. The pregnancy rate of dams receiving
KO Pn embryos was significantly reduced compared to KO Cyt
embryos (53%, n = 17 versus 88%, n = 17, respectively; Fig. 6F and
table S5). Of the dams that became pregnant, implantation rates
were variable but not significantly different (Fig. 6G and table S5).
However, the proportion of resorption sites and dying embryos was
markedly increased for KO Pn embryos with a mean of 60% resorptions
compared to just over 20% in KO Cyt embryos (Fig. 6H and
table S5). As a result, only half as many KO Pn embryos survived to
midgestation [24 of 178 (13%)] compared to KO Cyt embryos [49 of
182 (27%); P < 0.001, chi-square test; table S5]. These data support
the hypothesis that alterations to the maternal genome during
growth of Drp1/
oocytes is the primary reason for the high frequency
of developmental failures in Drp1mat/+
fetuses.
DISCUSSION
In this study, we reveal a marked maternal effect resulting from
the loss of Drp1 during oocyte development. The large majority of
peri- and postimplantation-stage embryos and fetuses derived from
Drp1 KO oocytes fail in utero, despite having sufficient DRP1
derived from the paternal allele for normal survival. Overall, our
findings support the conclusion that DRP1 in oocytes is essential
for normal mitochondrial function, which, in turn, is required for
establishing a maternal epigenome compatible with viable embryo
development.
A number of lines of evidence support this conclusion. Using
pronuclear transplantation, we demonstrate that WT ooplasm is
insufficient to completely rescue the developmental lethality seen
in Drp1mat/+
embryos and, conversely, that the cytoplasm of
Drp1mat/+
zygotes is compatible with higher rates of postimplantation
development. The fact that the embryonic lethality more
consistently follows the Drp1 KO maternal genome rather than the
cytoplasm is strong evidence that epigenetic programming of the
maternal genome is irreversibly disrupted during oocyte growth.
This is further supported by the observed reduction in DNA methylation
and H3K27me3 in the nucleus of Drp1 KO oocytes. These
findings directly demonstrate the maternal epigenome is altered
because of the loss of DRP1 during oocyte development. DNA
methylation and H3K27me3 in the female germ line are responsible
for regulating gene expression in resultant embryos (40, 41) and,
consistent with our findings, others have shown that disruption of
these epigenetic marks predominantly affects postimplantation
development with little effect on oocytes or preimplantation embryos
(41–43).
Single-cell oocyte RNA-seq and proteomics revealed genes involved
with epigenetic regulation to be among the most disrupted
in the absence of DRP1, including Got1, Dnmt3a, Mthfd1, Kdm5a,
Kdm5b, Tet2, Hist1h1c, and Hist3h2ba, providing a potential
mechanism for the observed marked recalibration of the oocyte
transcriptome and proteome. This analysis also revealed that Drp1
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
8 of 15
Fig. 5. Altered epigenetic marks in the nuclei of Drp1/
oocytes and Drp1mat/+
2-cell zygotes. Immunofluorescence (IF) showing 5mC (A, magenta), H3K27me3
(B, green), and merged (C) in Drp1fl/fl
oocytes. IF showing 5mC (D), H3K27me3 (E), and merged (F) in Drp1/
oocytes. (G) Quantification of 5mC fluorescence. (H)
Quantification of H3K27me3 fluorescence. IF showing H3K27ac (I, green), DNA (J, blue), and merged (K) in Drp1fl/fl
oocytes. IF showing H3K27ac (L, green), DNA (M,
blue), and merged (N) in Drp1/
oocytes. (O) Quantification of H3K27ac fluorescence. IF showing 5mC (P, green), H3K27me3 (Q, magenta) and merged (R) in Drp1fl/+
2-cell zygotes. IF showing 5mC (S), H3K27me3 (T) and merged (U) in Drp1mat/+
2-cell zygotes. (V) Quantification of 5mC fluorescence. (W) Quantification of H3K27me3
fluorescence. Results show mean ± SD and P values determined by two-tailed Student’s t test (G, H, O, V and W). Number of oocytes/zygotes analyzed are shown
(G, H, O, V, and W).
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
9 of 15
Fig. 6. Development after embryo transfer. (A to D) Embryonic lethality is due to defective Drp1/
oocytes. (A) Schematic representation of embryo transfer experiment.
Superovulated 5-week-old Drp1fl/fl
and Drp1CKO
females were mated with WT (+/+
) males to generate WT Drp1fl/+
embryos and heterozygous Drp1mat/+
zygotes, respectively.
One-cell zygotes harvested from the oviducts of plugged donor females were cultured in vitro for 24 hours until they reached the two-cell stage. Two-cell stage embryos
were transferred to the oviducts of different WT pseudo-pregnant recipients. (B) Recipients culled between E15.5 and E20.5 of pregnancy revealed that eight of nine
recipients receiving Drp1fl/+
embryos became pregnant compared to only four of eight recipients receiving Drp1Drp1mat/+
embryos. (C) Per-pregnancy implantation rate
of Drp1mat/+
embryos was markedly reduced than Drp1fl/+
embryos. (D) Within confirmed pregnancies, there was a higher percentage of embryo resorption in
Drp1cKO
females compared to Drp1fl/fl
females. (E to H) Embryo development after pronuclear swapping and embryo transfer. (E) Schematic representation of reciprocal
pronuclear swapping between one-cell zygotes, in vitro culture, and embryo transfer. (F) Comparison of pregnancy rates after implanting KO Cyt and KO Pn
zygotes. (G) Implantation rates of KO Pn and KO Cyt embryos. Within positive pregnancies, there was a higher percentage of resorption of KO Pn embryos compared
to KO Cyt embryos (H). P values determined by chi-square test (B and F) and two-tailed Student’s t test (C, D, G, and H). Results show means ± SD (C, D, G, and H). Number
of mice analyzed are shown (B to D and F to H).
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
10 of 15
deletion leads to disruption of the SCMC. Loss of this complex has
been shown to be associated with maternal effects leading to embryonic
arrest in mice [reviewed in (44)]. A number of SCMC components
are disrupted including Tle6, which primarily affects preimplantation
embryo development in mice (45), while TLE6 mutations are
linked with human embryonic lethality (46, 47). KHDC3L mutation
causes early embryonic failure, gestational abnormalities, and recurrent
pregnancy loss in human and causes widespread DNA
methylation deficiency (48), and NLRP5 mutations are associated
with zygotic imprinting (49). Nlrp2 in mouse oocyte is essential for
embryo development and postzygotic DNA methylation maintenance
(50). How loss of DRP1 causes changes in SCMC genes remains
unknown; however, given the decreased global DNA and histone
methylation in Drp1 KO oocytes, SCMC genes may be differentially
expressed because of the associated epigenetic modifications.
Drp1 deletion via Gdf9-Cre occurs during oocyte growth and is
coincident with the period during which the oocyte epigenome is
being established. The parental epigenetic profile needs to be erased
each generation before being reestablished largely during oocyte
growth (16, 17, 36). This is well established for classical genomic
imprinting (17, 42) but less well established for noncanonical
imprinting mediated by heritable histone modifications, including
H3K27me3. Nevertheless, this temporal association of Drp1 deletion
and establishing the oocyte epigenetic landscape provides additional
support for the link between mitochondrial function and the oocyte
epigenome. This conclusion is entirely consistent with the increasingly
prevalent link between mitochondrial metabolism and epigenetic
regulation in cancer, immunity, and cell differentiation
(51–54). Cell culture models show that mitochondrial dysfunction
can alter DNA methylation (55) and histone modifications (56) via
disrupted metabolic pathways. Furthermore, a recent study demonstrates
that metabolites can indeed alter DNA and histone methylation
in oocytes and embryos (57).
How the deletion of Drp1 in oocytes causes the changes in the
epigenome is likely to be a complex integration of impacts on mitochondrial
metabolism and downstream modification of the activity
of epigenetic regulators. Metabolic and epigenetic genes were the
most disrupted in single-oocyte transcriptomics and proteomics.
Mitochondrial structure is affected in Drp1 KO oocytes, consistent
with the known role of DRP1 in mitochondrial fission (3, 5), and
this is accompanied by a marked increase in levels of mitochondrial
FAD++
and NAD(P)H as well as small but significant changes in
MMP and mitochondrial ROS. It is unclear exactly how these mitochondrial
changes come about and whether they are all coupled, but
an increase in mitochondrial NADH and FAD++
is consistent with
a decrease in Complex I activity and a switch to Complex II. The
accumulation of NADH and FAD++
may affect normal TCA cycle
activity, in part because of the failure of the ETC to generate NAD+
needed for oxidation of mitochondrial metabolites, as well as its
availability for NAD+
-dependent epigenetic regulators (52). Thus,
the loss of DRP1 may ultimately disrupt the TCA cycle, the source
of epigenetic modifying metabolites such as S-adenosyl methionine,
succinate, and fumarate (11, 12, 33), thereby providing a potential
link between DRP1 loss and epigenetic modifications.
Despite the marked impact on postimplantation development,
Drp1 KO oocytes used in this study underwent growth, maturation,
fertilization, and preimplantation development. This minimal
impact on oocyte maturation and preimplantation development is
consistent with our observation that aneuploidy was not increased
because of the absence of DRP1. Mitochondrial mass has also been
shown to influence histone modifications and gene expression (58),
although we did not find any evidence for Drp1 deletion in oocytes
causing a decrease in mitochondria or mtDNA copy number.
In contrast, the decrease in MMP in Drp1 KO oocytes may contribute
to changes in gene expression as it is a known mediator of
mitochondria-directed nuclear gene expression (51). It remains
uncertain whether the relatively modest decrease in MMP and
increase in ROS seen in Drp1 KO oocytes is contributing to these
effects, particularly given that oocyte meiosis and subsequent preimplantation
development is not affected.
While Drp1mat/+
zygotes carrying a paternal Drp1 allele develop
to the blastocyst stage with only a minor change in TE cell number,
Drp1/
zygotes generated by parthenogenetic activation undergo
developmental arrest in the early cleavage stages. Given that the
development of control parthenotes and fertilized embryos is similar,
this demonstrates that at least one Drp1 allele is essential for preimplantation
development. The precise explanation for the catastrophic
failure of preimplantation development in the complete absence of
DRP1 is not known but may reflect cumulative mitochondrial
damage, building on compromised oocyte mitochondria. Alternatively,
DRP1 may be essential to support the developmental transition
of mitochondrial function as they undergo marked changes in
structure and function from the eight-cell stage, particularly in cells
destined to become the TE (59).
Our findings have important implications for understanding the
role of oocyte mitochondrial function in embryo development and
offspring health. We and others have shown that oocyte mitochondrial
function is disrupted in obesity (60, 61), metabolic disease
(62, 63), and maternal aging (64, 65). These conditions are also
often associated with heritable and even transgenerational changes
in offspring health (63, 66), but a direct link between oocyte mitochondrial
function and heritable epigenetic changes has not been
established. Our demonstration that a direct mitochondrial intervention,
deletion of Drp1, leads to changes to the oocyte epigenome
provides strong support that the heritable phenotypic changes are
mediated via perturbation of mitochondrial function.
Our findings are consistent with recent developments in other
systems, whereby mitochondria act as hubs, sensing environmental
changes and directing nuclear responses through epigenetic modulation.
The fact that mitochondria perform this function in oocytes
elevates them from mere providers of ATP to pivotal modulators of
embryonic development and offspring health.
MATERIALS AND METHODS
Mice
Drp1loxP/loxP
mice (3) were crossed with transgenic mice that carried
Gdf-9 promoter–mediated Cre recombinase (26). After multiple
rounds of crossing, homozygous Drp1cKO female mice lacking
Drp1 in oocytes (Drp1loxP/loxP
; Gdf9-Cre) or Drp1/
oocytes were
obtained. Mice that do not carry the Cre transgene are referred to as
Drp1loxP/loxP
and were used as controls. The mice were housed
under controlled environmental conditions with free access to
water and food. All animal experimentation was approved by
Monash University Animal Experimentation Ethics Committee
and was performed in accordance with Australian National Health
and Medical Research Council Guidelines on Ethics in Animal
Experimentation.
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
11 of 15
Collection of oocytes and early embryos
To obtain GV oocytes, 4 to 5-week-old female mice were intraperitoneally
injected with 8-IU pregnant mare’s serum gonadotropin
(PMSG; Intervet), and 44 to 48 hours later, the mice were euthanized.
Fully grown GV-stage oocytes surrounded by cumulus cells
were released by puncturing the ovaries in M2 medium (Sigma-­
Aldrich) containing 200 M 3-isobutyly-1-methylx-anthine (Sigma-­
Aldrich). Oocytes were freed from the attached cumulus cells by
repetitive pipetting through a narrow-bore glass pipette. Oocytes
were in vitro cultured in drops of M16 medium (Sigma-Aldrich)
under mineral oil (Sigma-Aldrich) at 37°C in a humidified atmosphere
of 5% CO2 in air or in M2 medium at 37°C. For collection of
ovulated MII-stage oocytes, mice were superovulated by sequential
intraperitoneal injections of 8-IU PMSG and 8-IU human chorionic
gonadotropin (hCG; Intervet) at an interval of 48 hours, and the
mice were culled 14 to 16 hours after hCG injection to harvest eggs
from oviducts. For collection of embryos, females were mated with
stud C57BL/6J males after hCG injection, and the mice were culled
14 to 16 hours after hCG injection to harvest one-cell zygotes from
oviducts. Zygotes were further cultured in SAGE 1-Step medium
(Cooper Surgical) up to blastocyst stage at 37°C in a humidified
atmosphere of 5% CO2 in air.
Detection of RNA synthesis
Two-cell stage zygotes were incubated with 2 mM EU (Thermo
Fisher Scientific) for 2 hours in SAGE 1-Step medium at 37°C in
a humidified atmosphere of 5% CO2 in air. Zona pellucida was
removed by Tyrode’s solution (Sigma-Aldrich) and then fixed in
4% paraformaldehyde (PFA) for 30 min at room temperature (RT)
and permeabilized with 0.5% Triton X-100 in phosphate-buffered
saline (PBS) for 20 min at RT. Zygotes were stained using the Click-iT
RNA imaging kit for nascent RNA (Thermo Fisher Scientific) and then
incubated with rabbit polyclonal RNA PolII (phospho-Ser2
) antibodies
(1:400) overnight at 4°C. This was followed by an incubation with
specific Alexa Fluor 647 (1:1000 diluted in blocking buffer; Invitrogen
Molecular Probes) for 1 hour at RT. DNA was labeled using 10-min
incubation in Hoechst 33342 (10 g/ml; Sigma-Aldrich).
mtDNA copy number assay
Individual oocytes were transferred into 10 l of lysis buffer [50 mM
tris-HCl (pH 8.0), 1 mM EDTA, 0.5% Tween 20, and proteinase K
(200 g/ml)]. Sample was incubated at 55°C for 2 hours, followed by
inactivation of proteinase K at 95°C for 10 min. Genomic DNA was
diluted five times for quantitative polymerase chain reaction (PCR).
A plasmid containing 12S ribosomal (r)RNA region of mtDNA was
used as quantification standards. The standard stock was serially
diluted to prepare the standard curve. Real-time fluorescence–
monitored quantitative PCR using the primer pair 5′-CGT TAG
GTC AAG GTG TAG CC-3′ and 5′-CCA AGC ACA CTT TCC
AGT ATG-3′ was performed in Stratagene Mx3000P (Agilent
Technologies). Each reaction of 20 l consisted of 10 l of SYBR
Green, 100 nM each of the forward and reverse primers and 2 l of
diluted DNA sample. Standard curves were created for each run,
and sample copy number was generated from the equation of Ct
value against copy number for the corresponding standard curve.
Parthenogenesis and culture
Ovulated Drp1fl/fl
and Drp1/
eggs were collected from oviduct
14 to 16 hours after hCG in M2 medium, and cumulus cells were
removed with hyaluronidase (0.3 mg/ml; Sigma-Aldrich) in M2
medium. After two quick washes in calcium-free CZB medium
[81.62 mM NaCl, 4.83 mM KCl, 1.18 mM MgSO4, 0.11 mM EDTA,
5.6 mM glucose, 37 mM sodium lactate, 1.2 mM KH2PO4, 25.1 mM
NaHCO3, 0.27 mM pyruvate, 1 mM l-glutamine, and bovine serum
albumin (BSA; 5 mg/ml), pH 7.4] containing cytochalasin B (5 g/ml;
Sigma-Aldrich) and 10 mM SrCl2 (Sigma-Aldrich), oocytes were
cultured in this medium for 2 hours. Oocytes were further cultured for
3 hours in M2 medium containing cytochalasin B (5 g/ml). Activated
eggs were cultured in SAGE 1-Step medium (CooperSurgical) up to
blastocyst stage at 37°C in a humidified atmosphere of 5% CO2 in air.
Timed mating and harvesting postimplantation
embryos and histology
Female mice at 5 weeks of age were individually housed with
C57BL/6J WT male (age, 3 to 4 months) and checked every morning
until vaginal plugs were detected. The day on which a vaginal plug
was detected was designated as E0.5. Pregnant females were euthanized
by cervical dislocation, and the uteri were collected in
ice-chilled PBS kept on ice. Embryos and placentae either were
snap-frozen in liquid nitrogen and stored at −80°C until further
analysis or were fixed in Bouin’s solution (Sigma-Aldrich) for histological
analysis.
Pronuclear swapping and embryo transfer
The protocol was previously described (67). Briefly, one-cell stage
Drp1fl/+
and Drp1mat/+
embryos were collected (24 hours after hCG
injection) in M2 medium containing hyaluronidase (0.3 mg/ml) to
remove cumulus cells. Embryos were then incubated in M16 medium
containing nocodazole (0.3 g/ml; Sigma-Aldrich) and cytochalasin
B (5 g/ml) in a humidified incubator with 5% CO2 in air for
15 to 45 min at 37°C. Both the pronuclei from a zygote were removed
by micropipette and drawn a small volume of inactivated Sendai
virus (Cosmo Bio, USA; ~3000 hemagglutinating units/ml) into the
enucleation/injection pipette. The virus and karyoplast were injected
into the perivitelline space of a second enucleated one-cell stage
embryo. The reconstructed embryos were cultured in inhibitor-free
M16 medium overnight to allow their cleavage. Two-cell embryos were
transferred into the oviducts of pseudo-pregnant C57BL6J/MARP x
BALBC F1 recipient females for further embryo development.
Determination of mitochondrial ROS
Mitochondrial ROS were measured using Mitosox Red (Thermo
Fisher Scientific, Waltham, Massachusetts, USA). Oocytes were
incubated with 5 M Mitosox in M2 media at 37°C for 30 min.
Epifluorescence and bright-field images were acquired using an SP8
confocal microscope (Leica, Germany), and fluorescence was
measured at an excitation wavelength of 488-nm and 520-nm longpass
filter. Images were acquired using a 40× water immersion
1.2–­numerical aperture (NA) objective with line averaging set to
2 to 4. Confocal sections obtained from the LAS AF Lite software
were analyzed using ImageJ [National Institutes of Health (NIH)].
Gray values were derived from the “analyze” function of ImageJ to
evaluate the fluorescent intensity of oocyte mitochondria.
Measurement of MMP
Oocytes were labeled with 50 nM TMRM (Sigma-Aldrich) and
MitoTracker Green (Thermo Fisher Scientific) in M2 medium for
30 min at 37°C. TMRM was excited using the 552-nm laser, and
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
12 of 15
fluorescence collected using a 560- to 630-nm band-pass filter.
MitoTracker Green was excited using the 488-nm laser and fluorescence
collected using a 495- to 540-nm band-pass filter. Confocal
sections obtained from the LASX software (Leica) were analyzed
using ImageJ for measurements of MMP. Gray values were derived
from the analyze function of ImageJ to evaluate the fluorescent
intensity of oocyte mitochondria. Ratio of TMRM and MitoTracker
Green was calculated to measure the MMP (21). Values were normalized
to the mean of control values in each experimental replicate
toavoidtheimpactofday-to-dayvariationsinthefluorescenceratio.
Measurement of oocyte ATP levels
Denuded GV oocytes (10 pooled per group) collected from PMSGprimed
mice were collected in 50 l of filtered ultrapure water and
stored at −80°C until use. To prepare standards, 10−7
M ATP standard
stock was obtained from ENLITEN ATP Assay System
Bioluminescence Detection Kit (Promega) and was diluted with
filtered ultrapure water. ATP levels were assayed using the ATP
bioluminescent somatic cell assay kit (FLASC, Sigma-Aldrich)
according to the manufacturer’s instructions. Briefly, 100 l of ATP
Assay Mix Working Solution was added to a 96-well plate (reaction
vial, M0187, Greiner) and allowed to stand at RT for 3 min. Somatic
Cell ATP Releasing Reagent (100 l), 50 l of filtered ultrapure
water, and 50 l of sample were added to a new tube, and 100 l was
transferred to the reaction vial. ATP concentration was measured
immediately using a luminometer (BMG, Clariostar, 76G58). Data
are normalized to the mean ATP levels of control group.
Autofluorescence imaging of NAD(P)H and FAD++
Autofluorescent signals of NAD(P)H and FAD++
were detected as
described previously (19). Briefly, the reduced nucleotides NADH
and NADPH were excited by ultraviolet light (405 nm), and
emission was collected using a 435- to 485-nm band-pass filter.
NAD(P)H autofluorescence is both detected in mitochondria and
cytosol. The oxidized flavoproteins (FAD++
) autofluorescence was
collected with a 505- to 550-nm band-pass filter after exciting with
488-nm lasers. FAD++
autofluorescence is exclusively localized into
the mitochondria.
Western blot
Oocytes were collected in SDS sample buffer and heated for 10 min
at 70°C. Proteins were fractionated at 200 V for 50 min using a 10%
NuPAGE bis-tris precast gel (Invitrogen) and Mops running buffer.
Proteins were blotted onto polyvinylidene fluoride membranes
(Invitrogen) for 1.5 hours at 30 V, and then the membrane was
blocked in tris-buffered saline-Tween 20 containing 5% skimmed
milk for 2 hours, followed by incubation overnight at 4°C with primary
antibodies, -actin (Abcam no. ab3280) and DRP1 (Cell Signaling
Technology no. 8570). For primary antibody detection, we
used goat horseradish peroxidase–conjugated anti-mouse or
anti-rabbit secondary antibodies (1:1000, 1 hour at RT, Dako).
Last, the membrane was processed using standard enhanced chemiluminesence
techniques (GE Healthcare) and processed using
the ChemiDoc MP imaging system (Bio-Rad).
Immunofluorescence and imaging
Oocytes or embryos were fixed in a solution of 4% PFA (Sigma-­
Aldrich) and 2% Triton X-100 in PBS for 20 min at RT. Blocking
was performed in PBS with 10% BSA and 2% Tween 20 for 60 min at
RT. For 5mC and H3K27me3 labeling, oocytes or embryos were
collected in M2 medium, and zona pellucida was removed by
Tyrode’s solution (Sigma-Aldrich) and then fixed in 4% PFA for
30 min at RT and permeabilized with 0.5% Triton X-100 in PBS for
20 min at RT. They were denatured with 4 N hydrochloric acid
for 10 min, neutralized with 100 mM tris-HCl (pH 8.5) for 15 min,
blocked in PBS with 10% BSA and 2% Tween for 60 min at RT, and
incubated with specific primary antibodies overnight at 4°C. The
primary antibodies used were DRP1 (Cell Signaling Technology no.
8570), 5mC (Bio-Rad no. MCA2201), H3K27ac (Cell Signaling
Technology no. 8173), H3K27me3 (Merck Millipore no. 07-449),
H3K4me3 (Active Motif no. 39159), CDX2 (Novus Biologicals no.
NB100-2136), OCT-4 (Abcam no. ab184665), H3K9me3 (Cell
Signaling Technology no. 13969), TOM20 (Santa Cruz Biotechnology
no. sc-11415), and RNA Pol II (phospho S2) (Abcam no. ab5095).
This was followed by an incubation with specific Alexa Fluors
(1:1000 diluted in blocking buffer; Invitrogen Molecular Probes) for
1 hour at RT. DNA was labeled using 10-min incubation in Hoechst
33342 (10 g/ml; Sigma-Aldrich). Serial Z sections of fixed oocytes/
eggs in PBS were acquired at RT in a glass-bottomed dish (World
Precision Instruments Inc., Sarasota, FL) using a laser-scanning
confocal microscope imaging system (SP8, Leica).
Images were analyzed using ImageJ software (NIH). The midsection
of each nucleus was identified by Hoechst staining and
determined by maximal area. Total intensities of DNA modifications
in the midsections were measured, and background signals
corresponding to the equal cytoplasmic area were subtracted. Within
each experiment, oocytes/zygotes were imaged using the same
setting on the microscope.
Analysis of oocyte mitochondrial morphology
Germinal vesicle stage oocytes were labeled with MitoTracker Orange
(25 nM) in M2 medium for 30 min at 37°C. Images acquired by an
SP8 confocal microscope were analyzed using MiNA (Mitochondrial
Network Analysis, https://github.com/StuartLab/MiNA), an ImageJ
(NIH)–based utility to aid in quantitatively describe the appearance
of mitochondrial morphology in fluorescence micrographs. The
mean number of attached lines used to represent each structure
(network branches mean) and the mean length of all the lines used
to represent the mitochondrial structures (branch length mean)
were calculated.
Chromosome spreads
For collection of four-cell embryos, females were mated with stud
C57BL/6J males after hCG injection, and the mice were culled 14 to
16 hours after hCG injection to harvest one-cell zygotes from
oviducts. Zygotes were further cultured in SAGE 1-Step medium
(Cooper Surgical) up to four-cell stage at 37°C in a humidified
atmosphere of 5% CO2 in air. Four-cell embryos were transferred
into SAGE 1-Step medium containing Colcemid (100 ng/ml;
Gibco) and were cultured overnight at 37°C to arrest at metaphase.
Embryos were exposed to Acid Tyrodes solution (Sigma-Aldrich)
for 1 min at 25°C to remove the zona pellucida. After a 2-min recovery
in M2 medium, the embryos were transferred into a 0.5% sodium
citrate (Sigma-Aldrich) drop for 2 min and then moved onto a clean
microscope slide that had been previously dipped in a solution of
1% PFA in MilliQ water (pH 9.2), containing 0.15% Triton X-100
and 3 mM dithiothreitol (68). The slides were allowed to dry slowly
in a humid chamber overnight. DNA on the slides was stained with
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
13 of 15
Hoechst 33342 (10 g/ml; Sigma-Aldrich) for 10 min, and slides
were mounted for observation using a Zeiss Imager M2 fluorescence
microscope linked to an AxioCam MRm charge-coupled device
camera system and processed using the Zen software 2011 (Carl
Zeiss Microscopy).
Oocyte meiotic spindle analysis
MII oocytes were fixed in 4% PFA (Sigma-Aldrich) and 2% Triton
X-100 (Sigma-Aldrich) in PBS for 30 min at RT. Oocytes were
blocked in PBS containing 10% BSA and 2% Tween 20 for 1 hour at
RT and incubated with Alexa Fluor 488–conjugated mouse anti–­tubulin
(1:200; Thermo Fisher Scientific no. 322588) for 1 hour at
RT. After three washes, chromosomes were labeled using Hoechst
33342 (10 g/ml; Sigma-Aldrich) for 10 min, and oocytes were
mounted in PBS drops on glass-bottomed dish (World Precision
Instruments, Sarasota, FL, USA) for imaging. Serial Z sections of
labeled oocytes were acquired using a laser-scanning confocal
microscope imaging system (SP8; Leica) and 40× water immersion
objective (1.2 NA) at 25°C. Fluorescein isothiocyanate was excited
using the 488-nm laser, and a band-pass of green-fluorescent
emission was 495 to 540 nm. Hoechst 33342 was excited with a
405-nm laser, and emission was collected with a band-pass filter of
438 to 458 nm. Images were analyzed using Las X software (Leica).
Histological analysis of ovary
Histological analysis of ovarian tissue was performed as previously
described (69). Briefly, ovaries were fixed in 4% PFA, dehydrated,
and embedded in paraffin. Paraffin-embedded ovaries were serially
sectioned at 5-m thickness and stained with hematoxylin for
morphological observation.
Single-oocyte RNA-seq
To perform single-cell RNA-seq analyses, GV-stage oocytes (18
oocytes/genotype) were individually placed in an Eppendorf tube
containing 10 l of nuclease-free water. Tubes were snap frozen in
liquid nitrogen and stored at −80°C until further experiment. The
whole lysate was supplied for complementary DNA synthesis and
RNA-seq library preparation with, respectively, a SMARTer Ultra
Low Input RNA Kit (Clontech, CA USA) and Nextera XT library
prep kit (Illumina, CA, USA), each according to the manufacturer’s
instructions. Indexed libraries were pooled (10 nM each) and sequenced
using an Illumina HiSeq 2500 System under single-end,
100–base pair conditions.
Read files are deposited: National Centre for Biotechnology
Information Gene Expression Omnibus (GEO) dataset (GSE180579).
Reads were aligned to mouse genome GRCm38 and quantified
against Ensembl 88 gene annotations. Tools were run using the
RNAsik pipeline (https://monashbioinformaticsplatform.github.
io/RNAsik-pipe/). PCR duplicates were marked with Picard (http://
broadinstitute.github.io/picard), alignments were performed with
STAR (70), and unstranded read counts quantified by gene using
featureCounts Subread version 1.5.2 (71). BAM file manipulation
was performed with SAMtools (72). The quality of single-cell
data was inspected using Scater version 1.4 (73) and differential
expression calculated with Monocle version 2.2.0 using the default
“negbinomial.size” method (74). One outlier cell (Drp1 KO sample
DG_10_S68) was removed because of low sequence diversity. Only
genes with at least 10 reads in 20% of cells were included in the
analysis. The DEGs of interest were thresholded with a P value
threshold of 0.05 (after multiple hypothesis correction with
Benjamini-Hochberg method). Expression values presented as log10
counts per million, as (fragment count +1) / total count for library
among analyzed genes × 1,000,000.
Oocyte proteomic analysis
MII-stage Drp1/
and Drp1fl/fl
oocytes (300 oocytes per sample)
were lysed in SDC lysis buffer [1% (w/v) sodium deoxycholate, 100 mM
Hepes, pH 8.1], heated at 95°C for 5 min and then sonicated before
measuring the protein concentration using the bicinchoninic acid
assay method. The lysed samples were denatured and alkylated
by adding tris(2-carboxyethyl) phosphine hydrochloride and
2-chloroacetamide to a final concentration of 10 and 40 mM,
respectively, and the mixture was incubated at 95°C for 5 min.
Sequencing grade trypsin was added at an enzyme to protein ratio
of 1:100 and incubated overnight at 37°C. The reaction was stopped
using 1% formic acid, and SDC was removed using 100% water-­
saturated ethyl acetate. The aqueous peptide–containing phase was
collected and concentrated in a vacuum concentrator.
Proteomic data acquisition
Using a Dionex UltiMate 3000 RSLCnano system equipped with a
Dionex UltiMate 3000 RS autosampler, an Acclaim PepMap RSLC
analytical column (75 m by 50 cm, nanoViper, C18, 2 m, 100 Å;
Thermo Fisher Scientific), and an Acclaim PepMap 100 trap column
(100 m by 2 cm, nanoViper, C18, 5 m, 100 Å; Thermo Fisher
Scientific), the tryptic peptides were separated by increasing concentrations
of 80% acetonitrile/0.1% formic acid at a flow of 250 nl/
min for 158 min and analyzed with a QExactive Plus mass spectrometer
(Thermo Fisher Scientific). The instrument was operated
in data-dependent acquisition mode to automatically switch between
full-scan mass spectrometry (MS) and tandem MS acquisition.
Each survey full scan [375 to 1575 mass/charge ratio (m/z)] was
acquired with a resolution of 70,000 (at 200 m/z), an AGC (automatic
gain control) target of 3 × 106
, and a maximum injection time
of 54 ms. Dynamic exclusion was set to 15 s. The 12 most intense
multiply charged ions (z ≥ 2) were sequentially isolated and
fragmented in the collision cell by higher-energy collisional dissociation
(HCD) with a fixed injection time of 54 ms, 17,500 resolution,
and an AGC target of 2 × 105
.
Proteomics data analysis
The raw data files were analyzed with the MaxQuant software suite
v1.6.2.10 (75) and its implemented Andromeda search engine (76)
to obtain protein identifications and their respective label-free
quantification (LFQ) values using standard parameters. These
data were further analyzed with Perseus (77) on the ProteinGroup.
txt file. First, contaminant proteins, reverse sequences, and proteins
identified “only by site” were filtered out. In addition, proteins that
have been only identified by a single peptide and proteins not
identified/quantified consistently in same condition have been
removed. The LFQ data were converted to log2 scale, samples were
grouped by conditions and missing values were imputed using the
“Missing not At Random” (MNAR) method, using random draws
from a left-shifted Gaussian distribution of 1.8 StDev (SD) apart
with a width of 0.3. Student’s t test was used to generate a list of
differentially expressed proteins for each pairwise comparison. A
cutoff of the adjusted P value of 0.05 (false discovery rate–adjusted)
along with a log2 fold change of 1 has been applied to determine
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
14 of 15
significantly regulated proteins in Drp1/
and Drp1fl/fl
comparison.
The enrichment analysis was performed using gprofiler (78) for
Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes,
and Reactome databases against mouse reference proteome. The
adjusted P value cut off of 0.05 was used to determine significantly
enriched GO term or pathway. All the figures were generated in R
using ggplot2 (79) and ComplexHeatmap (80) libraries.
Statistical analysis
Statistical analyses were performed in GraphPad Prism 8. P values
were calculated with Student’s t test or chi-square test as indicated in
respective figure or table legends. Not significant, P > 0.05; *P < 0.05;
**P < 0.01; ***P < 0.001; and ****P < 0.0001. All data are presented
as means ± SD of biological replicates, and error bars represent SD
of biological repeats unless otherwise stated in the figure legends.
SUPPLEMENTARY MATERIALS
Supplementary material for this article is available at https://science.org/doi/10.1126/
sciadv.abl8070
View/request a protocol for this paper from Bio-protocol.
REFERENCES AND NOTES
	 1.	 R. B. Hamanaka, N. S. Chandel, Mitochondrial reactive oxygen species regulate cellular
signaling and dictate biological outcomes. Trends Biochem. Sci. 35, 505–513 (2010).
	 2.	 P. Mishra, D. C. Chan, Mitochondrial dynamics and inheritance during cell division,
development and disease. Nat. Rev. Mol. Cell Biol. 15, 634–646 (2014).
	 3.	 J. Wakabayashi, Z. Zhang, N. Wakabayashi, Y. Tamura, M. Fukaya, T. W. Kensler, M. Iijima,
H. Sesaki, The dynamin-related GTPase Drp1 is required for embryonic and brain
development in mice. J. Cell Biol. 186, 805–816 (2009).
	 4.	 F. Kraus, M. T. Ryan, The constriction and scission machineries involved in mitochondrial
fission. J. Cell Sci. 130, 2953–2960 (2017).
	 5.	 N. Ishihara, M. Nomura, A. Jofuku, H. Kato, S. O. Suzuki, K. Masuda, H. Otera, Y. Nakanishi,
I. Nonaka, Y. I. Goto, N. Taguchi, H. Morinaga, M. Maeda, R. Takayanagi, S. Yokota,
K. Mihara, Mitochondrial fission factor Drp1 is essential for embryonic development
and synapse formation in mice. Nat. Cell Biol. 11, 958–966 (2009).
	 6.	 T. Uo, J. Dworzak, C. Kinoshita, D. M. Inman, Y. Kinoshita, P. J. Horner, R. S. Morrison, Drp1
levels constitutively regulate mitochondrial dynamics and cell survival in cortical
neurons. Exp. Neurol. 218, 274–285 (2009).
	 7.	 K. Itoh, K. Nakamura, M. Iijima, H. Sesaki, Mitochondrial dynamics in neurodegeneration.
Trends Cell Biol. 23, 64–71 (2013).
	 8.	 A. S. Rambold, E. L. Pearce, Mitochondrial dynamics at the interface of immune cell
metabolism and function. Trends Immunol. 39, 6–18 (2018).
	 9.	 M. D. Buck, D. O’Sullivan, R. I. Klein Geltink, J. D. Curtis, C. H. Chang, D. E. Sanin, J. Qiu,
O. Kretz, D. Braas, G. J. W. van der Windt, Q. Chen, S. C. C. Huang, C. M. O’Neill,
B. T. Edelson, E. J. Pearce, H. Sesaki, T. B. Huber, A. S. Rambold, E. L. Pearce, Mitochondrial
dynamics controls T cell fate through metabolic programming. Cell 166, 63–76 (2016).
	 10.	 M. Khacho, A. Clark, D. S. Svoboda, J. Azzi, J. G. MacLaurin, C. Meghaizel, H. Sesaki,
D. C. Lagace, M. Germain, M. E. Harper, D. S. Park, R. S. Slack, Mitochondrial dynamics
impacts stem cell identity and fate decisions by regulating a nuclear transcriptional
program. Cell Stem Cell 19, 232–247 (2016).
	11.	 T. Wai, T. Langer, Mitochondrial dynamics and metabolic regulation. Trends Endocrinol.
27, 105–117 (2016).
	12.	 N. S. Chandel, Mitochondria as signaling organelles. BMC Biol. 12, 34 (2014).
	13.	 D. Adhikari, K. Liu, Molecular mechanisms underlying the activation of mammalian
primordial follicles. Endocr. Rev. 30, 438–464 (2009).
	14.	 R. P. Jansen, Germline passage of mitochondria: Quantitative considerations and possible
embryological sequelae. Hum. Reprod. 15 (Suppl. 2), 112–128 (2000).
	15.	 L. Piko, K. D. Taylor, Amounts of mitochondrial DNA and abundance of some
mitochondrial gene transcripts in early mouse embryos. Dev. Biol. 123, 364–374 (1987).
	16.	 T. Kono, Y. Obata, T. Yoshimzu, T. Nakahara, J. Carroll, Epigenetic modifications during
oocyte growth correlates with extended parthenogenetic development in the mouse.
Nat. Genet. 13, 91–94 (1996).
	17.	 K. R. Stewart, L. Veselovska, G. Kelsey, Establishment and functions of DNA methylation
in the germline. Epigenomics 8, 1399–1413 (2016).
	18.	 R. Dumollard, K. Campbell, G. Halet, J. Carroll, K. Swann, Regulation of cytosolic and
mitochondrial ATP levels in mouse eggs and zygotes. Dev. Biol. 316, 431–440 (2008).
	19.	 R. Dumollard, Z. Ward, J. Carroll, M. R. Duchen, Regulation of redox metabolism
in the mouse oocyte and embryo. Development 134, 455–465 (2007).
	20.	 C. M. Dalton, J. Carroll, Biased inheritance of mitochondria during asymmetric cell
division in the mouse oocyte. J. Cell Sci. 126, 2955–2964 (2013).
	 21.	 U. Al-Zubaidi, J. Liu, O. Cinar, R. L. Robker, D. Adhikari, J. Carroll, The spatio-temporal
dynamics of mitochondrial membrane potential during oocyte maturation. Mol. Hum.
Reprod. 25, 695–705 (2019).
	22.	 J. Van Blerkom, M. N. Runner, Mitochondrial reorganization during resumption
of arrested meiosis in the mouse oocyte. Am. J. Anat. 171, 335–355 (1984).
	 23.	 R. Dumollard, P. Marangos, G. Fitzharris, K. Swann, M. Duchen, J. Carroll, Sperm-triggered
[Ca2+
] oscillations and Ca2+
homeostasis in the mouse egg have an absolute requirement
for mitochondrial ATP production. Development 131, 3057–3067 (2004).
	 24.	 T. Wakai, A. Mehregan, R. A. Fissore, Ca2+
signaling and homeostasis in mammalian
oocytes and eggs. Cold Spring Harb. Perspect. Biol. 11, a035162 (2019).
	 25.	 O. Udagawa, T. Ishihara, M. Maeda, Y. Matsunaga, S. Tsukamoto, N. Kawano, K. Miyado,
H. Shitara, S. Yokota, M. Nomura, K. Mihara, N. Mizushima, N. Ishihara, Mitochondrial
fission factor Drp1 maintains oocyte quality via dynamic rearrangement of multiple
organelles. Curr. Biol. 24, 2451–2458 (2014).
	26.	 Z. J. Lan, X. Xu, A. J. Cooney, Differential oocyte-specific expression of Cre recombinase
activity in GDF-9-iCre, Zp3cre, and Msx2Cre transgenic mice. Biol. Reprod. 71, 1469–1474
(2004).
	27.	 M. Manczak, H. Sesaki, Y. Kageyama, P. H. Reddy, Dynamin-related protein 1
heterozygote knockout mice do not have synaptic and mitochondrial deficiencies.
Biochim. Biophys. Acta 1822, 862–874 (2012).
	 28.	 V. Perez-Garcia, E. Fineberg, R. Wilson, A. Murray, C. I. Mazzeo, C. Tudor, A. Sienerth,
J. K. White, E. Tuck, E. J. Ryder, D. Gleeson, E. Siragher, H. Wardle-Jones, N. Staudt, N. Wali,
J. Collins, S. Geyer, E. M. Busch-Nentwich, A. Galli, J. C. Smith, E. Robertson, D. J. Adams,
W. J. Weninger, T. Mohun, M. Hemberger, Placentation defects are highly prevalent
in embryonic lethal mouse mutants. Nature 555, 463–468 (2018).
	 29.	 D. Li, J. Liu, X. Yang, C. Zhou, J. Guo, C. Wu, Y. Qin, L. Guo, J. He, S. Yu, H. Liu, X. Wang,
F. Wu, J. Kuang, A. P. Hutchins, J. Chen, D. Pei, Chromatin accessibility dynamics during
iPSC reprogramming. Cell Stem Cell 21, 819–833.e6 (2017).
	 30.	 A. M. Cervera, J. P. Bayley, P. Devilee, K. J. McCreath, Inhibition of succinate dehydrogenase
dysregulates histone modification in mammalian cells. Mol. Cancer 8, 89 (2009).
	 31.	 A. J. MacFarlane, C. A. Perry, H. H. Girnary, D. Gao, R. H. Allen, S. P. Stabler, B. Shane,
P. J. Stover, Mthfd1 is an essential gene in mice and alters biomarkers of impaired
one-carbon metabolism. J. Biol. Chem. 284, 1533–1539 (2009).
	 32.	 A. E. Beaudin, C. A. Perry, S. P. Stabler, R. H. Allen, P. J. Stover, Maternal Mthfd1 disruption
impairs fetal growth but does not cause neural tube defects in mice. Am. J. Clin. Nutr. 95,
882–891 (2012).
	33.	 J. P. Etchegaray, R. Mostoslavsky, Interplay between metabolism and epigenetics:
A nuclear adaptation to environmental changes. Mol. Cell 62, 695–711 (2016).
	34.	 C. Lu, C. B. Thompson, Metabolic regulation of epigenetics. Cell Metab. 16, 9–17 (2012).
	 35.	 U. Sharma, O. J. Rando, Metabolic inputs into the epigenome. CellMetab. 25, 544–558 (2017).
	 36.	 D. P. Barlow, M. S. Bartolomei, Genomic imprinting in mammals. Cold Spring Harb.
Perspect. Biol. 6, a018382 (2014).
	37.	 A. Inoue, L. Jiang, F. Lu, Y. Zhang, Genomic imprinting of Xist by maternal H3K27me3.
Genes Dev. 31, 1927–1932 (2017).
	 38.	 X. Liu, C. Wang, W. Liu, J. Li, C. Li, X. Kou, J. Chen, Y. Zhao, H. Gao, H. Wang, Y. Zhang,
Y. Gao, S. Gao, Distinct features of H3K4me3 and H3K27me3 chromatin domains
in pre-implantation embryos. Nature 537, 558–562 (2016).
	 39.	 H. Zheng, B. Huang, B. Zhang, Y. Xiang, Z. du, Q. Xu, Y. Li, Q. Wang, J. Ma, X. Peng, F. Xu,
W. Xie, Resetting epigenetic memory by reprogramming of histone modifications
in mammals. Mol. Cell 63, 1066–1079 (2016).
	 40.	 A. C. Ferguson-Smith, H. Sasaki, B. M. Cattanach, M. A. Surani, Parental-origin-specific
epigenetic modification of the mouse H19 gene. Nature 362, 751–755 (1993).
	 41.	 A. Inoue, Z. Chen, Q. Yin, Y. Zhang, Maternal Eed knockout causes loss of H3K27me3
imprinting and random X inactivation in the extraembryonic cells. Genes Dev. 32,
1525–1536 (2018).
	42.	 M. Kaneda, M. Okano, K. Hata, T. Sado, N. Tsujimoto, E. Li, H. Sasaki, Essential role for de
novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting. Nature 429,
900–903 (2004).
	 43.	 Q. Xu, Y. Xiang, Q. Wang, L. Wang, J. Brind’Amour, A. B. Bogutz, Y. Zhang, B. Zhang, G. Yu,
W. Xia, Z. du, C. Huang, J. Ma, H. Zheng, Y. Li, C. Liu, C. L. Walker, E. Jonasch, L. Lefebvre,
M. Wu, M. C. Lorincz, W. Li, L. Li, W. Xie, SETD2 regulates the maternal epigenome,
genomic imprinting and embryonic development. Nat. Genet. 51, 844–856 (2019).
	44.	 X. Lu, Z. Gao, D. Qin, L. Li, A maternal functional module in the mammalian Oocyte-ToEmbryo
transition. Trends Mol. Med. 23, 1014–1023 (2017).
	 45.	 X. J. Yu, Z. Yi, Z. Gao, D. Qin, Y. Zhai, X. Chen, Y. Ou-Yang, Z. B. Wang, P. Zheng, M. S. Zhu,
H. Wang, Q. Y. Sun, J. Dean, L. Li, The subcortical maternal complex controls symmetric
division of mouse zygotes by regulating F-actin dynamics. Nat. Commun. 5, 4887 (2014).
Adhikari et al., Sci. Adv. 8, eabl8070 (2022) 15 June 2022
SCI ENCE ADVANCES | RESEARCH ARTICLE
15 of 15
	 46.	 A. M. Alazami, S. M. Awad, S. Coskun, S. al-Hassan, H. Hijazi, F. M. Abdulwahab, C. Poizat,
F. S. Alkuraya, TLE6 mutation causes the earliest known human embryonic lethality.
Genome Biol. 16, 240 (2015).
	47.	 J. Lin, H. Xu, B. Chen, W. Wang, L. Wang, X. Sun, Q. Sang, Expanding the genetic and
phenotypic spectrum of female infertility caused by TLE6 mutations. J. Assist. Reprod.
Genet. 37, 437–442 (2020).
	 48.	 H. Demond, Z. Anvar, B. N. Jahromi, A. Sparago, A. Verma, M. Davari, L. Calzari, S. Russo,
M. A. Jahromi, D. Monk, S. Andrews, A. Riccio, G. Kelsey, A KHDC3L mutation resulting
in recurrent hydatidiform mole causes genome-wide DNA methylation loss in oocytes
and persistent imprinting defects post-fertilisation. Genome Med. 11, 84 (2019).
	49.	 L. E. Docherty, F. I. Rezwan, R. L. Poole, C. L. S. Turner, E. Kivuva, E. R. Maher,
S. F. Smithson, J. P. Hamilton-Shield, M. Patalan, M. Gizewska, J. Peregud-Pogorzelski,
J. Beygo, K. Buiting, B. Horsthemke, L. Soellner, M. Begemann, T. Eggermann, E. Baple,
S. Mansour, I. K. Temple, D. J. G. Mackay, Mutations in NLRP5 are associated
with reproductive wastage and multilocus imprinting disorders in humans. Nat.
Commun. 6, 8086 (2015).
	 50.	 S. Mahadevan, V. Sathappan, B. Utama, I. Lorenzo, K. Kaskar, I. B. van den Veyver,
Maternally expressed NLRP2 links the subcortical maternal complex (SCMC) to fertility,
embryogenesis and epigenetic reprogramming. Sci. Rep. 7, 44667 (2017).
	 51.	 D. E. Sanin, M. Matsushita, R. I. K. Geltink, K. M. Grzes, N. van Teijlingen Bakker, M. Corrado,
A. M. Kabat, M. D. Buck, J. Qiu, S. J. Lawless, A. M. Cameron, M. Villa, F. Baixauli,
A. E. Patterson, F. Hässler, J. D. Curtis, C. M. O’Neill, D. O’Sullivan, D. Wu, G. Mittler,
S. C.-C. Huang, E. L. Pearce, E. J. Pearce, Mitochondrial membrane potential regulates
nuclear gene expression in macrophages exposed to prostaglandin E2. Immunity 49,
1021–1033.e6 (2018).
	 52.	 I. Martínez-Reyes, L. P. Diebold, H. Kong, M. Schieber, H. Huang, C. T. Hensley,
M. M. Mehta, T. Wang, J. H. Santos, R. Woychik, E. Dufour, J. N. Spelbrink, S. E. Weinberg,
Y. Zhao, R. J. DeBerardinis, N. S. Chandel, TCA cycle and mitochondrial membrane
potential are necessary for diverse biological functions. Mol. Cell 61, 199–209 (2016).
	53.	 G. S. Ducker, J. D. Rabinowitz, One-carbon metabolism in health and disease. Cell Metab.
25, 27–42 (2017).
	 54.	 E. Ansó, S. E. Weinberg, L. P. Diebold, B. J. Thompson, S. Malinge, P. T. Schumacker, X. Liu,
Y. Zhang, Z. Shao, M. Steadman, K. M. Marsh, J. Xu, J. D. Crispino, N. S. Chandel, The
mitochondrial respiratory chain is essential for haematopoietic stem cell function. Nat.
Cell Biol. 19, 614–625 (2017).
	 55.	 O. A. Lozoya, F. Xu, D. Grenet, T. Wang, S. A. Grimm, V. Godfrey, S. Waidyanatha,
R. P. Woychik, J. H. Santos, Single nucleotide resolution analysis reveals pervasive,
long-lasting DNA methylation changes by developmental exposure to a mitochondrial
toxicant. Cell Rep. 32, 108131 (2020).
	 56.	 P. K. Kopinski, K. A. Janssen, P. M. Schaefer, S. Trefely, C. E. Perry, P. Potluri, J. A. Tintos-Hernandez,
L. N. Singh, K. R. Karch, S. L. Campbell, M. T. Doan, H. Jiang, I. Nissim, E. Nakamaru-Ogiso,
K. E. Wellen, N. W. Snyder, B. A. Garcia, D. C. Wallace, Regulation of nuclear epigenome
by mitochondrial DNA heteroplasmy. Proc. Natl. Acad. Sci. U.S.A. 116, 16028–16035
(2019).
	 57.	 L. Li, S. Zhu, W. Shu, Y. Guo, Y. Guan, J. Zeng, H. Wang, L. Han, J. Zhang, X. Liu, C. Li, X. Hou,
M. Gao, J. Ge, C. Ren, H. Zhang, T. Schedl, X. Guo, M. Chen, Q. Wang, Characterization
of metabolic patterns in mouse oocytes during meiotic maturation. Mol. Cell 80, 525–540.e9
(2020).
	58.	 R. Guantes, A. Rastrojo, R. Neves, A. Lima, B. Aguado, F. J. Iborra, Global variability in gene
expression and alternative splicing is modulated by mitochondrial content. Genome Res.
25, 633–644 (2015).
	59.	 M. N. Shahbazi, Mechanisms of human embryo development: From cell fate to tissue
shape and back. Development 147, dev190629 (2020).
	 60.	 L. L. Wu, D. L. Russell, S. L. Wong, M. Chen, T. Tsai, J. C. S. John, R. J. Norman,
M. A. Febbraio, J. Carroll, R. L. Robker, Mitochondrial dysfunction in oocytes of obese
mothers: Transmission to offspring and reversal by pharmacological endoplasmic
reticulum stress inhibitors. Development 142, 681–691 (2015).
	 61.	 N. Igosheva, A. Y. Abramov, L. Poston, J. J. Eckert, T. P. Fleming, M. R. Duchen,
J. McConnell, Maternal diet-induced obesity alters mitochondrial activity and redox
status in mouse oocytes and zygotes. PLOS ONE 5, e10074 (2010).
	 62.	 J. L. A. Ferey, A. L. Boudoures, M. Reid, A. Drury, S. Scheaffer, Z. Modi, A. Kovacs, T. Pietka,
B. DeBosch, M. D. Thompson, A. Diwan, K. H. Moley, A maternal high-fat, high-sucrose
diet induces transgenerational cardiac mitochondrial dysfunction independently
of maternal mitochondrial inheritance. Am. J. Physiol. Heart Circ. Physiol. 316,
H1202–H1210 (2019).
	 63.	 J. L. Saben, A. L. Boudoures, Z. Asghar, A. Thompson, A. Drury, W. Zhang, M. Chi,
A. Cusumano, S. Scheaffer, K. H. Moley, Maternal metabolic syndrome programs
mitochondrial dysfunction via germline changes across three generations. Cell Rep. 16, 1–8
(2016).
	 64.	 U. Al-Zubaidi, D. Adhikari, O. Cinar, Q.-H. Zhang, W. S. Yuen, M. P. Murphy, L. Rombauts,
R. L. Robker, J. Carroll, Mitochondria-targeted therapeutics, MitoQ and BGP-15, reverse
aging-associated meiotic spindle defects in mouse and human oocytes. Hum. Reprod. 36,
771–784 (2021).
	 65.	 R. Pasquariello, A. F. Ermisch, E. Silva, S. McCormick, D. Logsdon, J. P. Barfield,
W. B. Schoolcraft, R. L. Krisher, Alterations in oocyte mitochondrial number and function
are related to spindle defects and occur with maternal aging in mice and humans. Biol.
Reprod. 100, 971–981 (2019).
	66.	 Z.-J. Ge, H. Schatten, C.-L. Zhang, Q.-Y. Sun, Oocyte ageing and epigenetics. Reproduction
149, R103–R114 (2015).
	67.	 J. McGrath, D. Solter, J. Mann, Pronuclear transplantation in the mouse embryo. Cold
Spring Harbor Protoc. 2017, pdb.prot094417 (2017).
	68.	 C. A. Hodges, P. A. Hunt, Simultaneous analysis of chromosomes and chromosomeassociated
proteins in mammalian oocytes and embryos. Chromosoma 111, 165–169
(2002).
	 69.	 P. Reddy, L. Liu, D. Adhikari, K. Jagarlamudi, S. Rajareddy, Y. Shen, C. du, W. Tang,
T. Hämäläinen, S. L. Peng, Z. J. Lan, A. J. Cooney, I. Huhtaniemi, K. Liu, Oocyte-specific
deletion of Pten causes premature activation of the primordial follicle pool. Science 319,
611–613 (2008).
	70.	 A. Dobin, C. A. Davis, F. Schlesinger, J. Drenkow, C. Zaleski, S. Jha, P. Batut, M. Chaisson,
T. R. Gingeras, STAR: Ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21
(2013).
	71.	 Y. Liao, G. K. Smyth, W. Shi, featureCounts: An efficient general purpose program
for assigning sequence reads to genomic features. Bioinformatics 30, 923–930 (2014).
	 72.	 H. Li, B. Handsaker, A. Wysoker, T. Fennell, J. Ruan, N. Homer, G. Marth, G. Abecasis,
R. Durbin; 1000 Genome Project Data Processing Subgroup, The Sequence Alignment/
Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).
	 73.	 D. J. McCarthy, K. R. Campbell, A. T. Lun, Q. F. Wills, Scater: Pre-processing, quality control,
normalization and visualization of single-cell RNA-seq data in R. Bioinformatics 33,
1179–1186 (2017).
	 74.	 C. Trapnell, D. Cacchiarelli, J. Grimsby, P. Pokharel, S. Li, M. Morse, N. J. Lennon, K. J. Livak,
T. S. Mikkelsen, J. L. Rinn, The dynamics and regulators of cell fate decisions are revealed
by pseudotemporal ordering of single cells. Nat. Biotechnol. 32, 381–386 (2014).
	75.	 J. Cox, M. Mann, MaxQuant enables high peptide identification rates, individualized
p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotechnol.
26, 1367–1372 (2008).
	 76.	 J. Cox, N. Neuhauser, A. Michalski, R. A. Scheltema, J. V. Olsen, M. Mann, Andromeda:
A peptide search engine integrated into the MaxQuant environment. J. Proteome Res. 10,
1794–1805 (2011).
	 77.	 S. Tyanova, T. Temu, P. Sinitcyn, A. Carlson, M. Y. Hein, T. Geiger, M. Mann, J. Cox, The
Perseus computational platform for comprehensive analysis of (prote)omics data. Nat.
Methods 13, 731–740 (2016).
	78.	 U. Raudvere, L. Kolberg, I. Kuzmin, T. Arak, P. Adler, H. Peterson, J. Vilo, g:Profiler: A web
server for functional enrichment analysis and conversions of gene lists (2019 update).
Nucleic Acids Res. 47, W191–W198 (2019).
	 79.	 H. Wickham, ggplot2: Elegant Graphics for Data Analysis (Use R, Springer, 2009).
	80.	 Z. Gu, R. Eils, M. Schlesner, Complex heatmaps reveal patterns and correlations
in multidimensional genomic data. Bioinformatics 32, 2847–2849 (2016).
Acknowledgments: We thank Monash Histology Platform, Monash Bioinformatics Platform,
Monash Proteomics and Metabolomics Platform, and Monash Micro Imaging for enabling
many of the experiments described herein. We thank T. Kono, Tokyo University of Agriculture
for help with oocyte RNA-seq. Figures were created using BioRender.com. Funding: This work
was supported by National Health and Medical Research Council (NHMRC) Australia
1165627 (D.A. and J.C.), NHMRC Peter Doherty–Australian Biomedical Fellowship
APP1112766 (D.A.), NHMRC Research Fellowship 1117975 (R.L.R.), and NIH GM144103 (H.S.).
Author contributions: Conceptualization: D.A. and J.C. Methodology/Investigation: D.A.,
I.-w.L., J.L., U.A.-Z., Q.-H.Z., W.S.Y., L.H., Y.W., J.R.M., and R.L.R. Resources: H.S. provided Drp1
floxed mice. Funding acquisition: D.A. and J.C. Writing—original draft: D.A. and J.C. Writing—
review and editing: D.A., J.C., J.R.M., and R.L.R. Competing interests: The authors declare that
they have no competing interests. Data and materials availability: All sequencing data
have been deposited in the GEO under accession number GSE180579. All other data are
available in the main text or the Supplementary Materials. We obtained code MiNa from
https://github.com/StuartLab/MiNA.
Submitted 6 August 2021
Accepted 2 May 2022
Published 15 June 2022
10.1126/sciadv.abl8070