Mitochondrial defects triggered by amg-1 mutation elicit UPR^mt and phagocytic clearance during spermatogenesis in C. elegans

In most animals, sperm are produced by meiosis from the spermatocytes of males, whereas spermatogenesis in Caenorhabditis elegans occurs in males and hermaphrodites. Sperm are produced by spermatocytes after two meiotic divisions. The first meiotic division begins after the primary spermatocyte detaches from the syncytium rachis, followed by the separation of the secondary spermatocyte from the nucleus-free residual body (RB) via budding, which produces haploid spermatids. Therefore, haploid spermatids lack many cellular components, including most voltage-gated ion channels, microtubule proteins, actin and all ribosomes, as these components remain in the residual body (Machaca et al., 1996; Ward, 1986). The components isolated from sperm during budding are well defined and include the nucleus, multiple mitochondria and membranous organelles (MOs).

Mitochondria act as intracellular powerhouses and are crucial for sperm function and fertilization. The regulation of mitochondrial metabolism requires RNA-binding proteins, but how it regulates spermatogenesis remains largely unknown. Leucine-rich pentatricopeptide repeat (PPR) motif-containing protein (LRPPRC) is a member of the PPR protein family, localized in mitochondria, and enriched with several PPR motifs for recognition and binding of RNA (Mili and Piñol, 2003). Mutations in the LRPPRC gene cause a reduction in cytochrome c oxidase mRNA levels, ultimately leading to a deficiency of cytochrome c oxidase during oxidative phosphorylation (OXPHOS) within mitochondria (Xu et al., 2004; Debray et al., 2011). Similar to other PPR proteins, LRPPRC regulates mitochondrial gene expression at the transcriptional and post-transcriptional levels to promote OXPHOS (Liu et al., 2011; Akie et al., 2015; Lei et al., 2016). However, few studies have focused on whether LRPPRC mutations affect reproductive health. In addition, it is extremely important for mitochondria to recognize and degrade denatured or misfolded proteins generated in mitochondrial metabolism. They accumulate in mitochondria during mitochondrial dysfunction and trigger the mitochondrial unfolded protein response (UPR^mt). UPR^mt refers to the mitochondrial activation of nuclear genes that encode transcriptional activation programs for gene clusters, such as mitochondrial heat shock proteins and proteases, to maintain protein homeostasis within the mitochondria (Haynes et al., 2013; Jovaisaite et al., 2014). UPR^mt is activated when mitochondrial function is impaired, particularly in proteins associated with the respiratory transport chain complex (Durieux et al., 2011). Therefore, the UPR^mt reporter acts as a biological indicator of the functional state of mitochondria.

In our previous studies, we identified the LRPPRC homolog AMG-1 (abnormal mitochondria in germline 1; also known as C04E6.11) in the C. elegans germline, which binds the 12S and 16S ribosomal RNA (rRNA) involved in mitochondrial ribosome assembly and is essential for maintaining germ cell development (Wang et al., 2023a). Here, we found that amg-1 mutation resulted in defective mitochondrial capacity with a severe UPR^mt, and some morphologically abnormal mitochondria were discarded into the RB as spermatid budding occurred, ultimately inducing phagocytic clearance. Thus, AMG-1 is important in balancing the UPR^mt and phagocytic clearance during spermatogenesis.

Leigh syndrome is a severe neurological genetic disorder that affects the central nervous system. Very few studies have investigated the role of the PPR protein in animal reproductive development because 82% of patients with Leigh syndrome French Canadian type (LSFC), a mitochondrial disease caused by a mutation in the LRPPRC gene, die at a median age of 1.6 years (Debray et al., 2011). The most common mutation in patients with diseases caused by an LRPPRC mutation is a single base missense mutation from alanine to valine in exon 9 (A354V) (Debray et al., 2011; Mootha et al., 2003), which interferes with the expression of the LRPPRC protein, resulting in LSFC syndrome. Using C. elegans as a model, we reported for the first time that the PPR protein AMG-1 is involved in the regulation of spermatogenesis (Wang et al., 2023a). Although the C. elegans AMG-1 protein has low homology to the LRPPRC protein, the C. elegans AMG-1 protein shares a conserved amino acid site with the human LRPPRC protein, such as an alanine at position 356 (Fig. 1A). We converted the C to T at site 1067 in exon 6 of the amg-1 gene using CRISPR-Cas9 genome editing (Fig. 1B; Fig. S1), resulting in a single base mutation from alanine to valine at position 356 (ibp181), and it exhibited a sterile phenotype very similar to that of the amg-1(ibp80) strain (Fig. 1B; Fig. S1). Furthermore, the sterile phenotype was not rescued by mating with him-5 males (Fig. 1C).

The proximal arm of the him-5 gonad is sequentially distributed with oocytes, sperm-stored spermatheca and fertilized eggs in the uterus of adult hermaphroditic nematodes, whereas abnormal gonad morphology was observed in the amg-1(ibp80);him-5 and amg-1(ibp181);him-5 mutants, with no clear spermathecal structure. We also observed sperm-like cells distributed near the proximal arm of the gonad (Fig. 1D). The GFP::NKB-2 fluorescent protein is localized to the plasma membrane of spermatocytes and sperm and indicates the process of spermatogenesis (Wang et al., 2021, 2023b). We dissected out the gonads from hermaphroditic young adults to observe the formation and distribution of sperm and found that the him-5 sperm were distributed in the spermatheca and surrounding areas awaiting fertilization, whereas sperm in the amg-1(ibp181) mutant were in spermatogenesis at the same time (Fig. 1E), suggesting that the point mutation in amg-1 affects spermatogenesis and that its defective phenotype is consistent with that of the amg-1(ibp80) mutant. Thus, C. elegans AMG-1 might be an evolutionarily conserved protein with LRPPRC with some conserved functions.

Similar to spermatogenesis in mammals, spermatogenesis in C. elegans requires two meiotic divisions. C. elegans spermatocytes originate from the most distal germ cells of the gonads, and a primary spermatocyte divides into two secondary spermatocytes during meiosis I (Fig. 2A), with associated cytoplasmic divisions that may be complete or incomplete. Then, the secondary spermatocytes immediately trigger the initiation of the second meiosis. During meiosis II, each secondary spermatocyte buds to produce two haploid sperm and an RB, and haploid spermatids are produced when budding is complete. Sperm retain mitochondria, MOs and a haploid nucleus, whereas all ribosomes, nearly all actin and myosin, and most tubulin are packaged into RB (Ward et al., 1983; Ward, 1986; Machaca et al., 1996; Huang et al., 2012).

To investigate the effect of a mitochondrial mutation caused by amg-1 on defects in spermatogenesis, we observed sperm from tomm-20::gfp males. The GFP signal was on the outer mitochondrial membrane and indicates the morphology and distribution of mitochondria. We dissected out spermatocytes, budding spermatids and haploid spermatids to observe the distribution of mitochondria. We found that the distribution of mitochondrial fluorescence in amg-1 budding spermatids did not differ between the RB and the spermatids, nor did they show any distribution polarity (Fig. 2B,C). After budding, some mitochondria with the amg-1 mutation, as indicated by TOMM-20::GFP, entered the RB and displayed vesicular forms, suggesting that the mitochondria were defective. In addition, the fluorescence signal of the mitochondria in budded spermatids was unevenly and weakly distributed (Fig. 2B), implying that some defective mitochondria entered the RB during budding.

To observe whether defective mitochondria are discarded into the RB during budding, we analyzed him-5 and amg-1;him-5 spermatocytes undergoing budding, and budded RBs by electron microscopy. As a result, the swollen mitochondria in amg-1;him-5 mutant spermatocytes were distributed close to the RB during budding, and cristae-deficient mitochondria were present in the mature RB (Fig. 2D), indicating that damaged mitochondria were discarded during spermatogenesis through budding to ensure sperm quality.

amg-1 mutation results in mitochondrial dysfunction during C. elegans spermatogenesis (Wang et al., 2023a). In C. elegans, activation of the UPR^mt can be monitored by the expression of the transcriptional UPR^mt reporter hsp-6p::gfp (Yoneda et al., 2004). We determined that the mitochondrial malfunction caused by AMG-1 deficiency activated the expression of the hsp-6p::gfp reporter in hermaphrodites and males (Fig. 3A-D; Movies 1-4). A malfunction in the germ cell mitochondria of the amg-1 mutants induced the expression of the hsp-6p::gfp reporter primarily in gonadal sheath cells, particularly in the spermatheca of hermaphrodites and the vas deferens of males, but triggered little expression of the hsp-6p::gfp reporter in sperm (Fig. 3E; Fig. S2; Movies 1-8). This is distinct from the cell-non-autonomous UPR^mt caused by neuropeptide signals (Zhang et al., 2018).

The UPR^mt has been extensively described in C. elegans. When mitochondrial function is arrested, the UPR^mt triggers the transcription factor ATFS-1 and the co-regulatory factor DVE-1 to promote the transcription of nuclear-encoded molecular chaperones to maintain mitochondrial protein homeostasis, and the protease CLPP-1 is required to generate a mitochondrial-derived signal to activate the UPR^mt (Haynes et al., 2010; Nargund et al., 2012). Through transcriptome sequencing analysis of L4 stage gonads, we discovered that AMG-1 deletion resulted in significant upregulation of the transcription of atfs-1, dve-1 and clpp-1 (Fig. 3F-H), suggesting that the UPR^mt was activated. To test whether ATFS-1 is required for regulating the UPR^mt in gonadal cells activated by AMG-1 deletion, we knocked out the atfs-1 gene to examine the effect of the UPR^mt in amg-1 mutants. The results showed that there were almost no UPR^mt signals in the gonads of amg-1;atfs-1 double mutant hermaphrodites or males (Fig. 3A-D), indicating that the atfs-1 mutation strongly suppresses the activation of hsp-6p::gfp in amg-1 worms. Therefore, the UPR^mt of gonadal cells induced by the amg-1 mutation mainly relies on the ATFS-1-dependent pathway.

The UPR^mt is a transcriptional response triggered by a broad range of mitochondrial failures. Activation of the UPR^mt is responsible for the recovery of the mitochondrial protein network from stress to maintain normal mitochondrial function (Yoneda et al., 2004; Zhao et al., 2002). Intriguingly, malfunctions in germ cell mitochondria due to the loss of the AMG-1-induced UPR^mt in gonadal sheath cells, particularly in the spermatheca of hermaphrodites and the vas deferens of males but less so in germ cells, was monitored by the expression of the transcriptional reporter hsp-6p::gfp. Here, we discovered a novel UPR^mt elicited by mitochondrial stress in germ cells caused by loss of mtRBP AMG-1; however, this UPR^mt was observed in gonad but not in other tissues. The UPR^mt in amg-1 mutants depended on ATFS-1, a well-characterized transcription factor translocated from mitochondria to the nucleus during stress to induce the UPR^mt. Mature spermatids may lack the UPR^mt, as these cells are transcriptionally and translationally silent due to the elimination of organelles, such as ribosomes, but this does not mean that expression has been silenced in all germ cells, as we have also seen that the UPR^mt occurs in germ cells and oocytes. In addition, the strong UPR^mt in gonad sheath cells might be activated by their own mitochondrial dysfunction. Further studies of the UPR^mt caused by loss of AMG-1 will provide further insight into the mechanism of the UPR^mt and mitochondrial homeostasis and shed light on how animals enable organism-wide responses to tissue-specific mitochondrial stress.

Germline apoptosis usually occurs during oogenesis but rarely during spermatogenesis in C. elegans (Gumienny et al., 1999), and stress-induced apoptosis has not been observed in the male germline (Gartner et al., 2018). Interestingly, here, we detected spermatogenetic germ-cell phagocytic clearance in amg-1 mutant worms. The fusion protein CED-1::GFP has been used to visualize germ cell phagocytic clearance because it highlights the gonadal sheath cells clustered around apoptotic germ cells and marks early apoptotic corpses (Schumacher et al., 2005). A large number of apoptotic corpses accumulated in the proximal gonads of amg-1;him-5 adult hermaphrodites (Fig. 4A,C; Movies 9,10) due to the arrested differentiation of spermatogenesis. Phagocytic clearance, as detected by CED-1::GFP, also occurred frequently in the gonads of amg-1;him-5 males (Fig. 4B,C; Movies 11,12). In contrast, CED-1::GFP expression was mainly detected within the gonadal loop region where oogenesis took place in the control him-5 adult hermaphrodites.

To further investigate the apoptotic process in the gonads of hermaphrodites, we dissected out L4 stage gonads and used adult-stage hermaphrodites to observe spermatogenesis. The RBs produced during spermatogenesis in L4 stage him-5 gonads were highlighted by CED-1::GFP, whereas significantly more were highlighted in the amg-1;him-5 mutant (Fig. 4D). In the adult stage, him-5 spermatogenesis was replaced by oogenesis, whereas the amg-1;him-5 mutant gonads were distributed with aberrant sperm and very few oocytes, and the CED-1::GFP fluorescence signal was significantly stronger than the control (Fig. 4D). These data are consistent with the in vivo data (Fig. 4A,C and Movies 9,10), indicating that numerous phagocytic clearances might have occurred in the gonads of amg-1;him-5 mutant hermaphrodites.

We used transmission electron microscopy to reveal the ultrastructure of apoptotic cells in the proximal male gonad (Fig. 4E). The him-5 gonads had a compact distribution of sperm, whereas the sperm distribution in amg-1;him-5 gonads was sparse, and they mostly contained malformed mitochondria and MOs (Fig. 4E). We also observed the presence of apoptotic corpses encapsulated by engulfing cells (Fig. 4E, red arrow). CPS-6, the first mitochondrial protein identified to be involved in apoptosis in C. elegans (Parrish et al., 2001; Li et al., 2001) and apoptosis-induced DNA degradation further activate SCRM-1 to initiate externalization of phosphatidylserine on the cell membrane (Wang et al., 2007; Hsu and Wu, 2010). We collected gonads from L4 stage hermaphrodites for RNA-seq analysis. We found that the amg-1 mutation resulted in significant upregulation of cps-6 and scrm-1 transcription (Fig. 4F,G), suggesting that apoptosis might be triggered by mitochondrial defects. These data suggest that the AMG-1 mutation triggers a severe mitochondria-specific stress response.

Our previous studies revealed that AMG-1 binds most frequently to mtDNA-encoded 12S and 16S ribosomal RNAs and that AMG-1-mediated expression of 12S rRNA is crucial for germline mitochondrial proteostasis (Wang et al., 2023a). Based on new studies, our results indicate a previously undiscovered role for AMG-1 in regulating C. elegans spermatogenesis. AMG-1 is highly homologous to the LRPPRC protein, although it has evolved differently from mammals, and they share a similar PPR structural domain and a conserved amino acid site. amg-1 mutation causes defects in mitochondrial structure, resulting in delayed spermatogenesis in the proximal gonad, which in turn triggers a severe mitochondria-specific stress response, ultimately leading to the sterility of hermaphrodites. Therefore, AMG-1 protein is extremely important for reproduction and is a good model for studying Leigh syndrome.

The C. elegans strains were cultured at 20°C in a nematode growth medium with Escherichia coli OP50 (Brenner, 1974). All strains were obtained from the wild-type (WT) isolate Bristol N2. To ensure an adequate supply of males, strains carrying the him-5 (e1490) mutation were used as the WT (Hodgkin, 1983). The zcls13 strain [hsp-6p::gfp] was provided by Tian Ye (Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, China). The other strains were obtained from the Caenorhabditis Genetics Center (see Table S1 for strains and all key resources). If the amg-1 mutant allele is not indicated, it defaults to the amg-1(ibp80) mutant.

CRISPR-Cas9 genome editing was performed using plasmid pDD162 (Addgene plasmid #47549), which inserts the target sequence, and gRNA of the target sequence was designed. The pDD162 backbone was PCR amplified using Q5 high-fidelity DNA polymerase (New England Biolabs) before inserting the target sequence. The forward primer is 5′-N17GTTTTAGAGCTAGAAATAGC-3′, and the reverse primer is 5′-N17CAAGACATCTCGCAATAGGA-3′. N17 in the forward primer is homologous to the 17 nt 3′ end of the target sequence, while N17 in the reverse primer is homologous to the reverse complementary sequence at the 17 nt end of the target sequence. The PCR products were digested with DpnI (New England Biolabs) at 37°C for 2 h and then transformed into DH5α competent cells. Sequencing was used to determine whether the gRNA was successfully inserted into the plasmid backbone. After determining the concentration of the purified plasmid, microinjection was performed into the gonads of C. elegans.

To analyze the fertility of hermaphrodites, we selected L4 stage hermaphrodites and placed each worm on a plate. The hermaphrodites were transferred to a new plate every day. The progeny produced were counted until no eggs were laid.

Sperm were stripped from male worms (virgin males 72 h after the L4 stage) and placed in a drop of sperm medium (SM buffer: 50 mM HEPES, 50 mM NaCl, 25 mM KCl, 1 mM MgSO₄, 5 mM CaCl₂ and 10 mg/ml polyvinylpyrrolidone). Images were obtained using an Olympus FV1200 confocal microscope equipped with a 60×/1.35 NA oil immersion objective to examine gonads and sperm, or a 20×/0.75 NA objective (Olympus) to examine worms.

C. elegans were picked into SM buffer on a plastic Thermanox coverslip coated with poly-lysine to release the sperm or gonads 48 h after the L4 stage. A 100 µl aliquot of SM buffer containing 2.5% glutaraldehyde was added to fix the sperm or gonads. The tissues were fixed in 1% osmium tetroxide on ice for 2 h, stained with 1% uranyl acetate at 4°C overnight, dehydrated through an ascending ethanol gradient, followed by a transition with acetone. The samples were embedded in a flat mold with fresh 812 resin and polymerized in an oven at 60°C for 48 h. The coverslips were peeled off, leaving the sperm on the surface of the resin block. Ultra-thin sections (80 nm) were cut from the new blocks on a Leica EM UC7 ultramicrotome and observed on a FEI Tecnai Spirit 120 electron microscope at 100 kV. See Table S1 for resources used.

A total of 100 L4 stage hermaphrodite germlines were collected, washed three times in PBS, resuspended in 1 ml of TRIzol reagent, and stored at −80°C. The germlines were used for the RNA-seq analysis following the Smart-seq2 protocol (Picelli et al., 2014). See Table S1 for primers used in RNA-seq analysis.

We are grateful to Dr Fanxia Meng for critically reading the manuscript. We appreciate Drs Suhong Xu (Zhejiang University), Xiaochen Wang (Institute of Biophysics, Chinese Academy of Sciences) and Ye Tian (Institute of Genetics and Developmental Biology, Chinese Academy of Sciences) for providing strains or plasmids. Some strains were obtained from the Caenorhabditis Genetics Center, which is funded by the National Institutes of Health Office of Research Infrastructure Programs (P40 OD010440). We thank the staff members at the Center for Biological Imaging in the Institute of Biophysics, Chinese Academy of Sciences, for data collection.

The peer review history is available online at https://journals.biologists.com/dev/lookup/doi/10.1242/dev.202165.reviewer-comments.pdf