The mitochondrial trans-2-enoyl-CoA reductase is necessary for mitochondrial homeostasis in C. elegans

Abstract

Fatty acids function not only as signaling molecules and for energy storage but also as essential cofactors for mitochondrial enzymes. These fatty acid cofactors are produced by the mitochondrial fatty acid synthesis (mtFAS) pathway, the terminal enzyme of which is mitochondrial trans-2-enoyl-CoA reductase (MECR). Dysfunction of MECR prevents the synthesis of fatty acids and is the monogenic cause of Mitochondrial Enoyl-CoA Protein Associated Neurodegeneration (MEPAN) syndrome, a rare mitochondrial disease characterized by dystonia, basal ganglia degeneration, and optic nerve atrophy. Given the necessity of mtFAS products for mitochondrial function, MECR should be essential. Yet, evidence from MEPAN individuals and model organisms with MECR loss of function indicate that mitochondrial function is not as severely impaired as would be expected. However, many of these studies have been limited to single cells or cell types. To better understand the role of MECR and its products in a multicellular system, we used CRISPR/Cas9 to knock out its 2 orthologs in Caenorhabditis elegans, MECR-1 and MECR-2. We found that only MECR-1 is necessary for normal mitochondrial function, germline development, and neuromuscular function. We thus establish a model in which further studies of MECR/MECR-1 can clarify its biochemical, developmental, and physiological roles.

Introduction

Many mitochondrial enzymes require cofactors for their function, but lesser known than metallic cofactors (e.g. heme and iron–sulfur clusters) are the fatty acid cofactors whose importance is becoming increasingly appreciated (Kursu et al. 2013; Lambrechts et al. 2019; Kastaniotis et al. 2020; Wedan et al. 2024). Fatty acid cofactors derive from mitochondrial fatty acid synthesis (mtFAS), a cyclical pathway consisting of 4 enzymes and the acyl carrier protein (ACP) producing fatty acids with lengths from 6 to 16 carbons on ACP as a scaffold (Fig. 1a) (Hiltunen et al. 2009; Kursu et al. 2013; Venkatesan et al. 2014; Nowinski et al. 2018). The canonical product of this pathway is the eight-carbon (8C) octanoyl-ACP, itself the precursor of lipoic acid (LA). LA is an essential cofactor for the 4 oxoacid dehydrogenases (e.g. pyruvate dehydrogenase and alpha-ketoglutarate dehydrogenase) as well as the H-protein of the glycine cleavage system: all essential enzymes for catabolic metabolism (Feng et al. 2009; Ahting et al. 2015; Cronan 2024). Longer-chain mtFAS products, especially palmitoyl-ACP (14C) and myristoyl-ACP (16C), are important for the function of the electron transport chain (ETC) and iron-sulfur cluster biogenesis, although their exact role remains unclear (Feng et al. 2009; Kursu et al. 2013; Heimer et al. 2016; Vranken et al. 2018; Wehbe et al. 2019; Dutta et al. 2023; Rahman et al. 2023; Meisel et al. 2024). In fact, many functions of mtFAS, its regulation, and the downstream products remain enigmatic.

Fig. 1.

MECR orthology and necessity. a) Schematic of mtFAS with key products and downstream processes. b) Clustal Omega alignment of human MECR with C. elegans  MECR-1 and MECR-2. Conserved variant residues in MEPAN syndrome are boxed in red; the ATP- or NADH-binding sites are highlighted in green; the catalytic tyrosine is highlighted in blue; the internal (cytoplasmic) translation start site is marked with an arrow. The region deleted in mecr-1(av243) is underlined. c) AlphaFold structure predictions of human (Hs) MECR and C. elegans (Ce) MECR-1/2. d) Representative complementation assay with yeast (Sc) ETR1 or C. elegans mecr-1/2 on rich (YPDex) and glycerol (YPGly) medium. Cell density indicated above images. e) Kaplan–Meier survival curves of indicated mutants from the L4 stage (n = 59 to 172 per genotype; N = 2 to 3 independent assays). ****P ≤ 0.0001 by log-rank analysis with Bonferroni correction for multiple comparisons. ACP, acyl carrier protein; TCA, tricarboxylic acid; ISC, iron–sulfur cluster; OxPhos, oxidative phosphorylation.

Dysfunction of mtFAS is associated with rare mitochondrial diseases, the most common of which is Mitochondrial Enoyl-CoA Protein Associated Neurodegeneration (MEPAN) syndrome (Wedan et al. 2024). MEPAN syndrome is autosomal recessive and is caused by lost or reduced function of the mitochondrial trans-2-enoyl-CoA reductase (MECR) protein (Heimer et al. 2016). To date, over 30 cases of MEPAN syndrome have been diagnosed (Heimer et al. 2016; Gorukmez et al. 2019; Liu et al. 2021; Fiorini et al. 2023; Jia et al. 2024). Although some clinical features of MEPAN syndrome are consistent (e.g. dystonia, optic nerve atrophy, and basal ganglia abnormalities), the age of onset and severity of these phenotypes varies significantly between individuals. Genetic heterogeneity may explain some of this variability as many patients are compound heterozygous for variant MECR alleles, but an incomplete understanding of mtFAS and its products contributes to the poor understanding of this variability (Heimer et al. 2016; Gorukmez et al. 2019; Fiorini et al. 2023; Jia et al. 2024). It remains possible that some retention of MECR function results in differential outcomes for mtFAS and mitochondrial function more broadly.

Despite recent interest in mtFAS and MECR, most studies have been limited to single cells or cell types (Heimer et al. 2016; Nair et al. 2018; Kastaniotis et al. 2020; Nowinski et al. 2020; Dutta et al. 2023). To better understand the role of mtFAS and MECR in an intact, multicellular organism, we knocked out its orthologs (mecr-1 and mecr-2) in Caenorhabditis elegans. This system offers a multicellular model that is isogenic outside of the targeted mutations and one in which we can control for environmental and dietary effects. We found that mecr-1, but not mecr-2, is necessary for proper mitochondrial function, fertility, and motility. This novel model thus presents an opportunity to investigate complex roles of MECR/MECR-1 in the development and homeostasis of multicellular organisms, which will help with understanding MEPAN syndrome.

Results

MECR orthologs in C. elegans

The C. elegans ortholog of human MECR was previously identified as mecr-1 (Gurvitz 2009); however, a BLAST search of the human MECR amino acid sequence also returned the ORF Y48A6B.9, which we have named mecr-2. Sequence alignments reveal high conservation of key catalytic residues and those mutated in MEPAN syndrome in both MECR-1 and MECR-2 (Fig. 1b). Although MECR has 47% identity to MECR-1 and 42% identity to MECR-2, the 2 C. elegans paralogs only have 39% identity to each other (Fig. 1b). Yet, the predicted structure of all 3 proteins is similar (Fig. 1c). Notably, the internal methionine necessary for a cytoplasmic isoform of MECR in humans (Kim et al. 2014; Baker et al. 2022) is not conserved in either C. elegans ortholog (Fig. 1b), so all studies of C. elegans  MECR-1/MECR-2 are restricted to its mitochondrial function without potentially confounding noncanonical (cytoplasmic) roles.

To determine if mecr-2 is an ortholog of human MECR, we used a complementation assay in Saccharomyces cerevisiae whose MECR ortholog is ETR1 (Torkko et al. 2001; Miinalainen et al. 2003). S. cerevisiae carrying a deletion of ETR1 (ert1Δ) are viable on rich medium containing glucose but are inviable on nonfermentable carbon sources such as glycerol (Torkko et al. 2001). We expressed ETR1, mecr-1, or mecr-2 in etr1Δ yeast and tested their viability on glycerol-containing medium. As expected, plasmid-based expression of ETR1 restored growth of etr1Δ cells on glycerol as did expression of mecr-1; however, mecr-2 did not complement loss of ETR1 (Fig. 1d), suggesting that MECR-2 is not functionally redundant with MECR-1. However, we were unable to determine whether MECR-2 protein was synthesized or where it localized under these conditions, limiting our interpretation of these results.

To determine the necessity of both mecr-1 and mecr-2 in C. elegans, we used CRISPR-Cas9 gene editing to make independent deletion alleles of both genes as well as a double mutant. Deletion sequences were confirmed by Sanger sequencing and PCR (Supplementary Fig. 1). Deletion of either the entirety of mecr-1 (mecr-1(av238); Supplementary Fig. 1a) or only the C-terminus (mecr-1(av243); Fig. 1b; Supplementary Fig. 1a) resulted in both shortened lifespans (Fig. 1e; Table 1) and complete sterility (Table 2). Contrastingly, the 2 complete deletions of mecr-2 (knu1163 and knu1167) had normal lifespans (Fig. 1e; Table 1) and fertility (Table 2; Supplementary Fig. 2a to c). A double mutant (mecr-1(av238); mecr-2(knu1163)) had identical lifespan and fertility to the mecr-1(av238) single mutant, indicating that mecr-2 is not necessary in C. elegans and does not compensate for the loss of MECR-1 function (Fig. 1e; Tables 1 and 2). We chose to focus on the mecr-1(av238) allele to investigate the necessity of this enzyme in C. elegans development and homeostasis.

Table 1.

Summary of lifespan data from Fig. 1e.

Genotype	Median lifespan (days)	Sample size	Replicate assays
WT	12	172	3
mecr-1(av238)	9	82	2
mecr-1(av243)	9	95	2
WT	12	119	3
mecr-2(knu1163)	12	90	2
mecr-2(knu1167)	12	96	2
mecr-1(av238); mecr-2(knu1163)	9	59	2

Table 2.

Fertility of mutant strains at each indicated temperature.

16 °C
	# assessed	# fertile	% fertility	Replicate assays
WT	260	260	100	2
mecr-1(av238)	203	0	0	2
mecr-1(av243)	294	0	0	3
mecr-2(knu1163)	362	358	98.9	2
mecr-1(av238); mecr-2(knu1163)	348	0	0	3

20 °C
	# assessed	# fertile	% fertility
WT	275	275	100	3
mecr-1(av238)	247	0	0	3
mecr-1(av243)	226	0	0	2
mecr-2(knu1163)	301	300	99.67	2
mecr-1(av238); mecr-2(knu1163)	247	0	0	2

24 °C
	# assessed	# fertile	% fertility
WT	278	278	100	2
mecr-1(av238)	201	0	0	2
mecr-1(av243)	254	0	0	2
mecr-2(knu1163)	360	360	100	2
mecr-1(av238); mecr-2(knu1163)	264	0	0	2

mecr-1 loss impairs C. elegans fertility

Although mecr-1(av238) animals had grossly normal development (Supplementary Fig. 3a to c), 100% of homozygotes were sterile (Table 2). Shifting the culture temperature to 16 °C or 24 °C had no effect on the fertility of any analyzed strain (Table 2). Sterility appears to be caused by an absence of oocytes despite the presence of sperm (Supplementary Fig. 3d), a phenotype currently under further investigation. These data indicate that mecr-1 is necessary for germline development and function in C. elegans.

Mitochondrial function is impaired in mecr-1(av238)

Since MECR function is important for mitochondrial respiration in other organisms (Torkko et al. 2001; Nowinski et al. 2020; Dutta et al. 2023), we used Seahorse respirometry to measure the oxygen consumption rate (OCR) of WT and mecr-1(av238) animals. OCR is largely a measure of ETC activity, but other factors do contribute. Despite attempts to use day 1 adult (D1A) animals for this assay, they were unresponsive to stimuli, and we thus used L4 animals. Surprisingly, basal OCR was elevated in mecr-1(av238) compared to WT (Fig. 2, a and b). The ATP-linked OCR [the difference between basal OCR and the OCR following blockage of complex V (CV)/ATP synthase with dicyclohexylcarbodiimide (DCCD)] was not different (Fig. 2, a and c); however, the reduction in OCR following blockage of CV/ATP synthase was not as robust as expected. WT samples had only a 14% reduction in OCR, and mecr-1(av238) had only a 17% in OCR. This point is further discussed below. Finally, there was only a very modest decrease in the maximal OCR of mecr-1(av238) (Fig. 2, a and d). As expected given the basal OCR finding, the spare respiratory capacity, the ability to increase oxygen consumption from basal to maximal, was decreased in mecr-1(av238) (Fig. 2e). Collectively, these findings indicate that the loss of MECR-1 function impairs electron transport efficiency.

Fig. 2.

Mitochondrial homeostasis is disrupted in mecr-1(av238). a) Representative OCR traces from Seahorse respirometry (n = 74 to 75 per genotype). b to e) Normalized and averaged values for basal OCR b), ATP-linked OCR c), maximal OCR d), and spare respiratory capacity e). Each point represents a biological replicate. f) Representative images of hypodermal mitochondria and ΔΨ_m indicated with TMRE. g to i) Quantification of f) with the 488 nm [mitochondrial, g)], 561 nm [TMRE, h)] channels' signal intensity, and ratio of the 2 i). Each point represents a biological replicate (n = 23 to 24 per genotype). j) Relative mtDNA abundance. Each point represents a biological replicate (n = 4 per genotype). k) Relative mRNA expression of ETC components. gas-1 (CI), mev-1 (CII), clk-1 (ubiquinone), isp-1 (CIII), cox-4 (CIV), and asg-2 (CV/ATP synthase). Each point represents a biological replicate (n = 4 per genotype). (l) Representative western blot for LA. TBA-1: loading control. Molecular weights (in kD) shown on the left. E2 subunits (E2s) of the oxoacid dehydrogenases labeled on the right. Solid arrow heads indicate unidentified proteins. Open arrow head indicates nonspecific band. For a), *P ≤ 0.05, **P ≤ 0.01, and ****P ≤ 0.0001 by 2-way ANOVA. For b to e, g, h, j, and k), **P ≤ 0.01 and ****P ≤ 0.0001 by 2-tailed Student's t-test. DCCD, dicyclohexylcarbodiimide; FCCP, carbonyl cyanide-4-phenylhydrazone; NaN₃, sodium azide; OCR, oxygen consumption rate; ΔΨ_m, mitochondrial membrane potential; TMRE, tetramethylrhodamine, ethyl ester; ETC, electron transport chain; E2s, E2 subunits of oxoacid dehydrogenases; kD, kilodaltons.

To determine the effect of mecr-1 loss on mitochondria more directly, we generated a mecr-1(av238) strain with GFP-labeled mitochondria (Raiders et al. 2018) and used this strain to analyze both mitochondrial mass and mitochondrial membrane potential (ΔΨ_m). Mitochondrial mass (the pixel intensity of the GFP signal) was elevated in mecr-1(av238) (Fig. 2, f and g) suggesting a compensatory increase in mitochondria, which was corroborated by an increase in mitochondrial DNA (mtDNA) (Fig. 2j). Similarly, gene expression analysis of representative components of the ETC demonstrated increased expression of gas-1 (CI) and asg-2 (CV/ATP synthase) (Fig. 2k). These data all suggest that mecr-1(av238) had less functional mitochondria, resulting in compensatory increases in the ETC and mitochondria.

To further examine the impact of the altered OCR and ETC expression on mitochondrial homeostasis, we exposed the GFP-labeled strains to tetramethylrhodamine, ethyl ester (TMRE), a dye that accumulates in a ΔΨ_m-dependent fashion. Although the TMRE signal itself was elevated in mecr-1(av238) (Fig. 2h), when normalized against the mitochondrial mass, there was no difference in TMRE signal between the 2 strains (Fig. 2i), suggesting that the increased basal OCR and ETC expression does not translate into altered ΔΨ_m, possibly due to increased proton or electron leak across the membrane.

The mtFAS product LA is essential for catabolic processes upstream of, and including, mitochondrial respiration, so we also examined the production of LA in mecr mutants. Canonically, the lipoylated E2 subunits of the oxoacid dehydrogenases appear as a doublet at ∼42 kD, which we confirmed in WT samples (Fig. 2l). In mecr-1(av238), LA is reduced, but not absent on the E2 subunits. Surprisingly, we also observed the appearance of novel lipoylated proteins in mecr-1(av238) at ∼25 and ∼8 kD. The identity of these proteins is under current investigation. Although mecr-2 was unable to compensate for mecr-1 in gross phenotypic analysis, we tested whether it was necessary for LA production. However, lipoylation of the E2 subunits is unaffected in mecr-2(knu1163), and LA is not abolished in the mecr-1/2 double mutant, so we reaffirm that mecr-2 is not necessary for mitochondrial function.

MECR-1 loss alters neuromuscular function

Many mitochondrial diseases, MEPAN syndrome included, are characterized by neuromuscular dysfunction. We thus analyzed the neuromuscular function of mecr-1(av238) through analysis of swimming behavior with the WormLab analysis platform (Restif et al. 2014). Intriguingly, there were noticeable differences in mecr-1(av238) swimming behaviors at both L4 and D1A. Generally, swimming behaviors were unaffected or reduced at L4 and unaffected or increased at D1A. The most striking difference was in the swimming speed, which was reduced at L4 but increased at D1A (Fig. 3, b and h). Stretch, the depth of individual body bends during swimming, was increased at both ages (Fig. 3, e and k). At D1A, mecr-1(av238) also showed increased curling, but the rate of curling was clearly reduced from the L4 stage in both WT and mecr-1(av238) animals. These findings could suggest altered or delayed neuromuscular development, which results in subtle, but measurable, changes in swimming behavior.

Fig. 3.

Impaired motility in mecr-1(av238). a to f) Motility analyses at L4 (n = 44 to 48 per genotype). g to l) Motility analysis at D1A (n = 27 to 32 per genotype). a and g) Wave initiation rate, b and h) swimming speed, c and i) time reversing, d and j) time curling, e and k) stretch (depth of individual bends), and f and l) activity index. Each point represents an individual animal. **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001 by 2-tailed Student's t-test.

MECR-1 expression is highest in the gonad

To better understand how expression of MECR-1 affects the observed phenotypes, we sought to create a translational fusion of MECR-1 to a fluorescent protein. Multiple attempts to directly fuse a fluorescent protein to MECR-1 at either terminus, even with a flexible linker, failed to generate a construct that did not impair MECR-1 function. Therefore, we turned to including an SL2 trans-splicing sequence (Spieth et al. 1993) between the coding regions of mecr-1 and mCherry. This SL2 sequence allows mecr-1 and mCherry to be transcribed as an operon, but MECR-1 and mCherry are translationally separated, so we were able to use expression of mCherry for analysis of endogenous levels and locations of MECR-1.

Supporting the fertility data, MECR-1::SL2::mCherry expression is highest in the gonad of D1A animals. Expression appears universally in the germ cells themselves with noticeable foci of increased expression (Fig. 4). These foci include the spermatheca most clearly, although it is unclear if the expression is from sperm specifically or the somatic cells of the spermatheca. mCherry foci also correspond to the position of the distal tip cell and in sheath cells, the somatic cells surrounding the proximal gonad that provide the niche for differentiation of gametes. The high mCherry expression in both germ cells and the somatic gonad does not clarify if sterility is cell autonomous or not, but does suggest that it is caused by lost MECR-1 function in the gonad as a whole.

Fig. 4.

MECR-1 expression in adult C. elegans. Representative fluorescent images of the MECR-1::SL2::mCherry reporter at D1A. a) Stitched maximum intensity projections, each captured at 40× magnification. b) Annotated version of a) including markers for insets. c to e) Single 0.5-μm optical slices of indicated regions with phase contrast, mCherry, and merged channels. Arrows and labels indicate somatic tissues/cells with high MECR-1::SL2::mCherry expression. Spth, spermatheca.

MECR-1::SL2::mCherry was expressed in somatic tissues as well, with the highest expression in the intestine and head and tail neurons. The intestine is the sole digestive tissue of C. elegans, so expression of this important metabolic enzyme is not surprising. Foci of mCherry expression also appear throughout the head and tail, which is consistent with expression in neurons of the nerve ring and tail ganglia, respectively. Notably, mCherry expression was low in most muscle cells with the exception of the rear pharyngeal bulb. Therefore, a neuronal, not muscular, origin of the observed neuromuscular phenotypes seems likely.

Discussion

MECR orthology

We have confirmed that mecr-1 is the C. elegans ortholog of human MECR and identified mecr-2 (Y48A6B.9) as a putative paralog. However, mecr-2 fails to complement S. cerevisiae ETR1, mecr-2 null C. elegans are indistinguishable from WT, and a double mecr-1/2 knockout is no worse than the single mecr-1 knockout. We conclude that mecr-1 is the only functional ortholog of MECR in C. elegans.

MECR-1 has high homology to MECR including residues mutated in MEPAN syndrome. However, MECR-1 does not retain the internal translation start site following the mitochondrial localization sequence that is present in other eukaryotes (Kim et al. 2014; Baker et al. 2022). Therefore, MECR-1 should only be expressed in mitochondria of C. elegans and not the cytosol, making this model a faithful, multicellular model specifically for mitochondrial functions of MECR without complications from interactions with cytosolic pathways. To understand the absolute necessity of MECR-1, we have focused here on deletion models. However, many MEPAN cases are genetically compound heterozygous and carry missense alleles of MECR (Heimer et al. 2016; Gorukmez et al. 2019; Fiorini et al. 2023), so patient-specific variants of MEPAN syndrome should be included in future studies.

The viability of mecr-1 null C. elegans contrasts with other models such as mice and Drosophila where global nulls are embryonic lethal (Nair et al. 2018; Dutta et al. 2023). While other studies have instead focused on knockouts of MECR orthologs in single cell types, these approaches preclude the ability to investigate complex, intercellular phenotypes or tissue-specific compensation, which seems likely in C. elegans. Therefore, the C. elegans model developed here allows for investigation of these intercellular phenotypes and an opportunity to better understand the full spectrum of MECR/MECR-1 function in a multicellular organism.

MECR-1 impacts on fertility

Sterility is common among mitochondrial mutants with most completely failing to produce mature gametes (Rea and Johnson 2003; Rea et al. 2007). mecr-1 mutants (both av238 and av243) are intriguing because of the seemingly functional process of spermatogenesis but a near complete absence of oogenesis. High expression of MECR-1 in the germ cells suggests that this defect is cell autonomous, but MECR-1 expression in the somatic gonad prevents us from ruling out cell-nonautonomous effects on oogenesis. We hypothesize that sterility is due to specific dysfunction of mitochondria within germ cells as a consequence of the lack of fatty acid cofactors produced by mtFAS, but that cell-nonautonomous effects are possible.

MECR-1 necessity for mitochondrial function

Given the role of mtFAS products in mitochondrial function, MECR should be essential: Both the medium- and long-chain fatty acids are essential for respiratory processes (Clay et al. 2016; Heimer et al. 2016; Cronan 2020; Dutta et al. 2023; Rahman et al. 2023; Bick et al. 2024; Meisel et al. 2024). Yet, we found that LA synthesis was reduced, but not abolished, in mecr-1(av238). Additionally, novel lipoylated proteins are apparent in these samples, suggesting that a cryptic compensatory pathway may exist. This pathway is unlikely to rely on lipoate scavenging as the enzymes necessary for this process are not conserved in eukaryotes (Tort et al. 2016; Lavatelli et al. 2020; Bick et al. 2024; Cronan 2024). Regardless of the mechanism, partial retention of LA in mecr-1(av238) could explain the relatively mild mitochondrial dysfunction, including the mildly altered OCR and lack of significant change in ΔΨ_m.

There are multiple possibilities, compensatory or otherwise, that may explain the increased basal OCR. First, there could be a greater number of (but less functional) mitochondria. The elevated mitochondrial mass, increased mtDNA, and increased expression of ETC subunits support this possibility. Second, MECR is known to alter the lipid profiles of mitochondria (Clay et al. 2016; Dutta et al. 2023), and uncoupled respiration, the leak of electrons and/or protons across the inner mitochondrial membrane as a consequence of an altered lipid profile, could provide a possible explanation for increased OCR (Kanazawa et al. 2008). A mechanistic explanation remains elusive, but future work will investigate these possibilities.

In addition to the elevated basal OCR, we were surprised that blockage of CV/ATP synthase with DCCD did not have a robust effect on OCR. Treatment with DCCD typically results in a dramatic drop in OCR, and we had previously determined that 20-μM DCCD was effective to induce such an effect (Kropp et al. 2021). However, this concentration is lower than what other studies have utilized (Mello et al. 2024), and we may have obtained different results using a higher concentration of DCCD.

The fact that mitochondrial, and general organismal, dysfunction is less severe in mecr-1(av238) than expected suggests that there are likely compensatory mechanisms to bypass the loss of mecr-1. We have discussed some of the mitochondria-specific possibilities above, but it does not escape our notice that these compensatory mechanisms may be tissue-specific, which could also explain some of the discrepancies between organism-level and tissue-level analyses. One intriguing possibility not yet discussed, although one for which we have no direct data, is the shunting of lipids from catabolic processes into mtFAS. There is no known enzyme that would be capable of performing such a compensatory role, but lipids are known to be mobilized in response to mitochondrial stress (recently reviewed in Smolková and Gotvaldová 2025). We hope to explore these, and other, compensatory mechanisms in future work.

Neuromuscular impairments in mecr-1(av238)

Dystonia is common among MEPAN individuals, likely due to basal ganglia and motor neuron degeneration (Heimer et al. 2016; Dutta et al. 2023; Fiorini et al. 2023; Jia et al. 2024). Therefore, we investigated the neuromuscular function of mecr-1(av238) in both L4 and D1A animals since the presence of fertilized embryos in fertile adult animals is known to alter movement (Restif et al. 2014). The absence of these embryos in D1A mecr-1(av238) likely explains the increased swimming speed and stretch observed in these animals. Low expression of MECR-1 in body wall muscles suggests that swimming phenotypes are unlikely to be due to dysfunction in the body wall muscles. Rather, high MECR-1 expression in the head/tail neurons indicates a role for MECR-1 in neurons, motor neurons included. Although MECR-1 expression was not obvious in the nerve cords, future analysis of the necessity of MECR-1 in neurons should be carried out.

Collectively, we report here a new model system for investigating the role of MECR/MECR-1 and the mtFAS pathway more broadly. We find that the C. elegans mecr-1(av238) model faithfully replicates some of the known mitochondrial phenotypes observed in other studies of MECR and its orthologs while also presenting an exciting and new system to study previously unknown or unappreciated roles for MECR/MECR-1. This model will be useful for studying mitochondrial development and function in multicellular organisms and disease models moving forward.

Materials and methods

Specific reagents including catalog numbers are listed in Supplementary Table 1.

C. elegans maintenance

C. elegans strains were maintained on Escherichia (E.)coli  OP50 on MYOB (Modified Youngren's, Only Bacto-peptone) plates at 20 °C following standard procedures (Brenner 1974). Strains used in this study are reported in Table 3. Sterile mecr-1 strains were balanced with the mnC1 balancer chromosome. Synchronous populations were obtained via timed egg lay or sodium hypochlorite synchronization.

Table 3.

Strains used in this study.

Strain	Genotype	Origin
N2	WT	CGC
AG600	mecr-1(av238 [deletion])/mnC1 [dpy-10(e128) unc-52(e444) nIs190] II	This study
AG601	mecr-1 (av243 [C-terminal deletion])/mnC1 [dpy-10(e128) unc-52(e444) nIs190] II	This study
COP2547	mecr-2(knu1163 [deletion]) III	This study
COP2549	mecr-2(knu1165 [deletion]) III	This study
PAK1	mecr-1(av238)/mnC1 [dpy-10(e128) unc-52(e444) nIs190] II; mecr-2(knu1163) III	This study
JJ2586	cox-4(zu476 [cox-4::eGFP::3xFLAG]) I	(Raiders et al. 2018)
AG629	mecr-1(av238)/mnC1 [dpy-10(e128) unc-52(e444) nIs190] II; cox-4(zu476[cox-4::eGFP::3xFLAG]) I	This study
PHX9228	mecr-1(syb9228[mecr-1::SL2::mCherry]) II	This study

Generation of mutant alleles

The Bristol N2 strain was used as the WT parental strain for all CRISPR/Cas9 gene editing. The mecr-1 deletion alleles were generated with 2 (5′ and 3′) site-specific 20-nucleotide guide sequences that were manually designed and produced by Horizon Discovery (PerkinElmer) who also provided the tracrRNA. Repair template design followed established protocols (Paix et al. 2015, 2017). Injection mixes, including a dpy-10 co-CRISPR marker, were prepared following established protocols (Paix et al. 2015, 2016; Kropp et al. 2021). Guide RNA and repair oligo sequences are reported in Table 4. The mecr-2 deletion alleles were generated by InVivo Biosystems. The mecr-1::SL2::mCherry reporter allele was generated by SunyBiotech. All variant sequences were confirmed by Sanger sequencing (Eurofins Genomics).

Table 4.

Guide and repair sequences for mecr-1 mutant strains.

Oligo	Sequence (5′-3′)
mecr-1 5′ guide	GACUGAGAACUUUUAACAUC
mecr-1 3′ guide	GGAAUAACUCGCUCCGUAAU
mecr-1 deletion repair	TAAGAAAATCTGTGTAAACTGTGCGAATTCCGGATAATCGGGTATCATTGCTACAATTTAATGTGATGTG
dpy-10 guide	GCUACCAUAGGCACCACCACGAG
dpy-10 repair	CACTTGAACTTCAATACGGCAAGATGAGAATGACTGGAAACCGTACCGCATGCGGTGCCTATGGTAGCGGAGCTTCACATGGCTTCAGACCAACAGCCTAT

Protein alignments and structure predictions

Protein sequence alignments were carried out with EMBL-EBI (European Molecular Biology Laboratory-European Bioinformatics Institute) Clustal Omega alignment tools. AlphaFold2 (Jumper et al. 2021) was used to generate the predicted structures of MECR-1 and MECR-2.

Complementation assay

cDNAs for mecr-1 and mecr-2 were generated from N2 total RNA samples, and S. cerevisiae ETR1 was generated from BY4741 total RNA using Invitrogen Super Script III. cDNAs were cloned into the p416-GPD expression vector, which were subsequently transformed into the BY4741 etr1::KanMX deletion strain (Horizon) via the lithium acetate method (Mumberg et al. 1995; Brachmann et al. 1998). Transformants were selected on synthetic complete (Sunrise Scientific) media lacking uracil (SC-Ura) containing 0.67% yeast nitrogen base and 2% dextrose. Solid media contained 2% agar. Transformants were grown overnight in liquid SC-Ura and then serially diluted and spotted onto SC-Ura and identical media that contained 2% glycerol instead of dextrose. Plates were incubated at 30 °C and imaged after 3 to 5 d.

Lifespan assays

Synchronous populations were generated via egglay. Assays began at the L4 stage. Animals were transferred to a fresh plate daily during the fertile period and then a fresh plate every 3 to 4 d thereafter. Animals were checked daily for viability by scoring for response to gentle touch with an eyelash. Dead animals were removed from plates, and animals accidently killed during transfer were censored from the analysis. Results from 2 to 3 independently completed assays were aggregated.

Brood size and fertility assays

Synchronous populations were generated via egglay. Assays began at the L4 stage, and fertile animals were transferred to a fresh plate daily for the entire fertile period. Progeny were scored 24 h after laying to determine whether or not they were viable. For fertility assays, synchronous populations were generated via egglay and assessed at the 72 h postegglay timepoint. Data from 2 to 3 independent assays were aggregated.

Oxygen consumption

Oxygen consumption was measured as in Kropp et al. (2021) and described in Supplementary Methods.

Microscopy

Detailed methods for microscopy are described in Supplementary Methods.

Western blotting

Western blotting was carried out as in Kropp et al. (2021) and described in Supplementary Methods.

Motility analyses

Swimming analyses were carried out as previously described (Kropp et al. 2023). In brief, synchronous populations of L4 or D1A animals were individually imaged in a drop of M9 on a microscope slide. Video analysis was accomplished with WormLab software (Restif et al. 2014).

Statistical analyses

Data were plotted and analyzed with GraphPad Prism version 10. Unless otherwise stated, data are represented as the mean ± standard deviation. Specific analysis, post hoc corrections, and sample sizes are reported in figure legends.

Data availability

Strains are available upon request. The authors affirm that all data necessary for confirming the conclusions of the article are present within the article, figures, and tables and Supplementary Data 1.

Supplemental material available at GENETICS online.

Funding

This work was supported, in part, by the Intramural Research Programs of the National Institute of Diabetes and Digestive and Kidney Diseases (A.G.) and the National Heart, Lung and Blood Institute (J.W.). This work was also funded by Kenyon College (P.A.K.).