Mitochondrial DNA variant in COX1 subunit significantly alters energy metabolism of geographically divergent wild isolates in Caenorhabditis elegans

Abstract

Mitochondrial DNA (mtDNA) sequence variation can influence the penetrance of complex diseases and climatic adaptation. While studies in geographically defined human populations suggest that mtDNA mutations become fixed when they have conferred metabolic capabilities optimally suited for a specific environment, it has been challenging to definitively assign adaptive functions to specific mtDNA sequence variants in mammals. We investigated whether mtDNA genome variation functionally influences Caenorhabditis elegans wild isolates of distinct mtDNA lineages and geographic origins. We found that, relative to N2 (England) wild-type nematodes, CB4856 wild isolates from a warmer native climate (Hawaii) had a unique p.A12S amino acid substitution in the mtDNA-encoded COX1 core catalytic subunit of mitochondrial complex IV (CIV). Relative to N2, CB4856 worms grown at 20°C had significantly increased CIV enzyme activity, mitochondrial matrix oxidant burden, and sensitivity to oxidative stress but had significantly reduced lifespan and mitochondrial membrane potential. Interestingly, mitochondrial membrane potential was significantly increased in CB4856 grown at its native temperature of 25°C. A transmitochondrial cybrid worm strain, chpIR (M, CB4856>N2), was bred as homoplasmic for the CB4856 mtDNA genome in the N2 nuclear background. The cybrid strain also displayed significantly increased CIV activity, demonstrating that this difference results from the mtDNA-encoded p.A12S variant. However, chpIR (M, CB4856>N2) worms had significantly reduced median and maximal lifespan relative to CB4856, which may relate to their nuclear-mtDNA genome mismatch. Overall, these data suggest that C. elegans wild isolates of varying geographic origins may adapt to environmental challenges through mtDNA variation to modulate critical aspects of mitochondrial energy metabolism.

Keywords: CB4856; N2; adaptation; bioenergetics; mitochondria.

Fig. 1

C. elegans mtDNA genomes of N2 and CB4856 isolates differ by only a single non-synonymous coding variant, which alters COX1 protein conformation. (a) Mitochondrial genomes of N2 and CB4856 animals were manually resequenced and compared. Red and blue lines represent extent of CB4856 mtDNA genome sequence determined in this study and previously deposited in NCBI [8], respectively. Synonymous and non-synonymous SNVs in CB4856 relative to N2 are indicated by black dashes and black stars, respectively. Primer pair names and locations are indicated within the circle. (b) The SNV in CB4856 (right panel) results in the replacement of a hydrophobic alanine with a hydrophilic serine, which appears by protein modeling to alter the N-terminal loop structure conformation through hydrogen bonding with the COX1 protein core.

Fig. 2

in vitro analyses of isolated mitochondrial CIV activity and protein levels, in N2 and CB4856 isolates. (a) COX specific activity was determined in the absence of TMPD and dodecyl-β-maltoside by increasing the concentration of substrate cytochrome c (note that higher concentrations of exogenously added cow heart cytochrome c are required to reach maximal turnover due to the imperfect fit with C. elegans COX). K_m values of COX for cytochrome c are 9.4 μM [chpIR (M, CB4856 > N2)], 10.6 μM (CB4856), and 8.3 μM (N2). Maximal catalytic rates were 73, 154, and 108 nmol O₂/min/mg total mitochondrial protein for N2, chpIR (M, CB4856 > N2), and CB4856, respectively. Experiments were performed in triplicates and each sample was measured three times (error bars indicate standard deviation). (b) The activity of CIV in the respiratory chain supercomplexes was similar between N2 and CB4856. S1–S4 represents the amount of CIV associated with CI and CIII, where S1: CI, CIII, CIV; S2: CI, CIII, CIV₂; and so on. (c) The total amount of mitochondrial protein associated with each supercomplex was also similar between these two wild isolates. Molecular mass is shown on the y-axis in kilodaltons and V2 represents the CV dimer. (d) Western blot analysis showed similar COX1 content in isolated mitochondria from both wild isolates (molecular mass is indicated in kilodaltons).

Fig. 3

Integrated respiratory capacity in isolated mitochondria is similar between N2 and CB4856 isolates but showed a reduced trend in chpIR (M, CB4856 > N2) cybrid worms. Polarographic analysis of integrated respiratory capacity was performed with substrates that donate electrons at CI (malate), CII (succinate), or cytochrome c/CIV (TMPD/ascorbate). (a) Similar state 3 (near-maximal) and (b) state 4 (ADP depleted) respiratory capacity rates were seen with all substrates tested in CB4856 relative to N2. Values represent mean and standard error of raw values from 10 biological replicates of N2, 3 of CB4856, and 5 of chpIR (M, CB4856 > N2). (c) The relative coupling efficiency as assessed by calculating the state 3/state 4 respiratory control ratio (RCR) was statistically similar between N2 and CB4856, although it trended toward decreased for CI- and CII-based respiration. CI-based RCR was significantly increased in chpIR (M, CB4856 > N2) relative to CB4856. (d) No significant difference in the ratio of ADP/O was observed between the three C. elegans strains.

Fig. 4

Significant strain differences were observed in in vivo assays of relative mitochondrial membrane potential and mitochondria content. (a) Mitochondrial membrane potential (ΔΨ_m) was studied in both wild strains at three different temperatures, with peak ΔΨ_m occurring in both strains at 15 °C. Interestingly, while N2 (black bars) had higher ΔΨ_m than CB4856 (gray bars) at 15 °C and 20 °C, CB4856 had relatively increased ΔΨ_m at 25 °C. (b) No significant difference was seen in mitochondria content between CB4856 and N2, as assessed at 20 °C by MitoTracker Green FM mean fluorescence intensity. (c) chpIR (M, CB4856 > N2) worms had increased mitochondrial content and decreased membrane potential relative to N2. Statistical analyses were performed by student's t-test, where *p < 0.05, **p < 0.01, and ***p < 0.001. For the in vivo fluorescence experiments, total number of animals studied per strain and condition are indicated within bars. Bars and error bars represent mean and standard error of the mean, respectively.

Fig. 5

mtDNA content was reduced in CB4856 relative to N2 worms at both early larval and young adult stages. qPCR analysis of mtDNA-encoded (nd4) and nDNA-encoded (drs-1) genes showed that relative mtDNA depletion occurred in CB4856 worms relative to N2. Results were similar in both L1 and young adult stage comparisons, which suggested that the difference in mtDNA content cannot be attributed to differences in oocyte content in adult stage worms (from two replicate experiments).

Fig. 6

CB4856 demonstrated an increased matrix oxidant burden and increased sensitivity to oxidative stress relative to N2. (a) Significantly increased oxidant burden occurs in CB4856 relative to N2, as indicated by 31.3% increase in terminal PB mean MitoSOX fluorescence intensity. Bars and error bars indicate mean and standard error of mean, respectively. Statistical analyses were performed by student's t-test, where *p < 0.05, **p < 0.01, and ***p < 0.001. (b) Baseline levels of H₂O₂ were approximately the same in isolated mitochondria from the two wild isolates. However, CB4856 mitochondria were more sensitive to oxidative stress, evidenced by a 62% increase in the level of hydrogen peroxide measured in the presence of antimycin A (AA). Bars indicate mean and standard error of four biological replicates per strain. M, malate. S, succinate.

Fig. 7

Lifespan analysis of C. elegans wild strains at 20 °C. (a) CB4856 has significantly shorter median and maximal lifespan relative to N2 at 20 °C (p < 0.0001). (b) Repeat analysis of CB4856 relative to N2 at 20 °C confirmed the strain's decreased lifespan (p < 0.0001) and demonstrated that chpIR (M, CB4856 > N2) cybrid worms have an even further decreased lifespan relative to N2 (p < 0.0001).