Elevated mitochondrial DNA copy number found in ubiquinone-deficient clk-1 mutants is not rescued by ubiquinone precursor 2-4-dihydroxybenzoate

Abstract

Inside mitochondria reside semi-autonomous genomes, called mtDNA. mtDNA is multi-copy per cell and mtDNA copy number can vary from hundreds to thousands of copies per cell. The variability of mtDNA copy number between tissues, combined with the lack of variability of copy number within a tissue, suggest a homeostatic copy number regulation mechanism. Mutations in the gene encoding the Caenorhabditis elegans hydroxylase, CLK-1, result in elevated mtDNA. CLK-1's canonical role in ubiquinone biosynthesis results in clk-1 mutants lacking ubiquinone. Importantly, clk-1 mutants also exhibit slowed biological timing phenotypes (pharyngeal pumping, defecation, development) and an activated stress response (UPR^mt). These biological timing and stress phenotypes have been attributed to ubiquinone deficiency; however, it is unknown whether the mtDNA phenotype is also due to ubiquinone deficiency. To test this, in animals carrying the uncharacterized clk-1 (ok1247) mutant allele, we supplemented with an exogenous ubiquinone precursor 2-4-dihydroxybenzoate (DHB), which has previously been shown to restore ubiquinone biosynthesis. We measured phenotypes as a function of DHB across a log-scale range. Unlike the biological timing and stress phenotypes, the elevated mtDNA phenotype was not rescued. Since CLK-1's canonical role is in ubiquinone biosynthesis and DHB does not rescue mtDNA copy number, we infer CLK-1 has an additional function in homeostatic mtDNA copy number regulation.

Keywords: 2-4-Dihydroxybenzoate; CLK-1; COQ-7; Copy number; Mitochondrial DNA; Ubiquinone.

Figure 1.. The previously uncharacterized clk-1(ok1247) allele is a large deletion and does not produce protein.

A) Schematic of region containing C. elegans clk-1 gene and relevant alleles. Exons are numbered. Black arrow indicates location of point mutation; nucleotide substitution is indicated above the arrow. Dashed region indicates deletion. Allele colors in this figure correspond to allele colors in other figures within this manuscript: wild-type (grayscale); clk-1(e2519) red; clk-1(qm30) green; clk-1(ok1247) purple. Nucleotide positions of start and end of each exon are shown with numbering beginning at start of exon 1. B) PCR genotyping of clk-1(ok1247) mutation using previously published from the Caenorhabditis Genetics Center (CGC). Expected wild-type band is 2,648 base pairs, mutant band is 1,334 base pairs. First lane wild-type (N2), second lane mutant clk-1(ok1247), third lane no template control (ntc), final lane 1kb Plus DNA ladder. C) clk-1(ok1247) deletion breakpoint sequence as determined by Sanger sequencing. D) Due to the lack of a C. elegans-specific CLK-1 antibody, we generated a novel polyclonal antibody in rabbits. This antibody shows good specificity. Using this in-house generated antibody, we demonstrated the absence of CLK-1 protein in clk-1(ok1247). N2 and clk-1(ok1247) “undiluted” samples were diluted in “diluted” lanes. Mutant lanes at same dilution contain more total worm lysate than wild-type lanes. However, they appear to have less protein, which may be due to clk-1 mutant worms being smaller. E) Loading control gel visualization of total protein using TGX stain-free gel activation of the gel used for the blot in D.

Figure 2.. clk-1(ok1247) mutant copy number phenotype is reproducible across two primer pairs.

Relative mtDNA copy number measured by qPCR in individual worms and normalized to sample’s nuclear DNA copy number (actin). Wild-type (N2) gray circles (n=16), clk-1(ok1247) mutants purple triangles (n=11). Striped bars depict primer pair A, dotted bars depict primer pair B. Bars represent mean and SD. One-way ANOVA with two pairwise comparisons, primer pair A p= 0.0139, primer pair B p= 0.0080.

Figure 3.. Schematic representation of experimental design to determine which CLK-1 role is responsible for elevated mtDNA copy number.

Wild-type animals with CLK-1 protein and ubiquinone maintain typical mtDNA copy number (left). clk-1 null mutants do not produce ubiquinone and exhibit elevated mtDNA copy number (right). To determine if the clk-1 mutant elevated mtDNA copy number phenotype is caused by lack of ubiquinone or due to lack of CLK-1 protein, worms were treated with an exogenous ubiquinone precursor, 5-demethoxyubiquinone (DHB) (center). We reasoned that if the elevated mtDNA copy number in clk-1 worms is caused by lack of ubiquinone, copy number should be rescued upon DHB supplementation. If copy number is not rescued, we infer an additional role for CLK-1 in regulating mtDNA copy number. Solid lines indicate presence, dashed lines indicate absence.

Figure 4.. clk-1 mutant biological timing and stress phenotypes are rescued as a function of DHB treatment concentration.

A) Time to development as measured by counting and removing worms that have reached L4 or older at each given time point. For each genotype and treatment combination, three replicates were plated, and vertical lines represent the mean and standard error of the mean of those three plates. Wild-type (N2) shown as gray solid line, clk-1(ok1247) shown as purple dashed line. clk-1(ok1247) mutants exhibit slow growth on 0mM DHB but visibly hastened growth on 100mM DHB. Statistics were determined based on a multiple linear regression and Cox proportional hazards measurements. Hazard ratio: DHB dose 1.00, genotype 0.2233, DHB dose by genotype interaction 1.0070. B) Pharyngeal pump rate measured as pumps per minute. Wild-type (N2) shown as gray circles (n=20), clk-1(ok1247) shown as purple triangles (n=20). Bars represent mean. clk-1(ok1247) mutants on 0mM DHB exhibit slowed pump rate phenotype (one-way ANOVA comparison, p< 0.0001) but a smaller magnitude difference on 100mM DHB (one-way ANOVA comparison p= 0.0062). Linear regression: wild-type (slope= −0.0942) and clk-1 mutants (slope= 0.8600). Wild-type did not deviate from zero (p= 0.1988), while clk-1 mutants did deviate from zero (p< 0.0001). Slopes are significantly different from each other (p< 0.0001). C) Defecation cycle time measured as seconds between pBoc. Wild-type (N2) shown as gray circles (n=15), clk-1(ok1247) shown as purple triangles (n=15). Bars represent mean. clk-1(ok1247) mutants exhibit increased time between pBocs on 0mM DHB (one-way ANOVA comparison p< 0.0001), but normal time between defecations on 100mM DHB (one-way ANOVA comparison p= 0.8831). Linear regression: wild-type (slope = −0.03315) and clk-1 mutant (slope = −0.2990). Wild-type slightly deviated from zero (p= 0.0145), while clk-1 mutants strongly deviated from zero (p< 0.0001). Slopes are significantly different from each other (p< 0.0001). D) UPR^mt activation as measured by GFP fluorescence driven by the hsp-6 promoter. Wild-type (N2) shown as gray circles (n=6–9), clk-1(ok1247) shown as purple triangles (n=7–9). Bars represent mean. clk-1(ok1247) mutants exhibit UPR^mt activation on 0mM DHB (one-way ANOVA comparison p<0.0001) but no UPR^mt activation on 100mM DHB (one-way ANOVA comparison p=0.6548). Linear regression: wild-type (slope= −0.4936) and clk-1 mutants (slope= −8.423). Wild-type did not deviate from zero (p= 0.6035), while clk-1 mutants strongly deviated from zero (p= 0.0024). Slopes are significantly different from each other (p= 0.0114).

Figure 5.. clk-1(ok1247) mutant copy number is not rescued by DHB supplementation and appears to be caused by lack of CLK-1 protein, not lack of ubiquinone.

A) Relative mtDNA copy number measured by qPCR in individual worms and normalized to sample’s nuclear DNA copy number (actin). Wild-type (N2) depicted as gray circles (n=15–16), clk-1(ok1247) depicted as purple triangles (n=15–16). Bars represent mean. clk-1(ok1247) mutants exhibit elevated mtDNA copy number on 0mM DHB (one-way ANOVA comparison p= 0.0003), and on 100mM DHB (one-way ANOVA comparison p< 0.0001). Linear regression: wild-type animals (slope= 0.0003) and clk-1 mutant animals (slope= 0.0010). The slopes did not significantly differ from each other (p= 0.5469). Elevation of lines differ (p< 0.0001), indicating a strain difference in copy number. Primer pair A used for mtDNA quantitation. B) Mutants harboring the clk-1(ok1247) null allele, produce no CLK-1, and subsequently no ubiquinone. Worms treated with the exogenous ubiquinone precursor DHB exhibit rescue of other clk-1 mutant phenotypes, but do not exhibit rescue of the mutant elevated copy number phenotype (left). Thus, we infer that CLK-1 regulates mtDNA copy number through a ubiquinone-independent mechanism (right).