Sirt4 Modulates Oxidative Metabolism and Sensitivity to Rapamycin Through Species-Dependent Phenotypes in Drosophila mtDNA Haplotypes

Abstract

The endosymbiotic theory proposes that eukaryotes evolved from the symbiotic relationship between anaerobic (host) and aerobic prokaryotes. Through iterative genetic transfers, the mitochondrial and nuclear genomes coevolved, establishing the mitochondria as the hub of oxidative metabolism. To study this coevolution, we disrupt mitochondrial-nuclear epistatic interactions by using strains that have mitochondrial DNA (mtDNA) and nuclear DNA (nDNA) from evolutionarily divergent species. We undertake a multifaceted approach generating introgressed Drosophila strains containing D. simulans mtDNA and D. melanogaster nDNA with Sirtuin 4 (Sirt4)-knockouts. Sirt4 is a nuclear-encoded enzyme that functions, exclusively within the mitochondria, as a master regulator of oxidative metabolism. We exposed flies to the drug rapamycin in order to eliminate TOR signaling, thereby compromising the cytoplasmic crosstalk between the mitochondria and nucleus. Our results indicate that D. simulans and D. melanogaster mtDNA haplotypes display opposite Sirt4-mediated phenotypes in the regulation of whole-fly oxygen consumption. Moreover, our data reflect that the deletion of Sirt4 rescued the metabolic response to rapamycin among the introgressed strains. We propose that Sirt4 is a suitable candidate for studying the properties of mitochondrial-nuclear epistasis in modulating mitochondrial metabolism.

Keywords: Sirt4; TOR pathway; coevolution mtDNA/nDNA.

Figure 1

The effects of rapamycin on the whole-fly metabolic rate across Drosophila from divergent mitochondrial and nuclear lineages. (A) mtDNA phylogeny between the five mitochondrial haplotypes used in this experiment. The tree is grouped based on similarities in the composition of amino acid sequences. The red line demarcates evolutionarily divergent species. The name of the species is followed by the mitochondrial haplotype. In parentheses are the lines from which the mtDNA was isolated (modified and adapted from Montooth et al. 2010). Whole-fly oxygen consumption was measured in coevolved D. melanogaster mtDNA and nDNA: (B) OreR;OreR and (C) Zim⁵³;OreR, D. simulans mtDNA and nDNA: (D) C167.4;C167.4, and the introgressed lines harboring D. simulans mtDNA in D. melanogaster nDNA background: (E) sm21;OreR and (F) siI;OreR after treatment with rapamycin and starvation. For each strain identifier, the left side represents the mitochondrial haplotype, and the right side represents the nuclear background. For each pairwise comparison, the raw p-values, Holm-Bonferroni corrected p-values, and Hedges’ g effect sizes are presented in Table 1 and File S2. # = 0.0.5 < P < 0.1; * = 0.01 < P < 0.05; ** = 0.001 < P < 0.01; *** = P < 0.001.

Figure 2

The effects of rapamycin on the oxygen consumption during the presence (w¹¹¹⁸) and absence (Sirt4white⁺¹) of the Sirt4 gene. (A) and (B) represent the mitochondrial haplotypes grouped by strains. (C) and (D) represent haplotypes grouped by species of the mtDNA. For each pairwise comparison, the raw p-values, Holm-Bonferroni corrected p-values, and Hedges’ g effect sizes are presented in Table 1 and File S2. The significance bars reflect the raw p-values. # = 0.05 < P < 0.1; * = 0.01 < P < 0.05; ** = 0.001 < P < 0.01; *** = P < 0.001.

Figure 3

mtDNA species-dependent responses to the deletion of Sirt4 and rapamycin exposure. Grouped D. melanogaster mtDNA (OreR;OreR and Zim⁵³;OreR) is represented in (A) and grouped D. simulans introgressed mtDNA (sm21;OreR and sI1;OreR) in (B). For each pairwise comparison, the raw p-values, Holm-Bonferroni corrected p-values, and Hedges’ g effect sizes are presented in Table 1 and File S2. The significance bars reflect the raw p-values. # = 0.05 < P < 0.1; * = 0.01 < P < 0.05; ** = 0.001 < P < 0.01; *** = P < 0.001.

Figure 4

The effects of Sirt4 deletion on oxygen consumption among the standard and rapamycin treatments. Mitochondrial haplotypes grouped by strains are represented in (A) and (B). mtDNA haplotypes grouped by species are represented in (C) and (D). For each pairwise comparison, the raw p-values, Holm-Bonferroni corrected p-values, and Hedges’ g effect sizes are presented in Table 1 and File S2. The significance bars reflect the raw p-values. # = 0.05 < P < 0.1; * = 0.01 < P < 0.05; ** = 0.001 < P < 0.01; *** = P < 0.001.

Figure 5

Comparison of the effects of rapamycin on the consumption of oxygen (blue) and the production of carbon dioxide (pink) during the presence (w¹¹¹⁸) and absence (Sirt4white⁺¹) of the Sirt4 gene. (A) Mitochondrial haplotypes grouped by strains. (B) Haplotypes grouped by species of mtDNA. Error bars represent the 95% confidence interval. The number inside the bars represents the mean respiratory quotient (RQ): volume of CO₂/volume of O₂. All values were normalized to the reference sample (OreR w¹¹¹⁸ in standard treatment).

Figure 6

(A) Volcano plot depicting the spread of all of the pair-wise analyses. The y-axis scales the p-values (log10-transformed) that were adjusted for multiple comparisons using the Holm-Bonferroni method, and the x-axis scales the Hedges’ g effect sizes for each pair-wise comparison. The dotted line represents the alpha level of 0.05; green samples above the line are significant (P < 0.05), whereas the gray samples below the line are not significant (P > 0.05). The numbers pointing to each dot corresponds with the ID # on Table 1 and File S2. (B) Graphical representation summarizing the impact that rapamycin treatment has on modulating oxygen consumption for each condition used in this research. Each big circle represents a distinct cell that corresponds with a strain used in this experiment. The small circle within the cell represents the species of the mtDNA. The lines within the black circle (nucleus) represent the nuclear chromosomes. Blue represents DNA from D. melanogaster and red corresponds with D. simulans DNA. The deletion of Sirt4 is labeled as a darker blue chromosome.

Figure 7

Multiple sequence alignments of 183 Sirt4 orthologs. (A) Histogram of the K_A/K_S substitution ratio for all of the pairwise alignments. (B) The average K_A/K_S from each species was plotted as a histogram. (C) Histogram plot of the K_A/K_S ratio from all of pairwise alignments with D. melanogaster (Ensembl Peptide ID: FBpp0070817).

Figure 8

Phylogeny of human and D. melanogaster Sirt4 paralogs and proteobacterial Sirtuins. (A) Phylogeny generated from DNA sequences. (B) Phylogeny generated from protein sequences. At each node, the numbers represent the bootstrap support values as a percentage; the higher the value, the more reliable the grouping at the node. The multiple sequence alignments are displayed to the right of the phylogeny.