Purine Homeostasis Is Necessary for Developmental Timing, Germline Maintenance and Muscle Integrity in Caenorhabditis elegans

Abstract

Purine homeostasis is ensured through a metabolic network widely conserved from prokaryotes to humans. Purines can either be synthesized de novo, reused, or produced by interconversion of extant metabolites using the so-called recycling pathway. Although thoroughly characterized in microorganisms, such as yeast or bacteria, little is known about regulation of the purine biosynthesis network in metazoans. In humans, several diseases are linked to purine metabolism through as yet poorly understood etiologies. Particularly, the deficiency in adenylosuccinate lyase (ADSL)-an enzyme involved both in the purine de novo and recycling pathways-causes severe muscular and neuronal symptoms. In order to address the mechanisms underlying this deficiency, we established Caenorhabditis elegans as a metazoan model organism to study purine metabolism, while focusing on ADSL. We show that the purine biosynthesis network is functionally conserved in C. elegans Moreover, adsl-1 (the gene encoding ADSL in C. elegans) is required for developmental timing, germline stem cell maintenance and muscle integrity. Importantly, these traits are not affected when solely the de novo pathway is abolished, and we present evidence that germline maintenance is linked specifically to ADSL activity in the recycling pathway. Hence, our results allow developmental and tissue specific phenotypes to be ascribed to separable steps of the purine metabolic network in an animal model.

Keywords: ADSL deficiency; AICAR; SAICAR; SZMP; ZMP; metabolism; purine salvage pathway.

Figure 1

Purine metabolism in C. elegans. (A) Schematics of the purine biosynthesis pathways in C. elegans based on sequence homology. Enzymes subjected to functional analysis are represented in red, other enzymes in green, and metabolites in black. (B) Drop test to assess adenine auxotrophy in yeast S. cerevisiae mutants deficient for enzymes ADSL (ade13∆), ATIC (ade16∆ ade17∆), or GPPAT (ade4∆) expressing C. elegans genes adsl-1, atic-1b and ppat-1, respectively. Expression of S. cerevisiae genes ADE13, ADE17, and ADE4 was used as positive control, and expression vector without insert as negative control. A culture with adenine is shown as auxotrophy control. In all drop tests presented, four drops are shown per condition, corresponding to serial dilutions (1:10) of cellular suspensions, with deceasing cell concentrations from left to right. (C) Zoom in on HPLC chromatogram peaks of specific metabolites SAICAR (SZMP riboside form), SZMP, sAdo (succinyl-adenosine, SAMP riboside form), SAMP, ATP, and GTP, upon adsl-1(RNAi) and ppat-1(∆); adsl-1(RNAi). (D) Zoom in on HPLC chromatogram peaks of specific metabolites SAICAR (SZMP riboside form), SZMP, AICAR (ZMP riboside form), ZMP, ATP, and GTP, in atic-1(∆) and ppat-1(∆); atic-1(∆). Ade, adenine; Ado, adenosine; SAdo, succinyl-adenosine; Ino, Inosine; Hypox, hypoxantine; Xant, Xantine; Gua, guanine; Guo, guanosine. In (C) and (D), scales are adjusted differently among metabolites, in order to highlight differences between genotypes.

Figure 2

Developmental defects in purine mutants. (A) DIC whole body images of representative young adults for each of the genotypes studied (Bar, 200 μm). (B) Fluorescence micrographs of worms expressing H2B::GFP in the germline under the control of the pie-1 promoter. Arrows point at rare germ nuclei expressing H2B::GFP; spotty pattern throughout adsl-1(∆) and ppat-1(∆); adsl-1(∆) corresponds to intestinal auto-fluorescence (Bar, 25 μm). (C) Tukey box plot depicting body length of all purine mutants and glp-1(−) for comparison (means statistically different—one-way ANOVA—indicated P values refer to comparison with wild-type—Bonferroni multiple comparison t-test—otherwise comparison to wild type bears no significant difference; n = 30 for all genotypes).

Figure 3

Life traits in purine mutants. (A) Graph depicting duration of postembryonic developmental stages, white bars represent larval stages, from L1 to L4 (left to right), orange bars represent lethargus phases consecutive to each larval stage (mean durations of larval development, from t0 until exit from fourth lethargus phase, are statistically different—one-way ANOVA—indicated P values refer to comparison with wild type—Bonferroni multiple comparison t-test—otherwise comparison to wild type bears no significant difference; see detailed representation of the data in Figure S4). (B) Graph depicting fraction of live worms over time for all genotypes studied (P value indicated for the adsl-1(∆) vs. wild type comparison, all other genotypes are not significantly different from wild type are significantly different from adsl-1(∆)—log rank test with Bonferroni correction; total combined of three independent experiments for each genotype [wild type n = 106, ppat-1(∆) n = 91, atic-1(∆) n = 126, ppat-1(∆); atic-1(∆) n = 149, adsl-1(∆) n = 227, ppat-1(∆); adsl-1(∆) n = 113].

Figure 4

Locomotion of purine mutants. (A) Cartoon schematics of the locomotion behavior assay. (B) Graph depicting the average locomotion speed in all purine mutants [error bar, SEM; n = 30 in all genotypes except wild type and ppat-1(∆); atic-1(∆) n = 40; means statistically different—one-way ANOVA—indicated P values refer to comparison with wild–type—Bonferroni multiple comparison t-test—wild type, ppat-1(∆), and ppat-1(∆) adsl-1(∆) are not statistically different from one another].

Figure 5

Muscle structure in purine mutants. Confocal micrographs of muscle myosin immunostaining in young adult body wall muscle cells of all studied purine mutants (Bar, 10 µm). (A) Wild-type. (B) adsl-1(∆). (C) atic-1(∆). (D) ppat-1(∆). (E) ppat-1(∆); adsl-1(∆). (F) ppat-1(∆); atic-1(∆).

Figure 6

Summary of the functional analysis of the purine biosynthesis pathway in C. elegans reported in this study.