Metabolome and proteome changes with aging in Caenorhabditis elegans

Abstract

To expand the understanding of aging in the model organism Caenorhabditis elegans, global quantification of metabolite and protein levels in young and aged nematodes was performed using mass spectrometry. With age, there was a decreased abundance of proteins functioning in transcription termination, mRNA degradation, mRNA stability, protein synthesis, and proteasomal function. Furthermore, there was altered S-adenosyl methionine metabolism as well as a decreased abundance of the S-adenosyl methionine synthetase (SAMS-1) protein. Other aging-related changes included alterations in free fatty acid levels and composition, decreased levels of ribosomal proteins, decreased levels of NADP-dependent isocitrate dehydrogenase (IDH1), a shift in the cellular redox state, an increase in sorbitol content, alterations in free amino acid levels, and indications of altered muscle function and sarcoplasmic reticulum Ca(2+) homeostasis. There were also decreases in pyrimidine and purine metabolite levels, most markedly nitrogenous bases. Supplementing the culture medium with cytidine (a pyrimidine nucleoside) or hypoxanthine (a purine base) increased lifespan slightly, suggesting that aging-induced alterations in ribonucleotide metabolism affect lifespan. An age-related increase in body size, lipotoxicity from ectopic yolk lipoprotein accumulation, a decline in NAD(+) levels, and mitochondrial electron transport chain dysfunction may explain many of these changes. In addition, dietary restriction in aged worms resulting from sarcopenia of the pharyngeal pump likely decreases the abundance of SAMS-1, possibly leading to decreased phosphatidylcholine levels, larger lipid droplets, and ER and mitochondrial stress. The complementary use of proteomics and metabolomics yielded unique insights into the molecular processes altered with age in C. elegans.

Keywords: Aging; Ascorbate; C. elegans; Caenorhabditis; Lifespan; Metabolome; Metabolomics; Methionine; Methylation; Methyltransferase; Nitrogenous bases; Proteome; Proteomics; Purine; Pyrimidine; S-adenosyl methionine; SAMS-1; SERCA; Sorbitol.

Fig. 1. Principal component analysis of metabolites shows separation of young and aged samples

(A) When graphed as the 2D relationship between the two top explanatory factors (F1 = 59.92% of the variation; F2 = 10.20% of the variation), the scaled young day 4 samples (n = 5) and the scaled aged day 10 sample (n = 5) are separated primarily linearly along the F1 axis, showing that the major variation in the distribution of metabolite levels is explained by the difference in ages between the groups. (B) When similarly graphed using the same two factors, the identified metabolites (n = 186) visibly congregate at two poles along the F1 axis.

Fig 2. Top 10 changed pathways based on metabolome analysis

The total observed change for each identified pathway was calculated as the total of the absolute log₂-fold change for each identified metabolite found within the pathway (n = 50 pathways). The top 10 changed pathways are shown in this figure. The width of each pie wedge is equivalent to the percent change of that pathway out of the total absolute log₂-fold change of all pathways. Numbers shown on each pie wedge are the total absolute log₂-fold change for that pathway.

Fig. 3. Age-related changes in free amino acid levels as they correspond to amino acid hydrophobicity

The levels of the more hydrophobic amino acids decreased and the levels of the more hydrophilic amino acids increased on day 10 compared to day 4 (Pearson correlation −0.67; p-value = 0.05). The data point for Leu is obscured by the data point for Trp. At pH 7.0 the relative hydrophobicity (on a scale from −100 to +100 normalized to glycine) is as follows: Met = 74, Phe = 100, Leu = 97, Trp = 97, Val = 76, Ile = 99, Thr = 13, Gly = 0, Ala = 41.

Fig. 4. Age-related decreases in hypoxanthine and nitrogenous base levels

(A) By measuring hypoxanthine levels of N2 nematodes treated with FUdR using a commercially available kit (Xanthine/Hypoxanthine assay kit, BioVision, Inc.), we found the concentration of hypoxanthine fell from a mean of 100.37 pmol/mg of protein (SEM ± 3.48) on day 5 (n = 6 samples; ~100 nematodes/sample), to a mean of 70.22 pmol/mg of protein (SEM ± 2.65) on day 12 (n = 6 samples; ~100 nematodes/sample; unpaired two-tailed t-test p-value = 4.2 × 10⁻⁵). (B) The metabolomics data indicated that nitrogenous base levels decreased with age.

Fig. 5. Age-related changes in monoacylglycerol and fatty acid levels

Monoacylglycerols primarily decreased with age, while fatty acids primarily increased. For brevity, we’ve only listed fatty acids with a log₂-fold change above 1 or below −1. For a complete list, see Table S3. *16-methylheptadecanoic acid.

Fig. 6. Age-related changes in D-sorbitol content and nematode redox state

(A) D-sorbitol levels in FUdR-treated N2 nematodes increased ~360% from day 5 to day 10 (day 5 mean = 63.28 nmol/mg protein, SEM ± 8.29, n = 4 samples, ~100 nematodes/sample; day 12 mean = 229.95 nmol/mg protein, SEM ± 6.60, n = 4 samples, ~100 nematodes/sample; unpaired two-tailed t-test p-value = 4.19 × 10⁻⁶) as measured by a commercial D-sorbitol assay kit (Megazyme, Inc.). (B) Aged/young log₂-fold changes for pyruvate/lactate (log₂-fold change = −1.67; young vs. aged unpaired two-tailed t-test < 0.0001), dehydroascorbic acid/ascorbic acid (DHA/AA; log₂-fold change = 2.94; young vs. aged unpaired two-tailed t-test = 0.024) and erythronic acid (EA)/ N-acetylglucosamine (NAG). (C) Both NAD⁺ and NADH were determined in 5- and 10-day-old wild-type N2 C. elegans using a fluorescence-based assay (NAD/NADH Ratio Assay Kit, eEnzyme, LLC) after growth in liquid culture at 20 °C in the presence of FUdR added from the L4 larval stage onward. Fluorescence was normalized to total protein content at each age. NAD⁺ content significantly decreased and NADH content significantly increased in the older nematodes (p-value < 0.05). (D) There was a strong decrease in the NAD⁺/NADH ratio with age as calculated from the data in panel C.

Fig. 7. Proteomic analysis of young vs. aged C. elegans using stable isotope labeling

The y-axis represents the log₂ of the detected abundance, as determined by adding together the total detected intensities of both the heavy (¹⁵N₄-¹³C₆-arginine and ¹⁵N₂-¹³C₆-lysine) labeled proteins (young day 4 nematodes) and the light non-labeled proteins (aged day 10 nematodes). The x-axis indicates the relative fold-change in abundance of individual proteins from the aged nematodes as compared to the young nematodes. The unfilled circles represent proteins with a significantly changed abundance, as determined by the Significance B method (Benjamini-Hochberg corrected p-value) with a threshold of 0.05. The dark gray circles are proteins without a significant change in abundance.

Fig 8. The top 10 altered pathways with age based on proteomic analysis

The total observed change for each identified pathway was calculated as the total of the absolute log₂-fold change for each identified protein found within the pathway (n = 20 pathways). The top 10 changed pathways are shown in this figure. The width of each pie wedge is equivalent to the percent change of that pathway out of the total absolute log₂-fold change of all pathways. Numbers shown on each pie wedge are the total absolute log₂-fold change for that pathway.