Exceptional evolutionary divergence of human muscle and brain metabolomes parallels human cognitive and physical uniqueness

Abstract

Metabolite concentrations reflect the physiological states of tissues and cells. However, the role of metabolic changes in species evolution is currently unknown. Here, we present a study of metabolome evolution conducted in three brain regions and two non-neural tissues from humans, chimpanzees, macaque monkeys, and mice based on over 10,000 hydrophilic compounds. While chimpanzee, macaque, and mouse metabolomes diverge following the genetic distances among species, we detect remarkable acceleration of metabolome evolution in human prefrontal cortex and skeletal muscle affecting neural and energy metabolism pathways. These metabolic changes could not be attributed to environmental conditions and were confirmed against the expression of their corresponding enzymes. We further conducted muscle strength tests in humans, chimpanzees, and macaques. The results suggest that, while humans are characterized by superior cognition, their muscular performance might be markedly inferior to that of chimpanzees and macaque monkeys.

Figure 1. Metabolome data quantification, annotation, and variation analysis.

(a) The total number of metabolite peaks measured in our datasets. The bottom part of each bar indicates the number of peaks that were annotated in respective datasets. The total numbers of peaks and numbers of annotated peaks are indicated above the bars. (b) PCA plot based on normalized intensities of 10,615 peaks from all metabolic datasets measured in five tissues of the four species. Each circle represents an individual sample colored according to the tissue. (c) PCA plots for each of the five tissues: muscle, kidney, PFC, CBC and V1. Each circle represents an individual sample colored according to species. (d) Percentages of peaks with a significant proportion of variation (ANOVA, permutation p<0.01) explained by tissue, species, species–tissue interaction term (sp∶ts), sex, age, and sample RNA preservation. RIN, RNA Integrity Number.

Figure 2. Tissue-specific metabolite clusters.

(a) Dendrogram of 9,603 metabolite peaks after hierarchical clustering, based on concentration profile correlations across tissues in all four species. The colors represent 13 metabolite clusters colored according to the tissue-specificity of the cluster concentration profiles: shades of red indicating brain-specific clusters, shades of blue: muscle-specific, and shades of green: kidney-specific clusters. Clusters cumulatively containing less than 10% of all assayed peaks were removed from further analyses and are colored in gray. (b) Proportions of peaks specific to a given tissue based on the cluster analysis. Colors of the bars correspond to the colors of clusters in panel a. (c) Numbers of pathways with significant enrichment of metabolites in the 13 metabolite clusters. The red part of each bar shows pathways with significant overrepresentation of enzymes with matching tissue-specific expression profiles.

Figure 3. Species-specific metabolic change.

(a) Proportions of metabolite concentration changes on the four species evolutionary lineages. The phylogenetic tree above the bars shows the human (red) and chimpanzee (orange) evolutionary lineages, the lineage connecting the common ancestor of humans and chimpanzees with macaques (gray), and the lineage connecting the common ancestor of humans, chimpanzees, and macaques with mice (blue). The colors within the bars show proportions of metabolite peaks with concentration changes specific to the corresponding lineages. (b) Relationship between species-specific metabolite concentration divergence in each tissue and the corresponding phylogenetic distances in million years (MY). The linear regression between metabolite concentration divergence and phylogenetic distances represented by the dashed line was calculated based on the proportions of metabolite peaks with significant concentration divergence on the corresponding lineage for the three nonhuman species (represented by circles). Asterisks colored according to the tissues show the proportions of metabolite peaks with significant concentration divergence on the human lineage relative to the chimpanzee linage. (c) Heat map of relative metabolite concentration levels across three primate species. The plot is based on the 4,698 metabolite peaks that showed significant concentration differences in any tissue in the three primate species (t-test, permutation p<0.01). The columns represent metabolite peaks; the rows represent 190 tissue samples sorted by species and tissue identity. For each peak, concentration level in a given sample was normalized to the concentration across all samples of other primate species within the same tissue (base 2 logarithm-transformed ratios). The colors show positive (red) and negative (blue) concentration differences from the mean expression level. Concentration levels, p-values and annotation of the peaks shown in the Figure are listed in Table S7. (d) Proportions of enzymes with matching species-specific expression profiles linked to species-specific metabolites in a given tissue normalized by the number of enzymes with expression profiles specific to any species in this tissue. Tissues showing significant overrepresentation of enzymes with matching species-specific expression profiles are indicated with asterisks: **, p<0.01 and *, p<0.05.

Figure 4. Functions of the human-specific metabolites, environmental effects, and muscle strength experiments.

(a) Functional units of KEGG pathways containing significant excesses of metabolites with human-specific concentration profiles in PFC and skeletal muscle. Columns represent metabolites showing human-specific concentration profiles in these tissues. Rows represent KEGG pathways enriched in the respective metabolites. Colors show higher (red) and lower (blue) concentration levels in humans compared with the other two primate species. Only pathways supported by human-specific expression of enzymes are shown. Metabolites directly linked to the enzymes with a human-specific expression profile in these tissues are indicated by darker shades of red and blue. Numbers and borderlines of different shades of gray indicate metabolite groups showing correlated concentration profiles across multiple pathways. Plotted metabolites, pathways, and metabolite groups are listed in Tables S14, S15 and further illustrated in Figures S10, S11. (b) Numbers of metabolite peaks differentiating humans, chimpanzees, and macaques subjected to environmental condition 1 (C1: physical activity deprivation), and macaques subjected to environmental condition 2 (C2: high nutrition and physical activity deprivation) from the control macaques in any of the five tissues (t-test, p<0.01). Numbers of peaks showing significant concentration differences between control macaques and humans or chimpanzees were estimated using six randomly chosen human or chimpanzee individuals. The error bars show 0.05 and 0.95 quintiles of the significant peak number distributions calculated by randomly sampling six human or six chimpanzee individuals 1,000 times. (c) Pulling strength of humans, chimpanzees, and macaques normalized by individual body weight. The shape of the symbols represents the individual's sex: circles: females, diamonds: males. Empty symbols represent trials measured using the apparatus built in the Leipzig primate center, filled symbols those measured using the apparatus built in Shanghai. The size of the symbols is proportional to the individual's age normalized by the maximal life expectancy of the species: humans: 120 years, chimpanzees: 60 years, macaques: 40 years. Up to three of the best results per individual are shown. Asterisks indicate the significance of muscle strength difference between humans and the other two species (Wilcoxon test, p<0.001). The highest measurement of each individual was used in the p-value calculation.