Transcriptional neoteny in the human brain

Abstract

In development, timing is of the utmost importance, and the timing of developmental processes often changes as organisms evolve. In human evolution, developmental retardation, or neoteny, has been proposed as a possible mechanism that contributed to the rise of many human-specific features, including an increase in brain size and the emergence of human-specific cognitive traits. We analyzed mRNA expression in the prefrontal cortex of humans, chimpanzees, and rhesus macaques to determine whether human-specific neotenic changes are present at the gene expression level. We show that the brain transcriptome is dramatically remodeled during postnatal development and that developmental changes in the human brain are indeed delayed relative to other primates. This delay is not uniform across the human transcriptome but affects a specific subset of genes that play a potential role in neural development.

Fig. 1.

Expression variation during primate and rodent brain development. (A) The first 2 principle components of the human and chimpanzee DLPFC dataset. The numbers represent each individual's age in years. The first and second components explain 25% and 15% of the total variance and are significantly correlated with age (r = 0.86, P < 10⁻¹⁶) and species identity (r = 0.84, P < 10⁻¹⁶), respectively. Red, humans; blue, chimpanzees. (B) The mean proportion of the total variance explained by sex, species identity, and age across all expressed genes. The values for 39 humans and 14 chimpanzees (orange bars, left) are based on 7,958 genes. The values for rodents (yellow bars, right) are based on 8,362 genes measured in 18 individuals. The expected values are calculated as the median of 1,000 permutations of each factor. Note that the proportion of variance explained by sex does not exceed the random expectation in humans and chimpanzees, whereas in mice it is not estimated, because only males were used. (C) Proportions of age-related genes and genes showing significant expression differences between humans and chimpanzees in the DLPFC transcriptome. Age+/Age− represents genes showing/not showing a significant expression difference with age, and Sp+/Sp− represents genes showing/not showing a significant expression difference between species. The number of genes in each category is given in parentheses. (D) The percentage of global expression change relative to newborns. One-hundred percent change was designated as the difference between the youngest and oldest individuals (in humans or chimpanzees) in terms of the summary measure of global expression (see Materials and Methods). Each point represents an individual.

Fig. 2.

Developmental shifts between humans and chimpanzees. The expression changes with age of 4 exemplary genes representing 4 phyloontogenetic patterns: human neoteny (EZH1), human acceleration (HER3), chimpanzee acceleration (ERBB2IP), and chimpanzee neoteny (MTMR2). The y axis shows normalized log2 expression levels. The x axis shows age in years. Each dot represents an individual; red, humans; blue, chimpanzees; dark green, rhesus macaques. The curves are fitted to the points using polynomial regression.

Fig. 3.

Gene expression neoteny in the human brain. (A) The distribution of genes among phyloontogenetic categories in the 2 prefrontal cortex areas and their overlap. (B) The proportion of gray-matter-specific genes among human-neotenic genes (red) or among genes in the other 3 phyloontogenetic categories (green). The error bars indicate 95% confidence intervals estimated by bootstrapping across genes within a category 10,000 times. (C) (Upper) Changes in human–chimpanzee expression divergence with age. The solid lines indicate the mean normalized expression divergence between humans and chimpanzees across the age range for human-neotenic genes (red), genes in the other 3 phyloontogenetic categories (green), and all age-related and differentially expressed genes (Age+,sp+ genes, gray). The dotted lines indicate 95% confidence intervals estimated by bootstrapping across genes within a category 10,000 times. (Lower) Changes in human–chimpanzee divergence for human-neotenic genes (red) and for genes in the other 3 categories (green) relative to all age-related and differentially expressed genes. The shaded areas indicate the age range where the 95% bootstrap intervals of human-neotenic genes do not overlap with age-related and differentially expressed genes (pink) or with genes in the other 3 categories (orange). For Figs. 3B and 3C, as well as for the overlapping genes shown in 3A, we chose gene sets by using relaxed significance cutoffs (Table 1). Using other criteria yields the same principal results.