Research Article

Conservation, acquisition, and functional impact of sex-biased gene expression in mammals

See allHide authors and affiliations

Science  19 Jul 2019:
Vol. 365, Issue 6450, eaaw7317
DOI: 10.1126/science.aaw7317

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

The genetics of sexual dimorphism

In mammals, many species exhibit sex-specific phenotypes that differ between males and females. Although attention has been directed to the effects of the X and Y sex chromosomes, we do not understand how sex affects the rest of the genome. Naqvi et al. examined gene expression in 12 tissues in male and female humans, mice, rats, dogs, and cynomolgus macaques and identified diversity in gene expression between the sexes. Examining sex-biased gene expression in human height identified opposing male or female bias. Although conservation of differential sex-specific gene expression among species was observed, specific genes differed in the sexes among species and lineages suggesting the evolution of species- or lineage-specific sex-biased expression.

Science, this issue p. eaaw7317

Structured Abstract


Sex differences are widespread in humans and other mammals. For example, the distribution of height or body size is shifted upwards in males relative to females, and sex differences are found in the immune and cardiovascular systems as well as in metabolism. However, little is known about how gene expression differs between the sexes in a broad range of mammalian tissues and species. A catalog of such sex-biased gene expression could help us understand phenotypic sex differences. Assessing the extent to which sex-biased gene expression is conserved across the body could also have important implications for the use of nonhuman mammals as models of sex-biased human biology.


To identify both conserved and lineage- or species-specific sex differences in gene expression, we sequenced RNA from male and female samples in 12 tissues in each of four nonhuman mammals (cynomolgus macaque, mouse, rat, and dog) and analyzed these data jointly with publicly available data from postmortem male and female human tissues. To assess the impact of sex-biased gene expression on the sex difference in mean human height, we applied methods that integrate the effects of genetic variation on both gene expression and phenotype (height in this case). We sought to understand which transcription factors (TFs) contribute to evolutionary changes in sex bias by analyzing motifs gained or lost concurrently with lineage- or species-specific changes in sex bias.


Linear modeling revealed ~3000 genes with conserved (species-shared) sex bias in gene expression, most of which was tissue specific. The cumulative effects of conserved sex bias explain ~12% of the sex difference in mean human height, and cases such as that of LCORL, a TF with conservation of both female-biased expression and genetic association with height, suggest a contribution to sex differences in body size beyond humans. However, most sex-biased gene expression (~77%) was specific to single species or subsets of species, implying that it arose more recently during evolution. We identified 83 instances where TFs showed sex-biased expression in the same tissue, in which their motifs were associated with gain or loss of sex bias at other genes, accounting for a significant portion (~27%) of lineage-specific changes in sex bias.


By conducting a 12-tissue, five-species survey of sex differences in gene expression, we found that although conserved sex bias in gene expression exists throughout the body, most sex bias has been acquired more recently during mammalian evolution. Height is likely subject to opposing selective pressures in males and females; our study thus documents how such selective forces can result in sex-biased expression which, when layered upon genetic pathways acting identically in males and females, can lead to trait distributions shifted between the sexes. Our findings also suggest that, in many cases, molecular sex differences observed in humans may not be mirrored in nonhuman mammals.

RNA sequencing of male and female samples in 12 tissues and five species reveals the functional impact and mechanistic underpinnings of sex-biased gene expression.

A survey of sex differences in gene expression using RNA sequencing data (left) leads to the discovery of both conserved (species-shared) and lineage- or species-specific sex biases in expression across the genome. Genes with conserved sex bias contribute to the sex difference in mean height in humans and other mammals, whereas lineage-specific changes can be partially explained by gains and losses of motifs for sex-biased TFs.


Sex differences abound in human health and disease, as they do in other mammals used as models. The extent to which sex differences are conserved at the molecular level across species and tissues is unknown. We surveyed sex differences in gene expression in human, macaque, mouse, rat, and dog, across 12 tissues. In each tissue, we identified hundreds of genes with conserved sex-biased expression—findings that, combined with genomic analyses of human height, explain ~12% of the difference in height between females and males. We surmise that conserved sex biases in expression of genes otherwise operating equivalently in females and males contribute to sex differences in traits. However, most sex-biased expression arose during the mammalian radiation, which suggests that careful attention to interspecies divergence is needed when modeling human sex differences.

View Full Text