In many mammals, the gene Ostn is expressed in muscles and bones. The discovery that the primate OSTN gene has been repurposed to also act in neurons provides clues to how humans evolved their cognitive abilities. See Article p.242
The cognitive abilities that separate humans and our primate relatives from other mammals are the product of millions of years of evolution, and stem from differences in how our brains develop and function1. Although brain maturation in all mammals relies in part on experience-driven development of neuronal circuits, human cognition depends particularly heavily on the experiential learning that occurs during our prolonged period of growth, which lasts up to two decades after birth2. The structural and functional changes that shape neuronal circuits during this developmental period are mediated by genes whose transcription is regulated by neuronal activity3. On page 242, Ataman et al.4 describe an unbiased screen to identify genes activated by neuronal excitation in human and mouse neurons. They identify a gene expressed in the bones and muscles of mice and other mammals that, over the course of evolution, was repurposed to act in the neurons of primates.
Because of the importance of experiential learning in humans, identifying gene-expression changes induced by neuronal activity is particularly relevant for understanding the genetic basis of our species' brain evolution. With this in mind, Ataman et al. cultured human neurons in vitro. These cultures contain a mix of differentiated cell types found in the brain, including glial cells, which support neurons and promote formation of the synaptic connections between them, and several subtypes of neuron from the brain's cortex. The authors stimulated these cultures to mimic increased neuronal activity, thereby increasing calcium signalling within neurons and inducing activity-dependent gene transcription.
Next, Ataman and colleagues used RNA sequencing to identify upregulated transcripts and confirmed that several well-characterized5,6 'immediate early' genes were rapidly induced within one hour of increased neuronal activity, including the transcription factors NPAS4 and FOSB. They also identified a set of late-response genes induced within six hours, which included previously reported genes7 such as BDNF. At this late time point, the authors found a few transcripts specific to human neurons. The most highly induced of these genes was that for osteocrin (OSTN).
In mice, the corresponding gene Ostn encodes a secreted protein that is involved in glucose metabolism in muscles and bones8,9. Ostn is not expressed in the mouse brain, and Ataman et al. found that its expression could not be induced by neuronal activity in mouse neuronal cultures. Therefore, OSTN expression seems to be regulated by neuronal activity in human but not mouse neurons — a supposition that the authors corroborated through several lines of investigation.
Ataman and colleagues reported that OSTN was widely expressed throughout the human neocortex (which is involved in higher cognitive functions, including sensory processing) and was highly enriched in the mature neurons of the developing cortex. They also found that, in the primary visual cortex of macaques, OSTN expression was induced by increased sensory-evoked neuronal activity. Together, these lines of evidence point to activity-dependent expression of OSTN in primate neurons in vivo.
“In primates but not in other mammals, OSTN might regulate structural changes that neurons undergo during learning.”
At the genetic level, what underlies this shift from Ostn transcription in the bones of mice to activity-dependent regulation of OSTN transcription in primate neurons? To address this question, the authors first demonstrated that OSTN expression in human neurons is regulated by a 2-kilobase-long promoter region that lies immediately upstream gene. By engineering a series of truncations and point mutations in this region, the researchers identified a minimal 85-base-long region that mediates gene activation in response to neuronal activity. In primates, this region contains three short DNA sequences to which transcription factors of the myocyte enhancer factor 2 (MEF2) family, such as MEF2C, can bind. In mice, however, the Ostn promoter does not contain these MEF2-responsive elements (MREs).
MEF2 transcription factors are crucial for activity-dependent transcription in neurons and have a role in key aspects of neuronal development10,11. The researchers found that two of the three MREs are highly evolutionarily conserved between humans and anthropoid primates, but that there are several differences in these sequences in prosimian primates, rodents, dolphins and several other mammalian species that result in the absence of MREs in the Ostn promoter of these species. By replacing the three primate MREs with the equivalent mouse sequences and monitoring gene transcription, the authors convincingly demonstrated that the evolutionary switch to activity-dependent transcription of the OSTN gene in primates emerged through a few single-nucleotide mutations that created MREs (Fig. 1).
What is the function of OSTN in primate neurons? The authors overexpressed or repressed OSTN in human neuronal cultures, and discovered that expression of the gene regulates the shape of dendrites — the branched parts of neurons that receive and integrate synaptic information from other neurons. This result suggests that, in primates but not in other mammals, OSTN might regulate structural changes that neurons undergo during learning. Indeed, OSTN belongs to a family of genes that encode secreted protein fragments called natriuretic peptides, which have been shown12 to promote branching of neuronal projections called axons.
Further experiments will be needed to fully determine the impact of OSTN expression on primate brain development. Defining the roles of OSTN — and of the other activity-dependent genes identified in Ataman and colleagues' study — will improve our understanding of the evolutionary mechanisms that enabled the emergence of primate-specific features of brain development and function.Footnote 1
Carroll, S. B. Nature 422, 849–857 (2003).
Petanjek, Z. et al. Proc. Natl Acad. Sci. USA 108, 13281–13286 (2011).
Greer, P. L. & Greenberg, M. E. Neuron 59, 846–860, (2008).
Ataman, B. et al. Nature 539, 242–247 (2016).
Kim, T. K. et al. Nature 465, 182–187 (2010).
Lanahan, A. & Worley, P. Neurobiol. Learn. Mem. 70, 37–43 (1998).
Hughes, P., Beilharz, E., Gluckman, P. & Dragunow, M. Neuroscience 57, 319–328 (1993).
Moffatt, P. et al. J. Biol. Chem. 282, 36454–36462 (2007).
Subbotina, E. et al. Proc. Natl Acad. Sci. USA 112, 16042–16047 (2015).
Flavell, S. W. et al. Science 311, 1008–1012 (2006).
Mao, Z., Bonni, A., Xia, F., Nadal-Vicens, M. & Greenberg, M. E. Science 286, 785–790 (1999).
Zhao, Z. & Ma, L. Proc. Natl Acad. Sci. USA 106, 18016–18021 (2009).
About this article
Cite this article
Kupferman, J., Polleux, F. Genomic remodelling in the primate brain. Nature 539, 171–172 (2016). https://doi.org/10.1038/539171a