Ancient Jumping Genes Turn Out to Be Hidden Architects of the Human Cortex

One of the enduring puzzles of human neuroscience is why our cortex, the folded sheet of neurons responsible for our most sophisticated cognitive abilities, is so much larger and more elaborately structured than that of other primates. The genetic changes underlying this expansion have proven notoriously difficult to pin down, and even harder to test experimentally, particularly when those changes act not by altering proteins but by rewiring the regulation of genes. A new study highlighted in Nature Neuroscience by Enakshi Sinniah and Alex A. Pollen of the University of California, San Francisco, suggests that a surprising share of the answer has been hiding in plain sight, embedded within one of the most abundant and long-dismissed components of our genome: Alu transposable elements.

Alu elements are short stretches of DNA, roughly 300 base pairs long, that belong to a family of mobile genetic sequences known as retrotransposons. Over tens of millions of years of primate evolution, they have copied and pasted themselves throughout the genome with extraordinary success. As the classic review by Mark Batzer and Prescott Deininger noted, Alus now account for roughly one tenth of the human genome, numbering well over a million copies. For decades, they were largely written off as genomic parasites or, at best, molecular fossils. That view has been steadily revised. Work summarized by Rie Fueyo, Jared Judd, Cédric Feschotte and Joanna Wysocka has shown that transposable elements have repeatedly been co-opted by their hosts as raw material for new gene-regulatory circuits, supplying binding sites for transcription factors and contributing to the evolution of gene networks in ways that can shape tissue identity and development.

The new research, a study by Matoba and colleagues published in the same issue of Nature Neuroscience, takes this idea into the territory of human cortical biology. Writing in a News and Views commentary, Sinniah and Pollen explain that Matoba and colleagues identified genetic variants residing inside Alu elements that are linked to measurable differences in human cortical structure. In other words, the repetitive DNA that so often gets excluded from genomic analyses because it is difficult to map and annotate may actually carry functional variants that help explain why different people, and by extension different lineages, build cortices of different sizes and shapes.

The significance of this finding becomes clearer when one considers how the cortex is built. Cortical size depends fundamentally on the number of neural progenitor cells produced during development and on how many rounds of division they undergo before differentiating into neurons. As early as 2002, Anjen Chenn and Christopher Walsh demonstrated in Science that modulating a key cell-signaling pathway in the developing mouse brain could expand the progenitor pool and produce a dramatically enlarged, folded cortical surface. That work established a principle that has guided the field ever since: relatively subtle changes in developmental regulation can produce large changes in cortical architecture. The question has always been which genetic variants in the human lineage perform this kind of regulatory tuning, and how their effects can be tested.

This is where the technical challenge lies. Most studies of human brain evolution have focused on protein-coding changes, which are comparatively easy to identify and to model in experimental systems. But a large fraction of the genome’s functional variation lies in regulatory DNA, the non-coding sequences that determine when, where and how strongly genes are expressed. As A. Marie J. Kim and Joanna Wysocki reviewed in Molecular Cell, mapping regulatory variation in the human brain, and distinguishing functional variants from the vast background of neutral sequence differences, remains one of the hardest problems in functional genomics. Variants buried inside repetitive elements compound the difficulty, because standard genome-sequencing pipelines often cannot uniquely place short reads within repeated Alu copies, meaning that variation in these elements is systematically under-detected and under-studied.

What makes Alu elements especially interesting as carriers of regulatory variation is their internal architecture. Alus contain their own built-in regulatory signals, including internal RNA polymerase III promoter elements that originally allowed them to be transcribed by the retrotransposition machinery. Work dating back to Greg Vansant and William Reynolds in 1995 showed that Alu sequences can also interact with the transcriptional machinery of RNA polymerase II, the polymerase responsible for protein-coding genes, and can influence the expression of nearby genes. When a variant arises within an Alu element sitting near a developmental gene, it can therefore perturb an existing regulatory module rather than needing to build one from scratch. Sinniah and Pollen emphasize that Matoba and colleagues show precisely this: variants that shape human cortical structure co-opt the inbuilt machinery of Alu elements to rewire regulatory networks in the developing brain.

The broader context is the comparative genomics of cortical development, a field comprehensively surveyed by André Sousa, Kenneth Meyer, Gabriel Santpere, Forrest Gulden and Nenad Sestan in Cell. Their synthesis documented how the human cortex differs from that of other mammals in its progenitor cell types, its developmental timing and its gene-expression programs, and how many of these differences are likely to be regulatory rather than protein-coding in origin. Because Alu elements expanded explosively in the primate lineage after splitting from rodents, and continued to expand within the ape and human lineages, they represent a plausible substrate for primate-specific and even human-specific regulatory innovation. A variant in an Alu element near a cortical progenitor gene is, in evolutionary terms, a candidate for a change that is both recent enough to be human-relevant and positioned exactly where a regulatory effect could influence cortical expansion.

The approach taken by Matoba and colleagues, as framed by the commentary, also illustrates a shift in how such questions can now be asked. Rather than examining one candidate gene at a time, the study connects genetic variants linked to cortical traits with functional assays of gene regulation in the developing brain, closing the gap between population-level association and mechanistic biology. This matters because the genetic changes associated with the development of the human cortex, as Sinniah and Pollen put it, are difficult to test functionally, especially in the context of gene regulation. Demonstrating that Alu-harbored variants can be tied both to measurable cortical differences and to concrete regulatory mechanisms provides a template for future work on other repetitive elements and other aspects of brain development.

There are also implications beyond evolutionary biology. Variation in cortical size and structure is associated with a range of neurological and psychiatric conditions, and understanding which variants influence cortical development could sharpen the interpretation of genetic risk factors. If a meaningful portion of that variation resides in Alu elements, then the routine exclusion of repetitive regions from genome-wide analyses may be causing researchers to miss relevant variants. The commentary’s central message is effectively a call to stop treating repetitive DNA as unreadable noise. The tools for resolving variation within repeated elements are improving, and studies such as that of Matoba and colleagues demonstrate that the effort is repaid with biological insight.

The image that emerges is one of the genome as a layered archive rather than a finished blueprint. Alu elements, once viewed purely as selfish molecular hitchhikers, have been conscripted over evolutionary time into the service of gene regulation, and now appear to carry variants that contribute to one of the most distinctive features of our species: the size and structure of the cerebral cortex. As Sinniah and Pollen conclude in their commentary, the variants that shape human cortical structure are hiding inside Alu transposable elements, co-opting their inbuilt machinery to rewire regulatory networks in the developing brain. For a field that has spent decades searching the genome for what makes the human cortex human, the answer, at least in part, appears to have been written in the repeats.

Subject of Research: The role of Alu transposable element variants in human cortical development and gene regulation

Article Title: Alus harbor genetic variants that shape the human cortex

Article References: Alus harbor genetic variants that shape the human cortex. (n.d.). https://doi.org/10.1038/s41593-026-02492-w

Image Credits: AI Generated

DOI: 10.1038/s41593-026-02492-w

Keywords: Alu elements, transposable elements, human cortex, cortical development, gene regulation, functional genomics, retrotransposons, neural progenitors, brain evolution, regulatory variants, primate genomics, Nature Neuroscience