An ancient viral insertion silences LINC00662 in chimpanzee cells, while the gene is active during early human brain development. Removal of the insertion restored gene expression; Separate experiments with human organoids linked LINC00662 to features of developing nerve cell extensions. The findings reveal a mechanism of evolutionary divergence, without establishing effects on intelligence, overall brain size or disease. A genetic remnant from an ancient infection still helps determine whether a brain development gene is activated in chimpanzee cells. Removing that viral DNA restored the gene’s activity in laboratory experiments. In human cells, where the insertion is absent, the same gene appears to help regulate how young nerve cells mature. The findings, published in Science Advances by a team led by Lund University, connect an inherited viral sequence with differences in gene regulation. The LINC00662 gene produces a long noncoding RNA, a molecule that functions without serving as an instruction for a protein. Its activity distinguishes the human and chimpanzee neural models examined in the study. The experiments identify a specific molecular difference, rather than a single explanation for the evolution of the human brain. They combine comparisons of genomes, stem cell models and gene editing to test a mechanism. Neither intelligence nor behavior was measured, and the work does not show that this insertion determines overall brain size. Johan Jakobsson, Professor of Molecular Neurogenetics at Lund University. (CREDIT: Lund University) A lasting record of ancient infections Retroviruses can insert genetic material into the DNA of infected cells. If an insertion enters the cells that give rise to eggs or sperm, offspring can inherit it. These inherited sequences, called endogenous retroviruses, can remain in a species’ genome long after the original infection is gone. The team investigated Pan troglodytes endogenous retrovirus 1, or PTERV1, a family absent in the human genome. Its remains are found in chimpanzees and other African apes, reflecting a complicated history of infection and spread. An insertion within LINC00662 is shared by chimpanzees and bonobos, suggesting that it arose in their ancestral lineage between approximately 1.5 million and 5 million years ago. The researchers identified 158 PTERV1 elements in the chimpanzee reference genome. Almost all had extensive DNA methylation in the stem cells and neuronal organoids examined. Methylation adds chemical marks to DNA and can help suppress activity, including potentially disruptive viral sequences. The inherited insertion is a genomic fossil, rather than evidence of an ongoing infection in the organoids. Its sequence contains interruptions that likely prevent it from making functional copies of the virus. However, once embedded in ancestral DNA, a remnant can retain regulatory effects without producing infectious particles. Therefore, the study refers to a persistent consequence of the infection across generations. This protective response can also affect neighboring DNA. In the chimpanzee LINC00662 region, methylation covered the viral insertion and extended to surrounding sequences. The comparable human region showed much less methylation, suggesting that silencing the viral remnant could also silence its host gene. LINC00662 silencing does not alter the state composition of neural organoid cells. UMAP embeddings of unguided neural organoid nuclei from days 15 and 30, showing clustering using all samples (left) and splitting by condition (right). (CREDIT: Johan Jakobsson et al, Science Advances 2026) Comparison of early neuronal development The researchers grew neuronal organoids from induced pluripotent stem cells, which can develop into many cell types. Organoids are three-dimensional laboratory models that reproduce selected features of developing tissue. They provide access to early developmental processes that would otherwise be difficult to compare across species. The initial comparison used cells from two humans and two chimpanzees. By day 15, the organoids had similar sizes, visible structures, and broad cell type compositions. Most of the cells were neural progenitors, which generate nerve cells or newly formed neurons. Comparing similar developmental stages helped the team assess gene activity without attributing each difference to an unequal mix of cell types. In that broadly comparable context, LINC00662 stood out. It was active in all cell populations of human organoids, but was undetectable in chimpanzee models. Analysis of additional published data sets supported the difference beyond the original cell lines. “The gene was strongly expressed in human organoids, but not in chimpanzee organoids,” said Patricia Gerdes, a postdoctoral researcher in molecular neurogenetics at Lund. Human development data sets also showed LINC00662 activity in fetal brain tissue. Expression peaked around 10 to 12 weeks after conception before declining towards adulthood. Deletion of the viral sequence A comparison alone could not establish that PTERV1 caused the silencing. To test that possibility, the team used CRISPR-Cas9 to remove the viral insertion of about 7,000 letters from chimpanzee stem cells. After removal, those cells began to express LINC00662 again. LINC00662 silencing induces axonogenesis-related transcriptional programs in developing neurons. (A) Selected enriched GO terms from GSEA in early neurons (left and middle) and neurons (right) (CREDIT: Johan Jakobsson et al, Science Advances 2026) “This shows that it is in fact the viral insertion that is responsible for the gene being inactivated in chimpanzees,” Gerdes said. The experiment directly linked the insertion to suppressed expression in the cells analyzed. It did not convert chimpanzee cells into human cells or demonstrate a change in an animal’s brain. The surrounding methylation pattern supports a model in which viral silencing spreads to the regulatory region of the gene. However, researchers say more molecular work is needed to explain the precise mechanism. They have not yet identified the specific DNA binding repressor that targets PTERV1. Testing what human RNA does Next, the team examined the function of LINC00662 in human neuronal cells. Its RNA accumulated mainly in the cytoplasm, outside the nucleus, and interacted with proteins associated with the cell skeleton. Several of these proteins participate in the formation of neurites, the extensions that become axons and dendrites. The researchers then used CRISPR interference to suppress LINC00662 in human stem cells and derived organoids. Unlike the chimpanzee deletion experiment, this method represses gene activity without cutting the target sequence. Two independently targeted guides produced similar results, strengthening the evidence that deletion of LINC00662 caused the observed changes. Organoids still formed and grew, and their broad cell type composition remained similar to that of controls. On day 15, the researchers detected few changes. However, by day 30, deleted organoids showed reduced staining for TUJ1, a marker associated with the neuronal skeleton, in regions containing developing neurons and their extensions. Another neuronal marker, MAP2, showed a downward trend that did not reach statistical significance. The results support a role for features associated with neurite maturation, rather than demonstrating a general failure of brain development. They also leave the exact interactions of RNA with its associated proteins not fully understood. Bright field images of human (HS1 and HS2) and chimpanzee (PT1 and PT2) organoids at days 5, 10, and 15. Right: Immunofluorescence images of day 15 neural rosettes stained for PAX6 (green), ZO1 (red), and DAPI (blue). (CREDIT: Johan Jakobsson et al, Science Advances 2026) An evolutionary clue with important limits The gene expression results added an unexpected complication. Analysis of 119,674 cell nuclei found increased expression of several programs related to axon and dendrite development after deletion of LINC00662. The authors suggest that this could represent compensation for poor neurite development, although that explanation remains unproven. The distinction is important because genetic activity and the resulting cellular structures do not always change in the same direction. LINC00662 appears to act in part through proteins once the RNA has been produced. Therefore, measuring RNA levels alone could miss important differences between species. Johan Jakobsson, Lund professor of molecular neurogenetics, said the work highlights the importance of understanding gene regulation. The findings do not establish that the ancient infection caused the split between humans and chimpanzees, nor do they reconstruct its effects on the population. Organoids also capture only selected stages and features of brain development. The next questions concern the extent to which similar viral insertions influence neuronal regulation and whether any contribute to brain diseases. These links remain research objectives, rather than proven medical applications. Testing additional remains could reveal whether LINC00662 represents an unusual example or one of many inherited consequences of ancient infections. Delve deeper into viral DNA and brain evolution. These studies explain inherited viral sequences, comparative neural models, and approaches to investigating brain evolution. Related stories
