Linked OXTR Variants Are Associated With Social Behavior Differences in Bonobos ( Pan paniscus )

Single-nucleotide polymorphisms (SNPs) in forkhead box protein P2 ( FOXP2 ), oxytocin receptor ( OXTR ), and arginine vasopressin receptor gene 1A ( AVPR1A ) have been associated with linguistic and social development in humans, as well as symptom severity in autism. Studying biobehavioral mechanisms in the species most closely related to humans can provide insights into the origins of human communication, and the impact of genetic variation on complex behavioral phenotypes. Here, we tested the hypothesis that similar genetic factors underlie social communication differences in both bonobos ( Pan paniscus ) and humans. We analyzed Sanger sequencing results to determine if bonobos exhibit individual variation at 10 loci across FOXP2 , OXTR , and AVPR1A that have been implicated in human social development and behavior. We identified a novel variant in bonobo FOXP2 , as well as three novel variants in bonobo OXTR that were 19-184 base pairs away from the target human SNPs. We also found a linked SNP combination (TGA) across the 3 novel bonobo OXTR sites at high frequency (65%) in the study population, including 6 homozygous bonobos. When comparing the combined OXTR genotypes, we found significant group differences in social behavior; bonobos with two copies of the TGA combination were more social than bonobos with one copy and bonobos with zero copies of the TGA combination. Taken together, our findings suggest that these OXTR variants may influence individual-level social behavior in bonobos and support the notion that linked genetic variants are promising biomarkers for differences in human social communication. Revision Summary The original manuscript version underwent peer-review at Frontiers in Behavioral Neuroscience. As a part of this process, a reviewer requested that the sequence files be included in the manuscript. While preparing these files, Dr. Hudson identified several errors in his work regarding the individual bonobo genotypes reported in the original version of the manuscript. Upon discovery of the errors, Dr. Skiba and Dr. Taglialatela conducted a new genetic analysis using the raw files and subsequently corrected the misreported information - updating the editorial team and including the raw sequences with the revised manuscript. In addition to requested edits for improvement and clarity by the peer-reviewers, this revised manuscript corrects several errors in the human and bonobo reference genomes, the SNP locations, and the individual genotypes reported in the original manuscript. Information in the figures, main text, and supplementary materials has been corrected in this revised version. Dr. Hudson and his lab member, Mr. Hansen, were unable to identify how the errors in data reporting occurred, and neither party contributed to the revised version of this manuscript. The co-author list has been adjusted to reflect contributions to the corrected analyses and revised manuscript. We are deeply grateful to the editorial team and reviewers at Frontiers in Behavioral Neuroscience for their thorough peer-review of this work - improving the overall quality of the manuscript and leading to the discovery and correction of substantial errors in the original version.