Genetic Adaptation Of Brain To High-Altitude Hypoxia In Indigenous Tibetans

ObjectivesBrain is the most important and most oxygen consuming organ,which consumes 20%oxygen supply of the whole body to maintain normal brain function.Whether it is environmental hypoxia exposure or pathological tissue hypoxia,it will first affect the brain.The indigenous Tibetans living at high altitude not only adapt to the hypoxia environment,but also form a series of stable hypoxia adaptive phenotypes including brain adaptation in the process of natural selection.In this study,we combine the multimodal magnetic resonance imaging technology and single nucleotide polymorphism(SNP)analysis approach to investigate the relationship between brain structure and function in Tibetans and the high-altitude(HA)adaptive genetic variants,and revealed the genetic mechanism of brain adaptation to hypoxia in Tibetans from the perspective of imaging genetics.Materials and methodsWe recruited 135 Tibetan subjects living on the plateau and 65 matched plain Han subjects,and collected the blood samples of the two groups for genotyping at 60 candidate single nucleotide polymorphisms loci selected from three HA strongly selective genes EGLN1,EPAS1 and PPARA.We screened out HA adaptive alleles and HA adaptive haplotypes.At the same time,we collected the multimodal brain MRI data(including T1-weighted imaging,diffusion tensor MRI,resting-state fMRI),a number of physiological indicators(hematology,pulmonary function,heart rate,blood pressure,oxygen saturation),neuropsychological test data(the Chinese revised version of the Wechsler Memory Scale subset test and the Rey-Osterrieth Complex Figure test).We analyzed the association of HA genetic variants and the morphology of gray matter,white matter structure,brain structural network and resting-state functional network of Tibetan subjects,using the analysis methods based on single genetic locus and multiple-loci genetic scores.Finally,we also carried out comprehensive correlation analysis of physiological indicators,neuropsychological test scores and large-scale brain network attributes.Results(1)Cerebral cortex analysis:The analysis results based on the HA adaptive haplotype showed that the HA adaptive groups had larger cortical surface area in the right superior frontal gyrus and left lingual gyrus than the non HA-diplotype groups.The analysis of multiple-loci genetic score showed that the subjects with more HA alleles had thicker cortical thickness in the left insula and the left lateral occipital gyrus,larger cortical volume in the right lingual gyrus and left isthmus cingulate,and larger surface area in the left inferior temporal gyrus.Furthermore,the thickness of left insular cortex was negatively correlated with cardiopulmonary function.(2)White matter analysis:The CC(rs6735530)homozygote group had a significant increase in FA value in the bilateral superior longitudinal fasciculus,the left forceps minor,the right pons,the right anterior thalamic radiation and the right arnold's fasciculus compared with the T carrier group;Compared with GG homozygous group,the C allele carrier group(rs4253778)had a significant increase in FA value in the right superior longitudinal fasciculus,left forceps minor and the anterior thalamic radiation,and had a significant decrease in MD value in the left inferior longitudinal fasciculus and bilateral inferior fronto-occipital fasciculus.The white matter fiber bundles with positive correlation between FA value and EGLN1 multiple-loci genetic score were mainly located in the bilateral temporal lobe,right precentral gyrus and angular gyrus,left postcentral gyrus,left insula and frontal regions.(3)Brain network analysis:The analysis of independent SNP showed that 14 HA adaptive groups had significantly lower centrality than that of the non-adaptive groups in the left anterior orbital gyrus.And some HA adaptive groups showed lower erythrocyte level and heart rate,higher FEV1/FVC ratio and arterial oxygen saturation than the non-adaptive groups.Further analyses of multiple-loci genetic score revealed significant correlations with structural brain networks with respect to local interconnectivity of the whole networks,nodal centrality in several frontal regions,inter-module communication between right frontal and parietal module and left occipital module,and connectivity strength of a subnetwork predominantly involving in intra-module edges in right temporal and occipital module.The above results are obviously dependent on gene loci,network type or topological scales.There was no significant association between HA-adaptive genetic variants and neuropsychological performance or brain functional network attributes.ConclusionsThe HA-adaptive genetic variants have wide correlations with the morphology of cerebral cortex,the structure of white matter and the connection mode of brain structural network in indigenous Tibetan highlanders.This association has genetic loci dependence and brain region dependence.The correlated brain areas are mainly located in frontal lobe,insular lobe and visual function area.The subjects carried more HA adaptive alleles showed the optimized brain phenotypes such as thicker cortical thickness,larger cortical surface area,higher FA value of white matter and so on in local brain areas.All these brain structural characteristics are more protective against hypoxic brain injury,which represents the evolutionary direction of brain in the extreme environment at high altitude.The brain adaptive phenotypes are closely related to the unique adaptive physiological phenotypes of Tibetans.