Novel insight into the genetic basis of high-altitude pulmonary hypertension in Kyrgyz highlanders

Abstract

The Central Asian Kyrgyz highland population provides a unique opportunity to address genetic diversity and understand the genetic mechanisms underlying high-altitude pulmonary hypertension (HAPH). Although a significant fraction of the population is unaffected, there are susceptible individuals who display HAPH in the absence of any lung, cardiac or hematologic disease. We report herein the analysis of the whole-genome sequencing of healthy individuals compared with HAPH patients and other controls (total n = 33). Genome scans reveal selection signals in various regions, encompassing multiple genes from the first whole-genome sequences focusing on HAPH. We show here evidence of three candidate genes MTMR4, TMOD3 and VCAM1 that are functionally associated with well-known molecular and pathophysiological processes and which likely lead to HAPH in this population. These processes are (a) dysfunctional BMP signaling, (b) disrupted tissue repair processes and (c) abnormal endothelial cell function. Whole-genome sequence of well-characterized patients and controls and using multiple statistical tools uncovered novel candidate genes that belong to pathways central to the pathogenesis of HAPH. These studies on high-altitude human populations are pertinent to the understanding of sea level diseases involving hypoxia as a main element of their pathophysiology.

Fig. 1

Geographic location of Kyrgyz population (box) relative to other major populations from the 1000 Genome Project used in the Admixture analysis of current study (a). YRI Yoruba in Ibadan, Nigeria, CEU Utah Residents with Northern and Western European Ancestry, JPT Japanese in Tokyo, Japan, CHB Han Chinese in Beijing, China, SAS (PJL), Punjabi from Lahore, Pakistan. b Topography of Kyrgyz republic with > 90% of the area at altitude > 3000 m above sea level. c Mean pulmonary artery pressure (mPAP) in healthy Kyrgyz highlanders (No-HAPH) is significantly lower than the age-matched HAPH patients. Error bar represents ± standard error

Fig. 2

Admixture and PCA depicting the genetic relatedness of Kyrgyz Population (KGZ), which includes HAPH, No-HAPH, HR and NR groups, to other major populations. a Admixture analysis shows that Kyrgyz population consists of major genetic proportion from East Asian lineage with minute contributions from the European genetic ancestry. b PCA reveals that the Kyrgyz population is located between SAS and EAS but more closely related to EAS. c Within Kyrgyz cluster the subjects are randomly distributed. SAS South Asian, EAS East Asian

Fig. 3

Layout of genetic variation in the top selected interval-1 in the HAPH, No-HAPH and JPT (outgroup) populations (a). The haplotype frequencies among No-HAPH is higher compared with HAPH and JPT. b Frequency of the top selected haplotype (interval-1) among Kyrgyz highlanders and populations from the 1000 Genome Project. The y axis is frequency of one of the SNPs (out of 147 fully linked SNPs) of the selected haplotype. c A representative box plot showing the genotype of an eQTL SNP in interval-1 and the respective expression of gene in Lung (P = 1.5e-11)

Fig. 4

Layout of genetic variation in selected interval-2 in the HAPH, No-HAPH and JPT (outgroup) populations (a). Haplotype frequencies among No-HAPH are higher compared with HAPH and JPT. b Frequency of a SNP from a set of perfectly linked SNP of interval-2 among Kyrgyz highlanders and populations from the 1000 Genome Project (b). Box plot of SNPs, rs11637876 (c) and rs12913583 (d) from interval-2 that was identified as eQTL for the expression of TMOD3 (P = 5.5e–7 for both SNPs)