Demographic histories of four spruce (Picea) species of the Qinghai-Tibetan Plateau and neighboring areas inferred from multiple nuclear loci

• Published in: Molecular biology and evolution 2010-05, Vol.27 (5), p.1001-1014
• Main Authors: Li, Yuan, Stocks, Michael, Hemmilä, Sofia, Källman, Thomas, Zhu, Hongtao, Zhou, Yongfeng, Chen, Jun, Liu, Jianquan, Lascoux, Martin
• Format: Article
• Language: English
• Subjects: Bayes Theorem; Biologi; Biology; Cell Nucleus - genetics; divergence time; Ecosystem; effective population size; Gene loci; Genetic diversity; Genetic Loci - genetics; Genetics, Population; Geographical distribution; Geography; Haplotypes; introgression; Linkage Disequilibrium - genetics; Models, Genetic; Molecular biology; NATURAL SCIENCES; NATURVETENSKAP; Nucleotides - genetics; Picea; Picea - genetics; Plant populations; Polymorphism; Polymorphism, Genetic; Population Dynamics; Population growth; Population structure; Qinghai Tibetan Plateau; Recombination, Genetic - genetics; speciation; Species Specificity; Tibet; Trees
• ISSN: 0737-4038; 1537-1719; 1537-1719
• DOI: 10.1093/molbev/msp301

Nucleotide variation at 12-16 nuclear loci was studied in three spruce species from the Qinghai-Tibetan Plateau (QTP), Picea likiangensis, P. wilsonii, and P. purpurea, and one species from the Tian Shan mountain range, P. schrenkiana. Silent nucleotide diversity was limited in P. schrenkiana and high in the three species from the QTP, with values higher than in boreal spruce species, despite their much more restricted distributions compared with that of the boreal species. In contrast to European boreal species that have experienced severe bottlenecks in the past, coalescent-based analysis suggests that DNA polymorphism in the species from the QTP and adjacent areas is compatible with the standard neutral model (P. likiangensis, P. wilsonii, and P. schrenkiana) or with population growth (P. purpurea). In order to test if P. purpurea is a diploid hybrid of P. likiangensis and P. wilsonii, we used a combination of approaches, including model-based inference of population structure, isolation-with-migration models, and recent theoretical results on the effect of introgression on the geographic distribution of diversity. In contrast to the three other species, each of which was predominantly assigned to a single cluster in the Structure analysis, P. purpurea individuals were scattered over the three main clusters and not, as we had expected, confined to the P. likiangensis and P. wilsonii clusters. Furthermore, the contribution of P. schrenkiana was by far the largest one. In agreement with this, the divergence between P. purpurea and P. schrenkiana was lower than the divergence of either P. likiangensis or P. wilsonii from P. schrenkiana. These results, together with previous ones showing that P. purpurea and P. wilsonii share the same haplotypes at both chloroplast and mitochondrial markers, suggest that P. purpurea has a complex origin, possibly involving additional species.