Comparative Chloroplast Genomics of Gossypium Species: Insights Into Repeat Sequence Variations and Phylogeny

Cotton is one of the most economically important fiber crop plants worldwide. The genus <i>Gossypium</i> contains a single allotetraploid group (AD) and eight diploid genome groups (A-G and K). However, the evolution of repeat sequences in the chloroplast genomes and the phylogenetic relationships of <i>Gossypium</i> species are unclear. Thus, we determined the variations in the repeat sequences and the evolutionary relationships of 40 cotton chloroplast genomes, which represented the most diverse in the genus, including five newly sequenced diploid species, i.e., <i>G. nandewarense</i> (C_1-n), <i>G. armourianum</i> (D_2-1), <i>G. lobatum</i> (D₇), <i>G. trilobum</i> (D₈), and <i>G. schwendimanii</i> (D₁₁), and an important semi-wild race of upland cotton, <i>G. hirsutum</i> race <i>latifolium</i> (AD₁). The genome structure, gene order, and GC content of cotton species were similar to those of other higher plant plastid genomes. In total, 2860 long sequence repeats (>10 bp in length) were identified, where the F-genome species had the largest number of repeats (<i>G. longicalyx</i> F₁: 108) and E-genome species had the lowest (<i>G. stocksii</i> E₁: 53). Large-scale repeat sequences possibly enrich the genetic information and maintain genome stability in cotton species. We also identified 10 divergence hotspot regions, i.e., <i>rpl33-rps18, psbZ-trnG</i> (<i>GCC</i>), <i>rps4-trnT</i> (<i>UGU</i>), <i>trnL</i> (<i>UAG</i>)<i>-rpl32, trnE</i> (<i>UUC</i>)<i>-trnT</i> (<i>GGU</i>), <i>atpE, ndhI, rps2, ycf1</i>, and <i>ndhF</i>, which could be useful molecular genetic markers for future population genetics and phylogenetic studies. Site-specific selection analysis showed that some of the coding sites of 10 chloroplast genes (<i>atpB, atpE, rps2, rps3, petB, petD, ccsA, cemA, ycf1</i>, and <i>rbcL</i>) were under protein sequence evolution. Phylogenetic analysis based on the whole plastomes suggested that the <i>Gossypium</i> species grouped into six previously identified genetic clades. Interestingly, all 13 D-genome species clustered into a strong monophyletic clade. Unexpectedly, the cotton species with C, G, and K-genomes were admixed and nested in a large clade, which could have been due to their recent radiation, incomplete lineage sorting, and introgression hybridization among different cotton lineages. In conclusion, the results of this study provide new insights into the evolution of repeat sequences in chloroplast genomes and interspecific relationships in the genus <i>Gossypium</i>.