Premachandra T et al. (2023), Population genomics and subgenome evolution of ...

XB-ART-59661

G3 (Bethesda) 2023 Feb 09;132:. doi: 10.1093/g3journal/jkac325.

Show Gene links Show Anatomy links

Population genomics and subgenome evolution of the allotetraploid frog Xenopus laevis in southern Africa.

Premachandra T, Cauret CMS, Conradie W, Measey J, Evans BJ.

???displayArticle.abstract???
Allotetraploid genomes have two distinct genomic components called subgenomes that are derived from separate diploid ancestral species. Many genomic characteristics such as gene function, expression, recombination, and transposable element mobility may differ significantly between subgenomes. To explore the possibility that subgenome population structure and gene flow may differ as well, we examined genetic variation in an allotetraploid frog-the African clawed frog (Xenopus laevis)-over the dynamic and varied habitat of its native range in southern Africa. Using reduced representation genome sequences from 91 samples from 12 localities, we found no strong evidence that population structure and gene flow differed substantially by subgenome. We then compared patterns of population structure in the nuclear genome to the mitochondrial genome using Sanger sequences from 455 samples from 183 localities. Our results provide further resolution to the geographic distribution of mitochondrial and nuclear diversity in this species and illustrate that population structure in both genomes corresponds roughly with variation in seasonal rainfall and with the topography of southern Africa.

???displayArticle.pubmedLink??? 36524354
???displayArticle.pmcLink??? PMC9911082
???displayArticle.link??? G3 (Bethesda)

???attribute.lit??? ???displayArticles.show???

	Fig. 1. Homoploid hybridization—gene flow between individuals with the same ploidy level (which in this example are both polyploid individuals)—can lead to differential introgression in subgenomes, either due to random events or in association with natural selection against some admixed genotypes as shown above. The parental populations (populations 1 and 2) are assumed to have two subgenomes with disomic inheritance (indicated with light blue and dark blue or red and purple, respectively, with color differences within each subgenome representing population subdivision). Gametogenesis in a hybrid individual produces gametes with a mosaic of population variation in each subgenome, depicted here as blocks that include entire chromosomes (in reality, recombination would likely generate smaller blocks of population-specific variation within each chromosome). If some backcrossed individuals have low fitness (gray) and high fitness individuals successfully breed in one parental population, different proportions of each subgenome may introgress, here shown as an extreme example of introgression only in subgenome 2.
	Fig. 2. Geographical distribution of X. laevis sampling localities in (a) southern Africa used for RRGS (red labels) and Sanger sequencing of mitochondria (circular pies) and (b) sub-Saharan Africa with circles, triangles, squares, and diamonds for X. laevis, X. petersii, X. poweri, and X. victorianus, respectively). For X. laevis, colors correspond to mitochondrial clades in Fig. 5. In (a), solid and dotted lines indicate the margins of the winter rainfall zone to the southwest and the summer rainfall to the northeast, respectively (Chase and Meadows 2007) and a red dashed line indicates the Great Escarpment (see main text). The locality of the X. gilli sample, not shown, falls within the range of the winter rainfall (blue) clade of X. laevis, and is listed in Supplementary Table 2.
	Fig. 3. Principal components analysis of RRGS data from X. laevis for the (a) L and (b) S subgenomes. For each analysis, the percentages of genotypic variation represented by the first and second principal components (PC1 and PC2, respectively) are indicated. Patterns of population structure are similar in each subgenome and samples cluster by rainfall zone (indicated with color in the left panel) and by locality (labeled in right panel).
	Fig. 4. Admixture plot using only the L subgenome (top) or only the S subgenome (bottom) for 2–11 ancestry components (K) labeled on the left side. The order of the samples is from left to right matches the order of samples in Supplementary Table 1.
	Fig. 5. Evolutionary relationships among mitochondrial sequences (a) are different from (b and c) the L and S subgenomes. In (a) and (b), black, gray, and white nodes indicate bootstrap values as indicated on the scale; numbers inside clades indicate the sample sizes of individuals; samples sizes of the S subgenome (not shown) are identical to those for the L subgenome. The scale bar indicates substitutions per site for the mitochondrial phylogeny; branch lengths of the phenogram are scaled by only variable positions and are not comparable to mitochondrial branch lengths. Samples in each clade of the mitochondrial tree are listed in Supplementary Table 2. Samples with the brown mitochondrial lineage in (a) were not present in the RRGS data (b). In the Neighbor-Net networks in (c), black bars indicate 0.01 substitutions per site for each subgenome.
	Fig. 6. Parsimony network of partial X. laevis mitochondrial sequences. Colors correspond to Fig. 2; the size of the nodes is proportional to the number of samples with each haplotype. Inferred nodes are indicated with black nodes; hashes on branches indicate changes between nodes that are not represented by a sampled sequence.

References [+] :

Alexander, Fast model-based estimation of ancestry in unrelated individuals. 2009, Pubmed