Assembly And Evolutionary Analysis Of The Soybean Mitochondrial And Chloroplast Genomes

Cultivated soybean(Glycine max(L.)Merr.)is a major crop domesticated from the wild progenitor Glycine soja Sieb.& Zucc.approximately 5000 years ago,providing high-quality vegetable oil,protein and livestock feed.The architecture and genetic diversity of its nuclear genome are extensively studied,providing valuable resources for soybean evolutionary and functional genomic research.By contrast,researches on soybean cytoplasmic genomes are limited.The nuclear genome of cultivar ?Williams 82‘(Wm82)is the first assembled and commonly used reference in soybean,but its complete cytoplasmic genomes have not been reported.Here,we de novo assembled the mitogenome of the cultivar Wm82(Wm82＿mt DNA)with Illumina PE300 deep sequencing data,and verified it with polymerase chain reaction(PCR)and Southern blot analyses.Wm82＿mt DNA maps as two autonomous circular chromosomes(370,871 bp Chr-m1 and 62,661 bp Chr-m2).Its structure is extensively divergent from that of the mono-chromosomal mitogenome reported in the landrace?Aiganhuang‘(AGH＿mt DNA).Synteny analysis showed that the structural variations(SVs)between two genomes are mainly attributed to ectopic and illegitimate recombination,but the core gene composition and sequence between them are not very different.Moreover,Wm82＿mt DNA and AGH＿mt DNA each possess six and four specific regions,which are absent in their counterparts and likely result from differential sequence-loss events.Further analysis revealed that low-frequency ectopic recombination mediated by small repeats(SR)is the key to the interconversion of mono-and bi-chromosomal configurations between the two soybean mt DNA,and is also an important genetic basis for structural evolution of soybean mt DNA.Mitogenome SV was further studied in 39 wild and 182 cultivated soybean accessions distributed world-widely with PCR/Southern analyses or a comparable in silico analysis.The results classified both wild and cultivated soybeans into five cytoplasmic groups,named as GSa–GSe and G1–G5;Wm82 and AGH belong to G1 and G5,respectively.Notably,except for members in GSe and G5,all accessions carry a bi-chromosomal mitogenome with a common Chr-m2.Phylogenetic analyses based on mt DNA structures and chloroplast gene sequences both inferred that G1– G3,representing >90% of cultigens,likely inherited cytoplasm from the ancestor of domestic soybean,while G4 and G5 likely inherited cytoplasm from wild soybeans carrying GSa-and GSe-like cytoplasm through interspecific hybridization.Therefore,it was concluded that there are at least three maternal donors with different cytoplasmic types for modern cultivated soybean.Interestingly,the proportion of G3 and its derivative G1 in domestic soybean remarkably increased during selective breeding,from 58% in landraces(LR)to 86% in improvement cultivars(IM),indicating that the genetic architecture of the cytoplasm is closely related to soybean improvement.In addition,the length of 10 representative chloroplast genomes(cp DNAs)in the cultivated soybean range from 152,217 bp to 152,227 bp,and the GC content is 35.4%.Except for ?Medium Green‘(MG)of G2,which have a mutated psb M gene,all cp DNAs contain 128 genes.Genome comparison analysis within same cytoplasm show that the four cp DNA sequences of G1 and G3 are identical,and the two cp DNA sequences of G5 are identical,but the cp DNAs within G2 or G4 have genetic diversity.A total of 77 single nucleotide polymorphism(SNP)and 48 insertion-deletion(Indel)are identified in the large single copy(LSC)and small single copy(SSC)regions of these 10 cp DNAs.These polymorphic loci provide valuable information for designing molecular markers to distinguish different cytoplasmic groups of cultivated soybean.Based on the reported four soybean cp DNAs and new assembled 10 cp DNAs in this study,the nucleotide hypervariable regions of wild,cultivated and semi-wild soybean(Glycine gracilis)within subgenus Soja mainly appear in three genes and two intergenic regions,and the type and quantity of their chloroplast simple sequence repeat(cp SSR)are consistent with the hypothesis that multiple origins of the cytoplasm of modern cultivated soybenas.After multiple sequence alignment and RNAseq data analyses,the annotation information of soybean chloroplast ycf1 and ndh B genes are revised.The phylogenetic trees indicate that the reported semi-wild soybean is more closely related to cultivated soybean G4,and ?PI437654‘ is a member of G4.Based on the results of PCR and bioinformatics analyses,the four cleaved amplified polymorphism sequence(CAPS)markers developed from cp DNA can effectively screen the ?Zhongdou 19‘(ZD)-type cytoplasmic male sterility(CMS)cytoplasm.Moreover,the editing of soybean chloroplast RNA is similar to that observed in other flowering plants,and the ndh B gene has the most editing sites.The editing of C to U frequently occures at the second position of the chloroplast m RNA codon,and the resulting amino acid changes are more non-synonymous substitution.Chloroplast RNA editing in soybean has differences and specificity in different tissues and developmental stages,among which the number of RNA editing in mature leaves is the largest,and two specific U to C RNA editing sites appear in psb A and trn Q-UUG genes.These data provide high-quality reference sequences for the cytoplasmic genome of soybean,provide more comprehensive and accurate information for the study of soybean genome,expand our understanding of the genetic behavior and structure of soybean mt DNA.Overall,results from the present study provide evidence that cytoplasmic genetic structure may affect soybean improvement,and shed light on soybean mt DNA evolution and domestication history.