2.1 The leafy sweet potato cpDNA and mtDNA assembly and annotation
The leafy sweet potato full-length cpDNA sequence is 161387 bp long. It is divided into four sections: a large single-copy region (LSC, 87597 bp), a small single-copy region (SSC, 12052 bp), and two inverted repeat regions (IRA and IRB, 30869 bp). (Fig. 1a). Gene annotation results showed that the leafy sweet potato cpDNA had 132 functional genes, including 87 protein-coding genes, 8 rRNA genes and 37 tRNA genes. In the IR region, 18 genes were duplicated, including 7 protein-coding genes (ycf1, ycf2, ycf15, ndhB, ndhH, rps7, and rps15), 7 tRNA genes (trnI-CAU, trnL-CAA, trnV-GAC, trnI) -GAU, trnA-UGC, trn (rrn4.5, rrn5, rrn16 and rrn23). The total GC content of leafy sweet potato cpDNA was 37.54%, and the AT content was 62.46%.
Statistical analysis revealed that 20 genes in the leafy sweet potato cpDNA contained introns. There were 11 protein-coding genes and 7 tRNA genes with one intron and two protein-coding genes (pafI and clpP) with two introns (Table S1). There are two copies of rps12, each with three exons; the two copies share the first exon, located in the LSC region, and the other two exons in the IR region. The full-length mtDNA sequence of leafy sweet potato is 269578bp. According to the gene annotation results, the leafy sweet potato mtDNA contains 70 functional genes, including 41 protein-coding genes, 6 rRNA genes, and 24 tRNA genes (Fig. 1b). The coding genes include 4 ATP synthase genes, 8 NADH dehydrogenase genes, 5 Cytochrome c biogenesis, 3 Cytochrome c oxidase, 13 Ribosomal proteins, 1 Maturases, 1 Ubichinol cytochrome c reductase, and 1 Transport membrane protein, 1 Succinate dehydrogenase and 1 RNA polymerase, of which nad2, nad4, nad7, cox1, cox2, rps3 and rps10 contain introns (Table S2). The leafy sweet potato mtDNA had a total GC content of 44.1% and an AT content of 55.9%. A is 27.9%, C is 22.2%, G is 21.9%, and T is 28.0%.
Even if the codons encoding amino acids are the same for different protein structures of organisms, the frequency of use of synonymous codons for amino acids is not equal. This unequal use of synonymous codons is called codon preference. Sex (Relative Synonymous Codon Usage, RSCU). The total length of protein-coding genes in leafy sweet potato mtDNA is 34710bp, most of which begin with typical ATG codons. However, nad4L and rps10 begin with ACG, mttB and rps3 begin with TTG, and cob begins with GTG (Table S1). There are 32 codons in mtDNA with RSCU values greater than 1.00, the majority of which end with A or T, only three of which end with G (ATG, TTG, TGG), the number of codons encoding leucine (Leu) is the greatest, 1198, accounting for 10.54%; cysteine (Cys) appeared the fewest, 173, accounting for 1.52% (Table S3).
The chloroplast gene sequences (Accession number: OM808940) and mitochondrial gene sequences (Accession number: OM808941) were assembled, annotated, and submitted to GenBank.
2.2 SNP analysis of chloroplast genome in sweet potato varieties
Intraspecific SNP analysis of the chloroplast genome of sweet potato cultivar 'Fucaishu 18' showed that there were 199 SNP sites 118 of which were located in the coding region. The coding region contained matK, rpoC2, psaB, accD, psbL, rps8, ycf1, ycf2, ndhB, ndhC, ndhE and ndhF genes. Other SNP sites were located in the non-coding region. Compared with other sweet potato cultivars, the chloroplast genome of 'Fucaishu 18' had 121 deletions, of which 37 were SSRs, 146 insertions, of which 66 were SSRs, and 7 long substitutions (Supplementary File 1).
2.3Repeat sequences analysis and RNA editing analysis
SSR (Simple Sequence Repeats) is a tandem repeat sequence consisting of 1 to 6 nucleotide repeat units. The single-copy sequences flanking each SSR are generally relatively conserved. The leafy sweet potato organelle genome contained 69 SSR sites, including 54 SSR sites in cpDNA and 15 in mtDNA (Table S4).
RNA editing is the insertion, deletion, or substitution of nucleotides in the mRNA produced by gene transcription, resulting in a sequence that is not complementary to the gene's coding sequence, and the amino acid composition of the protein produced by translation differs from the gene sequence information. RNA editing occurs in organelles such as mitochondria and chloroplasts, primarily in the mitochondrial genome, and is closely related to organelle function. In nature, mitochondrial transcripts of flowering plants have 300–500 editing sites, whereas chloroplast transcripts have only 30–50 editing sites. According to this study, the mitochondrial genome of sweet potato contains 492 RNA editing sites but only 48 RNA editing sites in the chloroplast genome (Figure S1).
2.4 Organellar phylogenetic relationships
We downloaded 40 Ipomoea and two outgroup chloroplast genomes from GenBank (https://www.ncbi.nlm.nih.gov/genome/browse/) to determine I.batatas' phylogenetic position. As shown in Fig. 2, all the nodes in the generated tree had 100% bootstrap support values. The phylogenetic tree supports the close relationship between I.batatas, I.tabascana and I.trifida. Overall, our chloroplast genome analysis findings provide a valuable foundation for future studies of the phylogenetic affinities of Ipomoea species.
There are few mitochondrial genome data of Ipomoea in the NCBI database. We used mitochondrial genome data from several other plants to construct phylogenetic trees. As illustrated in Fig. 3, 3 of the 11 nodes in the generated tree had less than 40% support values. Argyreia velutina did not belong to Ipomoea but was clustered with other Ipomoea plants, indicating that the mitochondrial genome evolved faster than the chloroplast genome.
2.5 Sequence transfer between genomes
Sequence transfer can involve the transfer of genetic material from one cell to another: between different bacterial species via conjugation, transduction, and transformation; between different single-celled organisms, such as bacteria and yeasts; or between organisms that are even more distantly related, such as bacteria and insects, fungi and animals, or plants and fungi.
2.5.1 Length differences of transferred sequences
We evaluated the frequency and pattern of sequence transfer in organelles genomes by having complete and high-quality assemblies for I. Batatas' nuclear genomes, chloroplast genomes and mitochondrial genomes.
The transfer locations were classified into five categories based on the chloroplast genome's sequence content and annotations: exon, intron, rRNA, tRNA, and CDS regions. Reflecting the results from the length of transfers analysis, all transfer location types were predominantly occupied by sequences of chloroplast origin(Fig. 4). Table 2 shows that the length of gene transfer fragments in the mitochondrial genome ranges from 153 bp to 4013 bp, with only 3 fragments less than 1000 bp occurring in the CDS intervals of ycf2, ycf15, and psbZ. Sequence transfer of long segments also occurred in the sweet potato organelle genome species. Among them, two fragments of 3581bp are located in the CDS region of ycf2, and the longest fragment of 4013bp is located in the interval between psaA and psaB (Table 2).
By comparing the mitochondrial genome and nuclear genome of sweet potato, we can find that 171 mitochondrial genome sequences come from the horizontal transfer of nuclear genome (Supplementary File 2).These sequences cover all the chromosomal genome sequences of sweet potato.Among them, the most 19 sequences were from chromosome 8, accounting for 11. 1%, 18 sequences from chromosome 5, accounting for 10. 5%, and the least were from chromosome 4 and chromosome 14, with only 5 and 6 sequences respectively.
2.5.2 PCR Determining gene sequence transfer
We used the same upstream primer but different downstream primers to amplify the homologous sequence by PCR to confirm the gene transfer sequence. The amplified sequence is then sequenced to determine whether it is from the chloroplast or mitochondrial genome(Fig. 5).
The results demonstrated that the 6 pairs of primers could amplify the corresponding bands from sweet potato total DNA. After comparison, it was discovered that 3 were from the sweet potato chloroplast genome sequence and the other 3 were from the sweet potato mitochondrial genome. The results were in line with expectations, indicating that the chloroplast genome and mitochondrial genome of sweet potato. It shows a phenomenon of gene transfer(Supplementary File 3).