A comparative analysis of the complete chloroplast genome sequences of four peanut botanical varieties

DNA extraction and sequencing

Four representative A. hypogaea varieties (var. hypogaea, var. hirsuta, var. fastigiata, and var. vulgaris) were collected from Shandong Peanut Research Institute, Qingdao, China. China has become the largest producer of cultivated peanut in the world (Yu, 2008), and these four main botanical varieties have been cultivated in China for more than 500 years. The seedlings were grown using hydroponic methods. The cp DNA was isolated from fresh leaves (>5 g) of 3- to 4-week-old seedlings using the Plant Chloroplast DNAOUT Kit (Bjbalb, Beijing, China). The quality of cp DNA samples was checked by agarose gel electrophoresis with Super GelRed (US Everbright Inc., Suzhou, China). Libraries with an average length of 350 bp were constructed using the NexteraXT DNA Library Preparation Kit (Illumina, Shanghai, China). The quality of the libraries was checked by GeneRead DNA QuantiMIZE Assay Kit (Qiagen, Duesseldorf, Germany). Sequencing was performed on the Illumina HiSeq Xten platform (Illumina, Shanghai, China), and the average length of the generated reads was 150 bp.

Data assembly and annotation

The quality of the raw paired-end reads was assessed by FastQC v0.11.3 (Andrews, 2014). All raw data for four A. hypogaea varieties were filtered based on the following rules: (1) adapter trimming; (2) quality control; each read has <5% unidentified nucleotides and >50% of its bases with a quality value of >20. This filtration was carried out using Cutadapt v1.7.1 (Martin, 2011). The high-quality data were then assembled into contigs using the de novo assembler SPAdes v3.9.0 (Nurk et al., 2013), and these contigs were further assembled into complete cp genome using NOVOPlasty (Dierckxsens, Mardulyn & Smits, 2017). The assembled data were checked against the published complete cp genome of A. hypogaea (GenBank accession no. KX257487, Prabhudas et al., 2016). The cp genes were annotated using the DOGMA tool with default parameters (Wyman, Jansen & Boore, 2004). The cp genome images were drawn with OGDraw v1.2 (Lohse, Drechsel & Bock, 2007).

Variation detection and evolutionary relationship analysis

Multiple sequence alignment was generated using VISTA and Mauve v2.3.1 software (Frazer et al., 2004; Darling, Mau & Perna, 2010) and was checked manually when necessary. All alignments were visualized using the VISTA viewer program (Mayor et al., 2000). Single nucleotide polymorphisms (SNPs) were identified by Mauve v2.3.1. The insertions/deletions (InDels) were retrieved from the sequence alignments using the mVISTA package. An InDels image including 10 bp up- and downstream was then generated. Simple sequence repeats (SSRs) were isolated from all filtered InDels. Repeat sequences with repeating units of 2–6 bp that repeated no fewer than three times were considered as SSRs.

The genetic relationship of the four peanut cp genomes together with two available peanut cp genome sequences (GenBank accession no. KX257487 and KJ468094; Prabhudas et al., 2016; Choi & Choi, 2017) were examined by constructing a minimum evolutionary (ME) tree using MEGA v6 with default parameters (Tamura et al., 2011). The cp genome sequences from four other related species (Robinia pseudoacacia, Ceratonia siliqua, Leucaena trichandra, and Senna tora) of Fabaceae were used as outgroups (CSI-BLAST E-value < 10⁻⁶).

Assembly and validation of cp genomes

More than 1 GB raw sequencing data per sample was generated from high-throughput sequencing. After cleaning and trimming, 22,511,400 (var. vulgaris) to 62,087,400 (var. hirsuta) paired-end reads were acquired, which were then mapped separately to the reference cp genome, attaining coverage amounts of 143× to 396×. After performing de novo and reference-guided assembly with minor modifications, we acquired four complete cp genome sequences for A. hypogaea var. hypogaea, var. hirsuta, var. fastigiata, and var. vulgaris (Fig. 1; Table 1).

For each of the assembled cp genome sequences, a .sqn file that was generated by the Sequin software (https://www.ncbi.nlm.nih.gov/projects/Sequin/), submitted to GenBank and acquired the following accession numbers: MG814006 for var. fastigiata, MG814007 for var. hirsuta, MG814008 for var. hypogaea, and MG814009 for var. vulgaris. Users can download the data for research purposes only when referencing this paper.

Genetic structure of the peanut cp genome

These four acquired peanut cp genomes were found to have the classical quadripartite structure of land plant cp genomes that comprises a LSC, a SSC, and two IR (A/B) regions. The sequence lengths among the four cp genomes ranged from 156,354 to 156,878 bp. The size varied from 85,900 bp (var. hirsuta) to 86,196 bp (var. fastigiata) in the LSC region, from 18,796 bp (var. hypogaea, var. hirsuta, and var. vulgaris) to 18,874 bp (var. fastigiata) in the SSC region and from 25,806 bp (var. hypogaea) to 26,091 bp (var. hirsuta) in the IR (A/B) region (Table 1). A total of 110 genes were identified from the cp genome: four ribosomal RNA (rRNA) genes, 76 protein-coding genes, and 30 transfer RNA (tRNA) genes (Table 2). Among the 110 identified genes, six protein-coding genes, six tRNA genes, and four rRNA genes were distributed in the IR (A/B) regions. The cp genome consisted of 55.66% coding regions and 44.34% non-coding regions including both intergenic spacers and introns. The overall GC content of the cp genomic sequences was 36.3–36.4%, and the GC contents of the LSC, SSC, and IR (A/B) regions were 33.8%, 30.2–30.3%, and 42.8–42.9%, respectively (Table 2).

Variation among the cp genomes

Among the four acquired peanut cp genome sequences, there was no difference at the junction positions (Fig. 2). A total of 46 SNPs were found within the quadripartite structural region. VISTA-based identity plots illustrated the hotspot regions of genetic variation among the cp genomes (Fig. 3). As expected, non-coding sequences exhibited more variation than the coding sequences, and greater amounts of substitutions were found in the trnI-GAU intron (25 SNPs) and the ycf3-psaA spacer (eight SNPs) regions. The only identified non-synonymous was located within the psaA gene. The hydrophobic amino acid Tyr in var. hypogaea, var. fastigiata, and var. vulgaris was replaced by the hydrophilic amino acid Asn in var. hirsuta.

A total of 26 InDels were identified: 13 were located in spacers, nine were in introns, and four were in genes; 15 were in the LSC region, two were in the SSC region, and nine were in IR (A /B) regions (Fig. S1). Among these InDels, large InDels (>50 bp) were found in the psbK–trnQ intergenic spacer, the trnL intron, and ycf1. Meanwhile, we identified six SSR regions (sequence identity >90%): four A stretches and one T stretches ranging from 7 to 16 bp, as well as one with a CTAG repeat motif. No C or G stretches were identified. Moreover, InDels in the ycf1 and the ycf2 regions represent frameshift mutations: the 63 bp-insertion at the end of the ycf1 gene led to a longer amino acid sequence in var. fastigiata, while a 18 bp-deletion was found in the middle of IR (A/B) ycf2 gene regions in var. hypogaea.

Genetic relationship analysis

Due to low genetic diversity, the whole cp genome sequences were used to construct an evolutionary tree based on ME algorithms. The results showed that these peanut cp genomes clustered into a monophyletic branch, while the four outgroup species were clustered into another branch. Among the six analyzed peanut cp genomes, var. hirsuta is relatively different from the rest and constitute a basal clade (Fig. 4). The high-support values (>99%) were shown above the nodes.