The plant materials of both D. lactucoides and H. leptocarpum were harvested from the Tibet Autonomous Region, China (Table 1) and dried in silica gel for analysis. The leaves were identified by researcher Guoyue Zhong of Jiangxi University of Traditional Chinese Medicine (JJUCM) as Papaveraceae plants, and the voucher specimens were deposited in the herbarium of JJUCM (Fig. 1).

Fig. 1.

Primitive plant pictures of D. lactucoides and H. leptocarpum.

The genomic DNA was extracted from the leaves of D. lactucoides and H. leptocarpum using a plant genomic DNA kit (Tiangen Biochemical Technology (Beijing) Co., Ltd., Beijing, China). The quality of genomic DNA was detected by a spectrophotometer and 1% agarose gel electrophoresis. Samples with qualified DNA concentration greater than 20 ng$\cdot$µL${}^{-1}$ were lysed by an ultrasonic instrument, and subjected to fragment DNA purification and end repair. Thereafter, libraries with an insert size of 350 bp were prepared according to Illumina’s standard procedure for genomic DNA library preparation, and then sequenced on the Illumina NovaSeq 6000 high-throughput sequencing platform (Illumina, San Diego, CA, USA). The sequencing was performed by Novgene Biotech Co., Ltd. (NanJing, Jiangsu, China).

The raw sequencing data were filtered using Trimmomatic software [14] (Aachen and Institute of Bio- and Geosciences: Plant Sciences, Forschungszentrum Jülich, Leo-Brandt-Straße, Jülich, Germany) to remove splice sequences and low-quality base sequences. The filtered plastomes were assembled from scratch using GetOrganelle 1.7.5 software [15] (Chinese Academy of Sciences, Kunming, Yunnan, China) and were annotated using CPGAVAS2 software [16] (Faculty of Technology, University of Bielefeld, Bielefeld, Germany) (http://47.96.249.172:16019/analyzer/annotate/). The plastome physical structure mapping was performed using OGDRAW software [17] (Max-Planck-Institut für Molekulare Pflanzenphysiologie, Am Mühlenberg, Germany) (https://chlorobox.mpimp-golm.mpg.de/OGDraw.html).

Simple sequence repeat (SSR) was screened using the online tool MISA v2.1 (Leibniz Institute of Plant Genetics and Crop Research, Gatersleben, Germany) (https://webblast.ipk-gatersleben.de/misa/) software with parameters set to 10 for single nucleotide SSR repeats, 5 for dinucleotide SSR repeats, 4 for single and trinucleotide SSR repeats, and 3 for tetranucleotide and pentanucleotide SSR repeats. The REPuter software [18] (Faculty of Technology, Bielefeld, Germany) (http://bibiserv.techfak.uni-bielefeld.de/reputer/) was used to annotate the plastome repeats, including forward, reverse, complementary, and palindromic repeats.

CodonW 1.4.2 software [19] (Paul Sharp lab, Dept of Genetics, University of Nottingham, Nottingham, UK) was used to calculate the codon preference (Relative synonymous codon usage, RSCU), and PAML v4.973 software [20] (Department of Biology, Galton Laboratory, London, UK) was used to calculate the ratio between non-synonymous mutation frequency (Ka) and synonymous mutation frequency (Ks) for each gene.

The MAFFT v7 software [21] (Bioinformatics Center, Institute for Chemical Research, Kyoto University Uji, Kyoto 611-0011, Japan) was used to compare the plastome sequences of D. lactucoides and H. leptocarpum (with default parameters), while DnaSP v5.1 software [22] (Departament de Genètica, Facultat de Biologia and Institut de Recerca de la Biodiversitat, Universitat de Barcelona, Barcelona, Spain) was used to assess the nucleotide diversity (Pi) of the plastomes with a step size of 200 bp and a window length of 600 bp.

IR boundary analysis was conducted using the online tool IRSCOPE (https://irscope.shinyapps.io/irapp/) [23] to compare the gene types and the location in the border area. The contraction and expansion of the IR regions in the plastomes of the 2 species were also analyzed.

Covariance analysis of the whole genome sequences of the D. lactucoides and H. leptocarpum plastomes was performed using the online mVISTA software (http://genome.lbl.gov/vista/mvista/submit.shtml) [24] to determine whether there were changes in the order of gene arrangement (Shuffle-Lagan was used for the alignment).

Phylogenetic analysis was performed using 14 spliced plastome sequences of the family Papaveraceae downloaded from the National Center for Biotechnology Information (NCBI) database (Table 2) and 2 newly sequenced plastomes of the Papaveraceae species in Tibet. Chelidonium majus L. was used as an outgroup. The phylogenetic tree was constructed based on the maximum likelihood (ML) method using PhyloSuite v1.2.2 [25] (Key Laboratory of Aquaculture Disease Control, Chinese Academy of Sciences, Wuhan, China) with bootstrap values of 10,000 replicates.

The full-length plastomes of D. lactucoides and H. leptocarpum are 166,819 and 163,282 bp, respectively, and exhibit a typical tetrad structure (Fig. 2, Table 3) with the GC contents of 39.23% and 38.57%, respectively. The large-single copy (LSC) regions of the plastomes of D. lactucoides and H. leptocarpum are 82,927 bp and 90,852 bp, while their small-single copy (SSC) regions are 19,538 bp and 17,600 bp, respectively. The LSC and SSC regions of D. lactucoides and H. leptocarpum are separated by two reversed IR regions of 32,177 bp and 27,415 bp, respectively.

Fig. 2.

Plastome structure of D. lactucoides and H. leptocarpum. Note: The thick black line indicates that the two reverse repeat regions IRa and IRb will be the large single copy region (LSC), the small single copy region (SSC) separated; dark part of inner circle indicates GC content.

For D. lactucoides, we annotated 133 genes (Table 4), containing 89 protein- coding genes (PCGs), 36 tRNA, and 8 rRNA genes. Among the 89 PCGs,14 (ndhB, rpl2, rpl14, rpl16, rpl22, rpl23, rpl36, rps3, rps7, rps8, rps11, rps12, rps19, and infA) were double-copy genes. Similarly, seven tRNA genes (trnA-UGC, trnI-CAU, trnA-GAU, trnL-CAA, trnN-GUU, trnR-ACG, and trnV-GAC) and four rRNA genes (rrn4.5, rrn5, rrn6, and rrn23) had double copies. For H. leptocarpum, we annotated 120 genes containing 76 PCGs, 36 tRNA, and 8 rRNA genes. Five PCGs (ndhB, rpl2, rpl23, rps7, and rps12), nine tRNA genes (trnA-UGC, trnD-GUC, trnH-CUG, trnI-CAU, trnI-GAU, trnL-CAA, trnN-GUU, trnR-ACG, and trnV-GAC), and four rRNA genes (rrn4.5, rrn5, rrn6, and rrn23) had double copies. D. lactucoides and H. leptocarpum had 14 identical genes containing one intron and 2 identical genes containing two introns. However, D. lactucoides had 13 more genes than H. leptocarpum, with rpl16 containing one intron. rpl22, rpl32, rpl18, infA, trnS-GGA, and trnfM-CAU were unique genes.

We identified 72 simple repeat sequences (SSRs) from D. lactucoides (Fig. 3), of which 43 were single nucleotide, 14 were dinucleotide, 5 were trinucleotide, 7 were tetranucleotide, and 3 were hexanucleotide sequences. The SSR sequences were mostly distributed in the LSC region (49), followed by the SSC region (11), and the IRa and IRb regions which both had 6. H. leptocarpum had 43 SSRs, which contained 16 single nucleotides, 12 dinucleotide, 6 trinucleotide, and 9 tetranucleotide sequences. The distribution pattern was similar to that of D. lactucoides, with 24 SSRs located in the LCS, 7 in the SSC, 12 in the IRa and IRb regions (six each). Among the single nucleotide sequences, the A/T repeats were predominant. Thus, SSRs were widely distributed in the plastomes and are important for molecular marker development, which could aid hybrid breeding and variety identification in Papaveraceae plants [26].

Fig. 3.

Type and number of SSRs in the plastomes of D. lactucoides and H. leptocarpum (A); Number of SSRs in the LSC, SSC and IR regions (B). SSRs, simple repeat sequences.

The tandem repeat sequence analysis showed 258 and 91 long repeats in D. lactucoides and H. leptocarpum, respectively, including forward, palindromic, reverse, and complementary sequences (Fig. 4). These long repetitive sequences can be used to study the evolutionary history of species [27].

Fig. 4.

Analysis of long repeat sequences of plastomes of D. lactucoides and H. leptocarpum.

The RSCU value is the ratio of the actual frequency of codon usage to the desired frequency of use and is often used to measure the codon use preference. The RSCU value $>$1 indicates a high codon usage frequency [28]. D. lactucoides and H. leptocarpum encoded 65 codons, and both had 32 high-frequency codons with RSCU values $>$1. The comparative analysis of the RSCU values showed that the highest RSCU value was 1.9962 and the lowest was 0.0038, and there was an insignificant difference in RSCU values between the two species. The most frequently used codon was TTA encoding leucine (Leu) (Fig. 5), and the least used codon was CGC encoding arginine (Arg). All amino acids were encoded by multiple codons except tryptophan (Trp). Arg, Leu, and serine (Ser) have six synonymous codons, while alanine (Ala), glycine (Gly), proline (Pro), threonine (Thr), and valine (Val) have four synonymous codons. Isoleucine (Ile) has three synonymous codons, and the rest of the amino acids have two synonymous codons. The codons usually differ only at the third base.

Fig. 5.

Relative use abundance of synonymous codons in chloroplast genomes of D. lactucoides and H. leptocarpum.

Ka/Ks indicates the ratio between heterosense substitution (Ka) and synonymous substitution (Ks) and is often used to determine whether the coding gene is subject to selective pressure during evolution; thus, it is important to determine the Ka/Ks ratio during evolutionary analysis [29]. If Ka/Ks is $>$1, the coding gene is considered to be under positive selective pressure, and if Ka/Ks = 1, the gene is considered to be under neutral selective pressure. However, when Ka/Ks is 1, the gene is considered to be under purifying selection pressure. In this study, two genes (psJ and rsp7) had Ka/Ks ratios $>$1 (Fig. 6), indicating a positive selection effect and that the genes would be retained during evolution. The remaining genes had Ka/Ks ratios of 0 to 0.51, indicating a purifying selection effect and that the mutant genes would be eliminated during evolution. Six genes (atpH, petN, psbA, psbF, psbL, and psbN) had Ka/Ks ratios of 0 and were all photosynthesis-related genes with a synonymous mutation frequency of 0.

Fig. 6.

Gene selection pressure analysis of plastomes of D. lactucoides and H. leptocarpum.

A total of 828 polymorphic loci were detected in the plastomes of D. lactucoides and H. leptocarpum (Fig. 7), with nucleotide diversity (Pi) values ranging from 0 to 0.5096 with a mean value of 0.1947. The IR region was less variable and more conservative than the LSC and SSC regions in terms of sequence variation. In addition, there were seven polymorphic loci with Pi values $>$0.476, located between 40,701 and 80,901 bp, all distributed within the LSC region. These regions with very high variability provide a basis for evolution which could aid species identification and kinship analysis of the Papaveraceae.

Fig. 7.

Nucleotide diversity analysis of plastomes of D. lactucoides and H. leptocarpum.

The JBL boundary (the boundary between the LCS and IRb regions) of D. lactucoides is located within the rpoA gene (Fig. 8), while that of H. leptocarpum is located between the rsp19 and trnH genes. Moreover, the JSB boundary (the boundary between the IRb and SSC regions) of D. lactucoides is flanked by the trnN and ndhF genes, while that of H. leptocarpum is flanked by the trnN and trnD genes. Overall, the JSA boundary had the least difference variance, while the JLB boundary had the most variance, which may be caused by the expansion or contraction occurring in these regions.

Fig. 8.

Comparative analysis of expansion and contraction of plastome genomic boundary regions (LSC, SSC, IR) in D. lactucoides and H. leptocarpum.

Covariance reveals interlocking homologous genes in the genomes of different species and their order of arrangement in the genetic map. The genetic distance between two species is positively correlated with the degree of covariance [30]. The genomes of different species may have some regions of covariance, which may be interrupted by rearrangements and inversions during evolution, and such phenomena can be detected visually through covariance analysis. As shown in Fig. 9, rearrangements and inversions occurred roughly between 15,000 and 85,000 bp region. The covariance is good at both ends, indicating that the gene homology is higher in this region.

Fig. 9.

Covariance analysis of plastomes of D. lactucoides and H. leptocarpum. Note: Blocks of the same color connected by the same line represent similar gene segments of different species.

In this study, the plastomes of D. lactucoides and H. leptocarpum (obtained by sequencing) and 12 other published species of Papaveraceae were used to construct a maximum likelihood (ML) phylogenetic tree (Fig. 10), with the plastomes of C. majus as an outgroup. The results showed that the self-extension values of all branch nodes were greater than 70%, indicating the reliability of the constructed developmental tree. All species originated from seven genera, and species of the same genus clustered together. The Hylomecon japonica and the genus Coreanomecon hylomeconoides clustered as the first major branch, located at the base of the phylogenetic tree. The second major branch included four species of the genus Corydalis (C. saxicola, C. tomentella, C. conspersa, and C. bungeana), one species of Hypecoum (H. leptocarpum), three species of Papaver, (P. somniferum, P. rhoeas, and P. orientale), two species of Meconopsis (M. punicea and M. henrici), and one species of Dicranostigma (D. lactucoides). In the second major branch, the genus Dicranostigma clustered separately as a single group, and the remaining four genera clustered into a large group, within which Meconopsis and Papaver diverged into a separate group from that of Hypecoum and Corydalis. Thus, the genus Meconopsis is closely related to Papaver, while Hypecoum is closely related to Corydalis. The results of this study are consistent with the taxonomic results reported by Ma Wenming [31].

Fig. 10.

Phylogenetic tree of plastomes of 14 species in Papaveraceae. Note: The parameter is, model = GTR+G, method = maximum likelihood, bootstrap = 10,000.