DNA barcode reference library construction and genetic diversity and structure analysis of Amomum villosum Lour. (Zingiberaceae) populations in Guangdong Province

Plant material sampling

Sample collections were conducted with the permission of Research Department of the Second Clinical College, Guangzhou University of Chinese Medicine from August 2018 to November 2018. A total of 141 samples from seven A. villosum populations were collected in Guangdong Province using random strategy (Jin & Lu, 2003) that sampling in the largest possible range according to the population size. Six of them were collected in Yangchun City and one was collected in Maoming City. Population ZY was collected from a deserted germplasm garden containing A. villosum plants coming from Guangdong (Jinhuakeng), Yunnan, and Guangxi provinces and the country Myanmar. The sampled plants were identified by Zhihai Huang, the chief Chinese pharmacist of the Second Clinical College of Guangzhou University of Chinese Medicine. Fresh and healthy leaves were removed from the plants, dried, and preserved in silica gel immediately upon collection in the field, then stored in an ultra-low temperature refrigerator (Eppendorf, Hamburg, Germany) in the laboratory. For detailed information and the geographic locations of the samples collected see Table 1 and Fig. 1.

DNA extraction, PCR amplification and sequencing

The total DNA was extracted using a DP305 plant DNA kit (TIANGEN Biotech Co., Ltd., Beijing, China). The NanoDrop2000 ultra-micro ultraviolet spectrophotometer (Thermo Scientific, MIT, USA) was used to determine the DNA concentration and purity. The extracted DNA was finally diluted to 10 ng/L as the template that be subsequently amplified. The PCR amplification reaction system of the experiment contained 2× Taq PCR Mix 12.5 L, forward primer (2.5 M) 1.0 L, reverse primer (2.5 M) 1.0 L, genomic DNA 2.0 L, and added up to 25 L with ddH₂O. The primer sequences and amplification conditions of different DNA barcodes are shown in Table S1. All amplification reactions were completed on the ProFlex PCR instrument (Life Technologies, New York, USA). PCR products were sent to Shanghai Majorbio Pharmaceutical Biotechnology Co., Ltd Guangzhou Branch to be sequenced.

ISSR-PCR amplification system

100 ISSR universal primer sequences published by Columbia University were screened (Zhang & Luo, 2004). Six primers that produced clear and reproducible banding patterns were selected (Table 2). ISSR-PCR amplifications were performed on 38 individuals randomly chosen from the seven A. villosum populations (Table 3). The 20 μL ISSR reaction volume included 10 μL2X PCR Mix (containing dye, MgCl₂, dNTPs), 2 μL template DNA (20 ng), 1 μLISSR selected primer, and 7 μL ddH₂O. The ISSR PCR amplification was programmed in the ProFlex thermocycler as follows: pre-denaturation at 94 °C for 5 min, 35 cycles of denaturation at 94 °C for 45s, annealing at the 46.83–56.60 °C according to the primers for 45s, extension at 72 °C for 2 min, with a final extension at 72 °C for 5 min and preservation at 4 °C. ISSR-PCR products were separated on a 2% agarose gel stained with Goldview by electrophoresis in 1 ×TAE buffer at 80 V. The gels were visualized under UV light and photographed with a ChemiDoc imaging system (Bio-Rad, Hercules, CA, USA). The molecular weights of ISSR-PCR products were estimated using a 100 bp Plus DNA Ladder (TIANGEN Biotech Co., Ltd., Beijing, China).

Data analysis

The two-way sequenced peaks of DNA barcodes were evaluated and assembled using the CondonCode Aligner v8.0.1 software (Bonfield, Cristina & Rodger, 1998). Low-quality areas at both ends of the assembled sequences were removed. ITS2 barcodes were annotated by cutting off the conserved 5.8S and 28S motifs based on HMM (Keller et al., 2009) at the ITS2 database (Ankenbrand et al., 2015). Mega6.0 software was used to align DNA barcode sequences and calculate sequence statistics including the base composition ratio, GC content, heterotopic site information, conservative site, and parsimony informative sites (Tamura et al., 2013). Haplotype sequences for each barcode were exhibited in the quick response (QR) code picture. In the picture, each vertical line represented a nucleotide base and the QR code on the right could be scanned directly for the DNA sequence.

Reproducible ISSR-PCR bands were determined with the help of the GelPro32 software and manual correction. These clear bands were scored as either present (1) or absent (0), thus generating an ISSR phenotype data matrix that was imported in Popgene32 software to analyze genetic diversity and genetic structure (Yeh et al., 1999). Genetic diversity parameters including percentage of polymorphic sites (PPB), number of alleles (Na), effective number of alleles (Ne), Nei’s gene diversity index (H), and Shannon’s polymorphism information index (I) were calculated. Genetic structure parameters including Nei’s gene differentiation coefficient (Gst), total population genetic diversity (Ht), intra-group genetic diversity (Hs), and gene flow (Nm) were calculated. GenAlEx 6.502 software was used to estimate the components of genetic variance within and among populations by analysis of molecular variance (AMOVA) and to assess the correlation between population genetic distance and geographic distance by Mantel tests (Guillot & Rousset, 2013). Genetic distances among populations were calculated and a UPGMA dendrogram was constructed by using NTSYS 2.10e (Rohlf, 2000).

DNA barcode reference library construction

We extracted the genomic DNA from 141 samples of A. villosum. The OD260/280 was 1.76−1.98 for all the DNA samples and the concentration was 73.70–1294.80 ng/L. Five DNA barcodes of all the samples were amplified and sequenced bi-directionally. Finally, 531 sequences were gained and submitted to the Genbank database (Table S2). The PCR amplification and sequencing results showed that the success rate of sequencing for each barcode was rbcL (100.00%) > ITS (98.58%) > ITS2 (95.04%) > psbA-trnH (53.90%) >matK (29.08%) (Table 4). The ranking of the success rate of PCR amplification was consistent with that of sequencing. Thus, these sequences constructed a DNA barcode reference library for A. villosum in Guangdong Province.

We also analyzed the sequences of each barcode we obtained. The sequences of the five barcodes had no variation sites, showing a strong conservation. Among the five DNA barcodes, ITS2 had the shortest sequence length and the highest GC content. The sequences of each barcode are shown in Fig. 2.

ISSR polymorphism and genetic diversity

We established the ISSR-PCR system for A. villosum. The ISSR-PCR fragments of A. villosum populations ranged from 300 bp to 2000 bp. An example of amplified profiles is shown in Fig. S1. All the ISSR-PCR profiles were deposited in Figshare (DOI: 10.6084/m9.figshare.14622987). A total of 66 ISSR bands were generated from A. villosum populations using the selected six primers. Among them, 56 bands were polymorphic, thus the percentage of polymorphic bands was 84.85% for all seven A. villosum populations. For each primer, it amplified 6–14 bands with the polymorphic ratio of 64.29%–100.0% (Table 2).

Genetic diversity information of A. villosum populations analyzed by Popgene 32 were shown in Table 3. Among the seven populations of A. villosum, population ZY had the highest genetic diversity (PPB =71.21%, Na =1.7121, Ne =1.4240, H = 0.2536, I = 0.3816) while population TK had the lowest genetic diversity (PPB =28.79%, Na =1.2879, Ne =1.1848, H = 0.1117, I = 0.1658). At the species level, the values of Na, Ne, H, and I were 1.9394, 1.5662, 0.3281, and 0.4895, respectively.

Genetic structure of A. villosum populations

Genetic structure information of A. villosum populations analyzed by Popgene 32 is shown in Table 5. Total genetic diversity (Ht) of the seven populations was 0.3299, while within the population genetic diversity was 0.1819. The gene differentiation coefficient (Gst) was 0.4487, indicating that 55.13% of the genetic variation existed within populations. This result was similar to the molecular variance analysis (AMOVA), which showed 68.74% (P = 0.001) genetic variation was within populations while 31.26% (P = 0.001) was between populations (Table 6). Additionally, the gene flow (Nm) among different populations was 0.6143.

Genetic distance and cluster analysis

Genetic distance is the main indicator used to examine the degree of genetic differentiation and the relationship between groups. The genetic distances of the seven populations were between 0.0844 and 0.3347 (Table 7). Among them, the smallest genetic distance was between the ZJD and the TK populations (0.0844), and the largest one was between the XFC and the YC populations (0.3347). Mantel tests carried out with NTSYS-pc 2.0 indicated that genetic distance and geographical distance were not significantly correlated (r = 0.02698, P = 0.5504) (Fig. 3).

The UPGMA clustering map of populations based on the genetic similarity coefficient was constructed using Nei’s genetic distances of the seven populations (Fig. 4). The A. villosum populations were divided into three groups at the similarity coefficient of 0.84. Three populations, ZJD, TK, and ZY, formed one group. Three populations, GY, MM, and YC, formed another group. One population, XFC, formed a single group. The results of PCoA based on the unbiased matrix of Nei were consistent with UPGMA cluster analysis. The first two principal components accounted for 48.04% of the total variation of ISSR markers indicated from PCoA (Fig. 5).