Total genomic DNA for all samples was extracted using the standard CTAB procedure (Doyle & Doyle, 1987) from 30 mg of dried leaf tissue, and served as the template for the polymerase chain reaction (PCR). DNA quality and quantity were determined on 0.8% agarose gels stained with 2.5 μL Goldview (Aidlab Biotechnologies Co., Ltd., Beijing, China), with AL2000 DNA marker (Aidlab Biotechnologies).
One nuclear ribosomal DNA (nrDNA) sequence, the ribosomal inter-transcribed spacer (ITS) region comprising spacer 1, the 5.8S ribosomal gene and spacer 2 (White et al., 1990), and the two chloroplast DNA (cpDNA) intron-spacer regions trn L-trn F (Taberlet et al., 1991) and ycf 1b (Dong et al., 2015), were used in this study (Table 2). PCR reactions were set up in a volume of 25 μL, composed of 20 μL ddH2O, 2.5 μL 10 × buffer, 0.5 μL 10 mM dNTPs, 0.5 μL each 5 μM primer, 0.5 μL DNA template and 0.5 μL 5 U/μL Taq polymerase (Aidlab Biotechnologies). The PCR reactions were carried out on a 2720 Thermal Cycler (Applied Biosystems by Life Technologies, Singapore). The PCR program for ITS1/2 and trn L-trn F was designed with an initial denaturation of 5 min at 94 ºC, followed by 35 cycles of 1 min at 94 ºC, 1 min at 55 ºC, 1 min at 72 ºC, and with a final extension of 10 min at 72 ºC. Amplification of ycf 1b used the following protocol: 4 min at 94 ºC, 35 cycles of 30 s at 94 ºC, 40 s at 58 ºC, and 1 min at 72 ºC, ending with 10 min at 72 ºC. All the PCR products were verified based on size by gel electrophoresis. The amplicons were sequenced by an ABI 3730 DNA Analyzer in forward and reverse directions based on the BigDye Terminator Cycle Sequencing Ready Kit (Applied Biosystems, Foster City, CA, U.S.A.) in BGI (Beijing Genomics Institute, China).
2.2 | Genetic diversity
The chromatograms from both directions of the ITS1/2 and cpDNA sequences were checked visually and edited manually with the software BioEdit (Hall, 1999) for base confirmation and contiguous sequence editing; each base/character was equally weighted before analysis, and each indel/gap was represented as a single mutation. Three sequences were manually aligned and trimmed integrally where necessary using MEGA v.6.5 (Kumar et al., 2008) separately. The two no-coding cpDNA sequences were assembled as a single locus for subsequent analysis by SequenceMatrix v.1.7.8 (Vaidya et al., 2011), and the partition homogeneity test of cpDNA sequences vs. nrITS sequences was carried out with PAUP* v.4.0a164 (Swofford, 2002). Since non-homogeneity of both matrices was detected, nrITS and cpDNA were analyzed independently.
DNASP v.6.12.01 (Rozas et al., 2017) was used to compute the number of identified ribotypes/chlorotypes (Nn ), haplotype diversity (h ) within populations, polymorphic sites (S ), nucleotide diversity (π ), and the average number of nucleotide differences (Nd ) separately for nrITS and cpDNA matrices. The geographic distribution maps of ribotypes/chlorotypes at the population level, also separately for nrITS and cpDNA sequences, were constructed and visualized with ArcGIS v.10.8 (ESRI, Redlands, CA, U.S.A.).