Isolation and characterization of microsatellite loci in Psilopeganum sinense Hemsl (Rutaceae), an endangered herb endemic to Yangtze River valley

Psilopeganum sinense is a perennial herb endemic to Yangtze River valley, a biodiversity hot spot in south-central China (Ying & Zhang 1994). Being a widely used Chinese traditional medicinal herb, P. sinense has been severely suffering over-harvesting by local residents. The ongoing construction of the world's largest dam, the Three-Gorges Dam (TGD) also inundated many local populations of P. sinense because its natural range is mostly restricted to the Three-Gorges Reservoir Area (TGRA). Over-harvesting and habitat loss have resulted in a rapid decline of population size and local population extinction, and raised urgent concerns to the public and conservationists for protecting P. sinense. However, it has been widely recognized by conservation geneticists that understanding of the population structure, gene flow and mating system is a prerequisite for formulating any conservation programmes for endangered plant species. We also hope to make P. sinense a model species for investigation of the variation of mating system throughout the year and the evolution significance of long flowering time, because this species can flower and fruit from spring to early winter. Because of their codominant and hypervariable nature, microsatellite markers have proven highly efficient molecular tools. Here, we report the isolation and characterization of a set of polymorphic microsatellites from the genome of P. sinense for population and conservation genetic studies.

Genomic DNA was extracted from leaf tissue of P. sinense using a modified cetyltrimethyl ammonium bromide (CTAB) method (Doyle & Doyle 1987). A microsatellite dinucleotide-enriched genomic library of P. sinense was then constructed using a modified protocol of fast isolation by amplified fragment length polymorphism (AFLP) of sequences containing repeats (FIASCO, Zane et al. 2002). A total of 250 ng genomic DNA was digested with 3 U of MseI (BioLabs) in a 25-µL volume, and then 15 µL of digested DNA was ligated to 1 µm MseI AFLP adaptor (5′-TACTCAGGACTCAT-3′/5′-GACGATGAGTCCTGAG-3′) using 1 U of T4 DNA ligase (TaKaRa) in a volume of 30 µL. The digestion–ligation mixture was subsequently diluted 10 times, and directly amplified using AFLP adaptor-specific primers (5′-GATGAGTCCTGAGTAAN-3′, i.e. MseI-N) in a 20-µL reaction containing: MseI-N 0.5 µm, dNTPs 0.2 mm, MgCl₂ 1.5 mm, 1 U of Taq DNA polymerase (Fermentas) and 5-µL diluted digestion–ligation DNA. The polymerase chain reaction (PCR) was performed using a programme of 94 ˚C 30 s, 53 ˚C 1 min, 72 ˚C 1 min for 19 cycles.

After denaturation at 95 ˚C for 5 min, the amplified products were hybridized with 0.15 µm of 5′-biotinylated (AC)₁₅ probe in a total volume of 250 µL hybridization buffer containing SSC 4.2× and SDS 0.07% at 48 ˚C for 2 h. The hybridization products were selectively captured with 300 µL Streptavidin MagneSphere Paramegnetic Particles (Promega). Captured DNA fragments were released from the beads-probe-DNA complex by incubating for 5 min at 95 ˚C in 50 µL of TE (Tris-HCl 10 mm, EDTA 1.0 mm, pH 8.0).

Recovered DNA fragments were amplified for 23 cycles using MseI-N primers. After purification using E.Z.N.A Gel Extraction Kit (OmegaBiotek), PCR products were ligated into pMD 18-T plasmid vector (TaKaRa) and transformed into Escherichia coli strain (JM109, Promega). Recombinant clones were identified using blue/white screening on Luria-Bertani agar plates containing ampicillin, X-Gal and IPTG. Insert positive bacterial clones were amplified using M13 forward and reverse primers and visualized by agarose gel electrophoresis. One hundred and fourteen identified positive clones were sequenced with ABI BigDye Terminators Cycle Sequencing Kit (Applied Biosystems) in an ABI PRISM 3100 automated sequencer.

Primer pairs flanking microsatellite regions were designed using the software primer 3 (Rozen & Skaletsky 2000). PCR amplifications were performed in a 10-µL reaction containing 10 mm Tris-HCl pH 8.4, 50 mm (NH₄)₂SO₄, 1.5 mm MgCl₂, 0.2 mm dNTPs, 0.2 µm each primer, 1 U Taq polymerase (Fermentas), and 20 ng of genomic DNA. The amplification profiles included: 5 min at 94 ˚C; 35 cycles of 50 s at 94 ˚C, 50 s at 55–57 ˚C depending on the primer pair (Table 1), 90 s at 72 ˚C; and a final extension for 8 min at 72 ˚C. Amplified products were separated on 6% denatured polyacrylamide gels using silver staining. A 10-bp DNA ladder (Promega) was used to identify alleles.

A total of 66 (58%) of identified sequences were found to contain simple sequence repeats, 29 of these appeared to be suitable for designing primers. The 29 primer pairs designed were then tested using 24 individuals of P. sinense randomly sampled from an extant population in TGRA. Of the 29 primer pairs tested, 15 primer pairs successfully amplified target fragments, and 11 loci were polymorphic, while the other four loci were monomorphic. The program cervus 2.0 (Marshall et al. 1998) was used to calculate the observed and expected heterozygosities and null allele frequencies for the 11 polymorphic microsatellite loci. Tests for Hardy–Weinberg equilibrium and linkage disequilibrium were performed using genepop 3.4 (Raymond & Rousset 1995). Significance levels were adjusted using Bonferroni correction for multiple testing. The 11 polymorphic loci revealed a total of 41 alleles, ranging from two to eight alleles per locus (Table 1), with an average value of 3.7. The observed and expected heterozygosities ranged from 0.000 to 0.750 and from 0.365 to 0.800, respectively. Six of the 11 loci deviated significantly from Hardy–Weinberg equilibrium (P < 0.01) after Bonferroni correction (Table 1), which could be the result of null alleles and/or inbreeding effects in the natural population investigated. A significant linkage disequilibrium was found only in pair of loci Psh1 & Psh10 (P < 0.01). We hope these polymorphic loci will be useful in conservation genetic studies of P. sinense.