The Himalayan region and the adjacent Hengduan Mountains of southwestern China, known as the Himalaya–Hengduan Mountains (HHM) region, have been designated as two of the world's 34 most important biodiversity hotspots (Myers et al., 2000). The HHM region is considered to be the cradle of many endemic plant groups (Li and Li, 1993) and the center for rapid radiation of several large alpine genera, such as Primula L., Pedicularis L., and Rhododendron L., as well as the center of the Sino-Himalayan floristic subkingdom (Wu and Wang, 1983). Its high species endemism is a likely product of high net diversification rates in the region, as seen in páramo hotspots evaluated by Madriñán et al. (2013). A number of studies have been devoted to the differences between the two parts of the HHM region (the Himalayas and the Hengduan Mountains), such as the direction of the mountain ranges, the time scale of the Qinghai– Tibet plateau (QTP) uplift process, and the effects of climate oscillations during the Quaternary (Favre et al., 2015). Correspondingly, the Sino-Himalayan floristic subkingdom in the HHM region has been recognized as including at least four subregions (Wu et al., 2011). However, it is not clear whether these differences between the Hengduan Mountains and the Himalayan regions have resulted in deep intraspecific lineage divergences and/or cryptic speciation in plant groups.

Primula sikkimensis Hook. (Primulaceae) is an endemic species in the HHM region (Hu and Kelso, 1996) and is the only species in Primula sect. Sikkimensis that is widely distributed in the region. It therefore provides a good example to examine the hypothesis of deep lineage divergence between the Himalaya and Hengduan mountains (Gao et al., 2007). Here, we developed a set of variable microsatellite markers using 454 pyrosequencing technology and further tested its cross-amplification in closely related taxa. These microsatellite markers will be important tools for surveying genetic divergence and cryptic speciation events in P. sikkimensis and its relatives.

METHODS AND RESULTS

Leaf samples of 62 individuals were collected in three populations from Chayu, Galongla, and Luding in China (Appendix 1). One individual of P. sikkimensis (sampled from Jiulong, China; Appendix 1) was used to isolate the microsatellite loci. Voucher specimens have been deposited at the herbarium of the South China Botanical Garden (IBSC), Guangzhou, Guangdong, China. Total DNA extraction of all samples was performed using a modified version of the cetyltrimethylammonium bromide (CTAB) protocol of Doyle and Doyle (1987). Microsatellite markers were isolated using a high-throughput genomic sequencing method as described by Wang et al. (2015). A shotgun library shearing 1 µg of genomic DNA was built using the DNA Library Preparation Kit (Roche Applied Science, Indianapolis, Indiana, USA) following the GS FLX+ library preparation protocol. The library was further enriched by hybridization with biotinylated oligonucleotide probes (AG)₁₀, (AC)₁₀, (AAC)₈, (ACG)₈, (AAG)₈, (AGG)₈, (ACAT)₆, and (ATCT)₆ by Tóth et al. (2000) and Zane et al. (2002). The simple sequence repeat (SSR)–enriched libraries were then sequenced using a Roche 454 GS FLX DNA sequencing platform. In total, 61,755 unique reads were obtained with sizes ranging from 300 to 600 bp.

Microsatellite repeats in unique reads were identified by MISA software (Thiel et al., 2003). The SSR search was performed for di-, tri-, and tetranucleotides with a minimum of six, five, and five repeats, respectively, and a minimum product size of 100 bp. In total, 5377 unique reads with at least one microsatellite motif were obtained. Among these reads, 3112 unique reads, which had at least 50 bp in each flanking region for primer design, were chosen to filter the perfect SSR loci (sequences in these reads are available upon request). Then 29 loci were randomly selected to design primer pairs using Primer3 software (Rozen and Skaletsky, 1999). The minimum primer annealing temperature was set to 60°C, primer size was between 18–22 bp with an optimal size of 20 bp, and other settings were left at default values.

These primer pairs were initially tested for successful PCR amplification in three P. sikkimensis individuals from three separate populations. PCR reactions were performed on a PTC-200 Thermal Cycler (MJ Research, Watertown, Massachusetts, USA) with the following conditions: an initial denaturation at 94°C for 3 min; followed by 30 cycles at 94°C for 30 s, locus-specific annealing temperature (Table 1) for 45 s, and 72°C for 50 s; and a final extension at 72°C for 7 min. Amplicons were checked on 2% agarose gel stained with ethidium bromide.

In total, 20 primer pairs that generated specific amplification of corresponding PCR products were further resynthesized using fluorophore labeling (FAM or HEX) and used for amplification in the 62 individuals from the three populations. The same PCR conditions were used as described above. One microliter of the fluorescent PCR product was added into the mixture with 8.8 µL of formamide and 0.2 µL of GeneScan 500 LIZ Size Standard (Applied Biosystems, Life Technologies, Waltham, Massachusetts, USA). PCR products were subsequently run on an ABI PRISM 3130 Genetic Analyzer (Applied Biosystems). Genotypes were scaled by GelQuest software (version 3.2.1; SequentiX, Klein Raden, Germany). Seventeen of the 20 primers showed clear and robust genotype information. The microsatellite information and GenBank accession numbers are listed in Table 1.

Table 1.

Characteristics of 17 microsatellite loci developed in Primula sikkimensis.

Genetic diversity parameters, including allelic richness (A), observed and unbiased expected heterozygosity (H_o, H_e), and inbreeding coefficient (F_IS), were estimated by GenAlEx 6.5 (Peakall and Smouse, 2012). Deviations from Hardy– Weinberg equilibrium (HWE) at each locus were tested through GENEPOP 4.0.7 (Rousset, 2008) and are presented in Table 2. Numbers of alleles varied from one to eight; H_o and H_e ranged from 0.111 to 1.000 and 0.061 to 0.811, respectively; and F_IS ranged from −1.000 to 0.660. Twelve loci showed significant deviation from expectations under HWE (Table 2) because of an excess of homozygotes. Null alleles, inbreeding, Wahlund effect, and sampling effect (small population size) could all potentially cause deviations from HWE. Four loci with presence of null alleles were detected by MICRO-CHECKER (van Oosterhout et al., 2004).

In addition, we tested cross-amplification with two related species in sect. Sikkimensis (P. alpicola (W. W. Sm.) Stapf and P. florindae Kingdon-Ward). One individual of P. alpicola was sampled from Paizhen, Tibet (29° 19′N, 95° 19′E), and one individual of P. florindae was collected at Lulang, Tibet (29°42′N, 94°43′E) (Appendix 1). Primer transferability was considered successful when one clear distinct band in the expected size range was detected on 2% agarose. In total, 10 SSR loci could be successfully used in both P. alpicola and P. florindae, and only four loci could not be amplified in these two species. Specifically, the transferability values were 76.5% in P. florindae and 58.8% in P. alpicola, respectively.

CONCLUSIONS

In this study, 17 microsatellite markers were successfully developed for P. sikkimensis; these markers showed high polymorphism and could therefore be a powerful tool in population genetic studies. Cross-amplification of these microsatellite loci in two related Primula species (P. alpicola and P. florindae) was successful, which enables further studies to clarify underlying genetic introgression and cryptic speciation events between P. sikkimensis and its closely related taxa.

Table 2.

Estimated population genetic parameters (per nSSR locus) in three populations of Primula sikkimensis and cross-amplification in two related species.^a