Evergreen broadleaf forests (EBLFs) contribute to global biodiversity and ecosystem maintenance, but are rapidly degenerating and fragmenting due to anthropogenic activities (Song and Chen, 2007). EBLF flora comprise large trees (e.g., species from the Fagaceae family) and a large variety of shrubs including many Rhododendron L. (Ericaceae) species (Song and Chen, 2007). Generally, EBLF fragmentation is expected to induce genetic differentiation among populations as a result of decreased gene flow, increased inbreeding, and genetic drift. This is especially true for shrub and herb species with relatively short generation times and low population densities (Zhao et al., 2006). However, very few studies have addressed this hypothesis by investigating the genetic structure of evergreen shrubs.

Rhododendron ovatum (Lindl.) Planch. ex Maxim. var. ovatum (subgenus Azaleastrum Planch. ex K. Koch) is a monoecious, self-incompatible evergreen shrub endemic to China and is one of the most prevalent shrubs in EBLFs. For the conservation and management of EBLFs, it is important to characterize the possibly dwindling genetic variation of R. ovatum. However, most microsatellite markers previously developed for Rhododendron species (e.g., Tan et al., 2009) are not functional for R. ovatum.

In the current study, we isolated and characterized 16 polymorphic and seven monomorphic microsatellite loci to reveal genetic variation in R. ovatum and to shed light on the underlying mechanisms, such as limited gene flow and historical demographics. To further study the dynamics of interspecific hybridization within the genus, we also carried out cross-amplification in R. simsii Planch. (subgenus Tsutsusi (Sweet) Pojark.), another important shrub species in EBLFs (Zhuang, 2012), thus further contributing to forest management and conservation.

METHODS AND RESULTS

Plant materials of R. ovatum were collected from three populations located in Tiantong and Tianmu in Zhejiang Province, China, and Jinggang in Jiangxi Province, China (Appendix 1). Microsatellite loci were developed according to the method recorded by Tong et al. (2012). Total genomic DNA was extracted from leaves dried with silica gel using the Plant Genomic DNA Kit (Tiangen, Beijing, China). After digestion with the MseI restriction enzyme (New England Biolabs, Beverly, Massachusetts, USA), approximately 250 ng of DNA was ligated to an MseI-adapter pair (F: 5′-TACTCAGGACTCAT-3′, R: 5′-GACGATGAGTCCT-GAG-3′). The diluted (1:5) ligation-digestion mixture was amplified with MseI-N primers (5′-GATGAGTCCTGAGTAA-3′) in a 20-µL PCR reaction at: 95°C for 3 min, followed by 17 cycles of 94°C for 30 s, 53°C for 1 min, and 72°C for 1 min. To enrich DNA fragments containing microsatellites, the amplification products were hybridized with the 5′-biotinylated probe (AG)₁₅ and the hybridization products were captured by magnetic beads coated with streptavidin (Promega Corporation, Madison, Wisconsin, USA). The enriched fragments were PCR amplified using MseI-N primers for 30 cycles. After purification using the multifunctional DNA Extraction Kit (Bioteke, Beijing, China), the PCR products were transformed into Escherichia coli strain JM109 with the pMD 19-T vector (TaKaRa Biotechnology Co., Dalian, China) followed by transient thermal stimulation.

Of the 241 selected clones, 174 were positive after PCR with (AG)₁₀ and M13F/ M13R as primers. The positive clones were sequenced on an ABI 3730 DNA Sequence Analyzer (Applied Biosystems, Foster City, California, USA). A total of 51 sequences containing microsatellites were chosen for simple sequence repeat primer design using Primer Premier 5.0 software (PREMIER Biosoft International, Palo Alto, California, USA). Twenty-three R. ovatum individuals were randomly selected from the samples mentioned above (Appendix 1) and were used to test the performance and polymorphism for all loci. PCR was performed in a 10-µL reaction system containing 50 ng of genomic DNA, 1× PCR buffer (without Mg²⁺), 2.5 mM Mg²⁺, 0.2 mM of each dNTP, 0.1 µM of each primer, and 1 unit of Taq DNA polymerase (Sangon, Shanghai, China) at 95°C for 5 min; 35 cycles of 40 s at 94°C, 45 s at 45–65°C (depending on specific locus, Table 1), and 45 s at 72°C; and a final extension at 72°C for 8 min. PCR products were separated using 8% polyacrylamide denaturing gels and visualized with silver staining using pUC19 DNA/MspI (HpaII) (Thermo Fisher Scientific, Waltham, Massachusetts, USA) as the ladder. A total of 16 polymorphic and seven monomorphic loci were obtained (Table 1), none of which have been previously reported in the genus Rhododendron based on the results of BLAST searches in GenBank.

Table 1.

Characterization of 16 polymorphic and seven monomorphic microsatellite loci developed in Rhododendron ovatum.^a

The polymorphisms among all polymorphic loci were further surveyed with 89 individuals from the three R. ovatum populations (Appendix 1). We labeled the forward primers using a fluorescent dye (5′HEX, 5′ROX, or 5′6-FAM) (Sangon). PCR reactions were then performed in a 10-µL reaction system using the same thermocycling program described above. The products were scanned on an ABI 3730 automated sequencer using GeneScan 500 LIZ (Applied Biosystems) as the internal lane standard and were genotyped using GeneMapper 4.0 (Applied Biosystems).

When analyzed using the software TFPGA version 1.3 (Miller, 1997) and FSTAT 2.9.3 (Goudet, 1995), the 16 polymorphic loci displayed moderate to high levels of genetic variation in the three populations. The number of alleles ranged from four to 30 among the loci, with a mean value of 14.2, indicative of a high level of polymorphism (Table 1). The observed and expected (based on Hardy–Weinberg equilibrium) heterozygosities within the population varied from 0.1034 to 0.9333 and from 0.1016 to 0.9542, respectively (Table 2). We failed to detect any significant linkage disequilibrium for all pairs of loci in all populations. Significant deviation from Hardy–Weinberg equilibrium was only found at one locus (MYH14), and only in the Tiantong and Jinggang populations after sequential Bonferroni correction (Rice, 1989). Signs of null alleles in the loci MYH3, MYH7, MYH14, and MYH16 were detected using MICROCHECKER 2.2.3 (van Oosterhout et al., 2004).

Table 2.

Characterization of the 16 polymorphic microsatellite loci in three Rhododendron ovatum populations.^a

We also tested the performance of these primer pairs in R. simsii, a closely related species to R. ovatum but not in the same subgenus. After scanning the PCR products in 16 R. simsii individuals sampled in Shanghai, China (Appendix 1), 12 polymorphic loci (except MYH3, MYH7, MYH10, and MYH13) could be used in this congeneric species. These loci revealed high levels of polymorphism and observed and expected heterozygosities (Table 3), similar to those evaluated by the polymorphic microsatellite loci specifically developed for R. simsii (Tan et al., 2009). However, only two of the eight microsatellite loci developed for R. simsii (Tan et al., 2009) could be amplified in R. ovatum.

Table 3.

Characterization of the 16 polymorphic microsatellite loci developed for Rhododendron ovatum in R. simsii.^a

CONCLUSIONS

The 23 microsatellite loci developed in the current study provide an appropriate resource to delineate the genetic variation and genetic structure of R. ovatum populations, thereby contributing to the management and conservation of EBLFs. These markers can also facilitate future population genetic studies at a multispecies level within the genus Rhododendron.