Isodon (Schrad. ex Benth.) Spach (Lamiaceae) is composed of approximately 100 species (Li, 1988). The genus is distributed widely from the Far East to Africa. In Japan, seven species and six varieties are currently recognized (Murata and Yamazaki, 1993).
Suzuki et al. (2007) reported that most of the species belonging to section Isodon (four species and six varieties) are pollinated by bumblebees in Japan. The corolla lengths of these Isodon taxa generally correspond to the proboscis lengths of the bumblebee species visiting the flower of the taxa (Suzuki et al., 2007). Consequently, differences in the flower visitors could cause reproductive isolation between Isodon taxa in Japan.
A previous phylogenetic study on Isodon in Japan and Korea based on chloroplast DNA revealed that individuals from different populations in the same species were rarely monophyletic in the tree (Maki et al., 2010), suggesting that interspecific reproductive isolation might be incomplete and that extensive hybridization may occur in Isodon in Japan and Korea (Maki et al., 2010). To resolve whether natural hybridization is responsible for speciation and genetic diversity in these species, population genetic analyses using a combination of rapidly evolving genetic markers and extensive taxon sampling will be necessary (Buerkle and Lexer, 2008). In this respect, highly polymorphic microsatellite markers will provide valuable insights into natural hybridization among Isodon species. In this study, we isolated and characterized 10 microsatellite markers for I. longitubus (Miq.) Kudô and explored their applicability to congeneric species.
METHODS AND RESULTS
For developing microsatellite markers, an enriched genomic DNA library was constructed using the streptavidin-coated magnetic bead method described by Bloor et al. (2001). The brief procedure from library construction to primer design was presented in Yamashiro et al. (2013). Genomic DNA was extracted from leaf samples of I. longitubus collected at Mt. Ishizuchi, Ehime Prefecture, Japan (T. Yamashiro 11843; see Appendix 1 for voucher information), using the procedure of Doyle and Doyle (1987). We designed a total of 25 primer sets for cloned sequences, of which 10 successfully amplified the target regions. PCR was performed in 10 µL of reaction mixture containing 3 ng of template DNA, 0.5 µM each primer, 0.25 mM each dNTP, 1.5 mM MgCl2, 10 mM Tris-HCl (pH 8.0), 50 mM KCl, and 0.25 U of Ex Taq polymerase (TaKaRa Biotechnology Co., Otsu, Shiga, Japan). The PCR amplification conditions were as follows: initial denaturation at 94°C for 2 min; followed by 30 cycles of 20 s at 94°C, 15 s at the annealing temperature (see Table 1), and 45 s at 72°C; and final extension at 72°C for 10 min. Two I. longitubus populations (T. Yamashiro 11843 and T. Yamashiro 11845; Appendix 1) were used for initial polymorphism screening. We tested cross amplification in I. inflexus (Thunb.) Kudô (N = 10), I. japonicus (Burm. f.) H. Hara (N = 10), and I. shikokianus (Makino) H. Hara (N = 9). Voucher specimens representing the populations examined have been deposited in the herbarium at Tohoku University (TUS; Appendix 1). For nine of the polymorphic loci (Isodon 2, 6, 7, 9, 36, 47, 64, 69, and 70), PCRs were conducted using the above-mentioned reaction mixture volume and cycling conditions, with primers labeled with one of three Beckman WellRED dyes (D2–4: Sigma-Aldrich, St. Louis, Missouri, USA). To amplify locus Isodon3, we employed the single-tube nested PCR methods described by Schuelke (2000). All PCRs were conducted using a PC-818S program temperature control system (Astec, Fukuoka, Japan). The PCR product sizes were measured using CEQ 8000 Genetic Analysis System (Beckman Coulter, Pasadena, California, USA) and CEQ 8000 fragment analysis software (Beckman Coulter). We used the GenomeLab DNA Size Standard Kit 400 (Beckman Coulter) to determine allele size.
Characteristics of the 10 microsatellites developed in Isodon longitubus.
The basic genetic diversity parameters obtained from two I. longitubus populations are presented in Table 2. In the Mt. Ishizuchi population, all 10 loci were polymorphic, with five to 19 alleles per locus. The observed and expected heterozygosity values ranged from 0.350 to 0.950 and 0.525 to 0.918, respectively. In contrast, in the Toon population, all loci except Isodon70 were polymorphic, with two to 12 alleles per locus. The observed and expected heterozygosity values ranged from 0.273 to 1.000 and 0.483 to 0.872, respectively. Hardy–Weinberg equilibrium and linkage disequilibrium were analyzed using GENEPOP version 3.3 (Raymond and Rousset, 1995). Although significant linkage disequilibrium was not found for all 45 possible pairwise comparisons, significant heterozygote deficiency (P < 0.05) was detected only at Isodon36 in the Mt. Ishizuchi population. Results of the cross amplification of the three congeneric species examined are summarized in Table 3. For I. inflexus, all 10 loci were amplified, although two loci (Isodon6 and Isodon70) were monomorphic. For I. japonicus, eight loci were amplified and three were monomorphic. For I. shikokianus, seven loci were polymorphic, one locus was monomorphic, and amplification failed for two loci.
Results of initial primer screening in two Isodon longitubus populations.
Cross-species amplification of 10 microsatellite primers in three congeneric Isodon species.