Natural hybridization is a common phenomenon in plant species and plays an important role in plant evolution because it can rapidly generate novel gene combinations (Rieseberg, 1997). The genus Leucosceptrum Smith (Lamiaceae) is endemic to eastern Asia and consists of five species. Leucosceptrum japonicum (Miq.) Kitam. & Murata and L. stellipilum (Miq.) Kitam. & Murata, which are morphologically distinct, are endemic to Japan (Murata and Yamazaki, 1993). Leucosceptrum japonicum is distributed throughout the entire Japanese archipelago while L. stellipilum is restricted to the western part of the Japanese mainland. In localities where these two species co-occur, natural hybridization between them has been reported (Takahashi, 2001). However, this report was made merely on the basis of morphological characters, and no genetic approaches have been used to investigate the genetic structure of the hybrid zone. In this study, we developed microsatellite markers amplifiable in both L. japonicum and L. stellipilum that could be used to investigate the genetic structure of a hybrid zone between these two species.

METHODS AND RESULTS

Voucher specimens used in this study were deposited at the herbarium of Tohoku University (TUS), Sendai, Miyagi Prefecture, Japan. Genomic DNA was extracted from fresh leaves of L. japonicum collected from a single individual from a population at Sakunami in northeastern Japan (Miyagi Prefecture; 38°22′16″N, 140°34′48″E; voucher no. TUS 419135) using a modified cetyltrimethylammonium bromide (CTAB) method (Maki and Horie, 1999). An enriched library was constructed by a modified biotin-capture method (Fischer and Bachmann, 1998). Five 100-ng aliquots of DNA were separately digested with NdeII (Nippon Gene, Tokyo, Japan), Csp6I, TaqI, HinfI, and HpyF3I restriction enzymes (Fermentas, Vilnius, Lithuania). Digestions were carried out at 37°C for 1 h, except for TaqI, which was at 65°C for 1 h. Fragments of 400–1000 bp were excised from 1.2% agarose gel and purified with Quantum Prep Freeze ‘N Squeeze DNA Gel Extraction Spin Columns (Bio-Rad, Hercules, California, USA). The size-selected fragments digested by NdeII were ligated into the sticky-end adapter described by Bloor et al. (2001). Those fragments digested with Csp6I, TaqI, HinfI, and HpyF3I were separately ligated into modified adapters, which have sticky ends consisting of TA, CG, ANT, and TNA, respectively. PCR amplification was performed using oligo A as a primer (Bloor et al., 2001) following the amplification steps from Bloor et al. (2001) and used the ligated fragments as templates to test for successful ligation; the PCR products were then hybridized to the 5′ biotin-labeled oligonucleotide probes (GT)₁₂ and (CT)_12, respectively. DNA molecules hybridized to the biotin-labeled probes were subsequently captured by streptavidin-coated magnetic beads (Promega Corporation, Madison, Wisconsin, USA). After washing with buffer I (6× saline sodium citrate [SSC] and 0.1% sodium dodecyl sulfate [SDS]) and buffer II (1× SSC and 0.1% SDS), captured DNAs were eluted into 200 µL of sterile water by boiling for 5 min. The enriched single-stranded DNA fragments were amplified by PCR using oligo A as the primer following the amplification steps from Bloor et al. (2001). The PCR products were ligated into pGEM-T Easy Vector (Promega Corporation) for 16 h at 4°C and then transformed into competent cells (DH5α; Toyobo, Osaka, Japan).

In total, 493 recombinant clones were sequenced with the universal M13 primers (forward: 5′-GTTTTCCCAGTCACGAC-3′, reverse: 5′-CAGGAAACAGCTATGAC-3′) using the BigDye Terminator version 3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, California, USA) on an ABI 3100 Genetic Analyzer (Applied Biosystems). Microsatellite sequences were found in 73 clones, of which 46 with sufficient flanking regions were chosen for primer pair designs using the software Primer Premier 6.0 (Premier Biosoft International, Palo Alto, California, USA). A modified version of the method by Schuelke (2000) was used, employing three primers in the PCR reaction: the fluorescent-labeled universal primer, a locus-specific forward primer with a universal tail attached at its 5′ end, and a normal locus-specific reverse primer. Four primer sequences (tail A: 5′-FAM-GCCTCCCTCGCGCCATCAG-3′, tail B: 5′-VIC-GCCTTGCCAGCCCGC-3′, tail C: 5′-NED-CAGGACCAGGCTACCGTG-3′, and tail D: 5′-PET-CGGAGAGCCGAGAGGTG-3′) used by Blacket et al. (2012) were employed as universal primers.

TABLE 1.

Characteristics of the 10 microsatellites developed from Leucosceptrum japonicum.

Initial testing of amplification of the primer pairs and loci polymorphisms was performed on three to four individuals of each species. PCR reactions were performed separately for each locus in 3-µL volumes containing 15–30 ng of genomic DNA, 1× Type-it Multiplex PCR Master Mix (QIAGEN, Hilden, Germany), 0.075 µM tailed forward primer, 0.1 µM fluorescent-labeled universal primer, and 0.25 µM reverse primer. The reactions were started with an initial denaturation at 95°C for 5 min; followed by 32 cycles at 95°C for 30 s, locus-specific annealing temperature for 90 s (Table 1), and 72°C for 30 s; and 60°C for 30 min on the iCycler (Bio-Rad). Three post-PCR multiplex combinations (Table 1) were designed by considering the fragment size and dye. A mixture of 0.5 µL of each of the PCR products was diluted 1 : 20 with distilled water and run with GeneScan 600 LIZ Size Standard (Applied Biosystems) on an ABI 3100 Genetic Analyzer (Applied Biosystems). Allele sizes were assigned using GeneMapper version 3.7 software (Applied Biosystems).

Successful amplification of L. japonicum was detected at 15 loci, of which 10 were reliably amplified in L. stellipilum. These 10 loci were subsequently used for genotyping 25 individuals of L. japonicum collected from the Sakunami population and 11 individuals of L. stellipilum collected from the Kumano population (33°53′38″N, 135°52′12″E; voucher no. TUS 419136) (Table 2).

TABLE 2.

Results of primer screening of the 10 newly developed microsatellite loci in one population of Leucosceptrum japonicum and one population of L. stellipilum.

The number of alleles per locus ranged from one to 11 in both L. japonicum and L. stellipilum. A total of 50 and 51 alleles were detected at the 10 microsatellite loci among 25 individuals of L. japonicum and 11 individuals of L. stellipilum, respectively. All 10 loci were polymorphic among the 25 L. japonicum individuals, with the exception of one monomorphic locus (Leu4). Excluding Leu7 and Leu8, eight loci were polymorphic among the 11 L. stellipilum individuals. The observed heterozygosity ranged from 0.00 to 0.84 and 0.00 to 1.00 in L. japonicum and L. stellipilum, respectively. The expected heterozygosity ranged from 0.00 to 0.84 and 0.00 to 0.89 in L. japonicum and L. stellipilum, respectively. Tests for deviation from Hardy–Weinberg equilibrium (HWE) at each locus and the linkage disequilibrium (LD) of all combinations of the loci were conducted with GENEPOP 4.2 (Rousset, 2008). All 10 loci were not significantly deviated from HWE in either species, and no combination of the loci was in LD (P > 0.05) after sequential Bonferroni correction (Rice, 1989). Because the flowers of both L. japonicum and L. stellipilum were very frequently visited by bumblebees (Li and Maki, personal observation), these species appear to be predominantly outcrossing, which is consistent with their high genetic diversities and the lack of deviation from HWE.

CONCLUSIONS

Reliable amplification in both L. japonicum and L. stellipilum and the relatively high level of diversity demonstrate the potential application of the microsatellite markers described here to investigate the genetic structure and hybridization patterns in hybrid zones between L. japonicum and L. stellipilum.