Acer miyabei Maxim. (Sapindaceae) is a deciduous tree species that grows in temperate forests in East Asia. The species comprises three infraspecific taxa: A. miyabei Maxim. subsp. miyabei f. miyabei, A. miyabei subsp. miyabei f. shibatae (Nakai) K. Ogata, and A. miyabei subsp. miaotaiense (Tsoong) A. E. Murray. Each subspecies has a characteristic distribution (Ogata, 1965; van Gelderen et al., 1994). Acer miyabei subsp. miyabei f. miyabei grows in Hokkaido and northern and central Honshu, Japan. Its occurrence is strongly associated with river flood-plain ecosystems, and some of the isolated southern populations are considered a relic of glacial times. Acer miyabei subsp. miyabei f. shibatae is also endemic to Japan, although its range is restricted to parts of Honshu. Acer miyabei subsp. miaotaiense was found in 1954 in Shaanxi Province in northwestern China (Tsoong, 1954). The discovery of this taxon is important because its distribution is likely a biogeographic stepping stone to A. campestre L., a morphologically similar European species (Ogata, 1967). Yet, the phylogenetic relationships among the subspecies, forms, and their related species have not been examined at the molecular level. Because of their limited range and habitat decline, all three infraspecific taxa of A. miyabei are listed in national or IUCN Red Lists (Ministry of the Environment, Government of Japan, 2012; IUCN, 2014). Natural populations of A. miyabei in Japan are typically fragmented by urban and rural development, which affects seed production and gene flow (Hotta, 2004; Nagamitsu et al., 2014).
Here, we present 12 microsatellite markers for A. miyabei to facilitate evolutionary and conservation studies. These markers were developed from two forms of A. miyabei subsp. miyabei, and tested on two natural populations of A. miyabei subsp. miyabei f. miyabei and an individual of A. miyabei subsp. miaotaiense. We also examined the transferability of the markers to three species that belong to the same section (sect. Platanoidea) as A. miyabei (Renner et al., 2007; Grimm and Denk, 2014): A. campestre, A. platanoides L., and A. pictum Thunb.
METHODS AND RESULTS
Microsatellite markers were developed for A. miyabei with an Ion Personal Genome Machine (PGM; Life Technologies, Carlsbad, California, USA). Library preparation, PGM sequencing, and genotyping were conducted at the Sugadaira Montane Research Center, University of Tsukuba, Japan. Total genomic DNA was extracted from dried leaves of a single A. miyabei subsp. miyabei f. miyabei individual from Sugadaira with a DNeasy Plant Mini Kit (QIAGEN, Hilden, Germany). The voucher specimen was stored at the Herbarium of Sugadaira Montane Research Center (no. 05507). The concentration of genomic DNA was determined with a Qubit 2.0 Fluorometer (Life Technologies).
The genomic DNA (100 ng) was sheared into fragments of 350–450 bp with Ion Shear Plus Reagents (Life Technologies), and adapter ligation, nick repair, and purification of the ligated DNA were conducted with an Ion Plus Fragment Library Kit (Life Technologies). Fragments of 300–350 bp were selected with an E-Gel Agarose Gel Electrophoresis System (Life Technologies), followed by library amplification with an Ion Plus Fragment Library Kit. The library was assessed and quantified with a BioAnalyzer (Agilent Technologies, Palo Alto, California, USA), and then diluted to 26 pM for template preparation. The library was enriched with an Ion PGM Template OT2 400 kit (Life Technologies) and sequenced with an Ion PGM Sequencing 400 Kit (Life Technologies) by using 850 flows on Ion 314 Chip V2 (Life Technologies) according to the manufacturer's protocol. Single processing and base calling were performed with Torrent Suite 3.6 (Life Technologies), and a library-specific FASTQ file was generated. A total of 557,106 reads were obtained and registered in the DNA Data Bank of Japan (DDBJ) Sequence Read Archive (DRA001873).
Characteristics of 12 polymorphic microsatellite markers developed for Acer miyabei.
Genetic diversity of 12 microsatellite loci in two natural populations of Acer miyabei (Bibi and Kyouwa) in Hokkaido, Japan.
The data sets were collated and applied to the QDD bioinformatics pipeline (Meglécz et al., 2010) to filter sequences containing microsatellites with appropriate flanking sequences to define PCR primers. QDD detected 4909 loci, each containing a microsatellite consisting of at least five repeats. Based on this information, we chose 58 primer pairs for loci consisting of either di- or trinucleotide repeats. For initial primer screening by PCR, we used four DNA samples from three A. miyabei subsp. miyabei f. miyabei individuals from the Bibi, Kushiro, and Sugadaira populations and one A. miyabei subsp. miyabei f. shibatae individual from the Sugadaira population (Appendix 1).
Each forward primer was labeled with either FAM, HEX, or TAMRA fluorescent dye. We also prepared unlabeled forward primers and mixed them with fluorescent ones. The ratio was initially set at 1 (fluorescent) to 24 (unlabeled) but was changed later as described below, following Suyama (2012). All reverse primers were PIG-tailed by adding GTTTCTT to obtain consistent addition of adenine by Taq DNA polymerase (Brownstein et al., 1996). DNA (ca. 10 ng) was placed into wells of 96-well plates and dried at room temperature over several hours. Singleplex PCR was performed with a single pair of primers in 2 µL of 1× Type-It Microsatellite PCR Kit Master Mix (QIAGEN) and 0.2 µM of each primer, overlaid with 6 µL of mineral oil as described in Kenta et al. (2008). The thermal cycler program was 95°C for 5 min; followed by 35 cycles of 95°C for 30 s, 60°C for 90 s, and 72°C for 30 s; and 72°C for 30 min. PCR products were mixed with 0.25 µL of GeneScan 500 LIZ Size Standard (Applied Biosystems) and 9.25 µL of Hi-Di formamide (Applied Biosystems). Samples were run on an ABI3130 Genetic Analyzer (Applied Biosystems), and PCR products were examined in GeneMapper ver. 4.0 (Applied Biosystems). If fluorescent signal intensity was too high or too low, the ratio of the fluorescent forward primer to the unlabeled one was optimized (Table 1). However, even at high ratios of fluorescent forward primers, products labeled with TAMRA were relatively poorly detectable, and thus we excluded the corresponding loci. Screening resulted in 18 primer pairs that consistently amplified clear bands. Eight of these primer pairs were newly labeled with PET or NED fluorescent dyes for performing two sets of multiplex-PCR reactions. PCR mixtures and the thermal cycler program were as described above. Multiplex amplification was successful under such conditions, and we finally tested DNA of 44 samples from two natural populations of A. miyabei subsp. miyabei f. miyabei at Bibi and Kyouwa in Chitose city, Hokkaido, Japan (Appendix 1). In these samples, nonspecific bands with three or more peaks were detected for six microsatellite loci. Thus, we consider the remaining 12 primer pairs (Table 1) as useful. No known genes were detected around the corresponding markers by BLAST searches with read sequence data.
Cross-amplification of 12 microsatellite loci in species closely related to Acer miyabei.a
For those markers, the mean number of alleles per locus was 3.42 in the Bibi population and 3.67 in the Kyouwa population (Table 2). For the Bibi population, the mean observed and expected heterozygosities per locus were 0.58 (range: 0.09–0.91) and 0.55 (0.09–0.79), respectively; for the Kyouwa population, the corresponding values were 0.48 (0.00–0.86) and 0.48 (0.00–0.77). For the two populations combined, the number of alleles per locus ranged from two to eight, whereas the observed and expected heterozygosities per locus were 0.05–0.75 and 0.05–0.79, respectively. These statistics were computed by CERVUS 3.0.7 (Marshall et al., 1998; Kalinowski et al., 2007). Deviations from Hardy–Weinberg equilibrium were tested with GENEPOP software (version 4.2; Raymond and Rousset, 1995). Significant deviations after Bonferroni correction (P < 0.05) were detected for the loci Acmi45 in Bibi and Acmi38 in Kyouwa (Table 2). Null allele frequencies estimated with CERVUS 3.0.7 (Marshall et al., 1998; Kalinowski et al., 2007) were nearly zero or negative except for Acmi2 and Acmi45 in the Bibi population. Cross-amplifications were carried out to test marker transferability to closely related taxa. All of the 12 loci were amplified with clear bands in a sample of A. miyabei subsp. miaotaiense (Appendix 1). Polymorphic variation was consistently detected in 10 microsatellite loci in A. campestre, five in A. platanoides, and four in A. pictum (Table 3). The result agrees with a morphological similarity between A. miyabei and A. campestre as demonstrated by Ogata (1967).
Using next-generation sequencing with the Ion PGM system, we developed 12 microsatellite markers for the threatened maple A. miyabei. These markers will help to characterize the genetic structure and diversity of the species. They will also help to understand its spatial genetic variation, levels of inbreeding, and patterns of gene flow, thereby providing a basis for conservation. Some of the markers were successfully transferred to closely related species. High transferability to A. campestre agrees with its morphological similarity to A. miyabei.
 The authors thank S. Yamaguchi, Y. Yamaguchi, O. Harada, R. Oyama, Dr. S. Kondoh, Dr. T. Hiura, Dr. T. Nagamitsu, Dr. H. Matsumura, and Dr. B. V. Barnes for their valuable support. The leaf specimen of Acer miyabei subsp. miaotaiense was kindly provided by the University of British Columbia Botanical Garden. This research was funded by the Japan Society for the Promotion of Science (grant no. 25890002) and the Fujiwara Natural History Foundation.