Psychotria L. (Rubiaceae) has been recognized as an important model system for the study of heterostyly and its evolutionary transition on oceanic islands (Watanabe and Sugawara, 2015). Psychotria homalosperma A. Gray is an evergreen tree found only in the Chichijima and Hahajima island groups of the oceanic Bonin Islands in the northwest Pacific Ocean. Previous studies have reported that the species is distylous with self- and intramorphic incompatibilities (Watanabe et al., 2014). Revealing the mating system and the gene flow patterns in this species will help in understanding the evolutionary significance of heterostyly on oceanic islands. Meanwhile, the Red List of Threatened Plants of Japan and Red List of Threatened Species in Tokyo have described P. homalosperma as “vulnerable to extinction” (Tokyo Metropolitan Government, 2011; Ministry of the Environment, 2012). Recently, with the exception of some populations on Hahajima Island, natural populations of P. homalosperma did not regenerate successfully, apparently because of disturbances from human activities (Watanabe et al., 2009; Sugai et al., 2015). Therefore, genetic information (e.g., genetic diversity within populations and genetic differentiation between islands) will be important for the development of an effective conservation plan for this species.
Here, we developed 26 microsatellite (simple sequence repeat [SSR]) markers for P. homalosperma for use in evolutionary and conservation studies. These markers were tested on two natural populations of P. homalosperma because it is currently distributed only in the two island groups of the Bonin Islands. We also examined the transferability of these markers to four species of Psychotria (P. boninensis Nakai, P. rubra (Lour.) Poir., P. manillensis Bartl. ex DC., and P. serpens L.) that occur naturally in Japan and adjacent areas.
METHODS AND RESULTS
Total genomic DNA of P. homalosperma was extracted from a fresh leaf collected from Sekimon (26°40′11.3″N, 142°09′16.4″E) on Hahajima Island, using the DNeasy Plant Mini Kit (QIAGEN, Hilden, Germany). A voucher specimen of this sample was deposited in the Makino Herbarium (MAK) of Tokyo Metropolitan University, Japan (Appendix 1). The extracted DNA of P. homalosperma was pyrosequenced using a 454 GS Junior System (Roche, Basel, Switzerland). Multiplex Identifier (MID) tags were used for multiplexing of the abovementioned sample and the sample of another species in the Bonin Islands, i.e., Gynochthodes boninensis (Ohwi) E. Oguri & T. Sugaw. These samples were then combined. The raw data were demultiplexed and MID tags removed from the reads using Newbler (Roche). The identification of SSRs and design of primers from the above DNA sequences were performed using the QDD 2.1 program (Meglecz et al., 2010). This program is composed of the following steps to design PCR primers: (1) selection of sequences that contain SSRs, (2) elimination of redundant sequences, (3) primer design, and (4) contamination detection.
The de novo sequencing produced 148,586 reads with an average length of 422 bp. These reads were registered in the DNA Data Bank of Japan (DDBJ) Sequence Read Archive (DRA004086). SSR loci were identified as having bordered sequences with more than five repeats for di- to hexa-nucleotide motifs, and the length of one sequence was more than 80 bp. According to these criteria, a total of 5544 reads contained SSR loci. To eliminate redundancy, all sequences containing SSRs were subjected to an “all-against-all” BLAST with an E-value of 1E–40. Subsequently, 2384 reads were selected from whole sequences containing SSRs. PCR primers were designed using Primer3 (Rozen and Skaletsky, 2000) implemented in the QDD program. Finally, a total of 1475 SSR primer pairs were designed by Primer3.
Table 1.
Twenty-six SSR markers for Psychotria homalosperma.a
Amplification and polymorphism tests were performed for 48 selected primer pairs with consideration for the SSRs (single motifs of di-, tri-, tetra-, and pentanucleotides with 8–13 repeats) and the type of design (“A” in QDD 2.1). Four universal primers with different fluorescent tags, as designed by Blacket et al. (2012), were prepared. The 5′ end of each forward primer was attached to the same sequence as a tail. Additionally, all reverse primers were given a PIG-tail (5′-GTTT-3′, 5′-GTT-3′, or 5′-G-3′) at the 5′ end of the sequence to reduce stuttering due to the inconsistent addition of adenine by Taq DNA polymerase (Brownstein et al., 1996). PCR amplification was performed using the Type-it Microsatellite PCR Kit (QIAGEN). Multiplex PCRs were performed for each of the four primer pair sets. The thermal cycler program was as follows: 95°C for 5 min followed by 32 cycles of 95°C for 30 s, 57°C for 1.5 min, and 72°C for 30 s, as well as a final extension step of 60°C for 30 min. PCR products were mixed with a GeneScan 600 LIZ Size Standard (Life Technologies, Waltham, Massachusetts, USA) and loaded onto an ABI3130 Genetic Analyzer (Life Technologies). Fragment lengths were examined using GeneMapper 4.0 (Life Technologies).
We then tested two populations of P. homalosperma from Chichijima and Hahajima islands to evaluate their allelic polymorphism. A total of 48 individuals were tested: 24 from Higashidaira (27°04′35.7″N, 142°13′14.9″E) on Chichijima Island and 24 from Sekimon (26°40′26.3″N, 142°09′21.6″E) on Hahajima Island. In addition, transferability among four other Psychotria species occurring in Japan (P. boninensis [N = 8] from Chichijima Island of the Bonin Islands; P. rubra [N = 8], P. manillensis [N = 8], and P. serpens [N = 8] from the Ryukyu Islands) was tested using the same PCR conditions described above. Voucher specimens of the representative individuals were deposited in MAK (Appendix 1). To characterize each locus, the number of alleles per locus (A), observed heterozygosity (Ho) expected heterozygosity (He), and fixation index (FIS) were calculated using GenAlEx 6.501 (Peakall and Smouse, 2006). The Hardy–Weinberg equilibrium (HWE) at each locus of each population and the linkage disequilibrium (LD) between loci of each population were tested with FSTAT 2.9.3.2 (Goudet, 2002).
Of the 48 tested SSR markers, 26 primer pairs were successfully amplified and showed polymorphism among 48 individuals of P. homalosperma (Table 1). The mean A was 6.50 (1−19) in the Chichijima population and 6.81 (2−18) in the Hahajima population (Table 2). For the Chichijima population, the mean Ho and He were 0.547 (0.083−1.000) and 0.578 (0.080−0.905), respectively. For the Hahajima population, the corresponding values were 0.581 (0.043−0.917) and 0.606 (0.122−0.910), respectively (Table 2). None of the loci deviated significantly from HWE. No significant LD between markers was observed in either of the populations. Of the 26 SSR markers tested, 10 were successfully amplified for P. boninensis and eight for P. serpens (Table 3), while none could be amplified for P. rubra or P. manillensis (Table 3).
Table 2.
Characteristics of 26 SSR markers in the two populations of Psychotria homalosperma.
CONCLUSIONS
Twenty-six novel SSR markers were developed for P. homalosperma using a next-generation sequencing approach. These markers are likely to be useful for evaluating the genetic structure and gene flow of P. homalosperma, which will subsequently facilitate the development of a conservation strategy for this species. The developed markers are unlikely to be useful for the study of the other tested Psychotria species in Japan, most likely because P. homalosperma is assigned to a section (sect. Pelagomapouria Fosb.) that is different from those of the other tested species (Yamazaki, 1993); moreover, P. rubra and P. manillensis are polyploid (the former is tetraploid [2n = 42] and the latter octoploid [2n = 84]) (Nakamura et al., 2003). However, future studies should examine the applicability of these markers to critically endangered sect. Pelagomapouria species found in the Hawaiian Islands (U.S. Fish and Wildlife Service, 2015).
Table 3.
Transferability of the 26 SSR markers for the four species of Psychotria in Japan.a
ACKNOWLEDGMENTS
The authors would like to thank A. Hisamatsu and other members of the Department of Forest Genetics (Forestry and Forest Products Research Institute) for their technical support. The authors are also grateful to A. Mukai for collecting the plant samples. This study was supported by the Environment Research and Technology Development Fund of the Ministry of the Environment, Japan (4-1402), and a Grant-in-Aid for Scientific Research to K.W. (no. 26840130).