Figs (Ficus L., Moraceae) and their pollinating wasps (Hymenoptera: Agaonidae) are a textbook example of pollination mutualism and one of the best cases of coevolution known (Jousselin et al., 2003). With more than 750 species, figs are a common component of tropical and subtropical habitats and are completely dependent on their pollinator wasps for pollen dispersal and the production of viable seeds. In turn, the pollinating wasps gall a subset of the developing fig seeds and so are completely dependent on the fig for the completion of their life cycle. Ficus hirta Vahl is a dioecious shrub or small tree with a broad distribution at low elevations in Southeast Asia, where it plays an important role in forest succession and renewal in human-disturbed mountainous landscapes. Consequently, understanding the genetic diversity of this species is relevant to forest restoration and conservation, as well as co-evolutionary studies.

Microsatellite loci have become a powerful tool for analyzing population genetic structure, gene flow, and levels of inbreeding. Many microsatellite primers have been developed for genetic studies of figs (Zavodna et al., 2005; Crozier et al., 2006; Ikegami et al., 2009; Zhang et al., 2011). Moreover, microsatellite primers for F. hirta's pollinator have been developed (Tian et al., 2011). Six of the microsatellite primers developed for other fig species are polymorphic in F. hirta; however, two of these loci exhibit significant excesses of homozygotes within populations, suggestive of null alleles (Yu et al., 2010). Additional marker loci are needed for population genetic studies of F. hirta; therefore, we have used a next-generation transcriptome sequencing approach (RNA-Seq) to develop microsatellites specifically for this species.

METHODS AND RESULTS

Plant material and RNA/DNA extraction—Fresh, female flowers from pre-receptive and from receptive male syconia of F. hirta growing at the South China Botanical Garden (IBSC; Guangdong Province) were excised with tweezers and then immediately frozen in liquid nitrogen for RNA extraction. A voucher specimen (Huiyu 368) has been preserved in the IBSC herbarium. Total mRNA was extracted from each of the two samples using a modified cetyltrimethylammonium bromide (CTAB) method (Doyle and Doyle, 1987). For each sample, an RNA-Seq library was constructed using the Illumina TruSeq RNA Sample Preparation Kit (Illumina, San Diego, California, USA) following the manufacturer's recommendations, and then sequenced on an Illumina HiSeq 2000 platform. Both cDNA library construction and Illumina sequencing were carried out by Novogene (Beijing, China). In total, 56,047,406 (prereceptive) and 60,136,746 (receptive) raw paired-end reads at least 100 bp in length were obtained. For testing new microsatellite loci, we sampled 94 individuals of F. hirta from four distinct, natural populations: three in China (DA: 19°33′02″N, 110°21′84″E; MH: 18°47′42″N, 110°23′27″E; WN: 21°58′46″N; 100°27′01″E) and one in Vietnam (YC: 20°15′182″N, 105°42′725″E). All of these individuals had morphological features typical of F. hirta var. hirta. Genomic DNA of each individual was extracted from silica gel–dried leaves using the modified CTAB method (Doyle and Doyle, 1987), and one voucher specimen per population was preserved in the IBSC herbarium (Huiyu-DA1, Huiyu-MH1, Huiyu-WN1, and Huiyu-YC1).

Microsatellite marker development—The program MISA (Thiel et al., 2003) was used to identify and localize putative microsatellite motifs in the F. hirta transcriptome sequence library. To be identified as a potential microsatellite locus, mononucleotide repeats were required to extend at least 10 repeats, dinucleotide repeats were required to extend at least six repeats, and repeats of all other motif lengths were required to extend at least five repeats. MISA identified 35,558 putative microsatellite loci. With this detailed information, the program Primer3 (Rozen and Skaletsky, 1999) was then used to design 11,468 primer pairs containing di-, tri-, tetra-, penta-, and hexanucleotide repeat motifs, with a melting temperature (T_m) ranging from 57°C to 63°C and fragment size from 100 to 300 bp.

Microsatellite screening—Primer pairs were synthesized (Invitrogen Trading Shanghai Co. Ltd., Shanghai, China) for 56 loci. The 56 loci were randomly selected with the constraint of having expected product sizes between 150 and 300 bp. Dinucleotide repeat loci had at least nine repeats and trinucleotide repeat loci had at least six repeats. These primer pairs were subsequently tested for proper PCR amplification in F. hirta using genomic DNA. PCRs were performed in a 20-µL volume containing 3 µL of 10× reaction buffer (containing Mg²⁺), 2 µL dNTP mix (10 mM; 2.5 mM each), 0.4 µM primers, 1 µL of 50 ng DNA, and 1 unit of Taq polymerase (TaKaRa Biotechnology Co., Dalian, China). PCR cycling was performed in the following conditions: one cycle of denaturation at 95°C for 3 min; 35 cycles for 45 s at 94°C, 45 s annealing at 60°C, and 45 s extension at 72°C; followed by a final 10-min extension at 72°C. The PCR products were checked on 1.5% agarose gels. Sixteen of the 56 primer pairs tested were found to amplify successfully in F. hirta (Table 1). To assess polymorphism of the 16 microsatellite primers, genomic DNA templates from the 94 F. hirta individuals from the four natural populations were used. PCR amplification was performed using forward primers labeled with fluorescent dyes (TAMRA, HEX, and FAM) with the same PCR reaction and cycling conditions as above. The PCR products were detected on an ABI PRISM 3100 Genetic Analyzer (Applied Biosystems, Carlsbad, California, USA) using Genotyper 4.0 and GeneScan 500 LIZ (Applied Biosystems) as an internal lane standard. Fragment lengths were calculated using GeneMapper 4.0 (Applied Biosystems) software. Nine of the 16 microsatellites were found to be polymorphic in F. hirta (Table 1).

Table 1.

Characteristics of 16 microsatellite loci developed for Ficus hirta.

Microsatellite marker data analysis—Genotypes were scored by analyzing the electropherograms from the fragment analysis using GeneMarker version 1.95 (SoftGenetics, State College, Pennsylvania, USA). Descriptive statistics, including the number of alleles and the observed and expected heterozygosities for each F. hirta population, were estimated using GenAlEx version 6.1 (Peakall and Smouse, 2006) (Table 2). For each population, the number of alleles per locus at the nine loci ranged one to 11 with a total of 85 alleles summed across loci. Observed heterozygosity values ranged from 0.000 to 0.955, and expected heterozygosity ranged from 0.000 to 0.882. FSTAT 2.9.3 (Goudet, 2001) was used to test for deviations from Hardy–Weinberg equilibrium for each locus in each population and linkage disequilibrium between loci. Using a sequential Bonferroni correction for multiple comparisons (Rice, 1989), these tests revealed that two loci (FH1, FH21) were not in Hardy–Weinberg equilibrium (as expected for dioecious F. hirta), and that levels of linkage disequilibrium between loci were not significant.

Cross-species amplifications—The 16 primer pairs that amplified successfully in F. hirta were tested for amplification in one individual of each of 29 additional Ficus species, using the same procedures as above. All 16 were found to amplify in at least 11 of these species, with most amplifying in a majority of the species (Appendix 1).

CONCLUSIONS

The microsatellite loci described in this study provide an important resource for studying the factors affecting the distribution of genetic diversity in F. hirta and other Ficus species. We can use these loci to further investigate large-scale and fine-scale population genetic structure in F. hirta, building on our past work in this species investigating relative rates of pollen and seed migration (Yu et al., 2010) and the constraints to postglacial range expansion arising from restricted pollen or seed dispersal (Yu and Nason, 2013). Our results also demonstrate the application of RNA sequencing as a valuable tool for developing microsatellite loci in nonmodel species.

Table 2.

Nine microsatellite locus–specific measures of genetic diversity across four populations of Ficus hirta.^a