Cratylia (Desv.) Kuntze is a South American genus distributed eastward from the Andes and southward from the Amazon River Basin. Five species are currently accepted: C. argentea (Desv.) Kuntze, C. bahiensis L. P. Queiroz, C. hypargyrea Mart. ex Benth., C. intermedia (Hassl.) L. P. Queiroz & R. Monteiro, and C. mollis Mart. ex Benth. (Queiroz and Coradin, 1995). Cratylia argentea is the most widespread, occurring from western Peru to Bolivia and Brazil in habitats including cerrado and seasonally dry forests. The other species are restricted to a particular type of vegetation: C. hypargyrea is found in Brazilian coastal Atlantic Forest, C. bahiensis and C. mollis grow in seasonally dry forest of northeastern Brazil, and C. intermedia occurs to the west of Paraná and to the northeast of Misiones Province in Argentina (Queiroz and Coradin, 1995).
Cratylia mollis and C. argentea are multipurpose shrub legumes with high potential for animal nutrition due to their nutritive value, particularly for dry season supplementation and silage (Argel and Lascano, 1998), and also show also tolerance to drought and adaptation to acidic soils (Andersson et al., 2006). Recent studies revealed that C. mollis seeds are an important lectin source (known as Cramoll) with applications for medicine (Fernandes et al., 2010). In the current study, our aim was to develop microsatellite markers to assess the genetic variation of C. mollis, and to assess their potential transferability to C. argentea and cross-amplification in other species of the genus.
METHODS AND RESULTS
Genomic DNA was isolated from one individual cultivated on the campus of the Universidade Estadual de Feira de Santana (J. G. Rando 1257, see Appendix 1) and extracted using the cetyltrimethylammonium bromide (CTAB) 2× protocol. Microsatellite markers were developed using both microsatellite-enriched library and intersimple sequence repeat (ISSR)–based cloning methods. A microsatellite-enriched library was built using the biotin-labeled microsatellite oligoprobes (GT)8 and (CT)8 and streptavidin-coated magnetic beads (Promega Corporation, Madison, Wisconsin, USA) according to Billotte et al. (1999). For the ISSR method, we performed PCR reactions with the ISSR primers GGCC(AG)8, GGCC(AC)8, CCGG(AG)8, CCGG(AC)8, GCGC(AG)8, and GCGC(AC)8 following closely the protocol of Provan and Wilson (2007). The selected fragments obtained by both methods were linked into pGEM-T Easy Vector (Promega Corporation, São Paulo, Brazil) and then transformed and cloned in TOP10 competent cells (Invitrogen, Life Technologies, Vila Guarani, São Paulo, Brazil).
Positive clones were amplified with the T7 (5′-TAATACGACTCACTATAGGG-3′) and SP6 (5′-ATTTAGGTGACACTATAGAA-3′) primers using the following conditions: denaturation step of 94°C for 5 min, followed by 35 cycles of denaturation at 94°C for 15 s, annealing at 55°C for 35 s, and extension at 72°C for 90 s; final extension was at 72°C for 7 min. DNA sequencing was performed with Big Dye Terminator Cycle Sequencing Kit (Applied Biosystems, São Paulo, Brazil). The software Imperfect Microsatellite Extractor (Mudunuri and Nagarajaram, 2007) was used for identification of simple sequence repeats in the nonredundant sequences.
Primers were designed using Primer3Plus (Rozen and Skaletsky, 2000) using the following criteria: product size range 100–300 bp, primer melting temperature (Tm) 50–60°C, primer GC 40–60%, maximum Tm difference: 2°C. Genotyping reactions were carried out in 10 µL with the Top Taq Master Mix kit (QIAGEN, Hilden, Germany) with approximately 4 ng of genomic DNA, 0.15 µM of the forward primer (added M13 sequence: 5′-CACGACGTTGTAAAACGAC-3′), 0.30 µM of the reverse primer, and 0.60 µM of an M13-labeled tail (with one of four fluorescent dyes: 6-FAM, VIC, NED, or PET). For each primer pair, the optimal annealing temperature was established in gradient PCRs from 61–t5°C; however, a touchdown program (60–54°C) was needed in some cases. For polymorphism assessment, individuals of two C. mollis populations (CMPOP_1, N = 24; CMPOP_2, N = 25) and 20 individuals cultivated from seeds of CMPOP_2 were genotyped. To test the cross-amplification and potential transferability, we sampled individuals obtained from the Laboratory of Plant Molecular Systematics (LAMOL) DNA bank at the Universidade Estadual de Feira de Santana (see Appendix 1): C. argentea (N = 9), C. bahiensis (N = 2), C. hypargyrea (N = 2), and C. intermedia (N = 2). Microsatellite profiles were analyzed with GeneMapper 4.0 (Applied Biosystems). The estimates of allelic diversity and deviations from Hardy–Weinberg equilibrium were carried out using GenAlEx 6.5 (Peakall and Smouse, 2006).
We amplified and sequenced 180 of 192 positive clones derived from the microsatellite-enriched and ISSR libraries and found 78 nonredundant sequences containing microsatellite loci. Percentage of dinucleotide motifs was higher (62%) than trinucleotide motifs (34%) or tetra-, penta-, and octanucleotide repeats (4% combined). The number of repetitions in dinucleotides varied from three to nine, in trinucleotides from three to five, in tetranucleotides from three to four, and in pentanucleotides from four to five. The motifs found in the sequences of the enriched library had fewer repetitions (3–4) than the ISSR library, and for this reason most of these were not used to design primers. Therefore, 32 primers pairs were designed (30 corresponding to the ISSR library), from which 19 presented amplification patterns compatible with the expected fragment sizes (Table 1).
Characterization of 19 microsatellite markers developed for Cratylia mollis.
Twelve microsatellite loci were polymorphic and presented considerable variation in allele number (2–11), observed heterozygosity (0.040–0.792), expected heterozygosity (0.226–0.883), and polymorphic information content per locus (0.212–0.870). The markers that were monomorphic in CMPOP_1 (CM_7, CM_15, CM_20) exhibited polymorphism in CMPOP_2 and vice-versa (CM_10), reflecting population differentiation in this species. Cross-amplification in C. argentea was successful in 16 loci; 12 of these were polymorphic, displaying an allele range of 2–4 in only nine individuals evaluated. Four monomorphic loci in C. mollis showed polymorphism in C. argentea. Cross-amplification success rates were also high in C. intermedia (15 loci) and C. bahiensis (12 loci); however, in C. hypargyrea only two microsatellite loci were successfully amplified (Table 2). Significant values (P < 0.01) of deviation from Hardy–Weinberg equilibrium were found in two loci of CMPOP_1 and one locus of CMPOP_2, where levels of observed heterozygosity were lower than expected (Table 2).
This heterozygote deficiency could reflect patterns of the population distribution and reproductive biology of C. mollis. This species presents a facultative breeding system, and experimental tests showed similar success in allogamous and autogamous crosses (Queiroz et al., 1997). It is pollinated by the carpenter bees Xylocopa grisescens Lepeletier and X. cearensis Ducke, which visit flowers that are simultaneously open in the same inflorescence before moving to other inflorescences, thus promoting geitonogamous crosses (L. P. Queiroz, unpublished observations). In addition, the species forms patchy populations on sandy soils (L. P. Queiroz, personal communication), which together with the pollinator behavior favors autogamy and low heterozygosity.
Results of initial microsatellite marker screening in two populations of Cratylia mollis and cross-amplification in other species.
This is the first report of microsatellite markers in C. mollis and the genus as a whole. The observed polymorphism in 12 of 19 microsatellite loci in C. mollis, their successful cross-amplification in C. intermedia and C. bahiensis, and their transferability to C. argentea indicate promising applications for the study of genetic diversity and structure, gene flow, and mating system in the species. These studies can be followed by comparative studies to understand the mechanisms involved in population divergence and speciation in this neotropical genus.
 The authors thank G. Pereira-Silva (Empresa Brasileira de Pesquisa Agropecuária–Centro Nacional de Recursos Genéticos e Biotecnologia [EMBRAPA–CENARGEN]), A. R. Barbosa, C. Snak, J. G. Rando, and E. R. de Souza for supporting this research. This work is part of the project “Patterns of Leguminosae diversity in Brazilian biomes” (SISBIOTA, processes CNPq 563084/2010-3, FAPESB PES0053/2011, and PNX0014/2009).