Development and characterization of microsatellite markers in Campomanesia adamantium, a native plant of the Cerrado ecoregions of South America

Premise A novel set of nuclear microsatellite markers was developed and characterized for Campomanesia adamantium (Myrtaceae) and tested for cross‐amplification in the related species C. sessiliflora. Methods and Results Forty‐one primer pairs were designed for simple sequence repeat loci, of which 36 successfully amplified and were polymorphic. The number of alleles ranged from two to 14, with an average of 8.14 alleles per locus. Additionally, cross‐amplification was tested in C. sessiliflora; more than 55.5% of the microsatellite loci amplified, confirming the use of these microsatellite markers in a related species. Conclusions We developed a set of microsatellite markers that will be useful for future studies of genetic diversity and population structure of C. adamantium and a closely related species, which will aid in future conservation efforts.

and SSR primers from Eucalyptus spp., Eugenia uniflora L., and Melaleuca alternifolia Cheel (Fagundes et al., 2016). Analysis of genetic diversity using transferable molecular markers only reflects polymorphisms in conserved genomic regions among congeneric species or genera from the same family; however, transferable microsatellite markers from other species may be limited in the level of genetic diversity they can reveal in the target species (Queirós et al., 2015). Thus, the use of species-specific microsatellites could more accurately report the genetic variability in C. adamantium and detect aspects of biodiversity that could be omitted by the use of transferable markers.
Therefore, the goal of this study was to isolate and characterize microsatellite markers for C. adamantium by constructing a microsatellite-enriched genomic library and to test the cross-amplification of these markers in C. sessiliflora (O. Berg) Mattos, a species native to Brazil with medicinal and economic value.

Plant material and DNA extraction
Young leaf tissue samples from one accession of C. adamantium (voucher 4666, Herbarium of the Federal University of Grande Dourados [DDMS], Dourados, Mato Grosso do Sul, Brazil) were collected to develop the microsatellite markers, and 45 samples from three natural populations were collected to validate the microsatellite markers for this species (Appendix 1). Ten samples of C. sessiliflora were collected from a single population to test cross-amplification of the markers in a related species. Detailed information about all samples collected for this study is provided in Appendix 1. DNA was extracted from young leaf tissue using the cetyltrimethylammonium bromide (CTAB) method (Doyle and Doyle, 1987).

Development of SSRs and primer design
Genomic DNA of C. adamantium was used to develop a microsatellite-enriched genomic library using a protocol adapted from Billotte et al. (1999). The isolation procedure consisted of digestion of genomic DNA with the AfaI restriction enzyme (Invitrogen, Carlsbad, California, USA) and ligation of fragments with double-stranded adapters 5′-CTCTTGCTTACGCGTGGACTA-3′ and 5′-TAGTCCACGCGTAAGCAAGAGCACA-3′. The enrichment was based on hybridization capture using (CT) 8 and (GT) 8 biotin-linked probes and streptavidin-coated magnetic beads (MagneSphere Magnetic Separation Products; Promega Corporation, Madison, Wisconsin, USA). Microsatellite-enriched DNA fragments were amplified by PCR, cloned into a pGEM-T Easy Vector (Promega Corporation), and then inserted into Escherichia coli XL1-Blue competent cells (Promega Corporation) using an electroporation technique. Positive clones were selected via a culture medium containing ampicillin and β-galactosidase.
Ninety-six colonies were selected using blue/white screening, and the sequencing was performed on an automated ABI 3500xL Genetic Analyzer (Applied Biosystems, Foster City, California, USA) with T7 and SP6 primers and the BigDye Terminator version 3.1 Cycle Sequencing Kit (Perkin Elmer-Applied Biosystems). Microsatellite sequences were identified in 41 clones with good quality sequences for primer design using Primer3Plus software (Untergasser et al., 2012), under the following parameters: size of final amplification products 100-350 bp, GC percentage minimum 40% and maximum 60%, primer annealing temperature ranging from 57°C to 65°C, and the maximum difference in annealing temperature between primer pairs of 3°C.

Primer validation and microsatellite marker evaluation
Forty-one primer pairs were designed and tested for amplification in C. adamantium in three populations (n = 45). In addition, for markers CAMP01, CAMP03, CAMP04, CAMP08, CAMP13, CAMP17, CAMP24, CAMP25, CAMP28, and CAMP36, the genotyping was performed with n ≥ 45, using additional individuals sampled from Crispim et al. (2018). Cross-species amplification was tested using the 36 most polymorphic markers (Table 1). PCR amplifications were performed in a 25-μL reaction volume containing 50 ng of DNA, 7.5 μL of ultra-pure water (Fermentas, Waltham, Massachusetts, USA), 0.15 μM of forward and reverse primer, and 12.5 μL of PCR Master Mix (50 U/mL of Taq polymerase DNA, 400 μM of dNTP, and 3 mM of MgCl 2 ) (Fermentas). PCR cycling conditions consisted of an initial denaturation at 94°C for 5 min; followed by 35 cycles of denaturation at 94°C for 1 min, annealing at 60°C for 1 min, and extension at 72°C for 1 min; and a final extension cycle at 72°C for 15 min. Amplified products were verified visually on 2% agarose gels. Polymorphism evaluation and genotyping were performed using electrophoresis on a 7% polyacrylamide gel stained with silver nitrate (Creste et al., 2001). To determine fragment sizes, we used 10-bp and 50-bp DNA ladders (Thermo Fisher Scientific, Waltham, Massachusetts, USA). The fragments were estimated to within 5 bp.
To evaluate the differences between the genetic diversity parameters of C. adamantium resulting from the use of transferable microsatellite markers and specific species, we selected 10 markers developed in this study (CAMP01, CAMP03, CAMP04, Statistical analyses were performed using GenAlEx version 6.5 software (Peakall and Smouse, 2012) to calculate the number of alleles per locus, the level of expected heterozygosity (H e ) and observed heterozygosity (H o ), and the polymorphism information content. We used the statistical environment software R (R Core Team, 2018) to estimate the multilocus genotype using the poppr R package (Kamvar et al., 2014). Tests for deviation from Hardy-Weinberg equilibrium (HWE) for proportions at each locus for each population were performed using the diveRsity R package (Keenan et al., 2013). The Bonferroni correction was used to correct multiple applications of the same test. We also estimated inbreeding coefficients (F IS ) within each population using the same package. Confidence intervals were obtained with 10,000 bootstrap replicates. The test for null allele presence was performed using MICRO-CHECKER (van Oosterhout et al., 2004). The influence of null alleles was determined in FREENA (Chapuis and Estoup, 2007) by computing the genetic divergence parameter (F ST ) values using an excluding null alleles (ENA) correction. After accounting for null allele frequencies, loci with frequencies of ≥0.2 were considered potentially problematic for the calculations.
Overall, C. adamantium primers successfully amplified 36 SSR loci, in which all markers were polymorphic in the analyzed populations (n = 45); the number of alleles ranged from two to 14, with an average of 8.14 per locus. The H o and H e ranged from 0.00 to 0.91 (average 0.52) and from 0.09 to 0.89 (average 0.78), respectively. When we tested the previously mentioned set of 10 markers (CAMP01, CAMP03, CAMP04, CAMP08, CAMP13, CAMP17, CAMP24, CAMP25, CAMP28, CAMP36) using an increased number of individuals (n ≥ 45), the number of alleles ranged from 11 to 23, with an average of 17.10. The level of H o and H e ranged from 0.24 to 0.72 (average 0.51) and from 0.81 to 0.92 (average 0.87), respectively (Table 2). These results confirm the reliability of the designed set of markers to evaluate genetic diversity in further studies of this species.
Significant deviations from HWE based on Fisher's exact test (P < 0.05) in C. adamantium were detected for 11 loci in the Dourados population, 15 loci in the Bonito population, and 10 loci in the Cerro Corá population (Table 2). When overall populations were considered, only two markers (CAMP21 and CAMP30) did not significantly deviate from HWE. Several factors (e.g., insufficient sample size, seed dispersal [Hedrick, 2005]) contributed to the observed deviations from HWE. In all sampled regions, C. adamantium was observed as a branched tree, very often found in a group of bushes of plants of the same species with the possibility of kinship among sampled individuals; this also was verified by Nucci and Alves-Junior (2017). Although some individuals with identical multilocus genotypes were found, the observed deviations from HWE may be due to factors such as population subdivision and the presence of null alleles.
The test for null alleles indicated significant results in some loci in the populations ( Table 2). The most frequent loci were CAMP01, CAMP02, CAMP11, CAMP14, CAMP26, CAMP28, and CAMP37, which showed a significant possibility of the presence of null alleles in two of the three tested populations. After null allele correction (ENA), overall F ST changed only slightly (from 0.302 to 0.283).
In C. sessiliflora, 20 microsatellite loci cross-amplified (Table 3). The number of alleles ranged from two to five, with an average of 2.85. Significant deviations (P < 0.05) from HWE were verified in three loci.

CONCLUSIONS
We developed a panel of microsatellite markers that will be helpful for future studies of genetic diversity and population structure of C. adamantium. The microsatellite loci described in this study successfully cross-amplified in C. sessiliflora, suggesting that these markers could be used to support genetic conservation and breeding programs for C. adamantium and other species in the genus.  Crispim et al. (2018). *Fisher's exact test significant for Hardy-Weinberg equilibrium proportions after Bonferroni correction (P < 0.003).