New microsatellites for the Atacama Desert endemic Balsamocarpon brevifolium (Fabaceae)

Premise Algarrobilla (Balsamocarpon brevifolium, Fabaceae) is an endemic xerophytic shrub restricted to the Atacama Desert in northern Chile. Extensive utilization of the region for coal production has endangered this species. Conservation efforts are underway, with genetic diversity analyses being key to the restoration of these populations. Methods and Results Fifteen new microsatellite markers were developed for B. brevifolium and used to analyze three populations from the Atacama and Coquimbo regions in Chile. Microsatellites were highly polymorphic, with an average of 5.77 alleles per marker and an average level of expected heterozygosity of 0.72. These markers were evaluated and cross‐amplified on two related species (Senna cumingii and Caesalpinia angulata) with partial success. Conclusions The development of this set of markers permits an extensive study of B. brevifolium populations for conservation purposes.

13 plants were available for this population because of the poor preservation conditions of plant material]). The populations were selected based on the distinct geographic distribution of the species. Representative samples of each population were deposited at the Herbarium of La Serena University (ULS), La Serena, Chile (Appendix 1). Heavy deforestation and changing climatic conditions have decimated most populations of this species, making it difficult to obtain a larger number of samples per population. Plant material was also obtained from individuals of Senna cumingii (Hook. & Arn.) H. S. Irwin & Barneby (voucher ULS 4693) and Caesalpinia angulata (Hook. & Arn.) Baill. (ULS 4708) (Appendix 1), species that are related and cohabit with B. brevifolium (Luebert and Pliscoff, 2006), to test for the crossamplification of developed markers.
Total DNA was extracted from young leaves using the DNeasy Plant Mini Kit (QIAGEN, Hilden, Germany) following the manufacturer's instructions. DNA was quantified by using BioSpec-nano (Shimadzu, Kyoto, Japan) and evaluated by electrophoresis on 0.8% agarose gels stained with ethidium bromide. DNA (50 μL, 450 ng/ μL) from representative samples of B. brevifolium (vouchers ULS 14345, ULS 14346, ULS 14347; Appendix 1) was used for microsatellite identification at Ecogenics GmbH (Balgach, Switzerland). For this purpose, size-selected fragments from genomic DNA were enriched for the formation of an SSR-enriched library by using streptavidin-coated magnetic beads and biotin-labeled GATA and GTAT repeat oligonucleotides (Kijas et al., 1994; for a review, see Santana et al., 2009).
The SSR-enriched library was analyzed on an Illumina MiSeq (Illumina, San Diego, California, USA) at Microsynth AG (Balgach, Switzerland) using the Nano 2 × 250 v2 format. The resulting 284,104 sequences were stitched and assembled using MIRA 4.0.1 software (Chevreux et al., 1999), and candidate microsatellites were searched in 22,361 paired-end read assembled contigs using Finder version 4.09 (Benson, 1999). Of the candidate microsatellites, 5983 contigs contained a microsatellite insert with a tetra-or a trinucleotide of at least six repeat units or a dinucleotide of at least 10 repeat units. Finally, Primer3 (Rozen and Skaletsky, 1999) was used to design PCR primers for the selected microsatellites using standard default values. Suitable primer design was possible in 2564 microsatellite candidates. Of them, a total of 430 microsatellites were further analyzed with size ranges from 80 bp to 250 bp; di-, tri-, and tetranucleotide motifs were identified, with seven to 22 repetitions per motif. These 430 sequences were deposited in GenBank (accession numbers MH052690-MH053105 and MF136749-MF136763; see Table 1)  min was included. PCR products were separated in 6% polyacrylamide gels and visualized by silver staining as described by Narváez et al. (2001). Genetic diversity parameters including effective number of alleles (A e ) and observed and expected levels of heterozygosity (H o and H e , respectively) were estimated using GenAlEx 6.5 (Peakall and Smouse, 2012). For each SSR marker, A e and H e were estimated based on the frequency of each allele per population, and H o corresponded to the frequency of heterozygous individuals calculated for each population.
Of the 40 primer sets analyzed, 15 markers (10 polymorphic and five monomorphic) that showed clear amplification were analyzed in more detail (Table 1). Twenty-five primers showed no amplification, or a complex pattern, and were not further analyzed. Sequences of the 15 markers that showed clear amplification have been deposited in GenBank (Table 1); sequences of the remaining 25 primers are available upon request. For the 10 polymorphic primers, allele number ranged from three to nine (average of 5.77 alleles per marker in the populations), A e ranged from 2.27 to 6.00, and the expected and observed levels of heterozygosity varied from 0.56 to 0.83 and from 0.53 to 1.00, respectively (Table 2). These are the first polymorphic microsatellite markers developed for B. brevifolium, although not the first genetic markers from Leguminosae applied to the characterization of this species (Gagnone et al., 2016).
The new markers also amplified in C. angulata, a species belonging to the same tribe (Caesalpinieae), and in S. cumingii (caper shrub), a species within a different tribe (Cassieae; Ulibarri, 2008) of the Caesalpinioideae subfamily (LPWG, 2017). However, results showed only partial amplification of the new markers in S. cumingii and less amplification in C. angulata (Table 3).

CONCLUSIONS
We have identified 15 new microsatellite markers for algarrobilla, 10 of which had high levels of polymorphism, representing the first markers developed in B. brevifolium. Some of these markers were also useful in the related species C. angulata and S. cumingii, which share the same habitat as B. brevifolium. Analyses of B. brevifolium individuals from different locations in the Atacama and Coquimbo regions in Chile would help characterize algarrobilla populations for conservation purposes and be the basis for future genetic studies for this species and possibly other endangered and related legumes endemic to northern Chile.   The estimated allele sizes for each primer/species combination are indicated for sample 1 of each species.