Development of 15 microsatellite loci in the endangered Streptanthus glandulosus subsp. niger (Brassicaceae)

Premise of the Study The endangered Streptanthus glandulosus subsp. niger (Brassicaceae) is endemic to a single peninsula in California and threatened by fragmentation. We developed microsatellite markers to investigate genetic diversity in the two extant populations and the degree to which they have diverged from one another. Methods and Results We used Illumina HiSeq high‐throughput sequencing to develop 15 microsatellite markers, 14 of which were polymorphic. These di‐ and trinucleotide repeats yielded one to 11 alleles per locus in 61 plants across the two populations. Levels of observed and expected heterozygosities ranged from 0.108 to 0.946 and 0.257 to 0.839, respectively. We demonstrated cross‐amplification in a second rare subspecies, S. glandulosus subsp. secundus, and in the widespread congener S. tortuosus. Conclusions These are the first microsatellites reported for this subspecies, and they will aid in the inclusion of genetic information in conservation planning. Cross‐amplification was demonstrated in two related taxa, including one of conservation concern.

Hilary's Preserve population. Due to restrictions on collecting, our specimens were vouchered with existing herbarium collections (Appendix 1). DNA was sent to the QB3 Genomics Sequencing Laboratory at the University of California, Berkeley, for shotgun library preparation and 2 × 300-bp paired-end Illumina HiSeq genome sequencing. Approximately nine million FASTQ reads per sample were filtered for minimum length (50 bp) and trimmed for quality using Trimmomatic version 0.23.3 (Bolger et al., 2014) using a window size of 4 bp with a threshold average quality of Phred = 25, and a minimum score of 15 at the leading and trailing ends of the read. Reads were assembled de novo with ABySS (Simpson et al., 2009) using a K-mer size of 41 bp, a minimum bubble identity of 0.9, and a minimum contig size of 300 bp. To avoid acquisition bias, we selected loci that were identified as microsatellites in all three genomic DNA assemblies using misa.pl (Beier et al., 2017) in a search for dimeric and trimeric repeats with a minimum of six repeat units. Loci were pre-screened for polymorphism by "round-robin" mapping trimmed reads from each genomic DNA to the other two genome assemblies using Bowtie2 (Langmead and Salzberg, 2012) using pre-set default parameters, followed by variant calling using NGSEP version 3.0.3 (Perea et al., 2016) with a ploidy of two. Evidence from transcriptome sequencing indicates that a common ancestor to S. glandulosus subsp. niger and other members of the tribe Thelypodieae underwent a major (possibly whole genome) duplication event ~8-10 mya (Kagale et al., 2014;Hawkins et al., 2017). A diploid chromosome number of 2n = 28 is almost universally conserved across the tribe (Warwick and Al-Shehbaz, 2006). Rare reports of unusual karyotypes (e.g., 2n = 14, 2n = 56) show no phylogenetic pattern (Cacho et al., 2014) and thus appear to be sporadic in nature. In Caulanthus amplexicaulis S. Watson, a 2n = 28 sister taxon to S. glandulosus subsp. niger, 250 of 258 microsatellite markers (97%) produced a single amplification product from each of two distinct homozygous inbred lines. Considered together, these lines of evidence justify the selection of diploid as a ploidy parameter. Genomic coordinates identified as microsatellites by misa.pl were thus filtered for the presence of dimeric and trimeric repeats in all three assemblies that were in all cases called by NGSEP as "STR" and "INDEL" with a minimum confidence score of 100. Forty-one microsatellite loci met these criteria.
Primers for the 41 loci were designed to have 35-55% GC content, a target melting temperature of 64-65°C (salt-adjusted, 50 mM NaCl), and an amplicon size of 90-150 bp. A subset of 23 loci were screened for amplification by electrophoresing the products on a 4% SFR agarose gel (VWR Life Science, Philadelphia, Pennsylvania, USA) with an Invitrogen 100-bp ladder (Invitrogen, Carlsbad, California, USA) using nine S. glandulosus subsp. niger individuals. Of those 23 loci, 15 produced consistent and robust amplification products and thus were suitable for genotyping. Sequence library data were deposited into the National Center for Biotechnology Information Sequence Read Archive (BioProject ID PRJNA503999).
PCR conditions were optimized using a Bio-Rad T100 Thermal Cycler (Bio-Rad Laboratories). Amplification reactions were   (Table 1) and peaks were determined using Peak Scanner 2.0 (Thermo-Fisher Scientific). We used GenAlEx (Peakall and Smouse, 2012) to measure the number of alleles per locus and observed and expected heterozygosities. We used GENEPOP 4.7.0 (Rousset, 2008) to test for linkage disequilibrium and deviations from Hardy-Weinberg equilibrium using a sequential Bonferroni correction for multiple tests, and MICRO-CHECKER 2.2.3 (van Oosterhout et al., 2004) to assess the presence of null alleles.
The average number of alleles across all loci was 5.70 (±0.46 SE; range 1-11) ( Table 2). Three loci (Sn313, Sn715, and Sn803) may harbor null alleles. We detected significant linkage disequilibrium between loci Sn558 and Sn463. Of the 15 loci, 12 deviated from Hardy-Weinberg equilibrium at the Old St. Hilary's Preserve site and five deviated at the Middle Ridge Park site (Table 2).
We tested all 15 loci for amplification in another rare subspecies (S. glandulosus subsp. secundus Greene) and a congener (S. tortuosus Kellogg) using the same protocol. Streptanthus glandulosus subsp. secundus is restricted to approximately three dozen populations in two neighboring counties, whereas S. tortuosus spans nearly the entire 1200-km length of California. Eleven loci successfully amplified in both taxa (Table 2), producing one or two resolvable amplification products per individual, suggesting they amplified a single locus. Three loci produced no amplification products in either species, and one locus amplified more than one product in S. glandulosus subsp. secundus (Table 2).

CONCLUSIONS
We developed microsatellite markers for the endangered S. glandulosus subsp. niger and two congeners. Optimization of these markers will be useful in quantifying the genetic diversity in the two remaining populations of S. glandulosus subsp. niger and the degree to which these populations have diverged from one another. These markers may also be useful in comparative studies with S. glandulosus subsp. secundus, which is also of conservation concern.

ACKNOWLEDGMENTS
The authors thank Dr. Lisa Wallace and Dr. Adam Schneider for thoughtful comments that improved this manuscript.

AUTHOR CONTRIBUTIONS
S.M.S. conceived of the project, collected leaf tissue, supervised the lab and field work, conducted allele scoring and analyses, and wrote Note: -= no amplification; + = amplification of more than two amplification products; A = number of alleles; H e = expected heterozygosity; H o = observed heterozygosity; n = number of individuals sampled; N = number of successfully genotyped individuals. a Locality and voucher information are provided in Appendix 1. b Significant deviation from Hardy-Weinberg equilibrium (*P < 0.05, **P < 0.01, NA = not applicable). c Values represent allele size range in base pairs.