Characterization of microsatellite markers for the endangered Daphne rodriguezii (Thymelaeaceae) and related species

Premise The endangered shrub Daphne rodriguezii (Thymelaeaceae) is endemic to the Balearic island of Menorca, where fragmentation and severe population decline are ongoing threats to this taxon. We developed a set of microsatellite markers to analyze the fine‐scale genetics of its few extant populations. Methods and Results Fifteen microsatellite markers were obtained through Illumina high‐throughput sequencing and tested in two populations. Twelve of these loci showed no evidence of null alleles and were highly polymorphic, with a mean number of 8.3 alleles per locus. Levels of observed and expected heterozygosity ranged from 0.100 to 0.952 and from 0.095 to 0.854, respectively. Seven to nine of these loci were successfully amplified in five other Daphne species. Conclusions This set of markers provides a useful tool for investigating the factors driving fine‐scale population structure in this threatened species, and it represents a novel genetic resource for other European Daphne species.

polymorphism data revealed a strong genetic structure among ex tant populations, but the unavailability of codominant markers for D. rodriguezii limited genetic inferences at the population level (Calviño-Cancela et al., 2012). Generation of fine-scale information with microsatellite markers for this species will provide insights on general patterns of heterozygosity across island systems (García-Verdugo et al., 2015) and will help us understand how dispersal limitation impacts the genetic structure of small island populations (Saro et al., 2019).

METHODS AND RESULTS
Microsatellite development was conducted by Ecogenics GmbH (Balgach, Switzerland) from genomic DNA extracted from silicadried D. rodriguezii leaf tissue. An Illumina TruSeq nano DNA library (Illumina, San Diego, California, USA) was prepared fol lowing the manufacturer's recommendations and analyzed on an Illumina MiSeq sequencing platform using the Nano 2 × 250 v2 format. A total of 1,507,789 raw reads were processed. The pairedend reads that passed the Illumina's chastity filter were subject to

Characterization of microsatellite markers for the endangered Daphne rodriguezii (Thymelaeaceae) and related species
Carlos García-Verdugo 1,2,5 , Juan Carlos Illera 3 , and Anna Traveset 4 de-multiplexing and trimming of Illumina adapter residuals, and subsequently checked with the FastQC v0.11.7 software (Andrews, 2010). Following quality check, paired-end reads were analyzed with the software USEARCH v10.0.240 (Edgar, 2010), which resulted in 99,945 merged reads that were further screened with the software Tandem Repeats Finder v4.09 (Benson, 1999). Microsatellite se quences with a tetra-or trinucleotide of a minimum of six repeat units (or a dinucleotide of at least 10 repeat units) were detected in 4584 merged reads. Primers were designed for 2950 microsat ellite regions using Primer3 (Untergasser et al., 2012). From these candidate loci, and with the aim of optimizing marker characteriza tion, we performed an initial step of simple sequence repeat (SSR) polymorphism testing by amplifying 36 loci in a subset of seven D. rodriguezii individuals. These loci were screened based on the motif (i.e., trinucleotide SSRs were preferred over dinucleotide SSRs), the size of the amplified fragment (for optimization of the subsequent design of multiplexed reactions), and their successful amplifica tion in all of the samples. Considering the allele size ranges and the apparent lack of null alleles across the seven individuals tested, we finally used 15 out of the 36 loci to evaluate their usefulness in revealing polymorphism with larger (i.e., population level) sample sizes.
In examining the levels of variability revealed by each SSR locus, we were constrained by the conservation status (EN) of the study species. However, we were able to obtain permissions to sample leaf material from two populations representing size extremes (Calviño-Cancela et al., 2012): (1) the only population with more than 300 mature individuals (population A) and (2) a population with <50 individuals (population B) (Appendix 1).  To test cross-species amplification of D. rodriguezii primers, all 15 SSR loci were amplified in closely related Daphne species (Alonso and Herrera, 2011), including leaf material freshly col lected from one population of D. laureola L. and two to three rep licates from herbarium samples for D. cneorum L., D. gnidium L., D. mezereum L., and D. oleoides Schreb. (Appendix 1). Rather than testing multiple samples in a closely related species, our aim was to test the cross-amplification of loci using a broader taxonomic coverage. PCR followed the same conditions previously described for D. rodriguezii, but annealing temperatures were chosen based on a temperature gradient protocol ranging from 50-65°C for each primer and Daphne species (see Appendix 2). PCR products were visualized on a 1.5% agarose gel stained with GelRed (Biotium Inc., Fremont, California, USA) and sequenced on an ABI PRISM 3130xl Genetic Analyzer sequencer using a GeneScan 500 LIZ Size Standard (Applied Biosystems, Waltham, Massachusetts, USA). GeneMarker 2.4.0 software (SoftGenetics, State College, Pennsylvania, USA) was used for visualizing the electrophero grams and performing allele scoring.
The software GENETIX (Belkhir et al., 2001) was used to ob tain the number of alleles per locus and estimates of observed and expected heterozygosities. Tests for linkage disequilibrium and potential deviations from Hardy-Weinberg equilibrium following a sequential Bonferroni correction for multiple tests were run on GENEPOP 4.7.0 (Rousset, 2008). MICRO-CHECKER 2.2.3 (van Oosterhout et al., 2004) was used to assess the presence of null al leles at each locus and population.
At the population level, the number of alleles per locus ranged from one to 11 (Table 2). The level of observed heterozygosity ranged from 0.000 to 0.952, and the level of expected heterozygosity ranged from 0.000 to 0.854 (Table 2). Three loci (Dro025, Dro035, Dro073) were fixed, or nearly so, for a single allele per population. The remaining 12 loci showed substantial levels of polymorphism, with a mean of 8.3 alleles per locus. Only one locus (Dro078) showed significant deviation from Hardy-Weinberg equilibrium after sequential Bonferroni correction in population B, most prob ably because this was the only combination of locus and population for which null alleles were clearly identified by MICRO-CHECKER. Significant (P < 0.001) linkage disequilibrium was found between loci Dro046 and Dro124, but only for population B.
In addition, this panel of microsatellites rendered positive am plifications in a minimum of seven loci per species (Table 3). The limited availability of herbarium samples per species precluded a clear assessment of the levels of polymorphism detected with these markers, but for some species (D. laureola, D. cneorum, D. oleoides), even relatively low sample sizes revealed that at least half of the amplified loci exhibited more than one allele (Table 3).

CONCLUSIONS
The set of microsatellites characterized for D. rodriguezii is a pow erful, cost-effective tool for detecting substantial levels of genetic variation using a relatively low number of multiplexed reactions, even in small populations. Such a genetic resolution will allow us to assess parentage relationships in forthcoming studies on finescale genetic structure. Additionally, the successful rates of crossamplification of these loci suggest that population genetic studies with these markers could be easily extended to other closely related Daphne species.

AUTHOR CONTRIBUTIONS
C.G.-V. and A.T. planned the study and collected plant tissue, J.C.I. and C.G.-V. conducted laboratory work and allele scoring, and C.G.-V. performed the analyses and wrote the manuscript, with in put from J.C.I. and A.T.

DATA ACCESSIBILITY
The primers and microsatellite sequences developed in this study have been deposited in GenBank (accession numbers MK507747-MK507761; Table 1). Raw sequence library data were deposited in the Short Read Archive of the National Center for Biotechnology Information (NCBI) (BioProject accession number: PRJNA523502).