South Africa's plant diversity is remarkable; however, the actual number of species that comprise this diversity is unknown. Polyploidy plays a prominent role in plant diversity in that ∼15% of speciation events occur via polyploidy and ∼35% of plant genera contain polyploids (Wood et al., 2009). Recent work highlights the gap in polyploid research in South Africa and the need to assess the presence of polyploid species (Krejčíkova et al., 2013). Helichrysum odoratissimum (L.) Sweet is a native South African perennial herb that belongs to tribe Gnaphalieae in the Asteraceae (daisy family). The species is also part of Hilliard's taxonomic Group 4 in the genus, which includes three additional species, H. gymnocomum DC., H. infaustum J. M. Wood & M. S. Evans, and H. griseolanatum Hilliard (Hilliard, 1983). Helichrysum odoratissimum contains three different ploidy levels: diploid, tetraploid, and hexaploid (2n = 14, 28, 42, respectively; Galbany-Casals and Romo, 2008). Given that there are three ploidy levels in H. odoratissimum, polyploidy may contribute to the morphological and ecological diversity in this species, and subsequently to South Africa's remarkable floral diversity. We developed microsatellites for future work that will examine gene flow patterns among H. odoratissimum populations of known ploidies.
METHODS AND RESULTS
DNA was extracted from 125 H. odoratissimum diploid individuals that were collected from across South Africa (Appendix 1) using QIAGEN's DNeasy Extraction kits following the manufacturer's protocol (QIAGEN, Valencia, California, USA). Extracted DNA was visualized on a 1% agarose gel. Four DNA samples from four representative populations (Appendix 1) were sent to Inqaba Biotec (Pretoria, South Africa) for initial microsatellite enrichment. At Inqaba, a Fast Isolation by AFLP of Sequences COntaining repeats (FIASCO) approach was used to identify microsatellite regions (Zane et al., 2002). Briefly, DNA is digested using MseI, resulting amplicons are enriched for repeat regions using biotinylated beads, and identified enriched regions are spliced into vectors to separate out and amplify repeat-enriched regions. Inqaba then sequenced 10 vectors that ranged approximately 250–600 bp in size, to ensure that the colonies contained repeat regions and no regions were over-represented. Once enrichment was confirmed, the microsatellite library was sequenced using next-generation sequencing (NGS) techniques. Once compiled, NGS data were received from Inqaba and MSATCOMMANDER version 2.8 (Faircloth, 2008) was used to identify unique microsatellite repeat regions. From these data, we identified 30 potential regions (10 di-, 10 tri-, and 10 tetranucleotide repeats) and developed primer pairs for these regions using Primer3 (Koressaar and Remm, 2007; Untergasser et al., 2012; Table 1).
We tested these primer pairs for amplification success and possible polymorphism across 10 geographically distinct H. odoratissimum populations (10 individuals total; Appendix 1). Primers were diluted to 20 µM and tested using the following PCR protocols. Amplifications were performed at a 10-µL final volume of DreamTaq 2× Master Mix (ThermoFisher Scientific, Waltham, Massachusetts, USA) with 0.2 µM of labeled forward and reverse primers, and 0.5 µL of 1 µg/mL bovine serum albumin (BSA). Two primers (Ho12, Ho13) were amplified using the above recipe with the addition of 0.1 µM 5′ M13 forward primer labeled with 6-FAM dye, and touchdown PCR conditions were followed according to Schuelke (2000). All reactions included 1 µL DNA of varying concentrations dependent on the sample (5–20 ng estimated from agarose gel visualization). For the other eight primers, we followed the DreamTaq PCR program outline: 95°C for 3 min; 95°C for 40 s; a specified annealing temperature (Table 1) for 40 s; 72°C for 30 s; 72°C for 4 min. Gel electrophoresis was used to assess polymorphism of the primer pairs using a 4% MetaPhor agarose gel (Lonza, Sigma-Aldrich Pty. Ltd., Johannesburg, South Africa) with a 50-bp ladder (ThermoFisher Scientific). Genetic analyses only included microsatellites that were deemed to be polymorphic across the 10 populations from agarose gels. Of the 30 pairs, we identified 10 primer sets that amplified successfully and repeatedly. Of those 10 primer sets, seven were polymorphic for our target species. We then cloned putative polymorphic amplicons from 2–5 individuals using a QIAGEN PCR Cloningplus Kit (QIAGEN, Whitehead Scientific (Pty.) Ltd., Johannesburg, South Africa) per the manufacturer's instructions. Amplified clone inserts from three individuals per primer pair (1–5 inserts per individual depending on cloning success) were sent to the DNA Sequencing Unit at the Central Analytical Facility (CAF), Stellenbosch University (Stellenbosch, South Africa), to verify that the amplicons contained the correct repeat region. Once verified, the remaining 114 individuals were amplified. Each forward primer was given a different dye label (VIC, NED, FAM) to allow for pooling PCR products for fragment analysis at CAF. Fragment analysis was carried out on an ABI 3130x1 (Applied Biosystems, Foster City, California, USA) using a GeneScan 500 LIZ Size Standard (Applied Biosystems). Chromatographs were scored manually using PeakScanner version 1.0 (Applied Biosystems). Data were then formatted for population genetic analyses.
Characteristics of 10 microsatellite loci for Helichrysum odoratissimum.a
Genetic diversity indices across populations for the seven polymorphic microsatellite loci.
We used 125 individuals from 10 populations from across South Africa to establish the usefulness of the seven polymorphic primers in population genetic analyses. These individuals are known diploids from flow cytometry data (Glennon et al., in prep.) and published chromosome counts (Galbany-Casals and Romo, 2008). Seven of the 10 loci were polymorphic within diploid H. odoratissimum. The number of alleles (A) and observed (Ho) and expected heterozygosity (He) were estimated for each locus using GenoDive (Meirmans and Van Tienderen, 2004; Table 2). Furthermore, deviation from Hardy–Weinberg equilibrium (HWE) was tested for each population in GenoDive using a fixation index (FIS) estimate that is equivalent to Weir and Cockerham's (1984) F-statistics. We conducted a Bonferroni correction for HWE significance levels due to multiple comparisons of populations across loci. Linkage disequilibrium was tested using GenePop (Raymond and Rousset, 1995; Rousset, 2008).
The mean A per locus was 15.42 (Table 2). Ho values ranged from 0.172 to 0.848, and He values ranged from 0.367 to 0.870 across loci. Only one population deviated from HWE (Elliot, Table 2), and FIS values ranged from −0.463 to 0.294. This deviation may be due to potential substructuring of the population, as it spanned a large geographic area. Genetic diversity estimates across the 10 populations showed that the average A per population was 4.9 and that Ho values ranged from 0.411 to 0.795 (Table 2). There was linkage disequilibrium between three pairs of the primers (Ho1/Ho2, Ho2/Ho7, Ho1/Ho12), but only in three of the 10 populations (Buffleskloof, Elliot, and Piketberg). Lastly, we tested a total of 10 additional Helichrysum Mill. species found throughout South Africa to test for cross-species amplification (Appendix 1). PCR amplicons were run on a 4% MetaPhor agarose gel to determine if the loci were polymorphic among species. We found that the primers amplified across all of the taxa, but bands were relatively faint for H. dasycephalum O. Hoffm., H. glomeratum Klatt, and H. psilolepis Harv. The amplicons differed in size across all species ranging from approximately 150 to 350 bp. We also successfully amplified DNA from additional polyploid H. odoratissimum individuals (data not shown).
These seven loci will be useful in determining patterns of gene flow as well as other population genetic parameters for future studies of population divergence within H. odoratissimum. The polymorphic microsatellites amplified will be useful in studies regarding speciation patterns in the taxonomic Group 4 of Helichrysum, and together with the monomorphic markers may also be potentially useful for future taxonomic revisions of closely related Helichrysum species in South Africa.
Specimens were collected under permits from the Department of Economic Development, Environmental Affairs and Tourism (DEDEAT) of the Eastern Cape Province (KZN OP 5161/2012, CRO 29/13CR) and Cape Nature (AAA008-00081-0028). This work was funded by the National Research Foundation Integrated Biodiversity Information Program (grant no. 86964 to G.V.C.). K.L.G. was supported by grant B8749.R01 from the Carnegie Corporation of New York to the Global Change and Sustainability Research Institute. The authors thank J. Choiniere, B. Cohen, S. Cousins, and M. Goodman for field assistance.