Pulsatilla vulgaris Mill. (Ranunculaceae) is an early-flowering perennial herb of conservation concern and a flagship and specialist species of calcareous grasslands across central Europe, ranging from France in the south to Sweden at its northern limit (Wells and Barling, 1971; Pfeifer et al., 2002). Over the last century, P. vulgaris has witnessed rapid decline and local extirpation across its range, and is listed as “near threatened” by the International Union for Conservation of Nature (IUCN, 2014). Its decline is linked to the abandonment of traditional grazing practices, which has resulted in the severe loss and fragmentation of calcareous grasslands following afforestation (Butaye et al., 2005), and increased above-ground competition from coarse grasses (Walker and Pinches, 2011). Knowledge of the landscape-scale distribution of genetic variation is required to create effective management plans for fragmented populations, and evaluations of genetic diversity and inbreeding will allow the identification of populations that are at highest risk of extirpation. This, however, requires genetic markers with high resolution such as microsatellites. No such markers are yet available for P. vulgaris, and any potential genetic analyses are complicated as this species is suspected to be an allotetraploid (2n = 32; Böcher, 1934). We therefore de novo developed microsatellite markers for this calcareous plant species, specifically selecting loci that can be analyzed as diploid or double-diploids in downstream genetic analyses.
METHODS AND RESULTS
We extracted genomic DNA from homogenized leaf tissue from two populations in the Franconian Alb, Germany (A25: 48°57′38.7″N, 10°56′32.2″E; A03: 48°59′51.7″N, 11°3′35.4″E), using the QIAGEN DNeasy Plant Mini Kit (QIAGEN, Mississauga, Ontario, Canada) following the manufacturer's protocol. A voucher from the study area is deposited at the Herbarium at the University of Toronto Mississauga (TRTE; voucher no.: BR-07052015; locality: Laubenthal, Landkreis Weissenburg, Germany; 48°58′48.7130″N, 11°03′14.2254″E). DNA from five individuals was mixed, and a 4-µL sample was sent to LGC Genomics (Berlin, Germany) for 454 shotgun sequencing on 1/8 of a plate using GS FLX Titanium (Roche, 454 Life Sciences, Branford, Connecticut, USA). In total, 92,833 reads were returned and assembled into 4088 contigs. We used MSATCOMMANDER (version 1.0.8; Faircloth, 2008) to identify di-, tri-, and tetranucleotide microsatellites to develop primers in regions flanking the identified microsatellites. A total of 457 microsatellites were identified, and 76 suitable primer pairs could be designed. We amplified 75 of these pairs by PCR in an initial screen of 10 individuals, and of these, 18 showed clear peaks and 14 of these were polymorphic (Table 1). Eight markers showed evidence of disomic inheritance (e.g., had a maximum of two alleles per individual) and were retained for further analysis (Table 2).
We amplified these eight microsatellites in two multiplex reactions in 97 individuals with the QIAGEN Multiplex PCR Kit (QIAGEN) in 10-µL final reaction volumes with 0.2 µM of each primer, 4.6 µL of Multiplex Mix, 1.2 µL of DNase-free water, and 5–30 ng of template DNA. Thermocycling conditions for Multiplex A (Table 1) strictly followed the manufacturer's protocol. Multiplex B was amplified using a touchdown procedure with an initial denaturation at 95°C for 15 min; 10 cycles of denaturation at 95°C for 60 s, annealing at 65–55°C for 60 s, and extension at 72°C for 90 s; 25 cycles of 95°C for 60 s, 55°C for 60 s, and 72°C for 90 s; and a final extension at 72°C for 10 min. PCR products were diluted 20× and 1–2 µL of product was loaded to an ABI 3730x1 Capillary Sequencer (Applied Biosystems, Burlington, Ontario, Canada) using GeneScan 500 LIZ (Life Technologies, Burlington, Ontario, Canada) size standard for fragment analysis at the Centre for Applied Genomics (The Hospital for Sick Children, Toronto, Canada). Genotyping was performed in Gene-Marker 2.4.0 (SoftGenetics, State College, Pennsylvania, USA).
Characteristics of 18 de novo–developed microsatellite markers for Pulsatilla vulgaris.
Results from HWE exact tests for diploid markers developed for Pulsatilla vulgaris in two populations.a
All diploid markers except for PV65 amplified a maximum of two alleles per individual. PV65 amplified a maximum of four alleles per individual in two sets of size-separated fragments (PV65a and PV65b, Table 1). Loci showed high levels of polymorphism with seven to 17 alleles per locus, and observed heterozygosity from 0.41 to 0.83 per locus (Table 2). Exact tests performed in GENEPOP 4.2 (Rousset, 2008) revealed three loci with significant departures from Hardy–Weinberg equilibrium (HWE), but departures were not consistent across populations, except for locus PV64. It is possible that selfing in one or both populations, or the presence of population substructure (i.e., the Wahlund effect; Wahlund, 1928), led to this homozygote excess. Test for departures from linkage equilibrium expectations in GENEPOP showed that all pairs of loci were unlinked (results not shown).
We used open-pollinated progeny arrays to confirm that the identified “diploid” loci followed a disomic pattern of inheritance by amplifying the eight loci in 10 seeds collected from each of 15 mother plants. For PV65, all offspring genotypes had maternal contributions from both of the size-separated amplicons, suggesting that PV65a and PV65b represent independent genomic complements with disomic inheritance. For the remaining loci, disomic inheritance was confirmed by the observation of a maximum of two alleles for all offspring, with each offspring sharing an allele with the maternal genotype.
These new microsatellite markers are highly informative and will be used to quantify gene flow across fragmented populations of P. vulgaris in southern Germany. We will test whether the reintroduction of shepherding is a suitable conservation measure to improve genetic connectivity among populations of this species. The combination of diploid and tetraploid markers presented here will be useful in clarifying the polyploidization history of this and related species.
 The authors thank Nimesh Patel, Amaneet Lochab, and Qasim Muhammad for assistance with DNA extraction and genotyping. We also thank Juergen Boehmer and Mr. Dadrich for help in the field. Funding was provided by the Natural Sciences and Engineering Research Council of Canada (NSERC CGS-D and Michael Smith Foreign Study Supplement to M.F.D., and Discovery Grant to H.H.W.) and the Government of Central Franconia, Bavaria, Germany.