Silene L. (Caryophyllaceae) has been recognized as an important model system for plant ecology and evolution, and has been a study system for different sexual and mating systems (Bernasconi et al., 2009). Silene acaulis (L.) Jacq., an arctic/ alpine species, exhibits features that make it ideal as a model species (e.g., diploid, two genders, usually single individuals, almost linear size increase with age, widespread distribution). The development of molecular markers for this species is therefore of particular interest.
Silene acaulis is a long-lived perennial cushion plant, widely distributed in arctic/alpine tundra ecosystems in Europe, northern Asia, and North America (Elven et al., 2012). It commonly forms dense circular cushions with a central taproot (Jones and Richards, 1962), a feature that facilitates identification of single individuals in dense populations. Individuals of S. acaulis are sexually polymorphic because the species is gynodioecious; however, flowers are variable and populations are functionally dioecious or trioecious (Shykoff, 1988; Delph and Carroll, 2001). Silene acaulis is diploid (2n = 24) throughout its distribution range (Elven et al., 2012) and reproduces solely through sexual reproduction. Data from crossing experiments indicate that sex is inherited by a nuclear-cytoplasmic interaction (Delph et al., 1999). The gender proportions in natural populations are variable (Klaas and Olson, 2006), and in harsher conditions the sex ratio is typically biased toward females (Hermanutz and Innes, 1994; Alatalo and Molau, 1995). The well-defined, nearly circular growth form has been used for demographic studies (Jones and Richards, 1962; Benedict, 1989; Morris and Doak, 1998). To combine demographic information with fine-scale genetic investigations, high-resolution microsatellite markers have been developed. We present 40 primer pairs for simple sequence repeat (SSR) regions in S. acaulis, 14 of which resulted in polymorphic products and were included in a multiplex PCR system. This setup has been tested on 304 individuals collected in the high arctic archipelago of Svalbard, Norway.
METHODS AND RESULTS
Fresh, frozen, and silica gel–dried leaf material collected in Svalbard, Norway, was used for DNA extraction with the QIAGEN Plant Mini Kit according to the manufacturer's protocol (QIAGEN GmbH, Hilden, Nordrhein-Westfalen, Germany). The only minor deviation to the standard protocol was that elution buffer volume was adjusted to 25 µL or 100 µL depending on the amount of leaf material used. The DNA amount was measured with a spectrophotometer (NanoDrop, ND-2000; Thermo Scientific, Wilmington, Delaware, USA) and adjusted to 5 ng/µL by adding deionized Milli-Q water (Merck Millipore, Darmstadt, Hessen, Germany).
The microsatellite regions were identified using next-generation high-throughput genome sequencing (Abdelkrim et al., 2009). All sequences were isolated by ecogenics GmbH (Zürich-Schlieren, Switzerland). Extracted DNA and plant material from one plant collected in Endalen (Svalbard, Norway; 78°11′12″N, 15°45′39″E; voucher from the breeding population deposited at Tromsø Museum [TROM], University of Tromsø, Tromsø, Norway [voucher no. TROM-V-135413]) were sent to ecogenics GmbH, and 13 µg of genomic DNA was analyzed on a Roche 454 GS-FLX platform (Roche, Basel, Basel-Stadt, Switzerland) using a 1/16th run and the GS-FLX titanium reagents. The total 38,453 reads had an average length of 357 bp, and 106 of these reads contained a suitable microsatellite insert that was a tetra- or trinucleotide of at least five repeat units or a dinucleotide of at least nine repeat units. Primers for 50 microsatellite inserts were designed, and all of them were tested for amplification.
Out of 50 primer pairs, 10 did not result in a satisfactory amplification using another plant from Endalen (F0). The remaining 40 primers were tested for polymorphism using two additional individuals, one from Polheim (79°53′33″N, 16°01′24″E) and one from Kvartsittsletta (77°03′33″N, 15°07′38″E). Further tests were made with 27 primer pairs using two additional individuals from Polheim, one individual collected on Edgeøya (78°04′54″N, 20°48′38″E), and three individuals germinated from seeds (F1 collected from the F0 individual; the individual from Kvartsittsletta was dropped because of insufficient DNA quantity. For cost-effective testing of the selected primer pairs, the M13 tail approach was chosen with a 6-FAM (Integrated DNA Technologies, Coralville, Iowa, USA) fluorescent color tail to visualize and estimate the length of the amplification product on a capillary sequencer (ABI-PRISM-3100; Applied Biosystems, Foster City, California, USA) (Schuelke, 2000). All PCR reactions for the amplification of the microsatellite primers had a 10-µL volume. Single substances in the PCR mix were 1.0 µL PCR buffer (QIAGEN), 1.0 µL dNTPs (2.0 mM each, QIAGEN), 0.1 µL HotStarTaq DNA polymerase (QIAGEN), 0.2 µL 5 µM forward primer with M13 tail (biomers.net, Ulm, Baden-Württemberg, Germany), 0.6 µL 5 µM reverse primer, 0.6 µL 5 µM 6-FAM tail (biomers.net), 4.5 µL H2O, and 2.0 µL 5 ng/µL template DNA. The PCR conditions were the same as in Vik et al. (2012), except that the annealing temperature was set to 55°C. For fragment length determination of the amplification products, the PCR product was diluted 10× before each was mixed with 0.2 µL GeneScan 500 LIZ and 8.8 µL HiDi (both Applied Biosystems). Fragment length was determined on an ABI PRISM 3100 (Applied Biosystems) at the University of Tromsø. The fragments were visually inspected using PeakScanner 1.0 (Applied Biosystems). Fourteen of the 40 markers (Tables 1 and 2) showed promising profiles and polymorphism within the eight tested plant individuals in the collection region of the Svalbard archipelago (Norway). The three included F1 generation seedlings (EN S F1) inherited one allele from their mother (EN M F0) except for primer pair 3 (no profile for EN M F0 available). Of 14 amplified regions, seven were polymorphic between siblings and 11 were polymorphic between mother and offspring (Table 3).
Characterization of 40 microsatellite loci isolated from Silene acaulis, of which 14 were included in a multiplex PCR approach.a,b
The 14 polymorphic markers (Table 1) were selected for further testing using a multiplex PCR set-up with 304 S. acaulis individuals collected in Endalen and Hotellneset (Svalbard, Norway). The M13-tailed forward primers were exchanged with primers that were labeled with the four different fluorescent colors 6-FAM, VIC, NED, and PET (Applied Biosystems). This allowed multiplexing of up to four primer pairs with a different color-labeled forward primer (Table 1). For calculating mean error rates per allele (Table 2; Pompanon et al., 2005), 20 to 24 individuals were replicated per multiplex PCR; in addition, negative controls were run through the entire procedure to monitor contamination. The fragments were scored using Geneious 7.1.3 (Biomatters Ltd., Auckland, New Zealand).
The multiplexing in PCR mix 1 and mix 4 was satisfactory. However, despite reliable amplifications in nonmultiplexed PCRs, we initially had a high number of allelic drop-outs or samples that did not amplify, especially in PCR mix 2 and mix 3 (see high error rates in Table 2). Nevertheless, reducing the multiplexing to only two or three primers (resulting in the following primer mixes: Silaca 3 and 8; Silaca 7 and 8; Silaca 23, 44, and 34; and Silaca 29, 32, and 18) improved the results. Thus, we are confident that the reason for this was not fragmented or low-quality DNA material, but competing primers in the PCR multiplex mix.
Genetic properties of 40 microsatellite loci isolated from Silene acaulis.
Allele distribution of a female Silene acaulis individual (EN M F0) and three of its seedlings (EN S F1).
The arctic/alpine species S. acaulis has strong potential as a model species for population genetic studies, as genotyping can easily be combined with demographic parameters. To date, the microsatellites have been used for a small-scale genetic study in different size and gender cohorts on the arctic archipelago of Svalbard. The observed heterozygosity was found to be lower (0.04–0.58) than the expected heterozygosity, and no marked differences were found between genders (Svoen et al., in prep.).