Sagina nodosa (L.) Fenzl, commonly known as knotted pearlwort, is a small, diploid, perennial herb in the Caryophyllaceae family found predominantly in archipelagos and along exposed coastlines of the Northern hemisphere. The arctic-alpine plant is especially delicate, and on Isle Royale it usually inhabits cracks in rocks by the shore where it finds protection against cold waters, storms, and competition from other plants. Although stable populations exist in northern Canada and the arctic, the plant is classified as threatened in Michigan, USA (Slavick and Janke, 1993), and may be at risk or sensitive in the Canadian provinces of Alberta, New Brunswick, Saskatchewan, and Prince Edward Island (CFIA and NRCan/CFS, 2013).
Knotted pearlwort has two known modes of reproduction: it can cross- and self-pollinate to produce seeds for dispersal or it can propagate vegetatively (Crow, 1978; Magee and Ahles, 1999). However, to our knowledge, no genetic studies have been performed on the plants of this genus, and thus, our understanding of Sagina plants' life history is primarily observational. In addition, little is known about the population structure, genetic diversity, and the relative isolation of the North American subpopulations, particularly in the context of climate change and the potential disappearance of its habitats in the southern range its distribution.
We developed microsatellite markers to investigate the population structure and life history in isolated populations of S. nodosa at Isle Royale National Park in Michigan, USA. The Isle Royale population is particularly interesting as it provides a disjunct sample of individuals from the endangered southern population (Soper and Maycock, 1963). Furthermore, the numerous geographic barriers from the steep shoreline to the separation of islands provide natural ecological experiments that can provide insights on the life cycle and demographic history of S. nodosa (Given and Soper, 1981).
The identification of novel microsatellite loci traditionally requires the construction of a genomic library enriched for repeat motifs and the sequencing of individual clones to identify polymorphic loci (Zane et al., 2002). Recently new methods using next-generation sequencing increased the efficiency and speed of the process, while reducing costs (Santana et al., 2009). Here we performed a low-coverage 454 next-generation sequencing of enriched genomic fragments derived from one individual to generate a massive library of contigs containing potential polymorphic sequences (Abdelkrim et al., 2009). We present data for 10 novel microsatellite loci in S. nodosa, which will be used for landscape genetic analysis of the disjunct population of Isle Royale, Michigan, USA.
METHODS AND RESULTS
We collected fresh leaf material from S. nodosa individuals from eight islands located in the northeastern tip of Isle Royale National Park (Michigan, USA): the main island on Blake Point (3 individuals), Third Island (4 individuals), Long Island (5 individuals), North Government Island (6 individuals), Split Island (4 individuals), Edwards Island (3 individuals), South Government Island (3 individuals), and South Government Island Islet (1 individual). GPS coordinates for each individual can be found in Appendix 1. Small, nondestructive clippings of buds, stems, and leaves were placed into labeled air-tight plastic bags with Drierite, a strong desiccant. We then extracted genomic DNA with ArchivePure DNA Purification Kit (5 Prime, Gaithersburg, Maryland, USA) according to the manufacturer's protocol. A voucher specimen was deposited at the University of Michigan herbarium (MICH). The specimen is labeled “Sagina nodosa, Michigan, Keweenaw Co., Isle Royale National Park, Edwards Island, Edwards, 31 Jul 2012 (MICH–1474911).”
One DNA sample from Isle Royale was then subjected to shotgun sequencing (1/16th run) using a Roche 454 Genome Sequencer FLX (454 Life Sciences, a Roche Company, Branford, Connecticut, USA) at the Evolutionary Genetics Core Facility (Ithaca, New York, USA). The library construction followed a modified protocol based on Hamilton et al. (1999). Following contig assembly with SeqMan Pro (Lasergene version 8.1.1; DNASTAR, Madison, Wisconsin, USA), we obtained 20,472 reads in FASTA format with an average read length of 341.3 bp and a total of 8,219,964 bp. The contigs were aligned under highly stringent conditions, requiring a minimum sequence length of 126 bp and with no more than 15 mismatched bases.
Using MSATCOMMANDER, we screened all assembled contigs for di-, tri-, tetra-, penta-, and hexanucleotide repeats, and ensured that each candidate contig had at least 40 bp in its flanking region and a minimum repeat length of 6 bp for dinucleotides and 5 bp for all other motifs (Faircloth, 2008). A total of 2287 sequences, or 11.17% of the total reads, had tandem repeats that satisfied our stringent conditions, with 765 (33.4%) dinucleotide repeats, 1059 (46.3%) trinucleotide repeats, 334 (14.6%) tetranucleotide repeats, 91 (4.0%) pentanucleotide repeats, and 38 (1.7%) hexanucleotide repeats.
We designed primers for 41 of the 2287 candidate contigs using BatchPrimer3 (You et al., 2008) with the following restrictions: amplification products must be within a size range of 100 to 500 bp, optimal melting temperature of 60°C (must be between 57°C and 62°C), optimal GC content of 50%, possession of at least one GC clamp, and reasonably low levels of self- and pair-complementarity. We then inspected each of the individual primers to ensure they did not amplify the same target microsatellite loci. Primers located near the ends of the contigs or in close proximity of microsatellite repeats were discarded.
For most primer pairs that amplified consistently (all except M33 and M35), we used a single-reaction nested PCR method (TP-PCR) to embed fluorescent dyes into PCR fragments for later genotyping (Schuelke, 2000). The tail 5′-CGAGTTTTCCCAGTCACGAC-3′ was appended to the 5′ end of our forward primers (Schuelke, 2000), and the tail 5′-GTTTCTT-3′ to the 5′ end of our reverse primers (Brownstein et al., 1996). A third fluorescently labeled primer, 6-FAM-5′-CGAGTTTTCCCAGTCACGAC-3′, was added to every reaction to embed the fluorescent dye 6-FAM label into the reaction products (Integrated DNA Technologies, Coralville, Iowa, USA). TP-PCR was performed with the following 25-µL PCR reaction mix: 1 µL of ∼10–15 ng/µL genomic DNA, 11 µL of PCR-grade dH2O, 10 µL of 2.5× HotMasterMix (5 Prime), 2.5 µL forward and reverse primer mix with a final concentration of 0.05 µM and 0.2 µL, respectively, and 0.5 µL of universal FAM-labeled primer with a final concentration of 0.2 µL. For more sensitive reactions, where yield was lower, the PCR Master Mix was replaced with the appropriate concentration of fresh Taq polymerase. Here, the PCR volume of 25 µL contained 1 µL of ∼10–15 ng/µL genomic DNA, 15.85 µL of PCR-grade dH2O, 0.5 µL of 0.2 mM dNTP, 0.15 µL of HotMaster Taq (5 Prime), 2.5 µL of 10× PCR buffer, 2.5 µL forward and reverse primer mix with a final concentration of 0.05 µM and 0.2 µM, respectively, and 0.5 µL of universal FAM-labeled primer with a final concentration of 0.2 µM. For loci M33 and M35, we used a 6-FAM fluorescently labeled forward primer (Integrated DNA Technologies, see Table 1 for primer sequence) and the Type-it Microsatellite PCR Kit (QIAGEN, Valencia, California, USA). For loci M33 and M35, each reaction (total volume of 12.5 µL) contained 2.5 µL of DNA, 6.25 µL of Type-it Multiplex PCR Master Mix (2×), 2.5 µL of DNase/RNase-free H2O, and 1.25 µL primer mix (concentration of 0.2 µM).
Characteristics of 10 microsatellite loci developed in Sagina nodosa.
All amplifications were carried out in the MultiGene Gradient PCR Thermal Cycler (Labnet International, Edison, New Jersey, USA). For the TP-PCR, the cycling started with initial denaturing at 94°C for 2 min; followed by 30 cycles of 94°C for 30 s, 56°C for 45 s, and 72°C for 1 min; followed by eight cycles of 94°C for 30 s, 53°C for 45 s, and 72°C for 1 min; and a final extension of 72°C for 10 min. For loci M33 and M35 (Type-it Microsatellite PCR Kit, QIAGEN), the cycling started with initial denaturing at 95°C for 5 min; followed by 28 cycles of 95°C for 30 s, 60°C for 90 s, and 72°C for 30 s; and a final extension of 60°C for 30 min. Prior to genotyping, all PCR products were visualized using gel electrophoresis.
PCR products were diluted in Hi-Di formamide (Applied Biosystems, Carlsbad, California, USA) and analyzed at the Cornell Life Sciences Core Laboratory Center on an ABI 4200 or 3730xl DNA Sequencer (Applied Biosystems) using the GeneScan 500 LIZ internal size standard (Applied Biosystems). For fragment analysis, we used GENEMAPPER version 3.7 (Applied Biosystems). All 29 individuals collected in Isle Royale were genotyped and analyzed using GENEPOP (Rousset, 2008) to quantify the allele frequencies and observed and expected heterozygosities at each locus.
Ten loci, all in linkage equilibrium, were polymorphic and included a balance of different microsatellite repeat motifs: four dinucleotide repeats, three trinucleotide repeats, two tetranucleotide repeats, and one hexanucleotide repeat (Table 1). In the 29 individuals genotyped, each locus had between two and nine total alleles and an observed heterozygosity ranging from 0.074 to 0.483 (Table 2). When individuals from all of the Isle Royale islands were analyzed as a single population, all loci significantly violated Hardy-Weinberg equilibrium (HWE), displaying a very low level of observed heterozygosity (Table 2). This is likely due to population substructure (i.e., Wahlund effect [Wahlund, 1928]) within the archipelago. Indeed, when we analyzed each island population separately, we found no deviation from HWE (although our sample sizes were too small for a definite conclusion). Alternatively, the low observed heterozygosity could be caused by significant clonal reproduction, but a larger analysis is needed to demonstrate the cause of HWE deviation.
We developed and characterized 10 novel microsatellite markers for 29 S. nodosa individuals from the disjunct Isle Royale population using next-generation sequencing technology. The individuals are from eight different islands (subpopulations), and preliminary analyses suggest strong population substructure. To our knowledge, these are the first microsatellite loci for species in the Sagina genus. These novel markers will be used to genotype a larger population of individuals to provide insight on the population structure and life history of S. nodosa in Isle Royale and other North American populations.
Genetic properties of the 10 microsatellite loci of Sagina nodosa among the 29 individuals sampled.