Forget-me-nots (Myosotis L., Boraginaceae) are found in both the Northern and Southern Hemispheres, with a center of diversity in New Zealand. The M. pygmaea species group (Meudt et al., 2015) comprises M. antarctica Hook. f., M. brevis de Lange & Barkla, M. drucei (L. B. Moore) de Lange & Barkla, M. glauca (G. Simpson & J. S. Thomson) de Lange & Barkla, and M. pygmaea Colenso, all native to New Zealand. Questions persist regarding the delimitation of these morphologically similar species (de Lange et al., 2010), four of which appear on the New Zealand threatened species list (de Lange et al., 2013). Indeed, of the 44 endemic New Zealand Myosotis taxa, 32 are considered threatened or at risk (de Lange et al., 2013). A priority in the conservation management of members of this genus is to both accurately delimit species and understand the levels and structure of genetic diversity present. Low genetic diversity in New Zealand Myosotis, as evidenced by previous studies (Meudt et al., 2013, 2015), suggests that additional molecular markers are needed.

Here we report the development of 12 polymorphic microsatellite markers for the M. pygmaea species group, which will be used in future studies of species delimitation and population genetic research. Additionally, we evaluate the utility of these loci in 18 other Myosotis species.

METHODS AND RESULTS

Sibling individuals were selected from the type locality of M. drucei as the source DNA for marker development (WELT SP100445; Appendix 1). Genomic DNA was extracted from fresh young leaf tissue from 15 seedlings using a modified cetyltrimethylammonium bromide (CTAB) method (Shepherd and McLay, 2011). To generate sufficient template for the requirements of Illumina MiSeq library preparation, extracted DNA was pooled and amplified using a REPLI-g kit (QIAGEN, Hilden, Germany) following the manufacturer's protocol. DNA was quantified using a Qubit 2.0 Fluorometer (Thermo-Fisher Scientific, Waltham, Massachusetts, USA), and a genomic library was prepared using the TruSeq Library Preparation Kit (Illumina, San Diego, California, USA) by the Massey Genome Service (Massey University, Palmerston North, New Zealand). The indexed library was pooled with three other libraries in equal concentration and sequenced using the paired-end 250-bp chemistry on a MiSeq (Illumina) by the Massey Genome Service. The resulting 2.7 million sequences were trimmed of low-quality results using a 0.01 quality cut-off in Dynamic Trim in SolexaQA (Cox et al., 2010), which yielded 1,449,369 trimmed paired-end sequences with an average length of 380 bp, ranging in size from 11–492 bp. Paired-end sequences were joined using the program FLASH (Magoc and Salzberg, 2011).

The paired-end sequences were then imported into Geneious 6.1.5 (Biomatters, Auckland, New Zealand), where only sequences >400 bp were retained. Organellar sequences were removed by performing a local BLAST search of the M. drucei sequences against the phylogenetically closest relatives (Soltis et al., 2011) with the most complete mitochondrial and chloroplast sequences from GenBank. The chloroplast genomes used were: Nicotiana undulata Ruiz & Pav. NC_016068 (Solanaceae), Olea europaea L. subsp. maroccana (Greuter & Burdet) P. Vargas, J. Hess, Muñoz Garm. & Kadereit NC_015623 (Oleaceae), Coffea arabica L. NC_008535 (Rubiaceae), and Arabidopsis thaliana (L.) Heynh. NC_000932 (Brassicaceae). The mitochondrial genomes used were: N. tabacum L. NC_006581, A. thaliana NC_001284, and Vigna radiata (L.) R. Wilczek NC_015121 (Fabaceae). The remaining 397,224 sequences were split into four groups (due to computer memory constraints), and the first group of 99,999 sequences was searched for perfect di- to hexanucleotide microsatellite repeats with a minimum of seven uninterrupted repeat units using a search tool in Geneious (Phobos plugin; Mayer, 2010), which identified 484 repeats. Sequences were removed from consideration if the paired-end sequences were found to be overlapping only in the repeat region, if regions near the microsatellite contained other microsatellite loci or single base pair repeats >4 bp, or if there were greater than 14 repeats. After removing unsuitable loci, primers were designed for 147 microsatellite regions using Primer3 within Geneious (Untergasser et al., 2012). The default settings were used except for: product size = 100−400 bp with a 50-bp buffer on both sides of the target region; primer size = 18 bp (minimum)−20 bp (optimal)−22 bp (maximum); melting temperature (T_m) = 47−55−60°C; 3′ GC content = 40−50−60%; maximum T_m difference = 10°C; GC clamp = 1; max poly N = 4. An M13 tag (CACGACGTTGTAAAACGAC) was added to the 5′-end of the forward primer for each locus, and a PIG-tail sequence (GTTTCTT; Brownstein et al., 1996) was added to the 5′-end of each reverse primer.

Table 1.

Primer sequences and characteristics of 12 microsatellite loci developed in Myosotis drucei.

Table 2.

Summary statistics of microsatellite polymorphism determined by screening 53 Myosotis drucei samples from four populations; three from the South Island and one from the North Island of New Zealand.^a

For reasons of practicality, 48 primer pairs were chosen to trial a range of: uninterrupted number of repeats, types of microsatellites (e.g., di-, tri-, tetra-, penta-, and hexa-), and PCR product sizes. These 48 were initially trialed on seven individuals from five populations of four M. pygmaea group species (Appendix 1). Each locus was amplified individually in 10-µL PCR reactions that contained 1 µL of a 1:50 dilution of template DNA (5–50 ng), 0.02 µM forward primer, 0.45 µM reverse primer, 0.45 µM M13 primer (labeled with FAM, NED, or VIC), 1.5 mM MgCl₂, 1× buffer BD (Solis BioDyne, Tartu, Estonia), 250 µM of each dNTP, and 1 unit FIREPol Taq polymerase (Solis BioDyne). PCRs were carried out with the following cycling program: an initial denaturation of 95°C for 3 min; 40 cycles of 95°C for 30 s, 53°C for 40 s, and 72°C for 1 min; and a final extension at 72°C for 10 min. A volume of 0.75 µL of each PCR product for three loci, each with a different fluorophore, was added to 9 µL of Hi-Di formamide (Applied Biosystems, Carlsbad, California, USA) premixed with a ROX-labeled CASS ladder (Symonds and Lloyd, 2004) for subsequent fragment separation on an ABI 3730 Genetic Analyzer (Applied Biosystems) by the Massey Genome Service.

Table 3.

Cross-amplification of 12 novel microsatellite loci in 22 Myosotis species.^a

Alleles were visualized and scored using GeneMapper version 3.7 (Applied Biosystems). Of the 48 primer pairs tested, 25 were polymorphic, two were monomorphic, seven were unscorable, and 14 did not amplify. Twenty-four of the polymorphic loci were further tested using the above PCR conditions on 15 individuals from five Myosotis species. The 12 markers (Table 1) with the best amplification rates were selected for further investigation using four populations of M. drucei to demonstrate the utility of the markers in a population genetic framework. For these four populations, Table 2 shows the number of alleles, and observed (H_o) and expected (H_e) heterozygosities, which were determined using GenAlEx (Peakall and Smouse, 2012). The average number of observed alleles per locus was 3.75, and average H_o was 0.059 (Table 2). H_o was typically lower than H_e, which matches the hypothesized mostly selfing nature of the M. pygmaea species group (Robertson and Lloyd, 1991; Brandon, 2001). The 12 markers amplified well across the other four species (one population each) in the M. pygmaea group (voucher information in Appendix 1) and were also trialed in an additional 18 species of Myosotis, 14 endemic to New Zealand, one from Australia, and three introduced to New Zealand from Europe. Amplification rates and polymorphism are reported in Table 3.

CONCLUSIONS

We describe 12 polymorphic microsatellite loci that will be useful for exploring species limits within the M. pygmaea species group, as well as determining the population genetic variation within and among other species of Southern Hemisphere Myosotis.