Arthropodium R. Br. (Asparagaceae) is a genus of nine species (Heenan et al., 2004) of perennial, lily-like herbs found in Australia, New Zealand, New Caledonia, New Guinea, and Madagascar. Within New Zealand, there are three endemic species: A. candidum Raoul (small renga lily), A. cirratum (G. Forst.) R. Br., and A. bifurcatum Heenan, A. D. Mitch. & de Lange (the latter two species both have the common names rengarenga and New Zealand rock lily).

Arthropodium cirratum was cultivated as a food source for Māori and translocated beyond its natural range (Shepherd et al., 2016). A recent phylogeographic study of A. cirratum using chloroplast sequences revealed a very high level of structuring, with many populations fixed for unique chloroplast haplotypes (Shepherd et al., 2016). Microsatellite markers will be useful for testing whether the nuclear genome demonstrates the high genetic structuring found for the chloroplast genome (Shepherd et al., 2016). They will also aid in examining the origins of the populations that derive from translocation by Māori and testing proposed hybridization between A. cirratum and A. bifurcatum (Heenan et al., 2004).

METHODS AND RESULTS

We extracted DNA from leaf tissue of four A. cirratum individuals (Appendix 1), each from different populations, using a modified cetyltrimethylammonium bromide (CTAB) method (steps 1, 3–7 from table 1 in Shepherd and McLay, 2011). The extracted DNA was pooled and amplified using a REPLI-g kit (QIAGEN, Hilden, Germany) following the manufacturer's protocol to generate sufficient template for library construction. An Illumina paired-end genomic library was constructed using the TruSeq Nano DNA Library Prep Kit (Illumina, San Diego, California, USA) following the manufacturer's instructions. The library was sequenced in a single lane using the Illumina MiSeq platform to generate 2 × 250-bp reads at the Massey Genome Service (Massey University, Palmerston North, New Zealand).

We assembled the resulting 10,955,497 paired sequence reads using MEGAHIT (Li et al., 2015), as this software required sequence reads of the same length. A set of four assembly parameters was tried with the reads, and the resulting contigs were merged to make a set of longest unique contigs. This resulted in 1.589 Gb of assembled sequence, comprising 2,618,361 contigs, with a maximum length of 18,007 bp, average GC content of 34.74%, and an N50 of 1513 bp when analyzed using QUAST (Gurevich et al., 2013). The SSR_pipeline (Miller et al., 2013) was used to detect di- and tetranucleotide repeats on this contig set with a minimum of 250-bp flanking sequence on each side to allow for PCR primer design. We used WebSat (Martins et al., 2009) to develop primers for 33 loci, which had at least eight tetra- or 15 dinucleotide repeat units. An M13 tag (TGTAAAACGACGGCCAGT) was added to the 5′ end of the forward primer of each locus. These primer pairs were tested on five samples, which included three samples of A. cirratum and one sample each of A. candidum and A. bifurcatum. Each locus was initially amplified individually in 10-µL PCR reactions that contained 1 µL, of diluted template DNA, 0.02 µM forward primer, 0.8 µM reverse primer, 0.8 µM M13 primer (labeled with FAM, NED, PET, or HEX), 1× MyTaq mix (Bioline, London, United Kingdom), and 0.1 M betaine. PCR thermocycling conditions were an initial denaturation of 94°C for 5 min; 30 cycles of 94°C for 30 s, 55°C for 45 s, and 72°C for 45 s; followed by eight cycles of 94°C for 30 s, 53°C for 45 s, and 72°C for 45 s; and a final extension at 72°C for 15 min.

Of the 33 primer pairs tested, 16 amplified in at least two species and were polymorphic. These 16 loci were subsequently screened using 63 samples from three populations of A. cirratum and additional samples of A. bifurcatum and A. candidum (Appendix 1). For this trial, some loci were coamplified in the same PCR reaction (ArtCir13 with ArtCir18, ArtCir9 with ArtCir23, and ArtCir43 with ArtCir48). For these combined PCR reactions, 1 µL of diluted template DNA was combined with 0.02 µM each forward primer, 0.8 µM each reverse primer, 1.2 µM M13 primer (labeled with FAM, NED, PET, or HEX), 1× MyTaq mix (Bioline), and 0.075 M betaine. The PCR annealing temperatures are reported in Table 1. Genotyping was performed on an ABI 3130xl Genetic Analyzer (Applied Biosystems, Foster City, California, USA) at the Massey Genome Service. Alleles were sized using the internal size standard GeneScan 500 LIZ (Applied Biosystems) and scored using Geneious version 10.0.2 (Biomatters Ltd., Auckland, New Zealand).

Table 1.

Primer sequences and thermal cycling conditions for 16 microsatellite loci developed for Arthropodium cirratum.

The number of alleles and observed and expected heterozygosities for the three A. cirratum populations were determined using GenAlEx 6.5 (Peakall and Smouse, 2012). Observed and expected heterozygosities ranged from 0.000 to 1.000 and 0.044 to 0.544, respectively (Table 2). Alleles per locus ranged from one to five in A. cirratum (mean = 3). All three of the A. cirratum populations exhibited private alleles, and 14 of the loci had private alleles in at least one of the three populations. Tests of pairwise linkage disequilibrium were performed using GENEPOP 4.2 (Rousset, 2008). No significant linkage disequilibrium was detected among paired loci comparisons after sequential Bonferroni correction (Holm, 1979). Deviation from Hardy–Weinberg equilibrium was tested for each locus with GenAlEx 6.5. Following sequential Bonferroni correction, significant deviation from Hardy–Weinberg equilibrium was observed for five loci (Table 2). This is unsurprising for a species with delayed autonomous self-pollination (Zhou et al., 2012). ArtCir18 showed fixed heterozygote genotypes for all the screened individuals in the Maunganui Bluff and Hick's Bay populations, but each population was fixed for different alleles.

Table 2.

Genetic diversity measures for three populations of Arthropodium cirratum.^a

Table 3.

Cross-amplification of 16 Arthropodium cirratum microsatellites in A. bifurcatum and A. candidum, showing fragment sizes of each allele.

All 16 loci amplified in the closely related species A. bifurcatum, and 12 of these were polymorphic (Table 3). Eight loci amplified in the more distantly related A. candidum, but none of the three samples screened were polymorphic at these loci.

CONCLUSIONS

We developed 16 variable microsatellite markers for A. cirratum using Illumina MiSeq data. Although most of the markers had a low number of alleles, many showed fixed allelic differences between the populations examined. These markers will be useful for characterizing genetic diversity and structure in A. cirratum and for examining the translocation of this species.