The endangered Hawaiian coot and Hawaiian gallinule are endemic to the Hawaiian Islands and both underwent precipitous declines in the early 1900s likely due to the loss and degradation of wetland habitat and introduced predators. Historically both species were common and distributed across the main Hawaiian islands; surveys conducted in the 1950–1960s suggest that numbers were reduced to <1,000 (Hawaiian coot) and 60 (Hawaiian gallinule) individuals throughout Hawaii (USFWS 2011). Population size has increased for the Hawaiian coot (1,777 ± 310 individuals; Underwood et al. 2013). The Hawaiian gallinule’s current distribution is now restricted to Oahu and Kauai and recent surveys recorded low numbers (~400 individuals range-wide), though numbers are likely underestimated as Hawaiian gallinules are secretive (Underwood et al. 2013). The microsatellite markers described herein will provide much needed variable nuclear loci to assess the impact of the population declines on genetic diversity of these endangered waterbirds as well as inter-population genetic variation and phylogeographic studies.

Genomic DNA was extracted from blood samples of a Hawaiian coot (USFWS band number 1056-91345) and Hawaiian gallinule (USFWS band number 1196-87722) trapped at James Campbell National Wildlife Refuge, Oahu, Hawaii. We employed next generation sequencing following a shotgun sequencing protocol, using GS Junior Titanium rapid library preparation and sequencing chemistry (Roche, Branford CT). The barcoded libraries were pooled, clonally amplified and sequenced on a Roche 454 GS Junior. The Hawaiian coot library generated 43,955 reads (154 contigs with 37,866 unassembled) and the Hawaiian gallinule library resulted in 36,493 reads (139 contigs and 32,236 unassembled).

Msatcommander 0.8.2 (Faircloth 2008) searched contigs and unassembled reads for microsatellite repeats with ≥6 repeat units and designed primer sets. Microsatellite motifs were identified in 153 reads from the Hawaiian coot and 158 reads from the Hawaiian gallinule. Microsatellite loci with dinucleotide repeat motifs, 20 in both the Hawaiian coot and Hawaiian gallinule, were selected for development and loci were tested for variability and cross species amplification among ten individuals representing each species sampled from James Campbell National Wildlife Refuge. PCR amplifications and thermocycler conditions followed methods described by Sonsthagen et al. (2004). We used Arlequin (v.3.1; Excoffier et al. 2005) to calculate observed and expected heterozygosity and tests for linkage disequilibrium and deviations from Hardy–Weinberg equilibrium (HWE), applying Bonferroni corrections for multiple tests (α = 0.05). Of the 40 loci tested for variability, 23 were polymorphic in Hawaiian coot and/or Hawaiian gallinule (Table 1). Genetic diversity metrics for all loci are listed in Table 1. Twenty loci were variable for the Hawaiian coot; the number of alleles ranged from 2 to 7 (mean 3.8) and observed heterozygosity from 10 to 100 % (average 59.5 %). Twelve variable microsatellites were detected in the Hawaiian gallinule; the number of alleles ranged from 2 to 4 (mean 2.4) and observed heterozygosity from 10 to 100 % (average 47.5 %). All loci were in linkage equilibrium and only one locus (Fal006) deviated from HWE expectation. These loci are sufficiently variable to assess levels of genetic diversity, as well as inter-population genetic structure, and will be useful for conservation genetic studies to aid in the management of these endangered waterbirds.

Table 1 Characterization of 23 microsatellite loci developed for the Hawaiian coot and Hawaiian gallinule, including repeat motif, primer sequence (universal tail in parentheses), allele size range in base pairs (bp), number of alleles (A), observed (HO) and expected (HE) heterozygosity, and GenBank accession numbers
Full size table