Spanning 2,600 km, and ranging in age from a few hundred thousand to more than 60 million years, the isolated volcanic archipelago of Hawai‘i offers a unique opportunity to understand population connectivity in the marine environment. For sessile benthic marine organisms, the larval phase of the life history is the only part of the life cycle that allows for gene flow between populations, but it is difficult to track miniscule larvae across oceanic scales (Levin 2006). In order to gain insight into the historical patterns of coral colonization of the archipelago, microsatellite markers are incredibly useful tool for understanding population genetic structuring and inferring patterns of connectivity (Selkoe and Toonen 2006).

Total genomic DNA of Acropora cytherea, Fungia scutaria, Montipora capitata and Porites lobata were extracted in parallel according to the chloroform extraction protocol described in Concepcion et al. (2006). DNA from A. cytherea and P. lobata were digested with restriction enzyme BstU I, while DNA from F. scutaria and M. capitata were digested with restriction enzyme Rsa I. Using a slightly modified protocol from Glenn and Schable (2005), double stranded linkers were ligated to the 5′ end of restriction fragments. Enrichment probes consisted of two mixtures of 3′ biotinylated probes—Mix I: (AAGC)5, (AACC)5, (AACG)5, (ATCC)5, (AAGG)5; Mix II: (ATC)8, G(AGG)6, G(CCG)5, (AAT)10, (AAG)8, (ACT)8, (AAC)8, (ACG)6, (ACC)6, (AGC)6 (Toonen 1997). Fragments with sequences complementary to these probes were captured with Streptavidin M-280 Dynabeads (Invitrogen, Carlsbad, CA, USA). DNA enriched for these tri- and tetranucleotide microsatellite repeat motifs was ligated into T-tailed pZErO-2 plasmids using a Topo-TA cloning kit (Invitrogen). Escherichia coli α-select chemically competent cells (Bioline Inc., Springfield, NJ, USA) were transformed with the recombinant plasmids. Across all four species, 8,064 clones were picked using a VersArray Colony Picker (Biorad) and subsequently consolidated into and re-grown in 384 well sample plates. Colonies from the 384 well plates were then imprinted and grown directly on nitrocellulose membranes resting on LB agar plates and incubated overnight at 37°C. Following Toonen (1997), DNA was extracted and fixed on the membrane and was subsequently screened for microsatellite positive colonies by hybridization with the same biotinylated oligonucleotide probe mixtures as above with a modified colorimetric detection step using chemiluminescent CDP-star (#N7001S NEB Biolabs) and subsequent exposure to Kodak® X-Ray film. 1,423 colonies were identified as containing microsatellite repeats. 1,039 clones from A. cytherea, F. scutaria and M. capitata were amplified and submitted for sequencing at the Hawai‘i Institute of Marine Biology EPSCoR Facility, while 384 clones from P. lobata were amplified and sequenced at the Huck Institute for the Life Sciences, Pennsylvania State University. The Staden-Troll pipeline (Martins et al. 2006) was used to identify tandem repeats. Primers were designed using Primer3 for each of 50, 118, 140, and 149 sequences from A. cytherea, F. scutaria, M. capitata and P. lobata, respectively. Primer sequences as well as the genus, species, bitscore and e-value for the top BLASTn hit (Altschul et al. 1990) can be found in Supplemental Table 1.

Table 1 Locus name, repeat motif, primer sequences, annealing temperature (T a) and approximate size of expected product for one nuclear coding region and eight microsatellites for M. capitata and seven microsatellites from A. cytherea
Full size table

Due to time constraints and cost, primers were ordered and tested for only the first 50 putative loci from M. capitata and A. cytherea. Primers were screened with a low annealing temperature (48°C) PCR against DNA extracts from pure zooxanthellae cultures provided by RA Kinzie believed to represent clades A, B and C isolated from Cassiopea sp. (KB8), Aiptasia pulchella (HIAp) and Montipora verrucosa (Mv), respectively. They were subsequently screened against host genomic DNA with an annealing temperature of 55°C. We used the tailed three primer method described by Gaither et al. (2009). Tails were added to the 5′ end of all forward primers that successfully amplified host genomic DNA, but not symbiont DNA (Table 1). PCRs were as follows: each 10 μl reaction contained: 1 μl 10×NH4 reaction buffer, 0.6 μl 50 mM MgCl2, 0.4 μl 10 mM total dNTPs (2.5 mM each), 0.35 pmol tailed forward primer, 1.5 pmol reverse primer, 1.5 pmol oligonucleotide dye label, 2–25 ng of template DNA, 0.1 μl of Biolase polymerase (Bioline Inc.), and deionized water to volume. PCR amplification was performed on a BioRad MyCycler™ as follows: 95°C for 10 min (1 cycle), 94°C for 30 s, 55°C for 30 s, 72°C for 30 s (35 cycles), followed by a final extension of 72°C for 30 min (1 cycle).

Eight primer pairs for Montipora capitata, and two primer pairs for Acropora cytherea in addition to five primer pairs from van Oppen et al. (2007) successfully amplified polymorphic loci and were selected for further genotyping of samples (Table 1). Fragments were analyzed on an ABI 3130XL Genetic Analyzer at the Hawai‘i Institute of Marine Biology and sized using Genemapper v4.0 and GS500LZ size standards (Applied Biosystems, Inc.). Arlequin 3.11 (Excoffier et al. 2005) was used to calculate genotypic disequilibrium, heterozygosity, and probability of departure from HWE. Each primer pair was initially screened for variability against 25 individuals sampled from each of French Frigate Shoals and Johnston Atoll (Table 2). The number of alleles for each locus ranged from 2 to 9 for both species (Table 2). No linkage was detected in either population for A. cytherea, and M. capitata. After controlling for false discovery rate (Benjamini and Yekutieli 2001), significant departures from HWE were detected in six out of 32 comparisons (Table 2). MicroChecker (van Oosterhout et al. 2004) detected no evidence of scoring errors, or large-allele dropout in either species, although possible null alleles were likely at each locus that showed significant departure from HWE in a population (Table 2).

Table 2 From each sampling location for each of eight microsatellites and one nuclear coding region isolated for Montipora capitata and two microsatellites isolated from Acropora cytherea plus five additional microsatellites isolated from congener Acropora millepora, we report sample size (N), number of alleles (N a ), observed heterozygosity (H o ); expected heterozygosity (N e ); inbreeding coefficient (F is ), P-values for tests of departure from HWE (P hwe ), and Brookfield-1 estimator of null alleles (Null est ). Significant values for P hwe after correction with a false discovery rate α = 0.05 shown in bold. Corrected α for M. capitata (P < 0.006) and A. cytherea (P < 0.014)
Full size table

Additionally, forward and reverse primers for ATP synthethase β (atpsβ) (Jarman et al. 2002) were redesigned for specificity in the genus Montipora with PCR conditions following Concepcion et al. (2008). This locus was also sequenced directly for each individual from each population (n = 50). Computational methods for determining phase of diploid sequence data is more cost effective and can be as accurate as cloning (Harrigan et al. 2008). Therefore, we used Phase (Stephens et al. 2001; Stephens and Donnelly 2003) as implemented in DnaSP (Librado and Rozas 2009) for determining gametic phases (Table 2).

The use of microsatellites to study coral population genetics is still in its infancy. We hope this library of both characterized, and untested microsatellite loci will provide a wealth of population genetic tools to aid studies both inside and outside of Hawai‘i, in species of corals other than the four for which they were isolated, as well as providing possible microsatellite markers for the symbionts (Symbiodinium spp.)