Isolation and characterization of macaroni penguin (Eudyptes chrysolophus) microsatellite loci and their utility in other penguin species (Spheniscidae, AVES)

Many species of penguin are currently in decline (IUCN, 2008). Macaroni penguins, Eudyptes chrysolophus, are the most numerous species of penguin in the Southern Ocean but are now listed as Vulnerable as a result of marked decline in populations in areas such as South Georgia and Marion Island (IUCN, 2008). We report the isolation and characterization of markers suitable for investigating population genetic structure and levels of genetic diversity in the macaroni penguin, with application to a number of other penguin species.

Blood samples were taken from breeding adult macaroni penguins present during the incubation or brood guard stage in the colony located at a place called Macaroni Cwm (54°00′10″S, 38°01′10″W) on Bird Island, South Georgia. Blood sampling was performed as part of work undertaken by the British Antarctic Survey. For each individual, 0.1 mL of blood was taken from the brachial vein and immediately stored in 1.4 mL of absolute ethanol in rubber-sealed screw-topped microfuge tubes. All animal procedures were performed in accordance with UK Home Office guidelines and subject to ethical review. Genomic DNA was extracted using an ammonium acetate extraction protocol (Nicholls et al. 2000). A microsatellite-enriched library was constructed essentially using the method of Armour et al. (1994). The library was enriched for the following di- and tetranucleotide microsatellite motifs and their complements: (GT)_n, (CT)_n, (GTAA)_n, (CTAA)_n, (TTTC)_n and (GATA)_n, which had been bound to magnetic beads by following the procedure described by Glenn & Schable (2005). Transformant colonies were not screened for the presence of a repeat but were directly sequenced.

Two-hundred and eighty-eight bacterial transformant colonies were sequenced in the forward and reverse orientation and a consensus sequence created. A total of 132 unique sequences contained microsatellite motifs (EMBL accession numbers: FM878332–FM878463). Forty-seven sequences contained a minimum of eight repeats. Primer sets were designed for 32 of these sequences using primer3 software (Rozen & Skaletsky 2000).

All loci were amplified individually prior to multiplexing (see Table 1 for multiplex scheme). Primer sets were tested in up to 28 macaroni penguin individuals belonging to the colony located at Macaroni Cwm. Polymerase chain reaction (PCR) amplification was performed in 2-μL reactions (Kenta et al. 2008) containing 10 ng of genomic DNA, 1 μL master mix (Qiagen multiplex PCR kit) and 1 μL of each primer at 0.2 μm concentration. PCR cycling conditions were 15 min at 95 °C, followed by 40 cycles of 94 °C for 30 s, 60 °C for 90 s and 72 °C for 60 s, with a final step of 72 °C for 10 min. PCR products were diluted and allele sizes assigned using an ABI 3730 DNA Analyser (Applied Biosystems) and the size standard ROX 500. After initial primer testing, seven markers were discarded due to unreliable or failed PCR amplification (Ech023, Ech025, Ech049, Ech078, Ech087, Ech109 and Ech122).

Twenty-five loci were polymorphic when tested in 28 individuals sampled from the colony located at Macaroni Cwm (Table 1). Individuals were sexed using the Z-002D primer set (Dawson 2007). Loci were checked for sex linkage based on the genotyping of known-sex macaroni penguins [17 females (ZW) and seven males (ZZ)]. Twenty-four of the 25 polymorphic loci displayed heterozygotes in females and amplified in males indicating they were not sex linked. One locus, Ech020, displayed nine different alleles but was homozygous in all 20 individuals in which it amplified (13 females, four males, three unknown sex). The evidence of homozygous females (ZW) suggests that Ech020 is Z-linked. There was possible evidence for null alleles at this locus as eight individuals failed to amplify (four females, three males, one unknown sex), which may explain why all males were homozygotes. The sequence of locus Ech020 (FM878351) displayed significant homology to sequence located on the chicken Z chromosome, confirming its Z-linked status (homology detected by following the procedure described by Dawson et al. 2006).

Observed and expected heterozygosity and predicted null allele frequency were calculated using CERVUSv3 (Marshall et al. 1998; Kalinowski et al. 2007). Tests for departure from Hardy–Weinberg proportions and linkage disequilibrium were implemented in genepopv4 (Raymond & Rousset 1995; Rousset 2007) for up to 28 individuals sampled from the colony located at Macaroni Cwm. None of the loci showed genotypic disequilibrium after a sequential Bonferroni correction for multiple comparisons (Rice 1989).

The genotypes of six loci showed significant departures from Hardy–Weinberg proportions after a sequential Bonferroni correction for multiple comparisons (P = 0.002; loci Ech005, Ech012, Ech063, Ech091, Ech113 and Ech130; Table 1). Some of these loci may possess null alleles but others may simply require further PCR optimization. Several of these loci show potential utility for genotyping other penguin species (see below).

Seventeen of the polymorphic loci were selected and assessed for multiplexing potential. The selected loci showed no significant deviation from Hardy–Weinberg equilibrium and had a low estimated null allele frequency when amplified separately (as singleplexes). For these loci, the same 28 individuals were amplified individually as a singleplex PCR and then as part of a multiplex set. Four loci (Ech003, Ech024, Ech050 and Ech051) were difficult to score or did not amplify well when part of their assigned multiplex set. However, 13 loci were easily scorable and the same genotypes were obtained following singleplex and multiplex genotyping. The mean observed and expected heterozygosity of these 13 loci measured in up to 28 individuals sampled from the colony located at Macaroni Cwm was 0.64 and 0.76, respectively, and the combined first and second parent exclusion probability was better than 99.9%.

Prior to this study, a total of only 39 penguin microsatellite loci had been isolated and from few penguin species: Humboldt penguin (Spheniscus humboldti) (Schlosser et al. 2003), little blue penguin (Eudyptula minor) (Billing et al. 2007), yellow-eyed penguin (Megadyptes antipodes) (Boessenkool et al. 2008) and Adélie penguin (Pygoscelis adeliae) (Roeder et al. 2001, 2002). We therefore tested all 25 loci for their cross-species utility in other species of penguin (Table 2). Three unrelated individuals from king penguin (Aptenodytes patagonicus), gentoo penguin (Pygoscelis papua), chinstrap penguin (Pygoscelis antarctica) and Adélie penguin (Pygoscelis adeliae) were genotyped. Seventeen markers were polymorphic in at least one of the four other penguin species tested (Table 2). Eight loci (Ech008, Ech012, Ech030, Ech051, Ech060, Ech071, Ech081 and Ech091) were polymorphic in at least three of the four additional species tested (Table 2). Despite testing only three individuals, between eight and 12 new polymorphic microsatellite loci were identified for each species (Table 2). If necessary, amplicons could be sequenced and new species-specific primers designed. These macaroni penguin microsatellite loci are expected to be useful for population genetic studies in a variety of penguin species.