Immune genes are ideal for studying the evolutionary processes affecting genetic diversity of wild populations, as they represent the most rapidly evolving genes within the genome, due to selection pressure from a wide diversity of coevolving pathogens (Hedrick 1998; Piertney and Oliver 2006). While genes of the major-histocompatibility complex have been widely used for assaying levels of functional diversity, variation at other immunity genes also affects variation in individual immune responses, and are thus of interest to population geneticists (Acevedo-Whitehouse and Cunningham 2006; Vinkler and Albrecht 2009; Turner et al. 2012). One such gene family is toll-like receptors (TLRs), which recognise a wide diversity of pathogens and are responsible for initiating the innate and acquired immune responses (Uematsu and Akira 2008). TLR sequence variation in wild populations has been associated with variation in resilience to infections (Villaseñor-Cardoso and Ortega 2011), and may influence variation in survival of wild animals (e.g. Grueber et al. in press). TLRs also offer some technical advantages over molecular genotyping of MHC (Grueber et al. 2012). Most previous studies of avian TLR diversity have primarily focussed on common or domestic species (Alcaide and Edwards 2011). Expanding the conservation genetics toolkit by examining genic regions associated with a variety of crucial biological processes, including those of the immune system, can provide valuable data for effective management of wild populations (Acevedo-Whitehouse and Cunningham 2006). We aim to provide additional genetic resources (PCR primers and sequences) to facilitate the examination of TLR diversity in threatened birds.

Our study species included one member of Apterygiformes (North Island brown kiwi Apteryx mantelli), one member of Gruiformes (takahe Porphyrio hochstetteri), two members of Psittaciformes (kakariki [red-crowned parakeet] Cyanoramphus novaezelandiae and kakapo Strigops habroptilus) and five members of Passeriformes (kokako Callaeas wilsoni, hihi [stitchbird] Notiomystis cincta, mohua [yellowhead] Mohoua ochrocephala, South Island saddleback Philesturnus carunculatus and New Zealand rock wren Xenicus gilviventris). Samples were collected for previous (Grueber et al. 2008; Taylor and Jamieson 2008; Tracy and Jamieson 2011) or concurrent studies and were provided to us either as extracted gDNA, whole blood in ethanol or lysis buffer (Seutin et al. 1991), or feathers. We used two samples from all species (except one from X. gilviventris) as an internal control against sample handling errors. Where tissue samples were provided, DNA was purified using a modified Chelex (Bio-Rad) extraction [based on Walsh et al. (1991), Casquet et al. (2012)] (blood samples), followed by an additional ethanol precipitation using GenElute linearised polyacrylamide (Sigma) as a DNA carrier (feather samples).

We first attempted to amplify TLR sequences in all species using published primers (Alcaide and Edwards 2011; Grueber et al. 2012). If these primers failed to produce a clear product, we re-examined the avian alignment for the locus in question (Alcaide and Edwards 2011). Primers were then designed at conserved regions using Primer3 (Rozen and Skaletsky 1998); ambiguities in the alignment were resolved to the most taxonomically similar species for which sequence was available (either published sequences available on Genbank, or sequences obtained herein). Our priority was to amplify fragments of maximal length, so we only designed internal primers in cases where no product amplified, or the sequence data was of too low quality to reliably identify ambiguities. Where multiple primer pairs were successful, we report only the pair that provides that longest fragment.

Amplification, clean up and sequencing followed Grueber et al. (2012). Amplifications were performed in a total volume of 15 μl, containing 1 μl extracted DNA, 0.6 U MyTaq polymerase (Bioline Ltd), 1× MyTaq buffer (giving a final concentration of 3 mM Mg2+ and 1 mM dNTPs) and 500 nM each primer (Sigma-Aldrich). Thermocyling conditions (on an Eppendorf Mastercycler pro S) consisted of 94 °C for 3 min, followed by 35 cycles of 94 °C for 40 s, locus-specific annealing temperature (Alcaide and Edwards 2011; Grueber et al. 2012; Table 2) for 40 s, and extension at 72 °C for 80 s, with a final extension step of 72 °C for 10 min. Products were purified by excision from 1 % agarose, cleaned up using the MEGAquick-spin total fragment DNA purification kit (iNtRON Biotechnology Inc.), and sequenced in both directions on an ABI 3730xl Genetic Analyser (service provided by Genetic Analysis Services at Otago). Sequences were edited using Sequencher v5.0 (Gene Codes Corporation), using IUPAC ambiguity codes where double-peaks were observed. Correct amplification of the target sequence was confirmed using BLAST (Altschul et al. 1997) search.

We developed a total of 30 new primers (Table 1) which, along with cross-species amplification of previously-published primers, enabled us to successfully obtain on average 1,087 bp fragments (range 677–1,345 bp) for between five and eight TLR genes in nine bird species across four families (amplification conditions and Genbank accession numbers provided in Table 2). Although we have not amplified the full sequence of each gene, our primers target their extracellular domains: the variable region responsible for pathogen binding (as opposed to the intracellular domain, which is highly conserved to maintain the integrity of intracellular signalling cascades; Areal et al. 2011). Many of these primers will likely work well in related species. These primers will be used to survey TLR diversity in threatened populations, and for addressing more general questions regarding avian TLR evolution and the role of immunogenetic diversity after population bottlenecks.

Table 1 Sequences of PCR primers developed in this study
Full size table
Table 2 Primers and annealing temperatures (Ta) used to amplify partial Toll-like receptor sequences in nine threatened New Zealand bird species (lengths shown are total amplicon length, including primers, and are approximate)
Full size table