Characterization of nuclear microsatellites in New Caledonian Araucaria species

The 13 species of Araucaria on New Caledonia represent a closely related monophyletic assemblage of endemic species (Setoguchi et al. 1998; M. Kranitz and A. Ponge unpublished data). This radiation of a set of wind-pollinated species in a small geographical area raises the question as to the landscape conditions under which differentiation and diversification occurred. To tackle this issue, microsatellite loci from two Araucaria species (A. rulei and A. subulata) have been isolated to investigate population genetic structure within this group.

Microsatellites were isolated using an enrichment procedure based on Edwards et al. (1995) and Squirrell & Wolff (2001) as described by Hughes et al. (2002). Briefly, restricted, ligated and amplified DNA was hybridized to Hybond® N + membranes onto which (GA)₁₃ and (CA)₁₃ oligonucleotides had been fixed. Following membrane enrichment, the DNA was cloned using the PCR-Script™ AMP cloning kit (Stratagene). Plasmid DNA was extracted using Qiagen minipreps and insert regions were sequenced using M13 primers with the Thermosequenase II dye terminator cycle sequencing kit (Amersham) and an ABI 377 sequencer.

Seventy colonies in total were produced from the A. rulei library of which only two contained a microsatellite sequence. In contrast, 81% of the colonies contained microsatellites in the A. subulata library (126/155 sequenced inserts had a microsatellite sequence of 7 or more repeat units). The most common microsatellite motif was GA with 42 sequences, 38 of which were uninterrupted; 41 clones contained CA sequences, of which 22 were uninterrupted; 38 compound microsatellites were recorded, of which 31 were uninterrupted. The A. subulata enriched library had a sequence redundancy rate of 17% leaving 104 unique loci.

Of the 106 unique A. subulata and A. rulei microsatellite sequences, 50 were suitable for primer design which was performed using primer 3 (Rozen & Skaletsky 2000). The resulting 50 primer pairs were screened using fluorescently labelled dCTPs on six A. subulata individuals from a single population. Of the 50 primer pairs tested, 30 amplified a single or double band of a similar size to that of the original sequenced clone. The remaining primer pairs produced either multiple bands or showed intermittent amplification. Of the 30 potentially useful microsatellite loci, 10 were polymorphic. Five of these loci which showed the clearest gel-profiles were selected for further screening and PCR optimization.

These primers were end-labelled with either JOE or FAM dyes and analysed with the PCR conditions described below and in Table 1. The 20 µL PCR mix contained one unit of Amplitaq Gold DNA polymerase (Applied Biosytems), 1 × Amplitaq Gold reaction buffer [15 mm Tris-HCl (pH 8), 50 mm KCl] (Applied Biosytems), 1.5 mm MgCl₂ (Applied Biosytems), 0.2 mm of each dNTP (Bioline), 0.2 µm of each primer (MWG Biotech) and 20–50 ng of genomic DNA. PCR amplifications were carried out using a GeneAmp 9600 thermocycler (Perkin Elmer) using the following program: 1 step at 95 °C for 5 min, followed by 35 cycles of 30 s at 95 °C, 30 s at T_a °C, 1 min at 72 °C, and a final 25 min extension of 72 °C.

The applicability of these primers for large scale population genetic studies on New Caledonian Araucaria species was tested using two approaches: In the first case, amplification in 1–4 individuals of eight species was tested: A. rulei; A. muelleri; A. luxurians; A. subulata; A. nemorosa; A. scopulorum; A. laubenfelsii; and A. bernieri. Using the PCR conditions described above, in all cases one or two banded PCR products of the expected size were produced. A more detailed approach was taken involving 332 individuals of A. columnaris from 12 populations, again using the same PCR conditions and running the samples on an ABI 377. The five loci gave between seven and 36 alleles per locus (Table 1). For all of the loci except for AS167, there was a small but significant deficit of heterozygotes. This global deficit of heterozygotes was paralleled by significant multilocus F_IS estimates at the population level from 11/12 populations of A. columnaris, with significant single locus F_IS estimates for Aru1 in eight populations, for AS152 in six populations, for AS190 in seven populations, for AS110 in five populations, and for AS167 in 1 population. This deficit of heterozygotes may be due to null alleles, although given its recurrence across loci and populations in this monoecious species, inbreeding may also be responsible. Tests for linkage disequilibrium revealed no significant associations among loci.