Isolation and Characterisation of Microsatellites from Pinus Radiata

Abstract

Pinus radiata is one of the world's most commercially important conifers with significant plantations in New Zealand, Chile and Australia. An intensive breeding and improvement programme is underway aimed at developing trees with improved growth rate and form. The goal of this thesis research was to isolate microsatellites from P. radiata genomic DNA and analyse their utility as genetic markers for the breeding programme. Twenty-three dinucleotide microsatellites were initially isolated from an unenriched genomic library. Their frequency in the library confirmed an earlier report that microsatellites were scarce in P. radiata with one (CT)n repeat per 0.7 Mb and one (CA)n repeat per 2.2Mb isolated (where n > 8). None of these microsatellites (which had up to 18 dinucleotide repeats) was polymorphic in a 19-tree population. A novel strategy was developed using a PCR-based protocol to enrich three genomic libraries for either (CT)n or (CA)n repeats. The protocol involved PCR of genomic DNA with primer pools that contained microsatellite motifs at their 3' ends and degenerate "blocking anchors" at their 5' ends. The blocking anchors were designed to "anchor" the primers at the 5' ends of microsatellites during PCR. Control experiments were successfully used to show the range of PCR stringencies under which this anchoring would occur. Cloning of anchored genomic PCR products yielded libraries with as much as a 2,000-fold enrichment for (CT)n repeats and a 1,250-fold enrichment for (CA)n repeats (where n > 12). These microsatellites were found both at the terminus of the cloned fragment, where a genomic insert was adjacent to bacterial plasmid DNA (32 SSRs), or internal to the terminus (with locus-specific DNA flanking both sides of the repeat, l6 SSRs). Locus-specific primers were designed for use in PCR of 32 microsatellite-containing loci (two specific primers were required for internal microsatellites, whereas one locus-specific primer was required for terminal microsatellites). Eleven of these loci were polymorphic and segregated according to Mendelian rules in megagametophytes and/or progeny of parental trees. Eight of the eleven polymorphic loci were shown to be present in the genome in multiple copies and three loci were present in single copy. Both single- and multi-copy microsatellites were highly informative in fingerprinting and paternity analyses. Three single- and five multi-copy microsatellites were placed on a framework P. radiata genetic map which was being used for detection of quantitative trait loci (QTL). One of these five multi-copy markers mapped to two loci, resulting in a total of nine mapped loci from the eight markers. The other four multi-copy microsatellites mapped to one locus each, indicating that all of the repeat copies were tightly linked. Seven primer pairs were tested against other Pinus species. Amplification products were observed in other hard pine species (subgenus Pinus) for all seven primer pairs but PCR products were observed for only three of the primer pairs when soft pine (subgenus Strobus) genomic DNA was used. Although fewer of the isolated microsatellites were internal than terminal (see above), six of the eight mappable markers were derived from internal repeats. This suggested that markers assayed with two locus-specific primers were more efficiently mapped than those containing one locus-specific primer and one of the anchoring primers. The results showed that by enriching libraries for microsatellites, I was able to generate markers which were useful for genetic analyses and which will be important to the P. radiata tree breeding programme.