Traditionally, simple sequence repeat (SSR)markers have been developed from libraries of genomic DNA. However, the large, repetitive nature of conifer genomes makes development of robust, single-copy SSR markers from genomic DNA difficult. Expressed sequence tags (ESTs), or sequences of messenger RNA, offer the opportunity to exploit single, low-copy, conserved sequence motifs for SSR development. From a 20,275-unigene spruce EST set, we identified 44 candidate EST-SSR markers. Of these, 25 amplified and were polymorphic in white, Sitka, and black spruce; 20 amplified in all 3 spruce species tested; the remaining five amplified in all except one species. In addition, 101 previously described spruce SSRs (mostly developed from genomic DNA), were tested. Of these, 17 amplified across white,Sitka, and black spruce. The 25 EST-SSRs had approximately 9% less heterozygosity than the 17 genomic-derived SSRs (mean H=0.65 vs 0.72), but appeared to have less null alleles, as evidenced by much lower apparent inbreeding (mean F=0.046 vs 0.126). These robust SSRs are of particular use in comparative studies,and as the EST-SSRs are within the expressed portion of the genome, they are more likely to be associated with a particular gene of interest, improving their utility for quantitative trait loci mapping and allowing detection of selective sweeps at specific genes.

In the first large study of conifer expressed sequence tag-simple sequence repeats (EST-SSRs), two large conifer EST databases were characterized for ESTSSRs. One database was from "interior spruce" (white and Engelmann spruce in Southern British Columbia) and Sitka spruce, while the other was from loblolly pine. We found 475 and 629 unique EST-SSRs in loblolly pine and spruce, respectively. 3′ ESTs contained 14% more SSRs than 5′ EST reads in loblolly pine and 41% more in spruce. Conifer EST-SSRs differed conspicuously from angiosperm ESTSSRs in several aspects. EST-SSRs were considerably less frequent in conifers (one EST-SSR every ∼50 kb) than in angiosperms (one EST-SSR every ∼20 kb). Dinucleotide repeats were the most abundant repeat class in conifers, while in angiosperms, trinucleotides were most common. Finally, the AT motif was the dominant motif recovered in both conifer species, whereas AG was the most common dinucleotide repeat in angiosperms. Also, as these ESTSSRs in conifers could be developed into useful genetic markers, our work demonstrates the value of large-scale EST sequencing projects for in-silico approaches for marker development.

Chamaecyparis nootkatensis is an ecologically and economically important conifer of the north Pacific coastal forests. To aid in studies of clonal structure and genetic differentiation of this and related species, we isolated and characterized microsatellites from C. nootkatensis. A microsatellite-enriched library yielded 75 repeat-containing sequences for which primer pairs were designed. Only five showed reliable amplification and polymorphism, with an average of 13.7 alleles/locus and a mean expected heterozygosity of 0.592. In progeny tests with four families, few null alleles were directly detected and loci segregated according to Mendelian expectations. However, in one primer pair, high heterozygote deficiency was observed, suggesting the presence of a null allele. The ability of primer pairs to cross amplify was tested on 18 species of the Cupressaceae sensu lato; three primer pairs yielded polymorphic loci in Cupressus and Juniperus species, but not in other Chamaecyparis species. This also supports recent findings of a closer affinity of C. nootkatensis with Cupressus over other Chamaecyparis species.

No abstract