Conservation genetics of Kincaid's lupine: A threatened plant of western Oregon and southwest Washington grasslands.

摘要

Kincaid's lupine (Lupinus oreganus Heller) is a federally listed threatened species native to remnant grassland of western Oregon and southwestern Washington, and is the primary larval host plant of a once thought extinct butterfly, Plebejus icarioides fenderi Macy. Past studies concerning Kincaid's lupine reproduction suggested that populations may suffer reductions in fitness and progeny vigor due to inbreeding depression, but no direct investigation into range-wide patterns of genetic variation has been undertaken. I used nuclear DNA and chloroplast DNA simple sequence repeat (SSR) markers to determine genet size and patterns of non-adventitious rhizomatous lupine spread, to estimate the number of genets within Kincaid's lupine populations, and to assess whether seed transfer for the purpose of genetic rescue is an appropriate genetics management strategy for Kincaid's lupine.;Patterns of allelic diversity at nDNA SSR loci within study patches revealed that non-adventitious spread of rhizomes can extend to at least 27 m and may dominate a portion of a lupine patch or small population. However, genet spread and arrangement in study patches were sufficiently integrated such that interplantlet Bombus foraging flights exceeding 2 m had > 90% probability of occurring between different genets. Within-lupine patch genetic diversity was well-undersampled, refuting the supposition that Kincaid's lupine populations suffer from inbreeding depression due to small effective population sizes. Estimation of Kincaid's lupine abundance through leaf cover and inflorescence number was tightly correlated with plantlet number, a unit of vegetative and sexual growth, within lupine patches but the relationship was not consistent between patches within populations or between populations. We used genet to plantlet ratios (determined through genotyping) and plantlet density to estimate genet population size in Kincaid's lupine patches. Because of the strong correlation between cover and plantlet density, historically collected lupine abundance data could be used to estimate genet population size provided that plantlet density is calibrated to patch-specific cover measurements.;Within patches and populations across the range of Kincaid's lupine there was little DNA evidence suggesting severe inbreeding. Only one of 24 populations and five study patches had strong statistical evidence of a recent genetic bottleneck despite the range-wide fragmentation of lupine populations and habitat. Mean population fixation index values for nearly half of the populations were near Hardy-Weinberg equilibrium expectations and only one small lupine population had a F-value > 0.20, suggestive of high inbreeding levels. Half of the populations actually had an excess of heterozygotes, suggesting that genetic diversity is not being lost. Chloroplast DNA coincides well with the observation that genetic diversity is not being lost through inbreeding or genetic bottlenecking in Kincaid's lupine. The mean number of cpDNA haplotypes per population was approximately 4 maternal lineages, which is very high for an animal pollinated plant with heavy seeds that have limited dispersal. Even relatively small populations of Kincaid's lupine had 2 or more cpDNA haplotypes, indicating that populations are not severely inbred. Both nuclear and chloroplast DNA SSR genetic marker diversity suggests that Kincaid's lupine does not require genetic rescue for effective conservation. Due to the longevity of Kincaid's lupine and the apparently large amount of within population genetic diversity, the encouragement of natural recruitment from vegetation management that improves habitat conditions is likely to maintain the relatively large amount of genetic diversity within Kincaid's lupine populations.