Low levels of genetic variation within populations of the four rare orchids Gymnadenia cucullata, Gymnadenia camtschatica, Amitostigma gracile, and Pogonia minor in South Korea: indication of genetic drift and implications for conservation

Abstract

Many terrestrial orchids are relatively rare, and their populations are small and spatially isolated. Population genetics theory predicts that populations of such species, affected historically by random genetic drift, would maintain low levels of genetic diversity and exhibit a high degree of among-population divergence. To test this prediction, I used enzyme electrophoresis. Genetic diversity within populations of the four rare, terrestrial orchids Gymnadenia cucullata (four populations) and its congener G. camtschatica (four populations), Amitostigma gracile (four populations in one region and three in another region), and Pogonia minor (three populations each in two regions) was investigated in South Korea at the landscape level. As predicted, populations of the four species harbor low levels of genetic diversity within populations: the mean percentage of polymorphic loci, %P, the mean number of alleles per locus, A, and the average expected heterozygosity, H e, were 12.5%, 1.13, and 0.036 for G. cucullata, respectively; 18.2%, 1.18, and 0.067 for G. camtschatica; 3.0%, 1.04, and 0.009 for A. gracile; and 2.7%, 1.06, and 0.014 for P. minor. Except for G. camtschatica (F ST = 0.000), a significantly high degree of genetic divergence between conspecific populations was detected in the other three species: F ST = 0.081 for G. cucullata; 0.348 and 0.811 in two regions for A. gracile; and 0.469 and 0.758 in two regions for P. minor. In addition, individuals within populations are highly structured in the four species (overall F IS = 0.276 for G. cucullata; 0.308 for G. camtschatica; 0.758 for A. gracile; and 0.469 for P. minor), suggesting that selfing, biparental inbreeding, and/or consanguineous mating have occurred in populations of the four species. With the exception of G. camtschatica, an allele at a locus is fixed in a population, whereas alternative alleles with low or high frequencies are detected in another population across the landscape. My results suggest that evolutionary histories of G. cucullata, A. gracile, and P. minor are different from G. camtschatica. Historical genetic drift would be an important force shaping the genetic structure of the Korean populations of G. cucullata, A. gracile, and P. minor. For G. camtschatica on Ulleung Island, relatively higher levels of genetic variation within populations compared to its congener G. cucullata (H e = 0.067 vs. 0.036) and little evidence of population genetic structure among populations (F ST = 0.000) suggest that individuals were, presumably, once continuously distributed on Ulleung Island, and populations have recently been isolated by habitat fragmentation through natural succession (e.g,. probably the encroachment of woody vegetation on grasslands) or human-mediated disturbances (e.g., collections). Thus, conservation strategies for the four species should be differently developed in order to preserve genetic diversity in South Korea.