Researches On Core Collection Construction Of Ginkgo Biloba

Ginkgo biloba L, endemic to China and also called "Gongsun Tree", is one of the oldest remained tree species in gymnosperm. Known as "living fossil", Ginkgo biloba is the single member of its family with no close relatives, and therefore its germplasm resources is extremely precious. There are ginkgo populations of different sizes distributed in some places in China, which have rich and valuable genetic resources.In order to effectively protect and use these resources, it is a important work to construct core collection of ginkgo's germplasm. Using the 205 samples collected from 14 sites in Hubei Province as primary materials, this paper made some prelimilary researches on core collection construction strategy of ginkgo's germplasm to provide basis for later's large scale core collection construction. Main results were as following:(1)ISSR-PCR was conducted to genomic DNA extracted from ginkgo seed's endosperm by using 9 ISSR primers. A total of 158 bands were obtained, in which 144 bands were polymorphic, and the percentage of polymorphic bands (PPB) was 91.12%. Based upon these results, the molecular data of the 205 ginkgo samples were gained.(2) The scheme and stategy for core collection construction of ginkgo's germplasm was researched firstly.45 candidate core collections were at first constructed by applying different grouping principles,5 kinds of overall sampling ratio,4 kinds of group sampling strategies and 2 kinds of grouping sampling method; then tests on these candidate core collections were conducted by using the following indexes:number of polymorphic loci (NPL), percentage of polymorphic bands (PPB), observed number of alleles per locus (AO), effective number of alleles per locus (AE), Nei's gene diversity (HE) and Shannon's diversity information index (HI).Finally, the best construction strategy was screened out, i.e.,firstly classifying samples according to areas, secondly applying an overall sampling ratio of 25%,thirdly deciding sample size within groups applying logarithmic strategy, and lastly screening samples by using UPGMA clustering sampling method. Based upon this strategy, a core collection was constructed with 52 samples screened out from the 205 primary samples.(3)Comparative analysis on genetic diversity between core collection and original collection as well as reserve collection were made by applying POPGENE software. A t-test between core collection and original collecito n was made, too. The results showed that the core collection reserved 25.37% samples of original colleciton, and the saving ratio of the number of polymorphic loci (NPL) and percentage of polymorphic bands (PPB) was 98.09% which indicated that the core collection was much close to original colleciton; the saving ratio of observed number of alleles per locus(Ao), effective number of alleles per locus(AE),Nei's gene diversity (HE) and Shannon information index (HI) were 99.05%,100.98%,101.69% and 101.26% respectively, indicating that the behind three indexes of the core collection were slightly higher than those of the original collection. The t-test also proved there was no obvious difference in genetic diversity indexes between core collection and original collection.In addition, comparision of above genetic diversity indexes between core collection and reserve collection indicated that core collection's genetic diversity indexes were all obviously higher than the reserve collection's.All these results showed that the core collection could embody the genetic diversity of the original collection and has high representativeness.(4) The AMOVA (Analysis of Molecular Variance) showed that the similarity of total gene genetic diversity between core collection and original collection was 99.48%. Core collection's coefficient of population differentiation and percentage of genetic variation among populations were both slightly higher than those of the original collection's, indicating that more genetic variation of core collection existed among population. This was because samples with relatively big genetic differences were choosed as preferable as possible to construct core collection in order to represent the original collection with the least number of samples.The similarity of estimate gene flow between the core collection and the original collection was 83.21%, and similarity of mean gene diversity within populations of the two collections was 54.31%.These showed that core collection could not totally represent the original collection's genetic diversity, no matter how high the saving ratio is, the core collection could not substitute the original collection, and it could not represent all the relationship of evolution and genetic variation within populations.Therefore, the reserve collection should not be abandoned, and on the contrary it is obviously necessary to establish a dynamic mechanism of information changing between core collection and reserve collection.