Genetic Variation and Geographical Differentiation Revealed Using ISSR Markers in Tung Tree, Vernicia Fordii

c Indian Academy of Sciences

ONLINE RESOURCES

Genetic variation and geographical differentiation revealed using ISSR markers in tung tree, Vernicia fordii

LINGLING ZHANG1,2, SHIYOU LU1, DONGFA SUN3 and JUNHUA PENG4∗

1Key Laboratory of Plant Germplasm Enhancement and Specialty Agriculture, and Wuhan Botanical Garden, Chinese Academy of Sciences, Wuhan, Hubei 430074, People’s Republic of China 2Graduate University of Chinese Academy of Sciences, 19 Yuquan Road, Beijing 100049, People’s Republic of China 3College of Plant Science and Technology, Huazhong Agricultural University, Wuhan, Hubei 430071, People’s Republic of China 4Science and Technology Center, China Seed Group, Wuhan, Hubei 430075, People’s Republic of China

[Zhang L., Lu S., Sun D., and Peng J. 2015 Genetic variation and geographical differentiation revealed using ISSR markers in tung tree, Vernicia fordii. J. Genet. 94, e5–e9. Online only: http://www.ias.ac.in/jgenet/OnlineResources/94/e5.pdf]

Introduction Materials and methods

Tung tree, Vernicia fordii is an oil-bearing woody plant Plant materials species of Euphorbiaceae, native to China and cultivated for A set of 87 accessions of tung tree germplasm was col- over one thousand years (Potter 1959). The oil extracted from lected from the main distribution areas of China, including tung tree fruit, called as China wood oil or tung oil, is a 40 from Enshi prefecture, Hubei province and 47 from superior drying oil used in production of paints, varnishes Suining prefecture, Sichuan province (figure 1). These 87 and polymers. The eleostearic acid is the main components accessions were classified into two geographical populations, of tung oil, and contains three conjugated unsaturated dou- Enshi and Suining. Enshi population was further divided into ble bonds. This molecular structure endows tung oil with 13 subgroups and the Suining population into four subgroups active chemical property and thus distinctive from other according to the niches of collection sites (table 1). Genomic plant oils resulting in a wide-application prospect in many DNA was isolated using a modified recycling CTAB method fields, such as electronics, aerospace and advanced ink (Zhang et al. 2013). (Bhuyan et al. 2010). Recently, tung oil was discovered to be an excellent feedstock for biodiesel production after blend- ing with medium-chain fatty acid oils or 0# biodiesel (Shang ISSR genotyping et al. 2010;Chenet al. 2012). Evaluation of germplasm collection and genetic variation are essential for breeding A set of 37 ISSR primers was screened from the Univer- tung tree cultivars with high oil yield and quality. However, sity of British Columbia (UBC) primers to genotype the 87 there are a few reports on the genetic evaluation of tung tree accessions (table 2). The PCR reactions were performed in a germplasm. Li et al. (2008, 2009) investigated genetic diver- 25 μL reaction volume (20 ng template DNA, 0.6 μM ISSR sity of 64 tung tree cultivars from six provinces in China, primers, 1 U Taq DNA polymerase, 0.2 mM dNTP mixture, × 2+ and demonstrated some level of association between inter- 1 PCR buffer, 0.15 mM Mg ,1%PVP,0.1%BSA)using simple sequence repeat (ISSR) markers and the eleostearic Biorad-My Cycle Thermocyclers (Bio-Rad Laboratories, content in tung tree. In the present study, we analysed genetic Inc., CA, USA) with the following programme: predenatura- tion at 94◦C for 4 min; 45 cycles of denaturation at 94◦Cfor variation and differentiation between two geographical ◦ populations using ISSR markers. 30 s, annealing at 40–65 C (depending on the primers) for 45 s, extension at 72◦C for 45 s; an extension step at 72◦Cfor7 min and a final step of hold at 15◦C for 15 min. PCR prod- ucts were separated on 2% agarose gels and visualized with ∗ For correspondence. E-mail: [email protected]. ethidium bromide staining. Keywords. tung tree; genetic variation; biodiesel; Vernicia fordii.

Journal of Genetics Vol. 94, Online Resources e5 Lingling Zhang et al.

(a) (b)

(c)

Figure 1. Geographical distribution of tung tree accessions examined in the present study. (a) Sampling sites of 13 subgroups in Suining population. (b) Sampling sites of Enshi and Suining populations. (c) Sampling sites of four subgroups in Enshi population. The Arabic number refers to the code of subgroup.

ISSR data analysis components for ISSR-based genetic variation and to test differentiation between and within the two geographical The amplified DNA fragments (bands) were treated as domi- populations (Excoffier et al. 1992; Peakall and Smouse nant markers and were scored in terms of a binary code 2006). (1/0). The genetic indices such as the percentage of polymorphic loci (PR), actual number of alleles (Na), effective number of alleles (Ne), Shannon’s information index (I), Nei’s gene diversity (He) (Nei 1973) and genetic distance Results and discussion (D) were calculated using PopGene 1.32 software (Yeh and Genetic variation Yang 2000). Based on the Nei’s genetic distance, the dendrogram was drawn using unweighted pair-group method Figure 2 shows PCR profile of the ISSR primer UBC810 in with arithmetic average (UPGMA). The analysis of mole- all the 87 tung tree accessions. There were 212 reliable frag- cular variance (AMOVA) was used to examine variance ments observed in our study, of which 99 (46.48%) were

Table 1. Tung tree accessions and their places of origin.

Acc. codea Collection siteb North latitude (◦N) East longitude (◦E)

1-1...1-16 Jigongwei, ES 29.26 109.26 2-17...2-20 Baifusi, ES 29.21 109.22 3-21...3-27 Ranjiacui, ES 29.16 109.26 4-28...4-40 Niedongcui, ES 29.14 109.27 5-41...5-46 Daying Temple, SN 30.57 105.24 6-47...6-49 Gaoyakou, SN 30.57 105.25 7-50/51/87 Longtouqiao, SN 30.55 105.26 8-52 Xihua, SN 30.77 106.00 9-53...9-54 Yufeng, SN 30.62 105.18 10-55...10-56 Huanglian, SN 30.71 105.19 11-57...11-59 Jianzhong, SN 30.71 105.16 12-60...12-64 Xifeng, SN 30.79 105.12 13-65...13-70 Xingwu, SN 30.71 105.25 14-71...14-75 Longmen, SN 30.72 105.28 15-76...15-77 Xiangyang, SN 30.74 105.34 16-78...16-82 Wenfeng, SN 30.73 105.37 17-83...17-86 Lingquan Temple, SN 30.54 105.62

aThe number after hyphen is the code of accession and the number before hyphen is the code of a subgroup. bES, Enshi in Hubei province; SN, Suining in Sichuan province.

Journal of Genetics Vol. 94, Online Resources e6 Genetic diversity of tung tree germplasm

Table 2. ISSR primer selected to genotype the tung tree germplasm.

◦ ◦ Primer code Primer sequence Ta ( C) Primer code Primer sequence Ta ( C)

812 (GA)8A 43 808 (AG)8C50 814 (CT)8A 46 810 (GA)8T50 818 (CA)8G 46 822 (TC)8A46 826 (AC)8C 50 891 HVH(TG)7 50 835 (AG)8YC 50 890 VHV(GT)7 46 842 (GA)8YG 50 850 (GT)8YC 50 868 (GAA)6 53 857 (AC)8YG 43 874 (CCCT)4 46 841 (GA)8YC 50 878 (GGAT)4 46 879 (CTTCA)3 40 881 (GGGTG)3 50 807 (AG)8T47 889 DBD(AC)7 50 816 (CA)8T50 809 (AG)8G 53 827 (AC)8G50 815 (CT)8G 43 834 (AG)8YT 53 887 DVD(TC)7 50 845 (CT)8RG 50 811 (GA)8C 41 855 (AC)8YT 53 840 (GA)8YT 50 852 (TC)8RA 53 846 (CA)8RT 50 854 (TC)8RG 55 880 (GGAGA)3 53 844 (CT)8RC 55 856 (AC)8YA 55

Ta, annealing temperature.

Figure 2. PCR proﬁle of ISSR primer UBC810 in 30 accessions of tung tree. M lane is the DNA size ladder with 1000 bp for the top 1st enhanced bright band and 500 bp for the 2nd enhanced bright band in the lane. polymorphic. The observed number of alleles per locus (Na) was 1.465 for the whole population. The effective number Table 3. Population genetic parameters of the tung tree germplasm. of alleles (N ) ranged from 1.0 to 1.997 with an average e Genetic Enshi Suining Whole of 1.216 for the 212 loci detected in the panel of 87 tung parameter population population population tree accessions. Shannon diversity index (I) was estimated as 0.203 in the whole population. Nei’s gene diversity (He) was PS 40 47 87 estimated as 0.132 in the whole panel of collected germplasm NPL859299 of tung tree (table 3). The set of tung tree germplasm thus PR (%) 39.91% 43.19% 46.70% had a medium genetic diversity. Na 1.399 1.432 1.465 Two Enshi accessions, 1–4 and 1–13, have the short- Ne 1.184 1.224 1.216 est genetic distance (GD) (0.058). The longest GD (0.243) He 0.116 0.133 0.132 occurred between the two accessions, 1–12 from Enshi, I 0.180 0.203 0.203 Hubei province and 11–58 from Suining, Sichuan province. PS, population size; NPL, number of polymorphic loci; PR, poly- Thus a cross between accessions 1–12 and 11–58 can be morphism rate; Na, observed number of alleles; Ne, effective num- made to develop a diverse mapping population for multiple ber of alleles; I, Shannon’s information index; He, average Nei’s purposes. gene diversity.

Journal of Genetics Vol. 94, Online Resources e7 Lingling Zhang et al.

In this study, all the tung tree accessions were collected from the wild. Our results showed a medium level of genetic diversity in this batch of tung tree germplasm (table 3). The 46.48% of ISSR polymorphism rate is obviously lower than that reported by Li et al.(2008, 2009). Li et al. (2008)mainly analysed genetic diversity of 64 tung tree cultivars from six provinces in China. Further, Li et al. (2009) analysed association of ISSR markers with the eleostearic content in tung tree. Tung tree collections in present study covered two distant and relatively small areas in the whole distribution region of this species (ﬁgure 1). The relatively low genetic diversity (table 3) may be attributed to the relatively small sampling areas.

Geographical differentiation As shown in table 3, polymorphism rate in Suining population (43.19%) was higher than that in Enshi population (39.91%), and both lower than 46.70% of the whole population. The other genetic parameters showed the similar trend as polymorphism rate (table 3). Therefore, genetic diversity of Suining population is higher than that of Enshi population. To determine the interpopulation genetic variation, ISSR data of the tung tree population was subjected to AMOVA analysis (Excoffier et al. 1992; Peakall and Smouse 2006). The within-population and interpopulation variation accounted for 92.85 and 7.15% of the total variation, respectively (table 4). Genetic variation of the tung tree germplasm is mainly attributed to the within-population variance. Genetic differentiation between the two geographical populations (Fst = 0.071) was highly significant (P < 0.01) as indicated by permutation test (table 4). Therefore, the significant geographic differentiation has truly occurred in the tung tree germplasm. AMOVA analysis showed that there was relatively small but highly significant genetic variation between the Enshi and Suining tung tree populations. Tung tree had been cultivated in China for over 1000 years. In history, many cultivars were developed from the local tung tree germplasm and cultivated mainly in the same region (Pan et al. 2013). Therefore, many landraces have been generated during the long-term cultivation and breeding (Fang and He 1998). In this study, the spatial isolation and historically endemic breeding may be the reasons for the significant genetic differentiation of tung tree between the Enshi and Suining populations. However, driven by high

Table 4. Analysis of molecular variance (AMOVA) based on ISSR marker data.

Source Degree Sum of Variance % of of variation of freedom squares components variation

Between population 1 59.031 1.05016 7.15 Within population 85 1159.751 13.64413 92.85 Figure 3. Dendrogram of 87 tung tree accessions based on Nei’s Total 86 1218.782 14.69429 original genetic distance. The marker groups are described in the section of results and discussion. The codes are consistent to those Fst = 0.07147, P<0.01 shown in table 1.

Journal of Genetics Vol. 94, Online Resources e8 Genetic diversity of tung tree germplasm economic value of tung oil, many introduction events of tung Acknowledgements tree cultivars occurred in history of China (Long 1996;Fang and He 1998;Tan2006). For instance, Laifeng in Enshi is This work was supported by Hubei Academy of Forestry (grant no. 201304707-3) and the National Natural Science Foundation of the home of Jinsi tung oil, a well-known tung oil brand in China (NSFC) (grant nos 31030055 and 30870233) and Chinese China, where many endemic varieties are introduced and cul- Academy of Sciences under the Important Directional Program of tivated (Tian et al. 2008). Therefore, cultivar introduction Knowledge Innovation Project (grant no. KSCX2-YW-Z-0722). could have accelerated gene flow of tung tree and reduced genetic differentiation among different tung tree populations, References e.g., the Enshi and Suining populations. Bhuyan S., Sundararajan S., Andjelkovic D. and Larock R. 2010 Phylogenetic analysis Effect of crosslinking on tribological behavior of tung oil-based polymers. Tribol. Int. 43, 831–837. As shown in figure 3, the 87 tung tree accessions could Chen Y. H., Chen J. H. and Luo Y. M. 2012 Complementary be divided into 10 distinct groups. Group V (26 samples) biodiesel combination from tung and medium-chain fatty acid oils. Renew. Energ. 44, 305–310. and group X (31 samples) included most of the observed Excoffier L., Smouse P. E. and Quattro J. M. 1992 Analysis of germplasm. The other 30 accessions were divided into molecular variance inferred from metric distance among DNA eight small groups. Group V contained one individual from restriction data. Genetics 131, 479–491. Enshi population and most of the accessions from five Fang J. X. and He F. 1998 Tung tree in China. China Forestry neighbouring subgroups of Suining population (thirteenth to Publishing House, Beijing, China (in Chinese). Li P., Zhang X. P., Chen Y. C., Liu G. Q., Zhou G. and Wang sixteenth subgroup and tenth subgroup). Group X contained Y. D. 2008 Genetic diversity and germplasm resource research on one individual from Suining population and 30 accessions tung tree (Vernicia fordii) cultivars, investigated by inter-simple from Enshi population. Of the rest, 10 individual in Enshi sequence repeats. Afr. J. Biotechnol. 7, 1054–1059. population, six were clustered into group III, and the other Li P., Wang Y. D., Chen Y. C. and Zhang S. S. 2009 Genetic diver- four were clustered into four groups with some accessions sity and association of ISSR markers with the eleostearic content in tung tree (Vernicia fordii). Afr. J. Biotechnol. 8, 4782–4788. from Suining population. Long Z. X. 1996 Genetic analysis on the traits related to overwin- In general, clustering analysis showed that Enshi popu- tering of tung tree seedling introduce to the north. J. Northwest lation was basically distinguished from Suining population. Forestry College 11, 31–35 (in Chinese). But some of the tung tree accessions had obvious paradox Nei M. 1973 Analysis of gene diversity in subdivided populations. between geographical and genetic distances. For instance Proc. Natl. Acad. Sci. USA 70, 3321–3323. Pan Y., Pan L., Chen L., Zhang L. L., Nevo E. and Peng J. H. sample 10–56 was from Suining, but was clustered into 2013 Development of microsatellite markers in the oil-producing group X including most of the accessions from Enshi. Acces- species vernicia fordii (Euphorbiaceae), a potential biodiesel sion 1–4 and 1–16 collected from the same location were feedstock. Appl. Plant Sci. 1, 1–4. genetically distant. Also, it is worth noting that six acces- Peakall R. and Smouse P. E. 2006 GENALEX 6, genetic analysis sions in the 5th subgroup, collected from the site of an in Excel. Population genetic software for teaching and research. Mol. Ecol. Notes 6, 288–295. old Buddhism temple, were actually clustered into two dis- Potter G. F. 1959 The domestic tung industry. I. Production and tant clades, group I and group VII. The paradox may be improvement of the tung tree. Econ. Bot. 13, 328–342. related to tung tree seed dispersal by Buddhism believers Shang Q., Jiang W., Lu H. and Liang B. 2010 Properties of tung oil who used tung oil for worshiping. Human activities in the biodiesel and its blends with 0# diesel. Bioresour. Technol. 101, history inevitably have resulted in a mixture of native and 826–828. Tan X. F. 2006 Status and suggestion on development of Vernicia some exotic tung tree accessions, leading to some incon- fordii. Nonwood Forest Res. 24, 62–64 (in Chinese). sistence between genetic and geographic distances. How- Tian G. Z., Sun D. F., Liu J. L., Yin Z. L., Tao X. L., Zhou G. L. ever, it was often difficult to figure out the historical events and Liu H. Z. 2008 Study on resources, phenotypes and locality that should be responsible for the above mentioned paradox factors of tung tree in Laifeng county. Hubei Agric. Sci. 47, 71– due to lack of relevant literature records. To further under- 74 (in Chinese). Yeh F. C. and Yang R. C. 2000 POPGENE version 1.32. Univer- stand the genetic structure of tung tree germplasm and to sity of Albert and Center for International Research. Available at: draw a more solid and clear conclusion, a more comprehen- http://www.ualberta.ca/~fyeh/popgene.html. sive study involving a nationwide germplasm collection is Zhang L. L., Dai L. J., Gou J. B. and Peng J. H. 2013 An effective underway. protocol to solve the problem in genomic DNA isolation of tung tree. J. Plant Biochem. Biotechnol. 22, 492–497.

Received 20 August 2014, accepted 18 September 2014 Unedited version published online: 29 September 2014 Final version published online: 20 February 2015

Journal of Genetics Vol. 94, Online Resources e9