Nucleotide variation at the methionine synthase locus in an endangered tree species , Fokienia hodginsii ( Cupressaceae ) in Vietnam

1 University of Engineering and Technology, Vietnam National University Hanoi, 144-Xuan Thuy Road, Cau Giay Street, Hanoi, Vietnam. 2 School of Medicine and Pharmacy, Vietnam National University Hanoi, 144-Xuan Thuy Road, Cau Giay Street, Hanoi, Vietnam. 3 Faculty of Biology, Hanoi National University of Education, Vietnam. 4 Institute of Biotechnology and Microbiology, Vietnam National University Hanoi, 144-Xuan Thuy Road, Cau Giay Street, Hanoi, Vietnam.


INTRODUCTION
Fokienia hodginsii is a monotypic, ancient, 'living fossil' and a member of the Cupressaceae family.It is endemic to Laos, Vietnam and southern China where it is widespread and is currently listed as globally nearthreatened (IUCN, 2004).F. hodginsii is highly valued both economically and culturally.F. hodginsii is widely  Table 1.extinction.F. hodginsii was listed in Vietnam Red Book in 1996 and the Government decided to close the F. hodginsii forests.
Population genetics plays an important role in conservation biology and ecology in general.The assessment of genetic diversity in animal and plant populations, especially of endangered species is now pervasive.Good methods for DNA analyses are being increasingly used to estimate the extent and organization of genetic diversity within populations, to infer the causes of its spatiotemporal dynamics and to suggest strategies for conservation and the wise use of genetic resources.In this sense, in a period of dramatic human exploitation and consumption of natural biological resources and concomitant development of biotechnologies, the Quang et al. 15399 emerging field of conservation genetics can help to guide the necessary harmony between economic development and nature preservation.Conservation genetics hence provides important tools for the assessment of biodiversity according to the biodiversity convention of the United Nations (Rio convention).Molecular identification and populations genetics studies of F. hodginsii in Vietnam were recently performed by Nguyen et al. (2011Nguyen et al. ( , 2012)).These studies employed inter-simple sequence repeat (ISSR)S marker to measure genetic diversity in populations distributed in northern Vietnam and used 18S-rRNA sequence to identify Cupressaceae species in Vietnam.The research investigated phylogenetic relationships of nine Cupressaceae species and revealed that F. hodginsii is within the genus Calosedus.Population genetics study using nucleotide polymorphism has not been reported for F. hodginsii.In this study, we aim to investigate sequence polymorphism at a nuclear functional locus, methionine synthase (MetE) to study nucleotide variation within and among populations of F. hodginsii collected across Vietnam.Information on genetic variation at nucleotide level will be valuable for the conservation of F. hodginsii in Vietnam.

MATERIALS AND METHODS
F. hodginsii seeds were collected from 12 populations across Vietnam (Figure 1 and Table 1).We collected seeds from trees growing at least 20 m apart.Seeds collected were then stored in silica gel and carried to the laboratory in Hanoi.Seeds were sown immediately after they arrived at Hanoi.

DNA extraction
In gymnosperms, megagametophytes are of maternal origin and are haploid.We extracted DNA from megagametophyte of F. hodginsii for use in this study.Since DNA samples are haploid, direct sequencing is used to obtain nucleotide sequences and directly determine haplotype sequences.F. hodginsii seeds were sown on wet paper in a Petri plate.After germination, we removed all seed coats and embryos.Using a pestle, we collected a fresh megagametophyte in a 2-ml tube.Haploid genomic DNA was extracted from megagametophytes using DNsaesy plant mini kit (QIAGEN, Valencia, CA).Total DNA was determined using a fluorometer and diluted to 5 ng/μl.

Polymerase chain reaction amplification
Polymerase chain reaction was carried out in 25 μl solution consisting of 2.5 μl MgCl2 (25 mM), 2 μl dNTP (8 mM), 10 pmol of each primer, 1.25 units Tag DNA polymerase (Invitrogen) and 1.5 μl of template DNA.The reaction mixture was subjected to amplification in the Gene Amp PCR System 2400, under the following thermal cycler: an initial denaturing step at 94°C for 5 min, followed by 40 cycles consisting of 1 min at 94°C, 30 s at 52°C, 1 min extension at 72°C and 10 min at 72°C for a final cycle to complete the extension of any remaining products before holding the samples at 4°C until they were analysed.The sequence of primers used for PCR amplification and internal primers for sequencing the MetE locus in this study are listed in Table 2.

DNA sequencing
PCR products were sequenced using the BigDye terminator sequencing kit (Applied Biosystems, Foster City, CA) on an ABI310 automated sequencer (Applied Biosystems).

Data analysis
The obtained sequences were aligned manually using a sequence alignment editor, SE-Al v. 2.0a11 (Rambaut, 1996).Intron positions were determined on the basis of homologous genes in other plants and the GT-AG rule.All indels were excluded from the analysis.DnaSP 4.10.0(Rozas et al., 2003) was used to estimate the standard population genetic parameters (pairwise nucleotide diversity and nucleotide polymorphism).
FST statistics (Hudson et al., 1992) were used to measure the amount of genetic differentiation among populations.Significance levels of the nearest-neighbor statistic (Snn; Hudson, 2000) and a weighted measure of the ratio of the average pairwise differences within populations to the total average differences (KST*; Hudson et al., 1992b) were calculated using 10,000 permutation tests.
Tajima's D (Tajima, 1989) statistic (DT) was used to test for deviations from neutrality.This test measures skews in the frequency spectrum, where a negative DT suggests an excess of low-frequency polymorphisms and a positive DT indicates an excess of intermediate-frequency polymorphisms.
Linkage disequilibrium (LD) analyses were performed using DnaSP.The 2 indices of LD, D′ and R 2 (squared allele frequency correlation) were estimated across all informative sites.D′ has a potential range from -1 to 1, with the magnitude of disequilibrium being indicated by a departure from zero in either direction.
Furthermore, the percentage of pairs of sites in significant LD was also calculated; the statistical significance of each pairwise test was evaluated using the  2 test and statistical significance was determined at a level of P < 0.05.Pairwise analyses are not fully independent because of LD itself; hence, the proportion of significant pairwise tests and the mean |D′| and R 2 were compared for pairs of sites separated by different molecular map distances (<400 or >400 bp).Comparison of these values over the two separate distance ranges was carried out by  2 analysis.

Nucleotide variation
The length of MetE locus sequenced in this study varied from 1567 to 1559 bp.Four indels were found and excluded from all analyses.Estimates of nucleotide variation within populations and in the pool (the whole samples) are shown in Table 3.A total of 42 polymorphic sites were detected in 144 sequences generated from 12 populations of F. hodginsii collected across Vietnam.The estimate of overall nucleotide diversity was 0.00401 and 0.00519 at the total (π tot ) and silent sites (π silent ), respectively.Nucleotide diversity within populations varied from 0.00300 to 0.00543 at the total and from 0.00357 to 0.00666 at silent sites.On average, approximately 85 and 78% of the nucleotide diversity in the pool was found within populations at total and silent  sites, respectively.
The estimates of total nucleotide diversity for the four southern populations (9 to 12) ranged from 0.00300 to 0.00380 which were lower than those for the eight populations (1 to 8) (0.00399-0.00543)(Table 3, Figures 1 and 2).The difference was also observed in the estimates of nucleotide diversity at silent sites between the 2 groups of populations (Table 3).
are shown in Table 3.None of the D T estimates were significant, suggesting that there was no significant excess of low-frequency and intermediate-frequency alleles within populations of F. hodginsii.The ratios of nonsynonymous to synonymous polymorphisms ( a /s) were estimated to be 0.2445, indicating that the MetE locus is generally under strong purifying selection.

Linkage disequilibrium
The 3 statistics of linkage disequilibrium were estimated and are shown in Table 4. Linkage disequilibrium between polymorphic sites was estimated and compared between the 2 groups of sites <400 and >400 bp apart.The estimates of the means |D'| and R 2 among sites <400 bp were significantly higher than those among sites >400 bp apart (P < 0.05, t-test).The percent of significant pairwise tests was also significantly lower in sites >400 bp apart (P < 0.05,  2 -test).The results suggest that linkage disequilibrium at MetE locus declined significantly among sites >400 bp apart.Recombination plays a role in generating nucleotide diversity in MetE of F. hodginsii.

Population differentiation
The overall estimates of genetic differentiation among 12 populations were low; F ST = 0.093 and K ST * = 0.078, even though both values were highly significant (P < 0.001).Analysis of genetic clustering using BAPS provided the best support for all the 144 sequences belonging to the same genetic cluster.Pairwise analysis of genetic differentiation among the 12 populations is shown in Table 5.Among the populations analyzed, many pairs of populations showed significant genetic differentiation, as evaluated by F ST and S nn .However, only a few pairs showed significant differentiation when evaluated by K ST *.Most of the significant F ST and K ST * values were very low.

DISCUSSION
The estimates of nucleotide diversity at total and silent sites of MetE locus in F. hodginsii in the present study were comparable to those of pal1 locus in an European conifer, Pinus sylvestris (Dvornyk et al., 2002).The estimates were generally higher than several other conifers previously reported such as Pinus (Garcia-Gil et al., 2003;Neale and Savolainen, 2004;Pot et al., 2005;Pyhäjärvi et al., 2007), Cryptomeria (Kado et al., 2003), Chamaecyparis (Kado et al., 2008) and Picea (Heuertz et al., 2006) (Table 6).Nucleotide diversity in conifers was generally lower than that in broad-leaved tree species such as Populus (Ingvarsson, 2005), Quercus (Quang et al., 2008), Zanthoxylum (Kamiya et al., 2008) and Betula (Järvinen et al., 2003).The present study revealed a low level of among-population differentiation for all local populations of F. hodginsii in Vietnam.Previous studies employing nuclear sequence polymorphism also revealed low and very low levels of genetic differentiation among populations of outcrossing forest trees such as oaks and conifers (Quang et al., 2008;Järvinen et al., 2003;Dvornyk et al., 2002;Garcia-Gil et al., 2003;Neale and Savolainen, 2004;Pot et al., 2005;Pyhäjärvi et al., 2007;Kado et al., 2003, Kado et al., 2008, Heuertz et al., 2006).A recent study (Nguyen et al., 2011) using ISSR markers to assess the genetic diversity within and differentiation among populations of F. hodginsii across Vietnam revealed a low level of diversity in comparison with other tree species previously reported.Nguyen et al. (2011) showed the possibility of habitat fragmentation to explain the very low level of genetic diversity observed in populations of F. hodginsii in Vietnam.In accordance with their report, the present study also revealed lower level of nucleotide diversity in populations located in central and southern Vietnam.Those populations may have been strongly influenced by habitat fragmentation and small sizes.In Vietnam, F. hodginsii forests were greatly fragmented by human activities such as logging and clearing and they form small forest patches.In central Vietnam, most of the populations of F. hodginsii were especially small in size, varying from 50 to 150 adult individuals per forest (Nguyen et al., 2011).Small population size and forest fragmentation may have caused increased levels of inbreeding and diversity loss (Ellstrand and Elam, 1993;Barrett and Kohn, 1991).
Lower level of nucleotide diversity was found in the central and southern parts of Vietnam.F. hodginsii forests in Vietnam have been overexploited.These forest areas are typically home to the ethnic minority groups.These people often have a low living standard and education.They are heavily reliant on forest products both for subsistence and to provide income through trade.To secure genetic diversity of this species in Vietnam, it is necessary to establish a collection bank of seeds and seedlings of F. hodginsii from populations distributed across Vietnam and especially from the populations in central and southern Vietnam where F. hodginsii forests are more fragmented and heavily exploited.On the conservation point of view, the populations in the south are especially important because they are adaptive to the climate conditions in the southern distribution limit of the species and are sensitive to the loss of genetic diversity when habitat fragmentation occurs.

Figure 1 .
Figure 1.Location of the 12 sampled populations of F. hodginsii in Vietnam.Population codes (1 to 12) are those given inTable 1.

Figure 2 .
Figure 2. Comparison of total nucleotide diversity among populations.Values of nucleotide diversity were multiplied by 100 000.

Table 1 .
Geographical location of the sampled populations.

Table 2 .
List of primers used in this study.

Table 3 .
Estimates of nucleotide variation within populations.

Table 5 .
Analyses of population differentiation.

Table 6 .
Summary of nucleotide diversity in conifers.