Characterization Of NCED And DHN And Population Genetics Analysis

Drought is a major factor that limit the distribution of plants and the production of crops. Drought stress induces a range of physiological and bio-chemical responses in plants, also in molecular level. Drought tolerent is a quatative trait (QTL), an assortment of genes involved in this process. These genes can be classified into two groups according to signaling pathway: ABA-independent and ABA-dependent regulatory systems. Here, we examined two genes involved in ABA-dependent pathway: 9-cis-epoxycarotenoid dioxygenase gene (NCED) and dehydrin gene (DHN). NCED is a key enzyme in ABA biosynthesis, which is on the upstream of ABA-dependent pathway; DHN, downstream gene, can be induced by drought and ABA to stabilize cellular structures and membranes against dehydration.In this research, we first cloned cDNAs and DNAs of MdNCED and dehydrin gene Mddehydrin from apple leaves and fruits respectively; and studied its quantative expression under drought treatments. Then, we cloned LeNCED1 and dehydrin gene pLC30-15 from 6 populations of two wild tomatoes: Solanum peruvianum and Solanum chilense; And made a homologus analysis. Finally, we made population genetics analysis of sequences from wild tomatoes, and try to find evidence of local adaption. Main contents and results are as below:1. We cloned a cDNA and a DNA of MdNCED from apple fruits. cDNA is 1945 bp long,encoding 607 amino acid, and there is no intron in gDNA. Expression pattern of MdNCED during fruit development were detected by real-time RT-PCR. MdNCED mRNA level was very low in young fruits, increased markedly with fruit growth, reached maximum in mature fruit, and then maintained stable. Taken together, the results indicate that MdNCED might play a key role in the regulation of fruit ripeness.2. We cloned a cDNA and a DNA of dehydrin gene of Mddehydrin from apple leaves. This gene is 1170bp long, including 2 exones and one intron. Sequence analysis shows there are several characterized motifs in the sequence: a S-segment, two Y-segments and three K-segments, so it is a Y2SK4 type dehydrin gene. Under drought treatment of 0,2,4,6,8d, experiment by real-time PCR shows Mddehydrin expression is very low under no drought treatment; while Mddehydrin mRNA level increased markedly with extent of dehydration, and reached maximum at the sixth day, which is 200-fold higher than under no treatment. Then expression of Mddehydrin mRNA decreased to 25% of maximum。It is fully proved that drought can induced expression of Mddehydrin mRNA strongly.3. We also sequenced LeNCED1 from 30 individuals of 6 populations, and totally obtained 60 sequences. LeNCED1 is 1827 bp long consisting of a 5'flanking region and a single exon. The deduced amino acid sequence of LeNCED1 shared high identity with StNCED1 of S. tuberosum, 96.8%。4. Using mode plant wild tomatoes S. peruvianum and S. chilense as plant material, we sequenced pLC30-15 from 30 individuals of 6 populations. We totally obtained 60 sequences. pLC30-15 is 918 bp long consisting of two exons, one intron and a 5'flanking region. There are one S-segment and three K-segments, blonging to SK3 type dehydrin gene. The deduced amino acid sequence of SK3 dehydrins shows pLC30-15 shared 91% identity with potato (AY292655).5. We surveyed nucleotide diversity at LeNCED1 gene, observed that (1) the nucleotide diversity of LeNCED1 at all sites is low in both species, about twofold lower than at the reference loci. In contrast, levels of diversity at silent sites are comparable with those at the reference loci. (2) Most remarkably, S. chilense shows higher nucleotide diversity (at all sites) than S. peruvianum due to its higher nucleotide diversity at non-synonymous sites. Taken together, this indicates that LeNCED1 has been under purifying selection in both species and that selection pressure on LeNCED1 appears to be stronger in S. peruvianum than in S. chilense. This may be due to the effective population size, which is larger in S. peruvianum, or due to a difference in selection coefficients. (3) A very small dN/dS ratio (<0.10) for LeNCED1 in both S. chilense and S. peruvianum also indicates that LeNCED1 has been under purifying selection.6. pLC30-15 exhibits higher average levels of nucleotide diversity (at all sites) than LeNCED1 and the reference loci, except for the S-, and K-segments, where diversity is very low.7. The dN/dS ratio at pLC30-15 is smaller than 1 but almost 10 times larger than at LeNCED1, indicating that purifying selection is also operating on its coding region but to a lesser extent than on LeNCED1. This difference in dN/dS between LeNCED1 and pLC30-15 results from a threefold higher rate of synonymous substitution (dS) and an almost twofold lower rate of non-synonymous change (dN) at LeNCED1 than at pLC30-15.8. We estimated population differentiation using Fst between pairs of populations. (1) We observed that in S. chilense genetic differentiation between the Tacna and Moquegua samples is very low at both loci. This agrees with the results found at the reference loci. Therefore, we pooled the samples of Tacna and Mocquegua into one sample (called TacMoq) and measured differentiation between the pooled sample and Quicacha, which is high. (2) In particular, LeNCED1 shows twofold higher Fst values than the reference loci, which is consistent with the strong purifying selection on this gene. (3) Overall, genetic differentiation between populations of S. chilense is higher than between those of S. peruvianum. (4) Except for the pair Tacna-Mocquegua, both candidate genes are more differentiated than the reference loci in S. chilense. (5) The high Fst at pLC30-15 between TaqMoc and Quicacha indicates local adaptation.9. We used Tajima's DT and Fu & Li's DFL statistics to detect deviations from the standard neutral model. Most of the individual populations do not exhibit significant departures from neutral equilibrium expectations. Only the sample from Quicacha shows significant deviations in both the DT and DFL statistics at locus pLC30-15, which much higher than those of the reference loci. Since these statistics deviate in a positive direction from the expectation, some form of diversifying or balancing selection may have acted on this gene in the Quicacha population in the (recent) past.10. The Quicacha sample shows evidence for a haplotype structure at pLC30-15, with only three haplotypes of 10 alleles. While haplotypes 1 and 2 show differences at only two sites, haplotype 3 is quite diverged from the other two haplotypes. Application of the haplotype test shows that the observed haplotype diversity is significantly lower than expected (P=0.006). Consistent with the significantly high Tajima DT estimate and the much lower corresponding value at the reference loci, this suggests that the pattern of variation at pLC30-15 has been caused by positive (diversifying) selection rather than demographic effects.