Possible involvement of locus-specific methylation on expression regulation of leafy homologous gene (CiLFY) during precocious trifoliate orange phase change process

Abstract

DNA methylation plays an essential role in regulating plant development. Here, we described an early flowering trifoliate orange (precocious trifoliate orange, Poncirus trifoliata L. Raf) was treated with 5-azacytidine and displayed a number of phenotypic and developmental abnormalities. These observations suggested that DNA methylation might play an important role in regulating many developmental pathways including early flowering trait, and then the expression level of five key or integrated citrus flowering genes were analyzed. Our results showed that flowering locus T (CiFT) relative expression level was increased with the increasing concentrations of 5-AzaC. However, leafy (CiLFY), APETELA1 (CiAP1), terminal flower1 (CiTFL1), and flowering locus C (CiFLC) showed highest relative expression levels at 250 µΜ treatment, while decreased sharply at higher concentrations. In order to further confirm DNA methylation affects the expression of these genes, their full-length sequences were isolated by genome-walker method, and then was analyzed by using bioinformatics tools. However, only one locus-specific methylation site was observed in CiLFY sequence. Therefore, DNA methylation level of the CiLFY was investigated both at juvenile and adult stages of precocious trifoliate orange by bisulfate sequencing PCR; it has been shown that the level of DNA methylation was altered during phase change. In addition, spatial and temporal expression patterns of CiLFY promoter and a series of 5' deletions were investigated by driving the expression of a β-glucuronidase reporter gene in Arabidopsis. Exogenous GA3 treatment on transgenic Arabidopsis revealed that GA3 might be involved in the developmental regulation of CiLFY during flowering process of precocious trifoliate orange. These results provided insights into the molecular regulation of CiLFY gene expression, which would be helpful for studying citrus flowering.

Figure 1. Phenotypic characteristics of different concentrations of 5-AzaC treated precocious trifoliate orange seeds grew for 15 and 35 days, control: treated with distilled water.

After the seed coats of precocious trifoliate orange seed were peeled and sterilized, the embryos were imbibed at 23°C on filter paper soaked with fresh 5-AzaC solution (control, 250, 500, and 1000 µΜ). The seeds were transferred daily to new filter paper containing fresh 5-AzaC solution. After 15 days, the germinated seeds were planted; the seedlings were watered regularly with nutrient solution and grown in test tubes in artificially lit growth cabinets under long days (16 h light and 8 h dark at 23°C) with fluorescent lights at a photosynthetic photon flux density of 200 µmol m-2s-1. The treated seedlings were analyzed after the seedlings grew in the soil for 20 days.

Figure 2. The relative expression analysis ofCiFT, CiAP1, CiFLC, CiLFY, and CiTFL1 under different 5-AzaC concentrations by real-time PCR.

Total RNA was isolated from throughout the aerial part (including stem, leaf and shoot apical meistem) of 35 days after seed germination as shown in Figure 1. Data points represent mean values ± SE of at least four replicates for the relative expression, which were normalized by the amount of the β-actin control expression. The primers used for the analyses were listed in table S1

Figure 3. Relative quantities of CiLFY in various tissues and stages of precocious trifoliate orange.

(A): Relative quantities of CiLFY at juvenile and adult stage of precocious trifoliate orange. (B): Relative quantities of CiLFY in various tissues of juvenile (leaves, roots and stems) and adult (leaves, roots, stems, flowers at full bloom, whole fruits at 30 days after flowering, apex bud and lateral bud) phase of precocious trifoliate orange. AB: apex bud, LB: lateral bud. Data points represent mean values ± SE of at least four replicates for the relative expression, which were normalized by the amount of the β-actin control expression. The primers used for the analyses were listed in table S1

Figure 4. Statistical analysis of the cytosine methylation status in CpG island and 5′-UTR of CiLFY gene at juvenile and adult phase of precocious trifoliate orange.

Figure 5. Histochemical localization of GUS activity in transgenic Arabidopsis.

a–d: seedlings at 7, 10, 14 and 20 days, respectively; e: lateral flower bud just emerged; f: lateral flower bud; g: apex flower bud; h: young inflorescence; i: fully-opened flower; j: young leaves; k old leaves; l: roots from adult plants; m: fruit. bar: 1 mM.

Figure 6. GUS activity of the CiLFY promoter in transgenic Arabidopsis.

(A): CiLFY full-length promoter and deletion derivatives. Numbers inside the bars indicated the end position of each deletion, number 1 (+1) represents the first nucleotide on the 5′ side of the transcription start site (TSS). (B): Folds increase in the GUS activity for the transgenic plants with full-length and various truncated of the promoters. Three independent transgenic lines for each of the promoter fragments were used for the assay. Error bars represent standard error. C: Histochemical localization of GUS activity in under untreated (a and b) and 100 µM GA₃ treated (c and d) transgenic Arabidopsis. D: Analysis of GUS activity from CiLFY deletion constructs in 17-old-days transformed Arabidopsis, Leaves were collected the day after spaying 100 µM GA₃. Error bars represent standard error. The results from one representative experiment were shown herein, expressed as means and standard errors calculated using Microsoft Excel. The data were processed using one-way analysis of variance (ANOVA), and statistical differences were compared based on Student's t-test, taking P<0.05 as significant.