Differential fine-tuning of gene expression regulation in coffee leaves by CcDREB1D promoter haplotypes under water deficit

Abstract

Despite the importance of the DREB1D gene (also known as CBF4) in plant responses to water deficit and cold stress, studies analysing its regulation by transgenic approaches are lacking. In the current work, a functional study of three CcDREB1D promoter haplotypes (named HP15, HP16 and HP17) isolated from drought-tolerant and drought-sensitive clones of Coffea canephora was carried out in plants of C. arabica stably transformed by Agrobacterium tumefaciens by analysing their ability to regulate the expression of the uidA reporter gene in response to water deficit mimicked by polyethylene glycol (-2.0 MPa) and low relative humidity treatments. A deletion analysis of their corresponding 5'-upstream regions revealed increased specificity of β-glucuronidase activity in the polyethylene glycol and low relative humidity treatments, with high expression in leaf mesophyll and guard cells in full-length constructs. RT-qPCR assays also revealed that the HP16 haplotype (specific to clone tolerant to water deficit) had stronger and earlier activity compared with the HP15 and HP17 haplotypes. As most of the cis-regulatory elements involved in ABA-dependent and -independent networks, tissue specificity and light regulation are common to these haplotypes, we propose that their organization, as well as the nucleic acid polymorphisms present outside these boxes, may play a role in modulating activities of DREB1D promoters in guard cells.

Keywords: Coffee; DREB1D gene; genetic transformation; promoter haplotypes; uidA reporter gene; water deficit.

Fig. 1.

Consensus sequence of CcDREB1D promoter haplotypes. The consensus sequence contains all single nucleotide polymorphisms (SNPs, in gray) and insertion/deletions (INDELs, lower case underlined with horizontal brackets) found in the HP16 and HP17 haplotypes of CcDREB1D promoters. The nucleotides are numbered (on the right) in each lane using the first nucleotide (+1) of the CcDREB1D mRNA sequence (based on RNAseq data evidence available in the Coffee Genome Hub) as the start of numbering. CREs are identified by boxes with their corresponding name below except DOF/guard-cell (underlined), W-BOX (underlined and italics) and ABRE-LIKE (bold). Horizontal arrows indicate the primers (Table 1) used to amplify the full-length and truncated versions of coffee CcDREB1D promoter sequences (Fig. 2C). The TCA microsatellite GAAWTT unidentified motif and the putative TATA box are also indicated (bold and brackets). The 19 bp in italics upstream of the ATG (+174) of the β-glucuronidase corresponds to the multiple cloning site of the pBI121 vector.

Fig. 2.

Haplotypes of CcDREB1D promoters of C. canephora. (A) CcDREB1D haplotypes found in drought-tolerant (D^T) clone 14 (HP15/HP16) and drought-sensitive (D^S) clone 22 (HP15/HP17) of C. canephora. Color code for haplotypes: HP15 (white), HP16 (black) and HP17 (gray). (B) Graphic representation of nucleotide variability detected in CcDREB1D promoter haplotypes. The x-axis corresponds to the bases of the HP15 sequence (Fig. 1). The y-axis corresponds to the frequency of polymorphic sites (S) observed in the HP16 (continuous line) and HP17 (dotted line) haplotypes compared with the HP15 sequence used as a reference. (C) Schematic representation of the CcDREB1D haplotypes analysed in transgenic plants of C. arabica. The schematic map of CcDREB1D (Cc02_g03430) is given in the upper part together with the DREB primers (F1, R1 in brackets, see Table 1) used to amplify the CcDREB1D promoter haplotypes. Plasmid names used for stable transformation of C. arabica are given for each construct, indicating the haplotype studied with its corresponding length (L, long; M, medium; S, short). The fragments were amplified using the forward primers (including the HindIII [AAGCTT] restriction site) corresponding to L-DREB (L, white square), M-DREB (M, white star) and S-DREB (S, white triangle) and the reverse primer R-DREB (including the BglII [AGATCT] restriction site) indicated by a black point and further cloned in front of the uidA reporter gene. pBI121 (CaMV35S:uidA gene construct) and pBI101 (uidA promoterless gene) were used as positive and negative controls of GUS enzymatic activities, respectively.

Fig. 3.

Histochemical localization of GUS activity in different tissues of transgenic C. arabica subjected to dehydration. Columns 1: root; 2: stem; 3: meristem; 4: leaf. The tissues belong to plants regenerated from constructs pHP17S (A), pHP16S (B), pHP16M (C), pHP16L (D), pHP15L (E), and pHP17L (F). The histochemical localization of GUS activity in control plants (WT, pBI101, and pBI121) is given in Supplementary Fig. S2. The scale bars given for each image correspond to 30 µm, excepted in (E4) and (D4) (80 µm), and (E3) (300 µm). Arrows indicate GUS staining restricted to specific cells and tissues: leaf epidermis (le), guard cell (gc), parenchyma (p), vascular tissue (vt), and apical meristem (am).

Fig. 4.

Histochemical localization of GUS activity in guard cells of C. arabica subjected to low air relative humidity (RH 9%). Guard cells visualized by bright field microscopy (×20 magnification) on the abaxial detached epidermis of coffee leaves. The explants were from coffee plants transformed by pBI101 (A, negative control), pBI121 (B, positive control), pHP17S (C, D), pHP17L (E, F), pHP16S (G, H), and pHP16L (I, J). Unstressed conditions for (C, E, G, I). Stressed conditions (9% RH) for (A, B, D, F, H, J). The scale bars in each image correspond to 30 µm.

Fig. 5.

Proportions of GUS-stained guard cells: water deficit-induced regulation of GUS activity in mature leaf stomata driven by CcDREB1D promoter haplotypes. Change of GUS-stained guard cells’ proportions following 0, 3, 6, 12 and 24 h of exposure to PEG (equivalent to −2.0 MPa) and low air relative humidity (RH 9%) treatments in pHP15L-, pHP16L-, and pHP17L-transformed coffee plants. Box-and-whisker plots display variation in proportions of GUS-stained guard cells for each time assessed. The lower and upper hinges represent the first and third quartile respectively, the bold line represents the median, and the whiskers represent the smallest and the greatest values.

Fig. 6.

Expression profiles of the uidA and CaDREB1D genes in leaves of transgenic C. arabica during water deficit. Expression of the uidA (A) and CaDREB1D (B) genes was tested in leaves of coffee plants transformed by the pHP15L (white isobars), pHP16L (light gray isobars), and pHP17L (dark gray isobars) constructs and subjected to 0, 3, 6, 12, and 24 h osmotic stress (PEG treatment) by RT-qPCR experiments using the GUS-F/R and DREBA09-F/R primer pairs, respectively. Expression levels are indicated in relative quantification using the expression of the CaGAPDH gene as a reference. The results are expressed using T₀ samples as internal calibrators for each construct. The relative quantification values correspond to the mean of at least three biological repetitions analysed by three technical replicates±SD. The significance of expression level differences was evaluated using the pairwise Wilcoxon rank test (non-parametric test). Treatments sharing the same letter are not significantly different.