The grape berry-specific basic helix-loop-helix transcription factor VvCEB1 affects cell size

Abstract

The development of fleshy fruits involves complex physiological and biochemical changes. After fertilization, fruit growth usually begins with cell division, continues with both cell division and expansion, allowing fruit set to occur, and ends with cell expansion only. In spite of the economical importance of grapevine, the molecular mechanisms controlling berry growth are not fully understood. The present work identified and characterized Vitis vinifera cell elongation bHLH protein (VvCEB1), a basic helix-loop-helix (bHLH) transcription factor controlling cell expansion in grape. VvCEB1 was expressed specifically in berry-expanding tissues with a maximum around veraison. The study of VvCEB1 promoter activity in tomato confirmed its specific fruit expression during the expansion phase. Overexpression of VvCEB1 in grape embryos showed that this protein stimulates cell expansion and affects the expression of genes involved in cell expansion, including genes of auxin metabolism and signalling. Taken together, these data show that VvCEB1 is a fruit-specific bHLH transcription factor involved in grape berry development.

Fig. 1.

Phylogenetic analysis of VvCEB1. VvCEB1 (black circle) and its homologues (bold) belong to a distinct bHLH gene subfamily. The different bHLH gene subfamily numbers and Arabidopsis thaliana (At), Thellungiella halophila (Th), Petunia hybrida (Ph), Oryza sativa (Os), and Hordeum vulgare (Hv) GenBank accession numbers reported in this figure are as described by Pires and Dolan (2010). The phylogenetic tree represents a non-exhaustive list of bHLH transcription factors; each bHLH subfamily is generally illustrated by a known A. thaliana bHLH and its closest homologues in V. vinifera (Vv), S. lycopersicum (Sl), and Populus trichocarpa (Pt). Sequence data were obtained from: http://www.genoscope.cns.fr/externe/GenomeBrowser/Vitis/ (V. vinifera), http://planttfdb.cbi.edu.cn/family.php?sp=Sly&fam=bHLH (S. lycopersicum), and http://www.ncbi.nlm.nih.gov/genbank/ (P. trichocarpa). The phylogenetic tree was constructed with MEGA version 4 (Tamura et al., 2007) using the neighbour-joining method with 2000 bootstrap replicates.

Fig. 2.

Sequence analysis of VvCEB1. Full-length sequence comparison of VvCEB1 and its closest homologues SlbHLH1, CaUPA20, AtbHLH137, AtBIGPETAL and AtBEE1 using the ClustalW program with default parameters. Conserved residues are shaded in black; dark grey shading indicates conserved residues in at least four out of six of the sequences, and light grey shading indicates conserved residues in three out of six of the sequences. Basic residues that putatively function as a nuclear localization site for VvCEB1 are indicated by arrows above the alignment. Putative protein–protein binding domains and a bHLH domain for VvCEB1 are labelled above the alignment in black circles and open squares, respectively. Another conserved region with predicted secondary structure was identified for VvCEB1: an acidic α-helical domain, indicated by a black line.

Fig. 3.

qRT-PCR analysis of VvCEB1 expression patterns in grapevine cv. Cabernet Sauvignon. (A) VvCEB1 expression in various grapevine organs: roots (R), stem (S), leaves (L), inflorescences (I), and ripening berries (RB) at 80 d p.a. Results are shown as means ±SD for three independent experiments. Gene expression was normalized against VvEF1γ expression. (B) VvCEB1 expression at different stages of berry development, from inflorescences (I) to mature berries at 100 d p.a. (DPA). The arrow indicates the veraison stage. Results are shown as means ±SD for four replicates from two independent experiments (carried out in the summer of 2006 and 2009). Gene expression was normalized against VvEF1γ. (C) VvCEB1 expression in different tissues from ripening berries at 80 d p.a. Results are shown as means ±SD for three independent experiments. Gene expression was normalized against VvEF1γ.

Fig. 4.

Correlation between VvCEB1 expression and grape berry size. qRT-PCR analysis showing VvCEB1 transcript accumulation (black squares) in different grape varieties exhibiting different berry weight (grey bars). Berries were harvested 3 weeks after veraison and seeds were removed for this analysis. Gene expression was normalized against VvEF1γ. All data are means ±SD of four replicates from two independent experiments (summer 2009 and 2010).

Fig. 5.

Overexpression of VvCEB1 strongly affects grape embryo development. (A) Relative expression level of VvCEB1 transcript accumulation in transgenic grape 41B embryos. The VvCEB1 transcript level was quantified by qRT-PCR in control (pFB8 empty vector) and VvCEB1-overexpressing (35S::VvCEB1) lines. These embryos were collected 7 d after initiation of embryogenesis. Gene expression was normalized against VvEF1γ. Data are means ±SD of three independent experiments. (B) Control (pFB8 empty vector) and VvCEB1-overexpressing (35S::VvCEB1) embryos lines were observed with a stereomicroscope 32 d after the initiation of embryogenesis. Bars, 1mm. (C) Embryos sizes were measured at different stages of 41B embryo development, from 7 to 32 d after initiation of embryogenesis (DAE), both for control (pFB8 empty vector, black circles) and VvCEB1-overexpressing (35S::VvCEB1, open circles) lines. Thirty embryos were measured for each developmental stage in both lines. Results are shown as means ±SD. Kruskal-Wallis one way analysis of variance on ranks was performed and asterisks indicate statistically significant differences between lines at the same time point (P<0.001).

Fig. 6.

Overexpression of VvCEB1 increases cell size in grape 41B embryos. (A) Difference in cell size between control (PFB8 empty vector) and VvCEB1-overexpressing (35S::VvCEB1) embryos lines. Observations were assessed at 12 and 24 d after initiation of embryogenesis (DAE). Bars, 25 µm. (B) The cell area of control (pFB8 empty vector, black circles) and VvCEB1-overexpressing (35S::VvCEB1, open circles) lines was assessed at different stages of 41B embryo development, from 5 to 32 d after initiation of embryogenesis (DAE). Data are means ±SD of the ten largest cells measured in this area for each embryo and 20 embryos for each line were used to collect these data. Kruskal–Wallis one way analysis of variance on ranks was performed and asterisks indicate statistically significant differences between lines at the same time point (P<0.001).

Fig. 7.

Expression of VvCEB1 promoter in tomato. (A) GUS activity in various organs of tomato plants stably transformed with VvCEB1 promoter: GUS (VvCEB1pro:GUS) transgene. CaMV 35S promoter: GUS (35Spro:GUS) and Wild-Type (WT) plant lines were used as controls. Bars = 2.5mm. (B) GUS activity in tomato fruits stably transformed with VvCEB1promoter: GUS (VvCEB1pro:GUS) transgene at different development stages: 6 d post anthesis (DPA), 20 DPA, 30 DPA, and Breaker + 6 d (Br + 6 d). Wild-Type (WT) line was used as control. Bars = 2.5mm.