Expression profiling of tomato pre-abscission pedicels provides insights into abscission zone properties including competence to respond to abscission signals

Abstract

Background: Detachment of plant organs occurs in abscission zones (AZs). During plant growth, the AZ forms, but does not develop further until the cells perceive abscission-promoting signals and initiate detachment. Upon signal perception, abscission initiates immediately; if there is no signal, abscission is not induced and the organ remains attached to the plant. However, little attention has been paid to the genes that maintain competence to respond to the abscission signal in the pre-abscission AZ. Recently, we found that the tomato (Solanum lycopersicum) transcription factors BLIND (Bl), GOBLET (GOB), Lateral suppressor (Ls) and a tomato WUSCHEL homologue (LeWUS) are expressed specifically in pre-abscission tissue, the anthesis pedicel AZs. To advance our understanding of abscission, here we profiled genome-wide gene expression in tomato flower pedicels at the pre-abscission stage.

Results: We examined the transcriptomes of three tomato flower pedicel regions, the AZ and flanking proximal- (Prox) and distal- (Dis) regions, and identified 89 genes that were preferentially expressed in the AZ compared to both Prox and Dis. These genes included several transcription factors that regulate apical or axillary shoot meristem activity. Also, genes associated with auxin activity were regulated in a Prox-Dis region-specific manner, suggesting that a gradient of auxin exists in the pedicel. A MADS-box gene affecting floral transition was preferentially expressed in the Prox region and other MADS-box genes for floral organ identification were preferentially expressed in Dis, implying that the morphologically similar Prox and Dis regions have distinct identities. We also analyzed the expression of known regulators; in anthesis pedicels, Bl, GOB, Ls and LeWUS were expressed in the vascular cells of the AZ region. However, after an abscission signal, Bl was up-regulated, but GOB, Ls and LeWUS were down-regulated, suggesting that Bl may be a positive regulator of abscission, but the others may be negative regulators.

Conclusions: This study reveals region-specific gene expression in tomato flower pedicels at anthesis and identifies factors that may determine the physiological properties of the pre-abscission pedicel. The region-specific transcriptional regulators and genes for auxin activity identified here may prevent flower abscission in the absence of signal or establish competence to respond to the abscission signal.

Figure 1

Screening for genes differentially expressed in different tomato pedicel regions by expression microarray analysis. (A) Flower pedicel regions used in this study. Tomato abscission zones (AZs) form in an intermediate region in the pedicels and have a knuckle-like structure in which a groove forms for abscission. The pedicel region between the AZ and the main stem of the inflorescence is referred to as the proximal region (Prox) and the region between the AZ and the flower is referred to as the distal region (Dis). RNA from each region was extracted from five to twenty pedicels at anthesis. (B) A schematic of the genome-wide transcriptome screen for genes up-regulated in the AZ. Expression was compared between AZ and Prox or Dis at anthesis by microarray analyses with 3 independently prepared samples. Two circles in the Venn diagram indicate the number of probes showing higher signal from AZ than that from Prox or Dis, and the 112 probes in the overlap were further investigated in this study. By merging results for probes encoding the same gene, we found 89 independent genes up-regulated in the AZ. (C) A schematic of the screening for transcription factor and phytohormone-related genes differentially expressed between Prox and Dis. The expression profiles of Prox and Dis examined by microarray analyses were compared and genes for transcription factors or phytohormone-related activity were selected from the genes exhibiting significantly different expression patterns between the two regions.

Figure 2

Expression of Bl, GOB, LeWUS and Ls in tomato pedicels. (A) Expression of Bl, GOB, LeWUS and Ls during pedicel development. Expression of the genes in pedicels at 5 and 3 days before anthesis, at anthesis and at 2 days to 3 weeks after anthesis was examined by qRT-PCR. Levels of transcripts of each gene are shown as fold-change values compared to the sample harvested at anthesis. (B) Expression of Bl, GOB, LeWUS and Ls in response to an abscission stimulus. Anthesis flowers were removed from the pedicels to induce abscission and then the pedicel AZs were harvested at 0 hours, 12 hours, 1 day and 2 days after flower removal. The expression of Bl, GOB, LeWUS and Ls was examined by qRT-PCR. Error bars indicate standard deviation of biological triplicates. At 1 day after flower removal, approximately 10% of pedicels were abscised and then the rate increased to 36% at 2 days after (Additional file 8). Both attached and detached pedicels were examined simultaneously. As positive controls, expression analyses were performed for Cel5 and TAPG4, which are up-regulated after flower removal [27]. Levels of transcripts of each gene are shown as fold-change values compared to the 0 hour sample. Error bars indicate standard deviation of biological triplicates. (C) Expression of Bl, GOB, Ls, and LeWUS in a flower pedicel AZ at anthesis. Dig-labeled antisense probes were hybridized to serial sections of a pedicel AZ at anthesis. Transcripts of these genes, indicated by arrowheads, were found in several lines of cells within the vascular bundles. The upper left panel is a schematic of an AZ. The control section was stained with a hematoxylin and eosin stain solution. AZ; abscission zone, Co; Cortex, V; Vascular bundle, and Cp; Central parenchyma.