Abscisic acid (ABA) regulates grape bud dormancy, and dormancy release stimuli may act through modification of ABA metabolism

Abstract

In warm-winter regions, induction of dormancy release by hydrogen cyanamide (HC) is mandatory for commercial table grape production. Induction of respiratory stress by HC leads to dormancy release via an uncharacterized biochemical cascade that could reveal the mechanism underlying this phenomenon. Previous studies proposed a central role for abscisic acid (ABA) in the repression of bud meristem activity, and suggested its removal as a critical step in the HC-induced cascade. In the current study, support for these assumptions was sought. The data show that ABA indeed inhibits dormancy release in grape (Vitis vinifera) buds and attenuates the advancing effect of HC. However, HC-dependent recovery was detected, and was affected by dormancy status. HC reduced VvXERICO and VvNCED transcript levels and induced levels of VvABA8'OH homologues. Regulation of these central players in ABA metabolism correlated with decreased ABA and increased ABA catabolite levels in HC-treated buds. Interestingly, an inhibitor of ethylene signalling attenuated these effects of HC on ABA metabolism. HC also modulated the expression of ABA signalling regulators, in a manner that supports a decreased ABA level and response. Taken together, the data support HC-induced removal of ABA-mediated repression via regulation of ABA metabolism and signalling. Expression profiling during the natural dormancy cycle revealed that at maximal dormancy, the HC-regulated VvNCED1 transcript level starts to drop. In parallel, levels of VvA8H-CYP707A4 transcript and ABA catabolites increase sharply. This may provide initial support for the involvement of ABA metabolism also in the execution of natural dormancy.

Keywords: 9-cis-epoxycarotenoid dioxygenase; ABA 8′-hydroxylase; abscisic acid; bud; dormancy; grapevine..

Fig. 1.

ABA delays bud break in a concentration- and duration-dependent manner. Vines of Vitis vinifera cv. Early Sweet from a vineyard at Gilgal, located in the Jordan Valley, were pruned to three-node spurs. The detached canes were cut into single-node cuttings, randomly mixed, and groups of 10 cuttings were prepared. (A) Four treatments were carried out, each with nine groups of 10 cuttings. The bases of the cuttings were immersed in vases containing 10 μM ABA, 100 μM ABA (with 0.02% Triton), or only 0.02% Triton (for control and HC treatments). The vases were placed in a growth chamber and forced at 22 °C under a 14h/10h light/dark regime. After 48h, the solutions were replaced with tap water, and sprayed with 0.02% Triton instead, apart from the HC-treated buds which were sprayed with 3% ‘Dormex’ as detailed in the Materials and methods. The treated groups were forced under the above conditions for another 28 d. Bud break was monitored at 7, 11, 14, 18, 21, 25, and 28 d after spraying. Values are averages of the nine groups in each treatment ±SE. (B) Both ABA treatments were carried out using 100 μM ABA. In the ABA 96h treatment, the cuttings were returned to ABA solution for an additional 48h after spraying. All other details are as in (A).

Fig. 2.

ABA attenuates the enhancing effect of various dormancy release stimuli. (A) In the combined ABA–HC treatment, buds were treated with ABA for 48h prior to HC treatment. All other details are as in Fig. 1. The experimental scheme used in (B), (C), and (D) is identical to that described in (A) apart from the indicated changes. (B) HC treatment was replaced with HS treatment where cuttings were immersed in 50 °C water for 1h as detailed in the Materials and methods. (C) HC treatment was replaced with AZ treatment where cuttings were sprayed with 2% sodium azide and 0.02% Triton. (D) HC treatment was replaced with hypoxia treatment where cuttings were incubated for 48h in sealed jars which were flushed with N₂ to reduce O₂ level to 1%.

Fig. 3.

Dormancy release stimuli modulate transcription of regulators of ABA metabolism. Total RNA was extracted from control, HC-, HS-, and azide-treated buds sampled 12, 24, 48, and 96h after treatment. Relative expression levels of VvNCED1, VvXERICO, and VvA8H-CYP707A4 were determined by qRT-PCR as described in the Materials and methods and normalized against VvActin. The values represent the mean ±SE of three biological repeats, each with two technical repeats. Relative expression levels are presented for HC versus control buds (A–C), HS versus control buds (D–F), and AZ versus control buds (G–I) sampled 12, 24, 48, and 96h after treatment.

Fig. 4.

HC modulates transcription of central components of ABA signalling. Relative expression levels of VvPP2C2 (A), VvPP2C4 (B), VvPP2C9 (C), VvRCAR1 (D), VvRCAR5 (E), and VvRCAR6 (F) are presented in HC and control buds sampled at 12, 24, 48, and 96h after treatment. All other details are as described in Fig. 3.

Fig. 5.

HC modulates expression of the central ABA response mediators VvABF genes. Relative expression levels of VvABF1 (A) and VvABF2 (B) are presented in HC and control buds sampled at 12, 24, 48, and 96h after treatment. All other details are as described in Fig. 3.

Fig. 6.

Effect of HC on ABA and ABA catabolite contents in grapevine buds. ABA (A), PA (B), DPA (C), and neoPA (D) levels were determined in HC-treated and control buds sampled 48h and 96h after treatment. The homogenized samples (0.5g) were used for hormone extraction as detailed in the Materials and methods, and ²H-labelled ABA, PA, DPA, and neoPA were spiked as internal standards. Levels of ABA and its catabolites were analysed by liquid chromatography–tandem mass spectrometry. The levels of the analysed molecules were calculated from the peak area ratios of the endogenous molecule to the relevant internal standard. Values represent means ±SE of three biological repeats (10–12 buds per repeat).

Fig. 7.

An inhibitor of ethylene signalling attenuates the enhancing effect of HC on ABA down-regulation. Levels of ABA (A) and ABA catabolites (B) were determined as described in Fig. 6 in HC- and NBD–HC-treated buds sampled 48h after treatment. Levels of VvNCED1 (C) and VvA8H-CYP707A4 (D) transcript were determined in the same bud samples as described in Fig. 3.

Fig. 8.

Changes in dormancy status of the bud population throughout the dormancy cycle. Canes from the vineyard under study were sampled weekly during the dormancy cycle. Single-node cuttings were prepared and bud break was monitored as described in Fig. 1. Bud-break percentages at 21 d are presented to describe the seasonal changes in dormancy status of the bud population in the vineyard. Values are averages of nine groups of replications, consisting of 10 buds each ±SE.

Fig. 9.

Expression profile of VvNCED1 and VvA8H-CYP707A4 throughout the dormancy cycle. Buds were sampled from canes that were harvested weekly during the dormancy cycle as described in Fig. 8. Levels of VvNCED1 and VvA8H-CYP707A4 transcripts were determined as described in Fig. 3.

Fig. 10.

Changes in the contents of ABA and its catabolites throughout the dormancy cycle. The contents of ABA (A) and ABA catabolites (B) were determined as described in Fig. 6 in buds sampled at seven time points throughout the dormancy cycle. Sampling and dormancy status are described in Fig. 8.

Fig. 11.

Differential effects of ABA treatment on bud dormancy release during the natural dormancy cycle. Single-node cuttings were prepared from canes that were harvested at seven sampling dates throughout the dormancy cycle as described in Fig. 8. Cuttings were grouped as described in Fig. 1 and used to analyse the effect of ABA on dormancy release of control and HC-treated buds as described in Fig. 2A. In parallel with the actual bud-break data (A–G), calculated values are presented as the difference in bud-break percentages between control and ABA-treated buds (H), and between HC- and ABA–HC-treated buds (I). These values represent the mean of differences for seven monitoring time points (7, 11, 14, 18, 21, 2, and 28 d) for each sampling date.