Latent buds and the shoot apexes from 8-year-old Vitis vinifera cv. Thompson seedless grown at the experimental station at the Chilean National Institute of Agriculture Research (INIA) in Santiago, Chile (33°34′S latitude) were collected on the same dates 7 March because it has been previously reported that the transition from paradormancy (PD) to ED occurs in mid-January (Kühn et al., 2009). Grape somatic embryos (GSE) used as plant material were initiated from a leaf disk of in vitro micro-propagated Vitis vinifera cv. Thompson seedless plantlets. NB2 culture medium (Dhekney et al., 2009) was used for callus micro-propagation, and cultures were maintained in the dark at 24°C until the formation of callus, when GSE were separated from the callus tissue and transferred to X6 medium and sub-cultured monthly. All samples were frozen in liquid N₂ and stored at -80°C until used.

For ABA treatment, GSE, the shoot apexes and latent buds of grapevines cv. Thompson seedless were sprayed with ABA (Sigma–Aldrich, United States) at concentrations of 10 and 100 μM, respectively, and water was used as a control. For GSE experiments, 0.3 g (0.1 g for each replicate) of GSE, mostly in the globular to heart-shaped stage, were transferred from subculture of semi-solid growth regulator-free X6 medium to liquid X6 medium. The effect of ABA (10 μM) on GSE was tested in a final volume of 30 ml of liquid X6 medium. Samples were shaken at 115 rpm for 72 h, and afterward, the liquid medium was discarded, and the GSE were frozen at –80°C until used. Shoot apexes were obtained from one month shoots from single-bud cuttings grown in a growth chamber and sprayed with 100 μM ABA. Samples were collected 24 h after the treatment, frozen in liquid N₂ and stored at –80°C until used. Latent buds collected on March 23 were used to measure the content of ABA within the buds 48 h after the application of a 100 μM ABA solution, 2.5% HC solution and co-application of ABA and HC solutions.

The effect of the dormancy-breaking compound HC (Sigma–Aldrich, United States) on the expression of CCG and ABA metabolism-related genes was studied on single-bud cuttings of Thompson seedless grapevines. Canes collected on 8 June, 2015 were cut into single-bud cuttings. Cuttings (10–12 cm length) were sprayed with an aqueous solution containing 2.5% HC, and water was used as a control. Subsequently, the cuttings were placed in a growth chamber set at 23 ± 1°C with a 16 h photoperiod; 48 h after HC application, samples were taken for gene expression analysis.

Fresh plant material (10 buds and approximately 0.1 g of shoot apex for each biological replicate) was powdered in liquid nitrogen. The samples were extracted in a shaker for 1 h at 4°C and 10 min by ultra-sonication with 3 ml of 80% methanol containing 1% acetic acid and 3 ng of 2,3,5-triiodobenzoic acid (TIBA) as an internal standard (Sigma–Aldrich, United States). The extracts were centrifuged at 3000 × g for 10 min, and the supernatant was filtered through glass wool and a Sep-Pack C₁₈ cartridge (Waters Assoc., Milford, MA, United States) that had been prewashed with 5 ml of 80% methanol. The procedure was repeated twice, and the filtrate was evaporated to dryness. After evaporation, the residue was dissolved in 1.2 ml ethyl acetate, and 1.2 ml of 0.5 M KH₂PO₄ pH 3.0 was added. The mixture was agitated with a vortex and centrifuged at 3000 rpm for 3 min. After centrifugation, the ethyl acetate layer was removed and placed in a clean tube. The aqueous phase was extracted once more with 1.2 ml of ethyl acetate. The collected ethyl acetate layers (2.4 ml) were evaporated to dryness. The samples were dissolved in 3 ml of ethylic ether and methylated with diazomethane in a diazomethane-generator apparatus according to manufacturer’s instructions. After methylation, the ether solution was evaporated to dryness, and the residue was dissolved in 50 μl methanol.

A Shimadzu gas chromatograph, model GC-2014, equipped with an electron capture detector (ECD-2014, Shimadzu Corporation, Kyoto, Japan) and computer integrator was used for ABA determination (Vermeer et al., 1987). A CBP1 capillary column (25 m×0.25 mm I.D.) with helium as the carrier gas at a flux of 1.5 ml min^-1 was used. The initial temperature of the column was 80°C and after 1 min was raised to 270°C at a rate of 20°C per min and maintained for 5 min. The injector was operated in the “splitless” mode at 225°C, and the temperature of the detector was 300°C. A calibration curve for ABA methyl ester and TIBA methyl ester was constructed. Endogenous ABA content was calculated from the peak area ratios of endogenous ABA to internal TIBA standard (Vermeer et al., 1987).

For gene expression analysis, total RNA was isolated and purified from GSE (0.1 g fr.wt), shoot apexes (0.1 g fr.wt) and latent buds (0.5 g fr.wt) of Vitis vinifera cv. Thompson seedless. In all cases, total RNA was extracted and purified using a modification of the method of Chang et al. (1993), as described in Noriega et al. (2007). DNA was removed by treatment with RNase-free DNase (1 U/μg) (Thermo Scientific, United States) at 37°C for 30 min. First-strand cDNA was synthesized from 1 μg of purified RNA with 1 μL oligo (dT)_12-18 (0.5 μg × μL^-1) as a primer, 1 μL dNTP mix (10 mM) and Superscript^® II RT (Invitrogen, CA, United States).

Quantitative real-time PCR (RT-qPCR) was carried out in an Eco Real-Time PCR system (Illumina, Inc. San Diego, United States) using KAPA SYBR FAST (KK 4602) qPCR Master Mix (2 X). Primers for CDKs (VvCDKA, VvCDKB1, VvCDKB2), CDK inhibitors (VvICK5, VvICK7), ICK-like genes (VvICK3-like; VvICK7-like), type A cyclins (VvCYCA1, VvCYCA2, VvCYCA3), type B cyclins (VvCYCB), type D cyclins (VvCYCD3.1, VvCYCD3.2a, VvCYCD3.2b) and ABA metabolism-related genes (VvNCED1, VvNCED2, VvA8H3) were designed using the PRIMER3 software (Rozen and Skaletsky, 2000) (Supplementary Table S1). cDNA was amplified under the following conditions: denaturation at 94°C for 2 min and 40 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 45 s. Relative changes in gene expression levels were determined using the 2^-ΔΔCT method (Livak and Schmittgen, 2001). Each reaction was performed in at least three biological replicates, each with three technical replicates. VvUBIQUITIN and/or VvACTIN were used as a reference gene for normalization

Differences in gene expression at different times were analyzed by ANOVA, and multiple comparison analysis was carried out by Dunnett’s and Tukey’s methods (α = 0.05). For pairwise comparisons, the Student’s t-test (α = 0.05) method was used.

In grapevines, transcript levels of the ABA biosynthesis genes VvNCED1 and VvNCED2 were significantly more highly expressed in the latent buds than at the shoot apexes (Figure 1A), while transcript levels of ABA catabolic gene VvA8H3 showed no significant difference between either organ (Figure 1A). As expected, the content of ABA was higher in the latent buds than in the shoot apexes (Figure 1B).

The following cell cycle-related genes have been identified in the Vitis genome: three CDK, VvCDKA, VvCDKB1, and VvCDKB2; two CDK inhibitors, VvICK5, VvICK7; two ICK-like genes, VvICK3-like, VvICK7-like; three type A cyclins, VvCYCA1, VvCYCA2, and VvCYCA3; one type B cyclin, VvCYCB; and three type D cyclins, VvCYCD3.1, VvCYCD3.2a, and VvCYCD3.2b. Transcript levels of CDKs VvCDKB1 and VvCDKB2 were expressed more highly at the shoot apex than at the latent bud, while VvCDKA was expressed similarly in both organs (Figure 2A). Expression analysis of type A cyclins, VvCYCA1, VvCYCA2, VvCYCA3; type B cyclin, VvCYCB; and type D cyclins, VvCYCD3.1, VvCYCD3.2a, VvCYCD3.2b, showed that all of them were expressed more at the shoot apex than at the latent buds (Figure 2B). In contrast, the transcript levels of the CDK inhibitors VvICK5, VvICK7 and VvICK7-like were higher in the latent buds than at the shoot apexes, while VvICK3-like was expressed more at the shoot apex (Figure 2C).

To verify whether ABA affects the expression of CCGs in meristematic tissues of grapevines, we analyzed its effect on the expression of the CDKs VvCDKA, VvCDKB1, VvCDKB2 (Figure 3A), the type A cyclins VvCYCA1, VvCYCA2, VvCYCA3, the type B cyclin VvCYCB, and the type D cyclins VvCYCD3.1, VvCYCD3.2a, VvCYCD3.2b (Figure 3B) and on the CDK inhibitors VvICK5 and VvICK7 and two ICK-like genes, VvICK3-like and VvICK7-like genes (Figure 3C) in GSE. The results indicated that ABA inhibits cell cycle progression since most of the genes coding for CDKs (VvCDKs) and for cyclins (VvCYCs) were down-regulated by the plant hormone (Figures 3A,B), while the expression of a CDK inhibitor (VvICK5) was up-regulated (Figure 3C). Similar results were obtained when ABA was applied to other organs containing meristems, such as the shoot-apex (Supplementary Figure S1).

It has been reported that the content of ABA increases with the development of ED in grapevine buds (Zheng et al., 2015). To test whether this increase is consistent with a decrease in the expression of CCG, transcript levels of CCG taken from microarray data on the development of Tempranillo grape buds grown in the Northern Hemisphere (Díaz-Riquelme et al., 2012) were plotted throughout the ED period (Figures 4A–C). Furthermore, to verify the data of Díaz-Riquelme et al. (2012), we used RT-qPCR to analyze the transcript levels of CCG in buds of Thompson seedless grapevines grown in the Southern Hemisphere (Figures 4D–F). The results showed that in either cases, the expression of VvCDKs (Figures 4A,D) and VvCYCs (Figures 4B,E) decreased with the development of the ED, while the expression of VvICK5 increased, and the expression of VvICK7 decreased while VvICK7-like and VvICK3-like remained relatively unchanged (Figures 4C,F).

To test whether HC reduces the content of ABA in grapevine buds, single-bud cuttings of cv. Thompson seedless were sprayed with a 2.5% solution of HC, and 48 h after treatment, analysis was conducted for the ABA biosynthesis-related gene VvNCED2, the ABA catabolism-related gene VvA8H3 and ABA content. The results showed that HC down-regulated VvNCED2 (Figure 5A) and up-regulated VvA8H3 gene expression (Figure 5B) and ABA content was drastically reduced (Figure 5C).

To test whether HC up-regulates the expression of CCG in grapevine buds, single-bud cuttings of cv. Thompson seedless were sprayed with a 2.5% solution of HC, and 48 h after treatment, gene expression analysis was conducted. The results showed that HC up-regulated the expression of the CDK VvCDKB1 (Figure 6A) and of the A-type cyclins VvCYCA2 and VvCYCA3; however, the expression of VvCYCDs was inhibited (Figure 6B), and the expression of CDK inhibitors, VvICK5 and VvICK7, was down-regulated (Figure 6C).

To determine the content of ABA within the buds after spraying them with the following solutions; 100 μM ABA, 100 μM ABA + 2.5% HC, 2.5% HC, single-bud cuttings harvested on March 23 were sprayed with the above solutions and water as control, 48 h after the treatment, analysis was conducted to determine the ABA content. The results indicated that the application of 100 μM ABA doubled the concentration of ABA in relation to control buds, while the application of HC, reduced the concentration of ABA to half of the concentration of the control, and co-application of ABA and HC resulted in ABA content similar to control buds (Figure 7A). To test if the modifications produced by HC on the expression of CCG are due to their reducing effect on the ABA content, single-bud cuttings of grapevines were sprayed with the following solutions 2.5% HC; 2.5% HC + 100 μM ABA and water as control. Expression of genes that were up-regulated by HC were analyzed 48 h after treatment. The results showed that transcript levels of VvCDKB1, VvCYCA3, and VvICK5 remained similar to control buds after the co-application of HC and ABA (Figures 7B–D).