Changes in abscisic acid metabolism in relation to the maturation of grapevine (Vitis vinifera L., cv. Mencía) somatic embryos

Abstract

Background: Somatic embryogenesis in grapevines is a complex process that depends on many physiological and genetic factors. One of its main limitations is the process of precocious germination of the somatic embryos in differentiation medium. This process lowers plant conversion rates from the somatic embryos, and it is probably caused by a low endogenous abscisic acid (ABA) content.

Results: Precocious germination of the somatic embryos was successfully avoided by culturing grapevine cv. Mencía embryogenic aggregates over a semipermeable membrane extended on top of the differentiation medium. The weekly analysis of the endogenous ABA and ABA-glucosyl ester (ABA-GE) contents in the aggregates showed their rapid accumulation. The expression profiles of 9-cis-epoxycarotenoid dioxygenase (VvNCED1), 8'-hydroxylase (VvHyd2), UDP-glucosyltransferase (VvUGT) and β-glucosidase (VvBG2) genes in grapevine revealed that the occurrence of a first accumulation peak of endogenous ABA in the second week of culture over the semipermeable membrane was mainly dependent on the expression of the VvNCED1 gene. A second increase in the endogenous ABA content was observed in the fourth week of culture. At this point in the culture, our results suggest that of those genes involved in ABA accumulation, one (VvNCED1) was repressed, while another (VvBG2) was activated. Similarly, of those genes related to a reduction in ABA levels, one (VvUGT) was repressed while another (VvHyd2) was activated. The relative expression level of the VvNCED1 gene in embryogenic aggregates cultured under the same conditions and treated with exogenous ABA revealed the significant downregulation of this gene.

Conclusions: Our results demonstrated the involvement of ABA metabolism in the control of the maturation of grapevine somatic embryos cultured over a semipermeable membrane and two important control points for their endogenous ABA levels. Thus, subtle differences in the expression of the antagonistic genes that control ABA synthesis and degradation could be responsible for the final level of ABA during the maturation of grapevine somatic embryos in vitro. In addition, the treatment of somatic embryos with exogenous ABA suggested the feedback-based control of the expression of the VvNCED1 gene by ABA during the maturation of grapevine somatic embryos.

Keywords: ABA metabolism; Cellulose semipermeable membrane; Gene expression; Reference genes; Somatic embryogenesis; qPCR.

Fig. 1

Effect of the sucrose concentration in the DM1 differentiation medium on the developmental stage of grapevine cv. Mencía somatic embryos. Data are shown as percentages (mean ± standard error) of the total number of embryos. Different letters represent significant differences among concentrations of sucrose in the medium for the same developmental stage (Mann-Whitney U test, p < 0.05)

Fig. 2

Grapevine cv. Mencía somatic embryo aggregates cultured on DM1 differentiation medium. a Somatic embryo aggregates showing asynchronic development after 30 days of culture. Arrowhead: Precociously germinated somatic embryo. Bar: 3 mm. b Developmental stages of somatic embryos cultured on DM1 differentiation medium. From left to right, heart stage, torpedo stage, cotyledonary stage, and precociously germinated embryo. Bar: 1 mm

Fig. 3

Relative water content of grapevine somatic embryo aggregates cultured for 5 weeks on DM1 solid medium supplemented with different amounts of sucrose (1, 3 or 6%) and over a semipermeable membrane on DM1 medium with 6% sucrose. Values (means ± standard errors) are shown as percentages of water loss ((fresh weight-dry weight) x (100/fresh weight)). Different letters indicate significant differences (Kruskal-Wallis test with Bonferroni post hoc test, p < 0.05) among treatments in the same week of culture

Fig. 4

Differentiation of grapevine cv. Mencía somatic embryo aggregates cultured in DM1 medium over a semipermeable cellulose acetate membrane. a General view of a 9 cm-diameter Petri plate containing DM1 differentiation medium with somatic embryo aggregates just placed over the semipermeable membrane. b Detail of the somatic embryo aggregates cultured on DM1 differentiation medium over the semipermeable membrane. Bar: 0.5 mm. c Growth of the somatic embryo aggregates after 30 days of culture on DM1 differentiation medium over the semipermeable membrane. d Normal development of somatic embryos on DM1 differentiation medium over the semipermeable membrane. Bar: 1 mm

Fig. 5

Time course of endogenous concentrations (mean ± standard error) of ABA (a) and ABA-GE (b) in grapevine cv. Mencía somatic embryo aggregates cultured for 5 weeks on DM1 differentiation medium supplemented with 1, 3 or 6% sucrose and over a semipermeable membrane on DM1 medium with 6% sucrose. The insert in (b) shows the time course of the endogenous concentration of ABA-GE in grapevine somatic embryo aggregates cultured for 5 weeks over a semipermeable membrane on DM1 medium with 6% sucrose. Different letters indicate significant differences (Student-Newman-Keuls test, p < 0.05)

Fig. 6

Relative expression profiles of ABA metabolism genes (a, VvNCED1; b, VvHyd2; c, VvUGT; and d, VvBG2) in grapevine cv. Mencía somatic embryo aggregates cultured for 5 weeks on DM1 differentiation medium supplemented with 6% sucrose or over a semipermeable membrane on the same DM1 medium with 6% sucrose. Somatic embryo aggregates collected just at the time of transfer to DM1 medium (nontreated samples) were used as the calibration group for the relative expression analyses. GAPDH(m) and EF1-α(m) were used as the reference genes for normalization. Data are represented as the means (± standard error) of two independent experiments. Asterisks indicate statistically significant differences between the calibrator group and the analyzed group, whereas arrows indicate statistically significant differences between treatments with or without the membrane, which in both cases were calculated using REST-2009© software (p < 0.05)

Fig. 7

Effect of exogenous ABA on the relative expression of the VvNCED1 gene in grapevine cv. Mencía somatic embryo aggregates cultured for 2 weeks over a semipermeable membrane on DM1 medium supplemented with 6% sucrose and then treated by pipetting directly onto the aggregates either a solution of 30% sucrose or a solution containing 30% sucrose and 10 μM ABA. The expression levels were calculated relative to the VvNCED1 gene expression in somatic embryo aggregates cultured under the same conditions but untreated with any solution (calibrator group). GAPDH (m) and EF1-α (m) were used as the reference genes for normalization. Data are presented as the means of two independent experiments with standard errors. The asterisk indicates a significant difference between the untreated samples and the treated samples (p < 0.05), as calculated using REST-2009© software