Crosstalk Pathway between Trehalose Metabolism and Cytokinin Degradation for the Determination of the Number of Berries per Bunch in Grapes

Abstract

In grapes, the number of flowers per inflorescence determines the compactness of grape bunches. Grape cultivars with tight bunches and thin-skinned berries easily undergo berry splitting, especially in growing areas with heavy rainfall during the grapevine growing season, such as Japan. We report herein that grape cytokinin oxidase/dehydrogenase 5 (VvCKX5) determines the number of berries per inflorescence in grapes. The number of berries per bunch was inversely proportional to the VvCKX5 expression level in juvenile inflorescences among the cultivars tested. VvCKX5 overexpression drastically decreased the number of flower buds per inflorescence in Arabidopsis plants, suggesting that VvCKX5 might be one of the negative regulators of the number of flowers per inflorescence in grapes. Similarly, the overexpression of grape sister of ramose 3 (VvSRA), which encodes trehalose 6-phosphate phosphatase that catalyzes the conversion of trehalose-6-phosphate into trehalose, upregulated AtCKX7 expression in Arabidopsis plants, leading to a decrease in the number of flower buds per Arabidopsis inflorescence. VvCKX5 gene expression was upregulated in grapevine cultured cells and juvenile grape inflorescences treated with trehalose. Finally, injecting trehalose into swelling buds nearing bud break using a microsyringe decreased the number of berries per bunch by half. VvCKX5 overexpression in Arabidopsis plants had no effect on the number of secondary inflorescences from the main inflorescence, and similarly trehalose did not affect pedicel branching on grapevine inflorescences, suggesting that VvCKX5, as well as VvSRA-mediated trehalose metabolism, regulates flower formation but not inflorescence branching. These findings may provide new information on the crosstalk between VvSRA-mediated trehalose metabolism and VvCKX-mediated cytokinin degradation for determining the number of berries per bunch. Furthermore, this study is expected to contribute to the development of innovative cultivation techniques for loosening tight bunches.

Keywords: berry number; cytokinin oxidase/dehydrogenase; grapevine; inflorescence; sister of ramosa3; trehalose.

Figure 1

Number of berries per bunch. Bunches were collected in the 2014 and 2018 growing seasons. Bars indicate means ± standard deviations of ten bunches. Different letters above the columns indicate statistically significant difference (p < 0.05, Dunnett’s test). PN, Pinot Noir. CH, Chardonnay. RIE, Riesling. KOS, Koshu. Phenotypes of inflorescence are indicated below the graph.

Figure 2

Upregulation of AtCKX7 in VvSRA-overexpressing plants. Total RNA was isolated from 12-day-old seedlings and subjected to real-time RT-PCR. Actin was used as an internal control. Data were calculated as gene expression level relative to actin gene expression level. Bars indicate means ± standard deviations of three seedlings. * p < 0.01 compared with wild type (Dunnett’s test). VvSRA-OE1, -OE2, and -OE3 were VvSRA-overexpressing Arabidopsis plants prepared in our previous study [13]. WT, wild type.

Figure 3

Relationship between number of berries per bunch and VvCKX expression in juvenile grape inflorescences. (A) VvCKX expression levels. Total RNA was isolated from juvenile grape inflorescences (approximately 5–10 mm in the longitudinal axis) of each cultivar and subjected to real-time RT-PCR. Actin was used as an internal control. Data were calculated as gene expression level relative to actin gene expression level. Bars indicate means ± standard deviations of three independent experiments. nd, not detected. (B) Regression lines between the number of berries per bunch (obtained from the 2018 growing season in Figure 1) and VvCKX1-like, VvCKX5, VvCKX7, or VvCKX9 expression in the juvenile inflorescences were drawn, respectively. Scatter plots of the number of berries per bunch versus each gene expression in the juvenile inflorescences were created. Regression lines and R-squared values were calculated from the scatter plots. PN, Pinot Noir. CH, Chardonnay. RIE, Riesling. KOS, Koshu.

Figure 4

Exogenous trehalose upregulates VvCKX5 gene expression in VR cells and juvenile grape inflorescences. (A) Transcription of VvCKX5 in VR cells treated with 5% trehalose. (B) Transcription of VvCKX5 in VR cells treated with 0.5, 1%, and 5% trehalose at 6 h post treatment. (C) Transcription of VvCKX5 in VR cells treated with 5% of each disaccharide at 6 h post treatment. (D) Transcription of VvCKX5 in juvenile grape inflorescences treated with 5% trehalose. VR cells and juvenile inflorescences from field-grown grapevines were treated with trehalose or other disaccharides. Total RNA was isolated from VR cells and inflorescences and subjected to real-time RT-PCR. Actin was used as an internal control. Data were calculated as gene expression level relative to actin gene expression level. Bars indicate means ± standard deviations of five independent experiments. * p < 0.05 compared with control (Student’s t-test for D). ** p < 0.01 compared with control (Student’s t-test for A or Dunnett’s test for B and C). Control, no treatment.

Figure 5

Exogenous NAA upregulated VvCKX5 gene expression in VR cells. VR cells were treated with excess amounts of NAA or kinetin. Total RNA was isolated from VR cells and subjected to real-time RT-PCR. Actin was used as an internal control. Data were calculated as gene expression level relative to actin gene expression level. Bars indicate means ± standard deviations of five independent experiments. ** p < 0.01 compared with control (Student’s t-test).

Figure 6

Phenotypes of Arabidopsis plants overexpressing VvCKX5. (A) VvCKX5 expression levels in VvCKX5-overexpressing plants. Three independent Arabidopsis lines overexpressing VvCKX5 (VvCKX5-OE3, VvCKX5-OE4, and VvCKX5-OE6) were obtained. Total RNA was isolated from rosette leaves of 36-day-old plants and subjected to real-time RT-PCR. Actin was used as an internal control. Data were calculated as gene expression level relative to actin gene expression level. Bars indicate means ± standard deviations of three independent plants. ** p < 0.01 as compared with wild type (Dunnett’s test). (B) Photograph of transgenic and wild type plants on day 44 after sowing. Scale bar = 5 cm. (C) Phenotype of transgenic plants on days 44 (D44) and 57 (D57) after sowing. Three plants were observed for each transgenic plant. Scale bar = 5 cm. (D) Plant height. Plant height was measured on day 44 (wild type) or 57 (transgenic) after sowing. Bars indicate means ± standard deviations of three independent plants. ** p < 0.01 as compared with wild type (Dunnett’s test). (E) Number of secondary inflorescences on wild type and transgenic plants. The number of secondary inflorescences on wild type and transgenic plants was measured on day 42 (wild type) or 57 (transgenic) after sowing, respectively. Bars indicate means ± standard deviations of three independent plants. WT, wild type. OE3, VvCKX5-OE3. OE4, VvCKX5-OE4. OE6, VvCKX5-OE6.

Figure 7

Overexpression of VvCKX5 decreases the number of flower buds per inflorescence in Arabidopsis plants. (A) Number of flower buds per inflorescence in transgenic and wild type plants. The number of flower buds per inflorescence in each plant was measured on day 42 (wild type) or 57 (transgenic) after sowing. Bars indicate means ± standard deviations of three independent plants. ** p < 0.01 as compared with wild type (Dunnett’s test). (B) Phenotypes of flowers on day 35 (wild type) or 52 (transgenic) after sowing. Scale bar = 1 mm. (C) Flower size. Sizes of flowers on inflorescences of wild and transgenic plants were measured on day 35 (wild type) or 52 (transgenic) after sowing, respectively. Bars indicate means ± standard deviations of five independent flowers. ** p < 0.01 as compared with wild type (Dunnett’s test). (D) Phenotypes of siliques on day 42 (wild) or 69 (transgenic) after sowing. Scale bar = 5 mm. (E) Silique length. Lengths of siliques on inflorescences of wild and transgenic plants were measured on day 42 (wild type) or 69 (transgenic) after sowing, respectively. Bars indicate means ± standard deviations of five independent siliques. ** p < 0.01 as compared with wild type (Dunnett’s test). WT, wild type. OE3, VvCKX5-OE3. OE4, VvCKX5-OE4. OE6, VvCKX5-OE6.

Figure 8

Effects of trehalose injection into buds nearing bud break on bunch characteristics in the 2018 growing season. Ten grapevine buds nearing bud break (Eichhorn–Lorenz Stage 3) were treated with 10% trehalose or water using a microsyringe or an atomizer on 4 April 2018. Young shoots (Eichhorn–Lorenz Stages 5–7) from the buds were again treated with trehalose or water using an atomizer 10 days post first treatment. Twenty bunches from ten shoots (two bunches/shoot) were collected at harvest (Eichhorn–Lorenz Stage 38). (A) Photographs of bunches at harvest. (B) Bunch weight. (C) Bunch length. (D) Number of berries per bunch. (E) Weight of ten berries. Bars indicate means ± standard deviations of twenty bunches. * p < 0.05 compared with control (Dunnett’s test). ** p < 0.01 compared with control (Dunnett’s test). Control, non-treated grapevines. H₂O-I, water-injected grapevines. H₂O-S, water-sprayed grapevines. Tre-I, trehalose-injected grapevines. Tre-S, trehalose-sprayed grapevines.

Figure 9

Effects of trehalose injection into buds nearing bud break on bunch characteristics in the 2020 growing season. Ten grapevine buds nearing bud break (Eichhorn–Lorenz Stage 3) were treated with 10% trehalose or water using a microsyringe or an atomizer on 8 April 2020. Young shoots (Eichhorn–Lorenz Stages 5–7) from the buds were again treated with trehalose or water using an atomizer 10 days post first treatment. Twenty bunches from ten shoots (two bunches/shoot) were collected at harvest (Eichhorn–Lorenz Stage 38). (A) Photograph of bunches at harvest. (B) Photograph of inflorescences. (C) Bunch weight. (D) Bunch length. (E) Number of berries per bunch. (F) Weight of ten berries. (G) Number of pedicels per inflorescence. Bars indicate means ± standard deviations of twenty bunches. * p < 0.05 compared with control (Dunnett’s test). ** p < 0.01 compared with control (Dunnett’s test). Control, non-treated grapevines. H2O-I, water-injected grapevines. Tre-I, trehalose-injected grapevines.

Figure 10

Effects of trehalose injection into buds nearing bud break on berry characteristics. Juices were obtained from each bunch (bunches are shown in Figure 8) by hand-pressing. Skins of ten berries from each bunch were peeled off with tweezers. (A) Soluble solids content. (B) Total acids content. (C) Anthocyanin content. Bars indicate means ± standard deviations of twenty bunches. ** p < 0.01 compared with control (Dunnett’s test). Control, non-treated grapevines. H₂O-I, water-injected grapevines. H₂O-S, water-sprayed grapevines. Tre-I, trehalose-injected grapevines. Tre-S, trehalose-sprayed grapevines.