Functional Characterization of VvSK Gene Family in Grapevine Revealing Their Role in Berry Ripening

The glycogen synthase kinase 3/shaggy kinase (GSK3) is a serine/threonine kinase that plays important roles in brassinosteroid signaling, abiotic stress responses, cell division, and elongation, etc. In this study, we characterized seven grape GSK3 genes, showing high similarities with homologs from other species including Arabidopsis, white pear, apple, orange, and peach. Gene chip microarray data derived from an online database revealed very diverse developmental and tissue-specific expression patterns of VvSKs. VvSK3 and VvSK7 showed much higher expression levels in almost every tissue compared with other members. VvSK7 was highly enriched in young tissues like berries before the veraison stage, young leaves and green stems, etc., but immediately downregulated after these tissues entered maturation or senescence phases. Prediction of cis-elements in VvSK promoters indicated that VvSKs might be sensitive to light stimulation, which is further confirmed by the qPCR data. Constitutive overexpression of VvSK7 in Arabidopsis leads to dwarf plants that resembles BR-deficient mutants. The photosynthetic rate was significantly reduced in these plants, even though they accumulated more chlorophyll in leaves. Transient overexpression of VvSKs in tomatoes delayed the fruit ripening process, consistent with the observation in grapevine which blocks VvSKs by EBR- or BIKININ-promoted berry expansion and soluble solids accumulation. Data presented in the current study may serve as a theoretical basis for the future application of BRs or related compounds in quality grape production.


Introduction
Glycogen synthase kinase 3 (GSK3) is a highly conserved serine/threonine kinase, implicated in many vital signal transduction pathways in eukaryotes. It was initially found in mammals, having two forms of GSK3, namely, GSK3α and GSK3β, which are described as key enzymes involved in glycogen metabolism [1,2]. Later studies discovered that GSK3 homologs also play crucial roles in various physiological processes in animals, such as protein synthesis, tumorigenesis, regulation of transcription factor activity, determination of cell fate, and glycogen metabolism, etc. [3][4][5][6]. Additionally, GSK3 is an essential component in the animal Wnt signaling pathway [7,8].
Plant GSK3s have more homologs with much diverse functions compared to the corresponding genes in animal. In plants, GSK3s function as negative regulators in brassinosteroid (BR) signaling. GSK3s phosphorylate and inhibit the transcription factors BZR1 and its homolog BES1, which control the BR-responsive gene expression. Perception of BRs through the receptor complex BRI1-BAK1 leads to deactivation and degradation of GSK3s, thus alleviating their inhibition to BZR1/BES1 and inducing BR-responsive gene expression [9,10]. Through regulation of downstream BR signaling or direct binding and phosphorylation of key components in other biological pathways, plant GSK3s are involved in diverse biological processes, including flower and reproductive organ development, stomatal development and stress responses, etc. In Arabidopsis, AtSK11 and AtSK12 showed specific strong expression in early floral meristems and restricted to sepal primordia, petals, carpels and the pollen-containing regions of the anthers in the later stage. Antisense reduction in either AtSK11 or AtSK12 transcript levels results in disrupted cell division in the floral meristem and leads to abnormal sepals, petals, and carpel development. AtSK31 may also involve in reproductive organ development as AtSK31 proteins highly enriched in gametophytes, floral organs, and embryos [11][12][13]. Arabidopsis AtSK21 directly phosphorylate and inhibit MAPKKK YODA (YDA) and MKK4/MKK5, which are the key components of the MAPK signaling that controls stomatal development and patterning [14]. Interestingly, AtSK21 also phosphorylates and inhibits transcriptional factor SPCH, which controls the initiation of stomatal development [15]. Work in several plant species has implicated GSK3s in abiotic stress responses. Transcripts of multiple GSK3s from Arabidopsis, rice, and wheat can be induced under salt stress conditions [2]. Overexpressing AtSK22 in Arabidopsis plants induced a marked upregulation of several salt stress-responsive genes, even in the absence of high salinity [16]. In rice, overexpressing OsGSK5 increased salinity tolerance in part via priority carbon allocation to root starch [17]. Given the dual roles of BR in plant development, GSK3s may involve in more diverse and general biological processes that need to be elucidated in the future.
The grape is one of the world-wide largest fruit crops. Studies of GSK3s in the grape are very limited even though brassinosteroid-like hormones have already been commercially used in grape production for decades. In the previous studies, only one grape GSK3 gene VvSK1 was identified and characterized. VvSK1 was strongly expressed at the post-veraison stage of berries, when the berries start to accumulate glucose, fructose and abscisic acid. Overexpressing VvSK1 in grapevine cells highly induced the expression of four monosaccharide transporters, namely, VvHT3, VvHT4, VvHT5, and VvHT6, and significantly promoted glucose and sucrose accumulation [18]. Considering VvSKs may involve in the regulation of key flavor and quality traits in grape berries, and the wide application of brassinosteroid-like hormones in grape production, systematic identification and characterization of gape GSK3s is necessary and of great importance for grape researchers and farmers.
In the present study, we employed bioinformatics approaches to identify and characterize grape GSK3s on a genome-wide scale based on several publicly available database. Furthermore, we investigated their spatiotemporal expression profiles in different tissues and developmental stages, by which we selected VvSK7 as a study candidate to further elucidate the putative roles of VvSKs in plant development by constitutively over-expression in Arabidopsis plant. These comprehensive results will provide an insight for further studies and assist in a better understanding of the potential functions of VvSKs in grapevine.

Phylogenetic and Conserved Structural Analysis of VvSKs
There are 7 VvSK genes named VvSK1, VvSK2, VvSK3, VvSK4, VvSK5, VvSK6 and VvSK7 that were reported in the grapevine genome [18]. The basic information, including gene ID, genomic DNA size, amino acid numbers, and theoretical PI are listed in Table S1. The phylogenetic tree was constructed using SK coding sequences (CDS) from Arabidopsis and different fruit species including grape, pear, apple, and peach, etc. (Figure 1). According to the relative divergence times, grape SK members are clustered into three subgroups (relative divergence times were higher than 150). VvSK5 and VvSK4 are clustered into subgroup I and subgroup III, respectively, while most of the VvSKs, including VvSK1, VvSK2, VvSK3, VvSK6 and VvSK7, are clustered into subgroup II. GSK genes from different fruit species are closely correlated. Overall, grape SKs showed a closer evolutional relationship with GSKs from citrus and peach. VvSK5 and VvSK7 are closely clustered with CsSK5 and CsSK7 respectively. VvSK1, VvSK2, and VvSK3 showed relatively close kinship with PpSK2, and VvSK4 showed relatively close kinship with PpSK5 and PpSK6. (Figure 1). Conserved motif prediction revealed that SKs from different species exhibited high similarity in motif composition. Except for motifs 8 and 9, most of the motifs are found in all analyzed SKs, indicating the high conservation in protein structures among these SKs (Figure 1, Figure S1). and CsSK7 respectively. VvSK1, VvSK2, and VvSK3 showed relatively close kinship with PpSK2, and VvSK4 showed relatively close kinship with PpSK5 and PpSK6. (Figure 1). Conserved motif prediction revealed that SKs from different species exhibited high similarity in motif composition. Except for motifs 8 and 9, most of the motifs are found in all analyzed SKs, indicating the high conservation in protein structures among these SKs (Figure 1, Figure S1).

Figure 1.
The phylogenetic tree and conserved motif analysis of GSK genes. SKs from Vitis vinifera (red mark), Arabidopsis thaliana, Pyrus bretschneideri, Prunus persica, Malus domestica and Citrus reticulata were put together for comparison. The phylogenetic tree was generated by MEGA 6.0 using the maximum likelihood method (time tree). The axis numbers mean the relative divergence time. Ten motifs named motif 1-10 were analyzed using the online program "MEME".

Syntenic Analysis of VvSKs
In order to analyze collinearity of SK genes in different species, syntenic analysis of SKs from A. thaliana, V. vinifera, and C. sinensis was performed. There are 13 collinear gene pairs were generated among grapevine, citrus, and Arabidopsis plants ( Figure 2, Table S2). The result suggests that VvSK1 and ASK2 might share a common ancestor. VvSK2 originated together with VvSK3, and VvSK3 might have the same origin with CsSK3 and ASK3. VvSK4 might share the same ancestor with CcSK4 and ASK5. VvSK5 might have the same ancestor with CsSK5, CsSK6, ASK7 and VvSK6. VvSK6 might be derived from a common ancestor with CsSK5. The origin of VvSK7 might be same with ASK4 and ASK9. The study of the collinear relationship of SKs in these plants showed vital implications for the functional study of VvSKs. The phylogenetic tree and conserved motif analysis of GSK genes. SKs from Vitis vinifera (red mark), Arabidopsis thaliana, Pyrus bretschneideri, Prunus persica, Malus domestica and Citrus reticulata were put together for comparison. The phylogenetic tree was generated by MEGA 6.0 using the maximum likelihood method (time tree). The axis numbers mean the relative divergence time. Ten motifs named motif 1-10 were analyzed using the online program "MEME".

Syntenic Analysis of VvSKs
In order to analyze collinearity of SK genes in different species, syntenic analysis of SKs from A. thaliana, V. vinifera, and C. sinensis was performed. There are 13 collinear gene pairs were generated among grapevine, citrus, and Arabidopsis plants ( Figure 2, Table S2). The result suggests that VvSK1 and ASK2 might share a common ancestor. VvSK2 originated together with VvSK3, and VvSK3 might have the same origin with CsSK3 and ASK3. VvSK4 might share the same ancestor with CcSK4 and ASK5. VvSK5 might have the same ancestor with CsSK5, CsSK6, ASK7 and VvSK6. VvSK6 might be derived from a common ancestor with CsSK5. The origin of VvSK7 might be same with ASK4 and ASK9. The study of the collinear relationship of SKs in these plants showed vital implications for the functional study of VvSKs.

Transcriptional Profiling of VvSKs in Different Organs and Developmental Stages in Grapevine
To understand the spatiotemporal expression profiles of VvSKs, the global transcriptomic data of developmental phases of 24 different tissues and organs including berry, leaf, flower, stem, root, and seed were retrieved from NCBI (GSE36128) [19]. The results ( Figure 3) showed that different VvSK members have very diverse developmental and tissue-specific expression patterns. Transcription of VvSK3 and VvSK7 were highly expressed in almost every tissue. VvSK3 showed especially high expression levels in buds and reproductive tissues, including inflorescences and flowers. The expression of VvSK7 was more diverse in different tissues. Interestingly, this gene showed very high expression levels in most of the young tissues, such as berries (including pericarp and seeds) at fruit set, post-fruit set and veraison stages, leaves before maturation, swelling buds and green stems, but very low expression levels in mature tissues such as ripening berries and senescing leaves. VvSK2 and VvSK4 showed very weak expression levels in almost every tissue. VvSK1, VvSK5, and VvSK6 showed moderate expression levels in different tissues. VvSK6 showed complementary expression patterns with VvSK7, e.g., VvSK7 showed very low expression level in senescing leaves while VvSK6 was highly enriched in this tissue.

Transcriptional Profiling of VvSKs in Different Organs and Developmental Stages in Grapevine
To understand the spatiotemporal expression profiles of VvSKs, the global transcriptomic data of developmental phases of 24 different tissues and organs including berry, leaf, flower, stem, root, and seed were retrieved from NCBI (GSE36128) [19]. The results ( Figure 3) showed that different VvSK members have very diverse developmental and tissue-specific expression patterns. Transcription of VvSK3 and VvSK7 were highly expressed in almost every tissue. VvSK3 showed especially high expression levels in buds and reproductive tissues, including inflorescences and flowers. The expression of VvSK7 was more diverse in different tissues. Interestingly, this gene showed very high expression levels in most of the young tissues, such as berries (including pericarp and seeds) at fruit set, post-fruit set and veraison stages, leaves before maturation, swelling buds and green stems, but very low expression levels in mature tissues such as ripening berries and senescing leaves. VvSK2 and VvSK4 showed very weak expression levels in almost every tissue. VvSK1, VvSK5, and VvSK6 showed moderate expression levels in different tissues. VvSK6 showed complementary expression patterns with VvSK7, e.g., VvSK7 showed very low expression level in senescing leaves while VvSK6 was highly enriched in this tissue.

Cis-Elements in the Promoter of SK Genes
To further clarify the gene function and transcriptional regulation mechanism of SKs, cis-elements in the promoter region of SK genes from different species were predicted using an online database "PlantCARE". Cis-elements related to light response, MeJA response, anaerobic induction, and gibberellin response were the most abundant ones in these SKs promoters ( Figure 4). Photoreaction related cis-elements such as Box 4, GATA-motif, GT1-motif, AE-box, G-box, MRE, TCCC-motif and TCT-motif were highly enriched in most of the SKs and especially in VvSKs, indicating VvSKs would be involved in photosynthesis or in light responsive signaling.

Cis-Elements in the Promoter of SK Genes
To further clarify the gene function and transcriptional regulation mechanism of SKs, ciselements in the promoter region of SK genes from different species were predicted using an online database "PlantCARE". Cis-elements related to light response, MeJA response, anaerobic induction, and gibberellin response were the most abundant ones in these SKs promoters ( Figure 4). Photoreaction related cis-elements such as Box 4, GATA-motif, GT1-motif, AE-box, G-box, MRE, TCCC-motif and TCT-motif were highly enriched in most of the SKs and especially in VvSKs, indicating VvSKs would be involved in photosynthesis or in light responsive signaling.

VvSKs are Sensitive to Dark Treatment
Light-responsive cis-elements were highly enriched in promoter regions of VvSKs (Figure 4), indicating that these genes could be sensitive to light or dark stimulation. To verify these light-responsive cis-elements, transcriptional levels of VvSKs under dark treatment were analyzed. The results ( Figure 5) showed that all the VvSKs exhibited light-responsive patterns. Expression levels of VvSKs in subgroup II including VvSK1, VvSK2, VvSK3, and VvSK6 dramatically decreased upon 8-h dark treatment. In contrast, expression levels of VvSKs in subgroup I (VvSK5) and subgroup III (VvSK4) were induced upon dark treatment. These contradictory responsive patterns indicate that VvSKs in subgroup II may function complementary with VvSKs in subgroup I and III in light-related photoreaction or signaling pathways.

VvSKs are Sensitive to Dark Treatment
Light-responsive cis-elements were highly enriched in promoter regions of VvSKs (Figure 4), indicating that these genes could be sensitive to light or dark stimulation. To verify these lightresponsive cis-elements, transcriptional levels of VvSKs under dark treatment were analyzed. The results ( Figure 5) showed that all the VvSKs exhibited light-responsive patterns. Expression levels of VvSKs in subgroup II including VvSK1, VvSK2, VvSK3, and VvSK6 dramatically decreased upon 8-h dark treatment. In contrast, expression levels of VvSKs in subgroup I (VvSK5) and subgroup III (VvSK4) were induced upon dark treatment. These contradictory responsive patterns indicate that VvSKs in subgroup II may function complementary with VvSKs in subgroup I and III in light-related photoreaction or signaling pathways.

Constitutive Over-Expression of VvSK7 Affects BR Signaling and Inhibits Plant Photosynthesis
To further verify the function of VvSKs, constitutive over-expression of VvSK7 driven by cauliflower mosaic virus promoter (CaMV 35S) in Arabidopsis Col-0 was generated. Two individual transgenic lines of L1 and L2 with 13 or 30-folds over-expression of VvSK7, respectively, were examined ( Figure 6b). Four-week old plants of 35S:VvSK7-GFP_L1 showed roundish and compact rosette leaves, and shorter petioles compared with Col-0. 35S:VvSK7-GFP_L2 with higher expression of VvSK7 showed much severe phenotype than L1. Except for the roundish and compact rosette leaves, L2 plants are dwarf and showed stunted growth. These phenotypes resemble AtSK21 gain of function [20] or BR-deficient mutants like dwf5-7 [21] in Arabidopsis or AtSK gain of function mutantlike bin2-1 [22] (Figure 6a). Moreover, expression levels of BR-signaling marker genes DWF4 and CPD were all increased greatly in L1 and L2 plants (Figure 6c), indicating that BR signaling was blocked in VvSK7-OE plants. Besides, VvSK7-OE plants also showed dark green colored leaves. The quantification showed that their total chlorophyll content, especially in the L2 plant, was significantly

Constitutive Over-Expression of VvSK7 Affects BR Signaling and Inhibits Plant Photosynthesis
To further verify the function of VvSKs, constitutive over-expression of VvSK7 driven by cauliflower mosaic virus promoter (CaMV 35S) in Arabidopsis Col-0 was generated. Two individual transgenic lines of L1 and L2 with 13 or 30-folds over-expression of VvSK7, respectively, were examined ( Figure 6b). Four-week old plants of 35S:VvSK7-GFP_L1 showed roundish and compact rosette leaves, and shorter petioles compared with Col-0. 35S:VvSK7-GFP_L2 with higher expression of VvSK7 showed much severe phenotype than L1. Except for the roundish and compact rosette leaves, L2 plants are dwarf and showed stunted growth. These phenotypes resemble AtSK21 gain of function [20] or BR-deficient mutants like dwf5-7 [21] in Arabidopsis or AtSK gain of function mutant-like bin2-1 [22] (Figure 6a). Moreover, expression levels of BR-signaling marker genes DWF4 and CPD were all increased greatly in L1 and L2 plants (Figure 6c), indicating that BR signaling was blocked in VvSK7-OE plants. Besides, VvSK7-OE plants also showed dark green colored leaves. The quantification showed that their total chlorophyll content, especially in the L2 plant, was significantly higher than that in Col-0 plants (Figure 6d). Interestingly, even though chlorophyll was more accumulated in VvSK7-OE plants, photosynthetic rate in these plants was reduced compared with Col-0 plants (Figure 6e). These results suggested that the BR hormone could be involved in photosynthesis in grapevine under the modulation of VvSKs. . The asterisks indicated statistically significant differences compared with Col-0, which were determined by two-tailed Student's t-test. * represents p ≤ 0.05, ** represents p ≤ 0.01, ns represents no significant difference.

VvSKs Inhibit Fruit Ripening in Tomato
The BR hormone is involved in fruit ripening in strawberry [23], but whether this is also the case in grapevine and whether this process is mediated by VvSKs is yet to be verified. In order to corroborate these questions, Agrobacterium tumefaciens strain GV3101 cultures containing 35S: VvSKs-GFP or the empty vector pK7FWG2 (control) were infiltrated into mature green tomato fruits of Lycopersicon esculentum cv. Moneymaker. Tomato fruits were characterized 4 days after infiltration. Real-time PCR results showed that transcript levels of VvSKs showed at least a 5-fold upregulation compared with the control fruits 5 days after infiltration (Figure 7b). The surface color of control fruits  3). The asterisks indicated statistically significant differences compared with Col-0, which were determined by two-tailed Student's t-test. * Represents p ≤ 0.05, ** represents p ≤ 0.01, ns represents no significant difference.

VvSKs Inhibit Fruit Ripening in Tomato
The BR hormone is involved in fruit ripening in strawberry [23], but whether this is also the case in grapevine and whether this process is mediated by VvSKs is yet to be verified. In order to corroborate these questions, Agrobacterium tumefaciens strain GV3101 cultures containing 35S: VvSKs-GFP or the empty vector pK7FWG2 (control) were infiltrated into mature green tomato fruits of Lycopersicon esculentum cv. Moneymaker. Tomato fruits were characterized 4 days after infiltration. Real-time PCR results showed that transcript levels of VvSKs showed at least a 5-fold upregulation compared with the control fruits 5 days after infiltration (Figure 7b). The surface color of control fruits developed into fully red just 5 days after infiltration, while the changes in surface color from green to red were markedly delayed in the 35S: VvSK-GFP fruits, especially in 35S: VvSK1-GFP fruits, which only turned slight yellow 4 days after infiltration (Figure 7a). In addition, firmness of 35S: VvSK-GFP fruits was also much higher compared with that of control fruits (Figure 7c). Altogether, these results indicated that VvSKs inhibit fruit ripening in tomatoes. only turned slight yellow 4 days after infiltration (Figure 7a). In addition, firmness of 35S: VvSK-GFP fruits was also much higher compared with that of control fruits (Figure 7c). Altogether, these results indicated that VvSKs inhibit fruit ripening in tomatoes. The asterisks indicated statistically significant differences compared with control, which were determined by two-tailed Student's t-test. * represents p ≤ 0.05, ** represents p ≤ 0.01, *** represents p ≤ 0.001.

BR Promotes Grape Berry Expansion and Soluble Solids Accumulation
Over-expression of VvSK7 inhibited photosynthesis in Arabidopsis and transient expression of VvSKs in the tomato delayed fruit ripening in it. These findings provided a strong indication for the involvement of VvSKs and their corresponding BR hormones in grape berry development. In order to figure out this question, 24-epibrassinolide, brassinazole (biosynthesis inhibitor of brassinosteroid) and BIKININ (GSK3/Shaggy-like kinase inhibitor) were applied to grape berries of the Shine Muscat (Vitis labruscana Bailey× V. vinifera L. Shine Muscat) variety at veraison stages. Fruit quality-related parameters including fruit size, berry weight, and soluble solids were measured at maturity (17th August). Compared with the mock treatment, EBR and BIKININ all promoted grape berry ripening, the surface color of EBR-and BIKININ-treated berries turned white just 2 weeks after treatment, while the berries treated with BRZ (brassinazole) remained green (Figure 8a). Moreover, EBR-and BIKININ-treated berries significantly accumulated more soluble solids than the mock berries ( Figure  8b). Furthermore, EBR and BIKININ also promoted grape berry expansion in both vertical and transverse diameters, and the weight of their berries was significantly higher than the mock and BRZtreated berries. Taken together, these results demonstrated that EBR and BIKININ can promote grape ripening and berry expansion, which is consistent with the result that transient over-expression of Figure 7. Transient over-expression of VvSKs in tomato inhibit fruit ripening. The 35S:VvSK7-GFP construct was injected into the fruits of 12 days after flowering, and empty PK7FWG2 was used as a control. (a) Tomato fruits 4 days after infiltration, OE: overexpression; (b) Transcription levels of VvSKs in tomato fruits 4 days after infiltration. Housekeeping gene SlACTIN was used as an internal reference. Expression levels were normalized with control (tomato fruits infiltrated with GV3101 Agrobacterium cultures containing pK7FWG2 empty vector); (c): Firmness of tomato fruits 4 days after infiltration. Three independent fruits from three replicates were used for measurement (n = 3). The asterisks indicated statistically significant differences compared with control, which were determined by two-tailed Student's t-test. * Represents p ≤ 0.05, ** represents p ≤ 0.01, *** represents p ≤ 0.001.

BR Promotes Grape Berry Expansion and Soluble Solids Accumulation
Over-expression of VvSK7 inhibited photosynthesis in Arabidopsis and transient expression of VvSKs in the tomato delayed fruit ripening in it. These findings provided a strong indication for the involvement of VvSKs and their corresponding BR hormones in grape berry development. In order to figure out this question, 24-epibrassinolide, brassinazole (biosynthesis inhibitor of brassinosteroid) and BIKININ (GSK3/Shaggy-like kinase inhibitor) were applied to grape berries of the Shine Muscat (Vitis labruscana Bailey× V. vinifera L. Shine Muscat) variety at veraison stages. Fruit quality-related parameters including fruit size, berry weight, and soluble solids were measured at maturity (17 August). Compared with the mock treatment, EBR and BIKININ all promoted grape berry ripening, the surface color of EBR-and BIKININ-treated berries turned white just 2 weeks after treatment, while the berries treated with BRZ (brassinazole) remained green (Figure 8a). Moreover, EBR-and BIKININ-treated berries significantly accumulated more soluble solids than the mock berries (Figure 8b). Furthermore, EBR and BIKININ also promoted grape berry expansion in both vertical and transverse diameters, and the weight of their berries was significantly higher than the mock and BRZ-treated berries. Taken together, these results demonstrated that EBR and BIKININ can promote grape ripening and berry expansion, which is consistent with the result that transient over-expression of VvSKs in tomatoes delayed fruit ripening. In addition, VvSK7 is a negative regulator for photosynthesis, thus, inhibition of VvSKs by applying EBR and BIKININ can promote photosynthesis and accelerate grape berry expansion and soluble solids accumulation. For each treatment, 10 representative berries from 3 individual grape clusters were used for measurement (n = 10). The asterisks indicated statistically significant differences compared with the mock treatment (DMSO), which were determined by two-tailed Student's t-test. * represents p ≤ 0.05, ** represents p ≤ 0.01, ns represents no significant difference.

Discussion
Protein kinases form a large family of enzymes that mediate eukaryotic cell responses to external stimuli [24]. To date, several protein kinase family members have been identified, including GSK3, Pto-like protein kinase, MAPKKK and PP2C [25][26][27][28]. In these kinases, GSK3 is reported to regulate different physiological and developmental processes in mammals and plants [2,7,[29][30][31]. However, the research of GSK3-like kinases has been limited to several model plants. Recent studies have shown that the GSK3 family or their corresponding BR hormones have been emerging as an important genetic engineering targets or plant growth regulators in numerous crops like rice (Oryza sativa) [17,32] and maize (Zea mays. L) [33]. In grapevine, GSK3 homolog VvSK1 is reported as a sugar-inducible protein kinase that regulates hexose transport and sugar accumulation in cultured grapevine cells [18]. In contrast, their evolutional relationship between different fruit crops and their function in grape berry development has not been elucidated for decades.
In the current study, a phylogenetic tree using VvSKs and GSK homologs from other fruit crops was constructed, and conserved motif prediction and syntenic analysis were also carried out. These results suggested that GSK genes in different fruit crops are highly conserved as they shared similar motif composition. VvSKs are evolutional closer to GSK genes in citrus and peach crops.
BRs are reported to be involved in light-regulated processes [34]. The BR-signaling transcriptional factor BZR1 represses photoreceptors phytochrome B and phototropin 1 and induces For each treatment, 10 representative berries from 3 individual grape clusters were used for measurement (n = 10). The asterisks indicated statistically significant differences compared with the mock treatment (DMSO), which were determined by two-tailed Student's t-test. * Represents p ≤ 0.05, ** represents p ≤ 0.01, ns represents no significant difference.

Discussion
Protein kinases form a large family of enzymes that mediate eukaryotic cell responses to external stimuli [24]. To date, several protein kinase family members have been identified, including GSK3, Pto-like protein kinase, MAPKKK and PP2C [25][26][27][28]. In these kinases, GSK3 is reported to regulate different physiological and developmental processes in mammals and plants [2,7,[29][30][31]. However, the research of GSK3-like kinases has been limited to several model plants. Recent studies have shown that the GSK3 family or their corresponding BR hormones have been emerging as an important genetic engineering targets or plant growth regulators in numerous crops like rice (Oryza sativa) [17,32] and maize (Zea mays. L) [33]. In grapevine, GSK3 homolog VvSK1 is reported as a sugar-inducible protein kinase that regulates hexose transport and sugar accumulation in cultured grapevine cells [18]. In contrast, their evolutional relationship between different fruit crops and their function in grape berry development has not been elucidated for decades.
In the current study, a phylogenetic tree using VvSKs and GSK homologs from other fruit crops was constructed, and conserved motif prediction and syntenic analysis were also carried out. These results suggested that GSK genes in different fruit crops are highly conserved as they shared similar motif composition. VvSKs are evolutional closer to GSK genes in citrus and peach crops.
BRs are reported to be involved in light-regulated processes [34]. The BR-signaling transcriptional factor BZR1 represses photoreceptors phytochrome B and phototropin 1 and induces photo-morphogenesis negative regulators including CONSTITUTIVE PHOTOMORPHOGENEIC 1 (COP1) and SUPPRESSOR OF PHYTOCHROME A (SPA1), leading to defects in plant photo-morphogenesis [35]. The BR hormone is also involved in photosynthesis by modulation of φPSII (efficiency of PSII) [36,37] and rubisco activity [38]. The current study suggested that light-responsive cis-element was highly enriched in promoter regions of VvSKs, Moreover, qPCR results also indicated that VvSKs are sensitive to the light response. Collectively, these findings suggested that VvSKs could mediate a feedback regulation of BR signaling in light-regulated processes. On the one hand, the BR-signaling downstream transcriptional factor BZR1 directly regulate photomorphogenesis or photosynthesis by repressing or inducing important components in these processes. On the other hand, light signaling can in turn alleviate or enhance its effect by modulation of VvSK-mediated phosphorylation and inhibition of BZR1 [39].
VvSK7 was highly enriched in the young tissues such as berries before the veraison stage, young leaves, and green stems, etc. Interestingly, VvSK7 was immediately downregulated once these tissues entered maturation or senescence processes, its expression level showed obvious downregulation in ripening berries and senescencing leaves, as shown in Figure 3. This observation provides a strong indication that VvSK7 is important for early tissue development but could be negative for maturation processes. Constitutive overexpression of VvSK7 in Arabidopsis lead to weakened BR signaling and defect in plant growth, which is consistent with its homologs in Arabidopsis [22,39]. In addition to the BR-related phenotype, VvSK7-OE plants showed more chlorophyll accumulation in leaves, but lower photosynthetic rate compared to Col-0 (Figure 6d,e), which is inconsistent, as its commonly assumed that chlorophyll content is positively correlated with photosynthesis rate [40,41]. Another observation of Rubisco and Rubisco activase could explain this inconsistency, as we also found that Rubisco activity in VvSK7-OE plants was decreased when compared to Col-0 and a putative interaction between Rubisco activase (RCA) and VvSK7 (data not shown). Over-expression of VvSK7 could inhibit Rubisco activity through RCA and lead to lower photosynthesis efficiency in Arabidopsis [42,43]. A similar study in C. sativus showed that Rubisco activity and photosynthesis rate were enhanced upon brassinosteroid treatment, which is consistent with our hypothesis [37]. Transiently overexpression of VvSKs in the tomato lead to delayed fruit ripening (Figure 7), consistent with the assumption that VvSK7 could be a negative regulator for maturation. Brassinosteroids were found to be involved in strawberry ripening. BR hormone content and transcriptional level of the BR-signaling receptor FaBRI1 were all increased during the ripening process [23]. By contrast, we observed the downregulation of VvSK7 when the grape berry entered the ripening process; however, the exact role and the underlying molecular mechanism still need to be elucidated.
In this study, we applied 24-epibrassinolide, BIKININ (GSK3/Shaggy-like kinase inhibitor) and brassinazole (biosynthesis inhibitor of brassinosteroid) at the veraison stages. The results showed that EBR and BIKININ treatment accelerated grape berry ripening, promoted fruit expansion, and soluble solids accumulation. BR treatment leads to dephosphorylation and proteasome-mediated degradation of GSKs [44], while small molecular compound BIKININ directly blocks the GSKs' kinase activity though competing for the ATP binding pocket [45]. Blocking VvSKs in the grape berry by these compounds promotes berry ripening, which is consistent with the observation in the tomato where fruit ripening was delayed by overexpression of VvSKs. Previous studies reported that overexpression of VvSK1 in cultured grapevine cells leads to upregulation of monosaccharide transporters VvHT3, VvHT4, VvHT5, and VvHT6 and results in increased glucose and sucrose accumulation [18]. However, different results are observed in the current study, showing that blocking VvSKs by EBR and BIKININ promoted the accumulation of soluble solids in berries. Considering that overexpression of VvSK7 leads to weakened photosynthesis rate, blocking VvSK7 by EBR or BIKININ will, in turn, promote photosynthesis which will ultimately lead to the accumulation of soluble solids in berries. Moreover, the expression level of VvSK7 in berries was significantly higher than that of VvSK1 (Figure 3), which means that the positive effect of EBR or BIKININ through VvSK7 could cover or compensate their negative effects through VvSK1.

Plant Materials and Treatments
One-year-old potted Shine Muscat grapevine seedlings grown in the greenhouse at Nanjing Agricultural University, Nanjing, Jiangsu, China were used for dark/light treatment. Before treatment, the grapevine seedlings were transferred to a plant growth chamber at 25 • C with continuous light (5000 lx). Leaves at 4-6 nodes were used for treatment. Along the central vein, half of the leaf was covered with tinfoil paper as a dark treatment, the other half was exposed to the light as light treatment. The samples were harvested 8 h after treatment. The collected leaves were immediately frozen in liquid nitrogen and stored at −80 • C.
Five-year-old Shine Muscat Grapevines were grown in Tangshan Grape Test Base at Nanjing (N32 Tomato plants (Lycopersicon esculentum cv. Moneymaker) prepared for VvSKs transient over-expression assay were grown in a greenhouse at Nanjing Agricultural University, Nanjing, Jiangsu, China.

Gene Cloning and Construction of the Expression Vectors
VvSK gene sequence was queried from BLAST on NCBI. The specific primers were designed for PCR amplification (Table S3) and PCR amplification was conducted as follows: at 95 • C predenaturation for 5 min, 95 • C for 10 s, 55 • C for 15 s, 68 • C for 2 min, 35 cycles, and 68 • C for 10 min extension. All primers were subjected to normal PCR and the reaction products were separated on a 1% agarose gel to ensure the bands were of the expected size and that there were no primer dimers. The target fragment of PCR run-off product was purified, recovered, and linked to the entry vector PDONR221. It was subsequently subcloned into Gateway ready vector PK7FWG2 to generate 35S:VvSKs-GFP plant expression vector using the LR reaction of Gateway recombination-based cloning (Invitrogen, Carlsbad, CA, USA). The positive clones were screened and sent for sequencing to the biological engineering company, Shanghai, China.

Transgenic Arabidopsis Method
For gene transformation, a DNA construct was transformed into 5-week-old Arabidopsis using the Agrobacterium tumefaciens strain C58C1 (Weidi, Shanghai, China) and the floral-dip method [46]. Transformed seeds were selected on Murashige and Skoog agar medium containing the appropriate antibiotics: 50 µg/mL kanamycin (Sigma).

Tomato transient expression
Using the freeze-thaw method, 35S:VvSKs-GFP vectors were transformed into the Agrobacterium strain GV3101(Weidi, Shanghai, China) [47]. About 5 mL culture of each strain was grown overnight at 28 • C in Luria-Bertani (LB) medium (50 mg/mL kanamycin and 50 mg/mL rifampicin, 10 mM MES, 20 mL acetosyringone). The overnight cultures were inoculated into 50 mL of LB medium and grown at 28 • C overnight. The cells were harvested by centrifugation (3000× g, 5 min, 20 • C) and suspended in infiltration buffer (10 mM MgCl 2 , 10 mM MES, 20 mL acetosyringone), adjusted to an optical density (OD600) of 0.8-1.0 of PK7WG2 and its derivatives, and left to stand at room temperature for 4 h. About 1 mL of Agrobacterium was infiltrated into every tomato at mature green stage with a 1 mL syringe. Ten uniformly sized fruit was used in the infiltration experiment, and the experiment was repeated three times.

Determination of Photosynthetic Rate
Li-6800 (LI-COR, Lincoln, NE, USA) portable photosynthetic instrument was used to measure the photosynthetic rate. The second true leaf from 4-week old Arabidopsis plants were used for measurement. 3 leaves from 3 individual plants were measured. The measured temperature is (20 ± 2) • C, and the light intensity is 1000 µmol·m −2 ·s −1 . The relative humidity is 25-30%, the concentration of CO 2 is (380 ± 10) µmol·mol −1 , the flow rate is 400 µmol·s −1 , and the average value of three readings is taken as the measured value for each leaf.

Chlorophyll Measurement
Use a punch to cut about 1 cm 2 of leaves (avoiding relatively thick veins) and cut about 5 mm long and 1 mm wide filaments. Put the leaf filaments into the calibration tube containing 5 mL 80% acetone, and place in the dark after sealing the nozzle until the filaments are completely white (overnight). Pour the solution into the cuvette and use the two wavelengths of 662 nm and 645 nm as the peak absorbencies of chlorophyll-a and chlorophyll-b. Calculate the chlorophyll concentration (CV = chla + chlb) of the extract according to the formula of Arnon [48].

Determination of Fruit Quality
Ten representative fruits from 3 individual grape clusters were picked in each treatment at the fruit mature stage. The vertical and horizontal meridians of the fruits were measured by vernier caliper; the soluble solid content of the fruits was measured by a hand-held sugar meter(PAL-1, Atago, JPN); the titratable acid content was measured by acid-base titration.

Genome-Wide Identification and Annotation of Grape SKs Genes
We quoted seven VvSKs from a paper written by Fatma Lecourieux, and using VvSKs as a template, we found other Sks in Pyrus bretschneideri, Malus domestica, Citrus reticulata and Amygdalus persica L. All the obtained sequences were stored in the InterProScan database (http://www.ebi.ac.uk/ Tools/pfa/iprscan5/, accessed on 5. 7. 2019). Length of sequences, molecular weights and isoelectric points of deduced polypeptides were calculated by using tools provided at the ExPasy website (http://web.expasy.org/protparam/, accessed on 20. 7. 2019) [49]. The choice of candidate SKs in Prunus persica, Malus domestica, Pyrus bretschneideri, Citrus sinensis was also based on the E-value (1 × 10 −5 ) and the highest similarity scores to VvSKs in NCBI. All the names and GenBank accession numbers of SKs are shown in Table S4. 4.9. Phylogenetic Tree, Conserved Motifs, Syntenic Analysis, Transcriptional Profiling and Cis-Elements Analysis of SKs Family MEGA version 6 (Sudhir Kumar, Arizona State University, Temp, AZ, USA) was used to construct phylogenetic trees by the maximum likelihood (ML) methods and the bootstrap test carried out with 1000 replicates [50]. The conserved motifs were identified using the online MEME program (version 4.12.0) [20]. We set the motif number as 10 and chose motifs with E-values ≤ 1 × 10 −30 . MCscanX was used to analyze the gene synteny and collinearity of VvSKs among Vitis vinifera, A. thaliana and C. sinensis [51]. The synteny figures were drawn by Circos-0.69 [52] and those results with E-value > 1 × 10 −5 were filtered. Global transcriptomic data retrieved from NCBI (GSE36128) were used to generate the heatmap. The heatmap was constructed using R with "pheatmap" package (https://cran.r-project.org/web/packages/pheatmap/index.html, accessed on 10. 3. 2020) [19]. The 1500 bp upstream of the SK genes of each species was used to perform cis-elements analysis in PlantCARE [53].

RNA Isolation and RT-PCR Analysis
Total RNA of grape tissue was extracted with the plant total RNA isolation Kit Plus from Fuji (Chengdu, China) according to the manufacturer's protocol. The concentration of total RNA was measured using a NanoDrop 2000 UV-Vis spectrophotometer (Thermo Scientific, Waltham, MA, USA) after treatment of genomic DNA with RNase-free DNase I (Takara, Dalian, China). The PrimeScriptTM RT reagent Kit (Takara) was used to obtain cDNA according to the manufacturer's instructions. Synthesized cDNA concentration was diluted to 100 ng/µL and each reaction mixture contained 10.0 µL SYBR Premix Ex Taq TM (Takara), 0.5 µL of each primer (10 µM), 1 µL cDNA, and 8 µL ddH2O in a total volume of 20 µL. Reactions were performed under the following conditions: 95 • C preheating for 4 min, followed by 40 cycles at 95 • C for 20 s, 60 • C for 20 s, and 72 • C for 40 s. The primer pairs (shown in Table S3) were designed by Primer 3 (http://primer3.ut.ee/, accessed on 20. 5. 2019). The PCR experiment was carried out with at least three technical replicates. The relative transcript levels of selected genes were calculated using the 2 −∆∆Ct method [54].

Statistical Analysis
Significant differences mentioned in Figure 6 (chlorophyll content and photosynthetic rate of 35 s: Vvsk7-GFP-L1 and L2 compared with that of Col-0), Figure 7 (firmness of VvSK transient over-expression tomatoes compared with that of control) and Figure 8 (physiological parameters of EBR-, BIKININ-or BRZ-treated grape berries compared with that of DMSO-treated grape berries) were analyzed by two-tailed Student's t test using software SAS 9.2 (SAS Institute Inc., Cary, NC, USA). The chart was performed by Excel 2010.

Conclusions
We identified seven VvSK genes through bioinformatics analysis, and found that Vvsks are highly conserved with the homologous genes of other species. The photoreceptor-related elements in the VvSK promoter region are the most abundant, and the transcriptional levels of Vvsk7 were highly enriched in young tissues, but decreased rapidly with their maturation or aging. In addition, the over-expression of phenotype of Vvsk7 in VvSKs was similar to that in BR-deficient Arabidopsis. More chlorophyll was accumulated in the leaves of VvSK7-OE plants, but its photosynthetic efficiency was lower than that of wild type plants. Furthermore, transient expression of Vvsks delayed the ripening of the tomato fruit. In the grape, the application of EBR or BIKININ might block the activity of Vvsks and promote ripening processes such as fruit expansion and soluble solid accumulation. These findings could lay the theoretical foundation for the functional study of SKs and the further construction of grape BR regulation networks.