Abstract
Although coloration of grape berry skins is influenced by temperature, the details of its effects have not been reported. To find temperature sensitive stages for coloration and to clarify the mechanisms that underlie the effect of temperature on anthocyanin accumulation, two-week treatments at temperatures of 20°C and 30°C were carried out at four different stages of development and ripening using each of three potted vines of Aki Queen (Vitis labrusca x V. vinifera). Anthocyanin accumulation in the skins was significantly higher at 20°C than at 30°C after the temperature treatment, and the most sensitive stage for the temperature treatment was from one to three weeks after coloring began (stage III). Furthermore, at harvest, the grapes treated at 20°C in stage III contained the highest concentration of anthocyanin. After temperature treatment in stage III, the concentration of abscisic acid (ABA), a plant hormone related to anthocyanin accumulation, in the berry skins was 1.6 times higher at 20°C than at 30°C. The copy numbers of accumulated mRNA of anthocyanin biosynthetic enzyme genes and a myb-related regulate gene, VvmybA1, were also higher at 20°C than at 30°C. These results and previous reports indicate that the high and low temperatures during ripening, especially in stage III, likely affect the production and/or degradation of ABA in berry skins and that the endogenous ABA level affects the expression of VvmybA1; the product of VvmybA1 then controls the expression of the anthocyanin biosynthetic enzyme genes.
The color of grape berry skins is determined by the accumulation of anthocyanins. The quantity and composition of anthocyanins influence skin color in black and red cultivars (Mazza and Miniati 1993, Shiraishi and Watanabe 1994). Temperature is known to influence the accumulation of anthocyanins in berry skins (Spayd et al. 2002). In grapes grown in southwestern Japan, including Kyushu Island, with high temperatures during the ripening season, skin color is poorer than in grapes grown in cool regions of Japan (Naito et al. 1986). The coloration of red grapes, such as Aki Queen (Vitis labrusca x V. vinifera), is severely inhibited by high temperatures, resulting in lower market prices. Temperature-control experiments have also shown that exposing whole vines or clusters to high temperature (30°C) inhibited anthocyanin accumulation (Kataoka et al. 1984, Kliewer 1970, Mori et al. 2004, Tomana et al. 1979). In these reports, however, temperature treatments were carried out throughout the ripening period (from the onset of coloring to harvest); the differences of temperature sensitivity for skin color among stages during the ripening period are unknown. Finding the temperature-sensitive stage would enable us to control the temperature of greenhouses effectively to improve skin color and to determine more precisely the best regions in which to grow the grapes. Our aim was to determine the optimal temperature-sensitive stage for anthocyanin accumulation.
In the control of anthocyanin accumulation by plant hormones, endogenous abscisic acid (ABA) concentration in the skins is closely correlated to the accumulation of anthocyanins (Inaba et al. 1976, Kataoka et al. 1983, Pirie and Mullins 1976). High temperature (30°C) inhibited anthocyanin accumulation with a reduction of endogenous ABA concentration (Tomana et al. 1979). However, spraying ABA to the clusters restored the level of anthocyanin accumulation in high-temperature treated grapes (Kataoka et al. 1984). These results suggest that ABA plays a key role in the anthocyanin biosynthesis in grapes.
The biosynthetic pathway of anthocyanin has been well studied using model plants, and several enzyme genes for anthocyanin biosynthesis have been isolated from grapes (Sparvoli et al. 1994). The expression of the gene of UDP-glucose:flavonoid 3-O-glucosyltransferase (UFGT) is critical for anthocyanin biosynthesis in grapes (Boss et al. 1996a,b, Kobayashi et al. 2001, 2002). Anthocyanin biosynthesis is controlled by Myb- and Myc-like transcription regulators (Dooner et al. 1991). Kobayashi et al. (2002) isolated myb-related regulatory genes, VlmybAs, from the Kyoho grape (V. labrusca x V. vinifera) and showed that these genes were involved in the regulation of anthocyanin biosynthesis via UFGT gene expression as well as enhanced expression of other enzyme genes of the biosynthetic pathway. Kobayashi et al. (2004) also showed that a homologue of VlmybAs, VvmybA1, was expressed in the berry skins of colored cultivars of V. vinifera but not in white ones. The expressions of seven anthocyanin pathway genes (PAL, phenylalanine ammonialyase; CHI, chalcone isomerase; CHS, chalcone synthase; DFR, dihydroflavonol 4-reductase; F3H, flavanone 3-hydroxylase; LDOX, leucoanthocyanidin dioxygenase; and UFGT) were much higher in the red-skinned sports than in the white-skinned progenitors (Kobayashi et al. 2001).
Ban et al. (2003) reported that ABA treatment of Kyoho grapes at veraison enhanced the accumulation of anthocyanin and the expressions of PAL, CHS, CHI, DFR, LDOX, and UFGT genes in the berry skins. Jeong et al. (2004) showed that ABA treatment of Cabernet Sauvignon grapes enhanced the expression of VvmybA1, which coincided with the enhanced expression of anthocyanin synthetic enzyme genes and anthocyanin accumulation. These findings suggest that VvmybA1 product induces the expression of enzyme genes of the anthocyanin biosynthetic pathway of grapes and that ABA probably has an influence on the expression of VvmybA1 via signal transduction. However, no reports have shown that temperature affects expression of anthocyanin biosynthetic genes, including VvmybA1, in the berry skin. Thus, the mechanism of the effect of temperature on the anthocyanin accumulation has not yet been fully clarified.
In order to find the temperature sensitive stage for coloration, we examined the effects of high (30°C) and low (20°C) temperature treatments at different stages of development and ripening on the accumulation of anthocyanin in berry skins using a red table grape, Aki Queen. We also determined the mRNA levels of VvmybA1 and anthocyanin biosynthetic pathway genes using a real-time quantitative polymerase chain reaction (Q-PCR) and endogenous ABA concentration in the berry skins to determine the mechanisms that underlie the effect of temperature on anthocyanin accumulation.
Materials and Methods
Plant materials and temperature treatments.
Two-year-old red table grapevines Vitis labrusca x V. vinifera cv. Aki Queen grafted on 5BB rootstocks were used in this study. These vines were grown in 60-L plastic containers with a sand and compost mixture (9:1 v/v) in a vinyl greenhouse in Akitsu, Hiroshima, Japan. Compound fertilizer containing 4.5 g, 3 g, and 4.5 g of N, P, and K, respectively, per vine was applied before budbreak. The vines were automatically watered with 4 L per container when the soil water potential decreased to −9.8 kPa, which is optimal for grapevine growing (Imai et al. 1991a,b). Vines were trained to a flat-top trellis system with a single trunk. Each vine developed six to nine shoots, some with one or two clusters. Three weeks after budbreak, shoots were thinned to three per vine. Each primary shoot was pinched to retain 12 nodes before blooming; after that, each lateral shoot was pinched once a week to retain only one lateral leaf. Cluster numbers were adjusted to three clusters per vine two weeks after blooming. The number of berries was also reduced to 30 berries per cluster following a standard cultivation method for this variety.
The temperature treatment was conducted using two glass growth cabinets kept at 20 ± 2°C and 30 ± 2°C throughout the day and night. In each treatment, three potted vines were transferred from the vinyl greenhouse to either the 20°C or the 30°C cabinet for two weeks, after which the vines were again moved to the vinyl greenhouse. Both the glass cabinets and the vinyl greenhouse were under natural sunlight conditions. Light transmissions of the vinyl and the glass were almost the same, but the UV-A transmittance through the vinyl was slightly less than through the glass. Treatments were carried out at four different stages of development and ripening in 2003: stage I, 24 June to 7 July; stage II, 8 to 22 July; stage III, 23 July to 4 Aug; and stage IV, 5 to 18 Aug. Mean air temperature of the vinyl greenhouse was 16.1°C, 17.6°C, 20.1°C, and 21.3°C, in each stage, respectively. In the control vines, date of full bloom, veraison, and coloring start were 2 June, 14 July, and 16 July 2003, respectively. In all treatments, grapes were harvested on 18 Aug.
Anthocyanin concentration and berry composition.
Three berries from each cluster (three clusters per vine, three vines per treatment; the average of three clusters per vine was used as a replicate) were randomly taken at the end of each treatment and at harvest, except for unusually large or small berries. After the berries were weighed, a skin disk (1.33 cm2/disc) was removed from the side of each berry using a cork-boring tool and weighed. Anthocyanin was extracted from three discs of the fresh berry skins with 10 mL of 50% (v/v) acetic acid for 24 hr at 4°C. Absorbance of the extract was then measured at 520 nm in a cuvette with a 10-mm path length using a Ubest-30 spectrophotometer (Jasco, Tokyo, Japan). Anthocyanin concentration was expressed as a milligram of cyanidin 3-monoglucoside chloride (Extrasynthese, Genay, France) equivalent per gram of fresh skin weight. Remaining berries were crushed in a polyethylene bag and filtrated. Obtained juice samples were immediately analyzed for total soluble solid (TSS; Brix) with a hand refractometer and for titratable acidity by titration with 0.1 N NaOH to a phenolphthalein end point.
ABA analysis.
From the vines treated with the high and low temperatures in stage III and the control vines, three berries per vine were taken at the end of treatment (4 Aug). Berry skins were immediately frozen in liquid nitrogen and stored at −80°C until use. Free ABA concentration in the skin sample of each vine was determined by gas chromatography-mass spectrometry according to Koshita et al. (1999). ABA was extracted from the homogenized sample to which 2H6-ABA was added as an internal standard, purified with high-performance liquid chromatography, and dried. The purified ABA was dissolved in methanol, methylated by addition of a few drops of ethereal diazomethane, and dried under nitrogen gas. After redissolving with 5 μL ethyl acetate, 1 μL was injected into a gas chromatograph equipped with a mass spectrometer (QP-5000, Shimadzu Inc., Kyoto, Japan) using the splitless technique and a fused silica capillary column (CBP, 1.25 m x 0.22 mm i.d., 0.25 μm film thickness; Shimadzu). The carrier gas (helium) had a flow rate of 54.7 cm/sec. The injector temperature was 250°C, and the column temperature was maintained at 120°C for 2 min and then increased at 20°C/min up to 280°C. Retention times of both the methyl esters of cis-ABA and 2H6-ABA were 8.25 min. Mass range monitored was between m/z 40 and 350. The concentration of ABA in the original extract was determined from the ratio of peak areas for m/z 190 (2H0)/194 (2H6).
Quantification of mRNA.
For the determination of the mRNA levels of anthocyanin biosynthetic enzyme genes and VvmybA1, Q-PCR was performed using the GeneAmp 5700 sequence detection system (Applied Biosystems, Foster City, CA) and Quantitect SYBR Green PCR kit (Qiagen, Hilden, Germany) as described in the manufacturers’ manuals. Total RNA was extracted according to Loulakakis et al. (1996) from the skin samples used for ABA analysis after mixing the samples of three vines from each temperature treatment and treated with DNase I. cDNAs of all samples were prepared by reverse transcription of 50 ng/μL of total RNA with 0.25 μM of an oligo dT primer using reverse transcriptase. For Q-PCR, 1 μL of the cDNA template per 20 μL of reaction mixture was used. Final concentration of each primer was 0.5 μM (for CHS2, CHS3, CHI1, F3H2, DFR, and LDOX) or 0.25 μM (for UFGT, VvmybA1, and VvUbiquitin1). The Q-PCR was performed under the following conditions: 95°C for 15 min, followed by 40 cycles at 95°C for 15 sec, at the annealing temperature for 20 sec and at 72°C for 20 sec. mRNA of VvUbiquitin1 was quantified as an internal control of the Q-PCR (Downey et al. 2003) using forward (5′-TGCAG GAAAAAGAAGTGTGG) and reverse (5′-CAATGCTCC CCGTGTGTAAC) primers at the annealing temperature, 56°C. The primer sequence and annealing temperature for the other primer set used were the same as those described in Jeong et al. (2004). The Q-PCR was carried out at four replicates per cDNA sample, and the average data were presented as the molar ratio to VvUbiquitin1.
Statistical analysis.
The F test was used to evaluate significant differences in skin anthocyanin concentration and composition of the berries after each temperature treatment and at harvest and ABA concentration after the temperature treatment in stage III. Where significant differences were found, multiple comparisons were carried out using the Tukey test except for harvest date. Dunnett’s test was used for a comparison with the control at harvest. No statistical analysis was applied for mRNA quantification because there was no biological replication. Excel software (Microsoft, Redmond, WA) and Excel statistical package (Esumi Inc., Tokyo, Japan) were used for all statistical calculations.
Results and Discussion
Anthocyanin accumulation, fruit composition, and ABA concentration.
After treatment, low temperature (20°C) treatment in stages II and III resulted in significantly higher anthocyanin concentration than that of high temperature (30°C) treatment (Table 1⇓). In particular, the low temperature treatment in stage III (from one to three weeks after the onset of coloring) was effective for anthocyanin accumulation compared to the controls. On the other hand, anthocyanin concentrations after the high temperature treatment in stages II, III, and IV were almost the same as the control of the former stage, indicating that the high temperature severely inhibited anthocyanin accumulation.
Anthocyanin accumulation at harvest was also affected by the treatments (Table 2⇓). The low temperature treatment in stage III significantly enhanced anthocyanin concentration at harvest, while the high temperature at the same stage significantly decreased it compared to the control berries, although vines were kept in the same vinyl greenhouse under natural conditions except for the treatment time. Thus, the low temperature treatment enhanced the anthocyanin accumulation compared to the high temperature treatment and the control; however, the effect of temperature varied greatly among the stages in which treatment was conducted.
The treatments in stage I and II started before veraison, which affected the date of veraison. The low temperature in stage I delayed veraison for three days, while the low temperature in stage II brought it forward by five days. The high temperature in stage I did not affect the date of veraison, while the high temperature in stage II delayed it four days compared with the control. However, these differences in the veraison date did not significantly affect anthocyanin concentration at harvest compared with the control.
Berry weight and TSS after treatment, as well as at harvest, were not affected by temperature. On the other hand, the decrease of titratable acidity was inhibited by the low temperature in stages III and IV (Table 1⇑). At harvest, the berries treated with low temperature at stage IV showed higher titratable acidity than the control berries.
ABA concentration in the skins of berries after treatment in stage III was 1.6 times higher at 20°C than at 30°C (Table 3⇓). This result agreed with Tomana et al. (1979). Thus, low temperature may facilitate the biosynthesis of ABA and/or reduce its degradation.
mRNA accumulations.
The mRNA accumulations of anthocyanin biosynthetic enzyme genes and VvmybA1 in the berry skins after treatment in stage III are shown in Table 4⇓. The mRNA levels of all the tested enzyme genes and VvmybA1 were higher at 20°C than those at 30°C. Thus, temperature conditions influenced the expression of anthocyanin biosynthetic genes, including VvmybA1.
These results as well as previous reports indicate that the high and low temperatures during ripening, especially in stage III, likely affect the production and/or degradation of ABA in skins and that the endogenous ABA level influences the expression of VvmybA1. The products of VvmybA1 then control the expression of anthocyanin biosynthetic enzyme genes. Thus, at least a part of the difference in anthocyanin accumulation between high and low temperature conditions was likely caused by the difference of transcription levels of the anthocyanin biosynthetic pathway genes.
However, mRNA still accumulated in the berry skins at the end of the high temperature treatment in stage III even though anthocyanin accumulation was almost completely inhibited during the high temperature treatment. Thus, another mechanism—such as anthocyanin degradation, inhibition of enzyme activities of the anthocyanin biosynthetic pathway, and/or low translocation of substrates for anthocyanin biosynthesis—may also contribute to the inhibitory effect of high temperatures on anthocyanin accumulation. Mori et al. (2004) reported that high temperature resulted in low enzyme activities of PAL and UFGT as well as in low concentration of phenylalanine, which is the substrate of PAL, in the berry skins. Our study also showed the possibility that temperature conditions influenced anthocyanin accumulation via plural mechanisms.
Conclusions
This study showed that stage III (one to three weeks after the onset of coloring) is the most sensitive for anthocyanin accumulation in the berry skins of Aki Queen. Low temperature treatment (20°C) during stage III significantly enhanced coloring. This finding will help determine the optimum temperature for greenhouses in which Aki Queen and its related red or black table grapes are grown. This study also showed that low temperatures caused a high level of endogenous ABA, which resulted in high expression of VvmybA1 and the anthocyanin biosynthetic genes, although it is possible that another mechanism also contributed to the accelerative effect of low temperatures.
Footnotes
Acknowledgments: We would like to thank Dr. S. Imai for valuable suggestions. We also wish to thank M. Miyoshi and K. Shibayama for helpful discussion and N. Karata and M. Kanameda for technical assistance.
Part of this work was supported by NARO Research Project no.166, Establishment of Agricultural Production Technologies Responding to Global Warming.
- Received May 2005.
- Revision received November 2005.
- Copyright © 2006 by the American Society for Enology and Viticulture