Ohio State University Extension Bulletin

Fruit Crops: A Summary of Research 1998

Research Circular 299-99


Cluster Thinning Effects on Fruit Weight, Juice Quality, and Fruit Skin Characteristics in 'Reliance' Grapes

Yu Gao and Garth A. Cahoon

Abstract

Fruit weight, juice quality, fruit skin color, the concentration of total anthocyanin and individual anthocyanins in fruit skin, and the percentage of individual anthocyanins in 'Reliance' grape (Vitis hybrid) were investigated after cluster-thinning treatments were imposed. Cluster-thinning treatments were 60 (control), 40, and 20 clusters per vine, and applied when the berries were 2-3 mm in diameter. Twenty clusters per vine produced fruits of the best weight, juice quality, and color among three cluster-thinning treatments. Fruit cluster thinning decreased vine yield significantly, following a quadratic relationship. Juice soluble solids concentration (SSC) was increased significantly by cluster thinning treatments. Individual berries were heavier with 20 clusters per vine than with 60. Juice pH was not affected by cluster thinning. Juice titratable acid concentration was lower at 20 clusters per vine than with 60 clusters per vine. Twenty clusters per vine produced fruits darker and less yellow than 60 clusters per vine. Fruit red pigmentation was increased quadratically by cluster thinning. Total anthocyanin concentration in berry skin was increased linearly by cluster thinning. The concentration of individual anthocyanins, including cyanidin-3-glucoside, peonidin-3-glucoside, and acylated cyanidin derivative, was increased linearly by cluster thinning. However, the concentration of delphinidin-3-glucoside, or petunidin-3-glucoside, or malvidin-3-glucoside, was not significantly affected. Cluster thinning linearly increased the percentage of cyanidin-3-glucoside and decreased the percentage of the acylated cyanidin derivative. The percentages of delphinidin-3-glucoside, petunidin-3-glucoside, peonidin-3-glucoside, and malvidin-3-glucoside were not affected by cluster thinning.

Introduction

Fruit color is an important quality attribute in table grapes. The pigments responsible for the attractive red, blue, purple, and black color are anthocyanins, a class of water-soluble flavonoid pigments. Fruit color development in grapes has been studied extensively due to its importance in the table and wine-grape industry. Fruit-cluster thinning has been shown to improve pigmentation of certain grape cultivars (4, 8, 10).

The detailed study of individual anthocyanins is essential to the understanding of anthocyanin metabolism and fruit-color improvement. Most published studies on fruit colora-tion in grapes have dealt with changes in total anthocyanin content (7, 12, 14). Anthocyanin profiles of most pigmented grape cultivars are known to be very complex. 'Cabernet Sauvignon' (Vitis vinifera L.) and 'Concord' (Vitis labruscana L.) have potentially up to 20 different kinds of anthocyanins (6, 16, 17). Advances in instrumental analysis have made the simultaneous investigation of all anthocyanins in a grape cultivar possible.

C18 reverse-phase high-performance liquid chromatography (HPLC) has been the method of choice in simultaneous separation, identification, and quantification of anthocyanins in grapes and other colored fruits (5, 11, 17). HPLC studies of cultivars and the effects of climatological factors on changes in individual anthocyanins during fruit ripening (2, 11) have produced results significant to our understanding of anthocyanin metabolism in field-grown pigmented grapes.

'Reliance' (Vitis hybrid), a red seedless table grape, was used in this study because of its fruit-color variation at maturity, economical importance in Ohio, and simple anthocyanin profile (3). C18 reverse-phase HPLC analysis of anthocyanins from 'Reliance' revealed seven components. With the joint use of HPLC, paper chromatography, thin-layer chromatography, and spectral measurement, three components were identified as delphinidin-3-glucoside, cyanidin-3-glucoside, and peonidin-3-glucoside. Four other anthocyanin components were tentatively identified as cyanidin-3,5-diglucoside, petunidin-3-glucoside, malvidin-3-glucoside, and an acylated cyanidin derivative. The authors' preliminary studies on fruit-color development in 'Reliance' showed that it was enhanced by cluster thinning, or the application of chelated nutrients, or ethephon.

The objective of this study was to investigate the effects of fruit-cluster thinning on 'Reliance' grape berry weight, color, juice quality, the concentration of total anthocyanin and individual anthocyanins in the berry skin, and the percentages of individual anthocyanins.

Materials and Methods

Grapevines, Treatments, and Statistics

'Reliance' grapevines were planted in 1985 at The Ohio State University/Ohio Agricultural Research and Development Center's Wooster campus. Grapevines were trained to the single-curtain high-cordon system. Row and vine spacings were 3.1 m x 2.4 m. In 1991, grapevines were pruned to 60 buds per vine. Shoot count was adjusted to 50 per vine one week after full bloom. Cluster-thinning treatments were applied when berries were about 2-3 mm in diameter. Treatments were 60 (control), 40, and 20 clusters per vine. The design was a randomized complete block where a whole vine served as an experimental unit. Treatments were replicated eight times. Mean separation was conducted by orthogonal contrasts at P < 0.05 or 0.01.

Fruit Juice Quality and Color

Fruit juice was obtained by pressing 100 berries per sample through a Garden-Way Squeezo strainer (Lemra Products, Boca Raton, Florida). Soluble solids concentration (SSC) of the fruit juice was measured with an ABBE-3L Refractometer (Baush & Lomb, Inc., Rochester, New York). Juice pH was measured with a Beckman pH meter (Model PHI 45, Beckman Instruments Inc., Fullerton, California). Titratable acidity (TA%) was measured by titration of 5 ml of juice with 0.1 N NaOH to a pH of 8.2 (1). Fruit skin color was measured with a Minolta Chroma Meter (Model CR-100, Minolta Camera Co., Ltd., Higashi-Ku, Osaka, Japan) as CIE (Commission Internationale de I'Eclairage, translated as the International Commission of Illumination) 1976 L*, a*, and b*. L* represents bright to dark as L* values increase from negative to positive; a* represents green to red as a* values increase from negative to positive; b* represents blue to yellow as b* values increase from negative to positive. Fruit skin color measurements were taken on six clusters per vine, from the top, middle, and bottom portion of the east-facing side of each cluster. The mean of readings from these six clusters was used.

Sample Preparation for Anthocyanin Analysis

Six fruit clusters per vine were collected on August 18, 1991, when SSC in most fruits reached 18-20%. The clusters were immediately stored at 4ºC. They then were weighed and berries per fruit cluster were counted the following day. Clusters were then frozen and stored at -20ºC for future analyses. Berry skins were removed and collected by first thawing the frozen berries in a refrigerator at 4ºC for 20 min. The berry skins were then peeled with tweezers and kept in an ice-chilled beaker. Finally, the berry skins were freeze-dried and ground with a coffee mill (Oster Model 663-06, Sunbeam Corporation, Milwaukee, Wisconsin).

Anthocyanin Extraction and Concentration

One g of ground berry skin was placed in 100 ml of 1% 12 N HCl in methanol. The anthocyanin extraction was carried out overnight in a refrigerator at 4ºC. Extracts were then filtered through Whatman No. 1 filter paper in a Buchner funnel. Twenty ml of deionized distilled water were added to each anthocyanin extract. The extracts were concentrated with a rotary evaporator under vacuum at 30ºC. Concentrates were transferred to a 25 ml volumetric flask and then brought to volume with deionized distilled water. Five ml of each anthocyanin concentrate was then passed through a 0.2 µm syringe membrane filter that had been equilibrated with 1 ml of the respective anthocyanin concentrate to avoid anthocyanin dilution by syringe membrane filters. Each filtrate was stored at room temperature for less than 20 min in a screw-capped sample vial before HPLC analysis.

HPLC Analyses

HPLC analyses were performed on a Model SP4000 pump (Spectra-Physics, San Jose, California) equipped with a 20 µl Rheodyne sample loop. The analytical column was a pH stable RP-18 Spherisorb (Merck, Darmstadt, Germany) (150 mm x 4.6 mm I.D.) packed with 5 µm particles by Alltech (Deerfield, Illinois). A Spectra-Physics UV1000 variable wavelength detector and Spectra-Physics Model 4600 integrator were used.

The following conditions were used for the analyses of anthocyanins: Solvent A was 10% formic acid in water; solvent B was high-purity acetonitrile. These solvents were filtered through 0.2 µm membrane filters and sparged with helium. Solvent flow rate was 1 ml/min. The solvent program used for anthocyanins was 95% A initially, decreased from 95% A to 72% A in 20 min. following a linear slope. Detection was carried out at 520 nm. The detector was set at one absorption unit full scale.

Determination of Anthocyanin Concentration

To convert the peak areas into pigment concentration per gram of dry berry skin, a solution of cyanidin-3-glucoside (a generous gift from Dr. Geza, Hrazdina, NYSAES, Geneva, New York) in 0.1 N HCl was prepared to establish a standard curve. The solution was filtered through a 0.2 µm membrane filter and absorbance determined at 520 nm. The concentration (g/100 ml) of this solution was calculated based on the extinction coefficient of cyanidin-3-glucoside in 0.1 N HCl. A dilution series was then made. Twenty µl of samples were injected into HPLC under identical analytical conditions as for the samples.

A standard curve was established between the concentration of diluted cyanidin-3-glucoside solutions and their peak areas. The concentration of cyanidin-3-glucoside was calculated based on the following equation where the correlation coefficient was 0.998:

Equation 1.mg/100 ml = (peak area) * (0.00000391)

Since one g of dried berry skin was dissolved in a final volume of 25 ml, the concentration of cyanidin-3-glucoside was calculated as:

Equation 2.mg/g dried berry skin = (peak area) * (0.00000391) * (25ml)/100ml

The concentration of each individual anthocyanin was expressed as cyanidin-3-glucoside equivalent based on their respective peak area, since cyanidin-3-glucoside was present at the highest concentration in the 'Reliance' anthocyanin profile (3). Total anthocyanin concentration was calculated as the sum of the concentration of all the individual anthocyanins. The percentage of each individual anthocyanin was calculated as:

Equation 3.percentage = (conc. of individual anthocyanin x 100)/total anthocyanin conc.

Results and Discussion

Fruit Characteristics

Yield and SSC increased quadratically following fruit-cluster thinning (Table 1). The juice pH value was higher for the 40 clusters per vine treatment than 20 or 60 clusters per vine. Juice titratable acidity (TA) was decreased by 20 clusters per vine in comparison with 40 and 60. Berry weight was increased by cluster thinning following a quadratic relationship. Cluster weight was higher with 40 clusters per vine in comparison with 20 or 60. This nonsignificant difference in cluster weight might have been partially caused by variation in berry counts which the authors did not adjust. The authors hypothesized that the drought during veraison may have stopped the berries from reaching their potential maximum weight. The effect of cluster thinning on cluster weight might also have been reduced by a drought during veraison in 1991. Van Zyl and Webber (13) found that berry growth is most sensitive to water stress at veraison, followed by the period just after flowering.

Table 1. Effects of Cluster Thinnin on Grape Yield, Quality, and Fruit Skin Color (CIE 1976 L*, a* and b*).

  Color Reflectance Parameterz
Treatment (cluster number/vine) Yield (kg/vine) Soluble Solids (oBrix) pH Titratable Acidity (%) Berry Weight (g) Cluster Weight (g) L* a* b*
60
12.7 ay
18.5 c
3.25 a
0.50 b
2.0 b
269.3 a
32.6 a
6.2 a
4.5 a
40
8.9 b
20.0 b
3.35 a
0.49 ab
2.1 b
294.4 a
31.3 ab
7.0 a
3.2 b
20
5.4 c
21.6 a
3.29 a
0.48 b
2.2 a
277.2 a
30.0 b
7.2 a
2.5 b
LSD
1.7
0.7
0.23
0.2
0.1
39.7
1.7
1.1
0.9
Linear Contrast
**x
**
NS
**
NS
NS
*
NS
**
Quadratic Contrast
**
**
**
**
**
**
**
**
**

z L* represents light color to dark color; a* represents red color to green color; b* represents yellow color to blue color.
y Mean separation by least significant difference (LSD) at P = 0.05 level.
x Orthogonal contrast at P = 0.05 or 0.01 level; *, **, NS = significant at 0.05 or 0.01 level, and not significant, respectively.


Cluster-thinning treatments produced darker L* berries following a quadratic relationship (Table 1). Forty clusters per vine increased fruit red color characterized as a* compared to 60. Cluster-thinning treatments produced less yellow berries (Table 1). Differences in fruit color density were also shown in L* and b*. This lack of linear relationship in fruit red color among three cluster-thinning levels might be due to that fact that 'Reliance' is a purplish red grape instead of a true red. Furthermore, the difference in fruit color may have also been partially reduced by the variation in berry count and/or the drought.

Anthocyanins

'Reliance' berries contain anthocyanins only in their berry skins. As measured with HPLC, total anthocyanin concentration of the berry skin was increased linearly by cluster-thinning treatments (Table 2). Similar results were found in other pigmented table and wine grapes. Kliewer and Weaver (8) showed that 18.7 clusters per vine (pruned and thinned) increased the percentage of coloration by 57% in comparison with 31.8 clusters per vine, and by 85% in comparison with 120 clusters per vine. Light crop loads in several red wine grape cultivars ('Alicante Bouschet,' 'Carignane,' 'Petite Sirah,' 'Pinot Pernand,' and 'Zinfandel'), were shown to produce more highly colored fruits than heavy crop loads (15).

Table 2. Effect of Cluster Thinnin on Individual Anthocyanins and Total Anthocyanin Content in Fruit Berry Skin.

Treatment (cluster number/vine) Total Anthocyanin Content (mg/g) Dp-3-gz (mg/g) Cy-3-g (mg/g) Pt-3-g (mg/g) Pn-3-g (mg/g) Mv-3-g (mg/g) Acylated Cy Derivative (mg/g)
60
3.83 by
0.59 a
2.64 b
0.06 a
0.15 b
0.03 b
0.50 b
40
4.60 ab
0.61 a
3.29 ab
0.06 a
0.17 b
0.03 b
0.58 b
20
5.99 a
0.76 a
4.35 a
0.08 a
0.22a
0.04 a
0.71
LSD
1.65
0.27
1.30
0.03
0.07
0.01
0.12
Linear Contrast
**x
NS
**
NS
*
NS
**
Quadratic Contrast
NS
NS
NS
NS
NS
NS
NS

z Abbreviations for individual anthocyanins: Dp-3-g = delphin-3glucoside; Cy-3-g = cyanidin-3-glucoside; Pt-3-g = petunidin-3-glucoside; Pn-3-g = peonidin-3-glucoside; Mv-3-g = malvidin-3-g = malvidin-3-glucoside; acylated Cy derivative = acylated cyanidin-glycoside.
y Mean separation by least significant difference (LSD) at P = 0.05 level.
x Orthogonal contrast at P = 0.05 or 0.01 level; *, **, NS = significant at 0.05 or 0.01 level, and not significant, respectively.


Cluster thinning affected the concentration of individual anthocyanins in 'Reliance' grape unequally. The concentration of cyanidin-3-glucoside, peonidin-3-glucoside, and the acylated cyanidin derivative were all increased linearly by cluster thinning (Table 2). However, the concentration of delphinidin-3-glucoside, petunidin-3-glucoside, and malvidin-3-glucoside, was not significantly affected. There was no quadratic relationship between cluster thinning and the concentration of individual anthocyanins. Cacho et al. (2) reported that the concentration of all five anthocyanidins in such grape cultivars as 'Tempranillo,' 'Moristel,' and 'Garnacha' increases similarly as total anthocyanin from veraison to full maturity. Without conducting a developmental study of the concentration of individual anthocyanins in 'Reliance,' it is hard to know why the synthesis of only cyanidin-3-glucoside, peonidin-3-glucoside, and acylated cyanidin-glycoside were increased by cluster thinning. The authors hypothesized that carbohydrate allocation and complexicity of individual anthocyanins may have something to do with this phenomena.

The percentages of two anthocyanins, in relation to total anthocyanin content, were also affected by cluster-thinning treatments (Table 3). The percentages of cyanidin-3-glucoside and the acylated cyanidin derivative were increased linearly by cluster thinning. The percentages of delphinidin-3-glucoside, petunidin-3-glucoside, peonidin-3-glucoside, and malvidin-3-glucoside were not affected by cluster thinning. There was no quadratic relation between the percentages of individual anthocyanins and cluster thinning.

Table 3. Effect of Cluster Thinnin on the Percentages of Individual Anthocyanins.

Treatment (cluster number/vine) Dp-3-gz (%) Cy-3-g (%) Pt-3-g (%) Pn-3-g (%) Mv-3-g (%) Acylated Cy Derivative (%)
60
12.5 ay
67.3 b
1.4 a
4.4 a
0.8 a
13.6 a
40
12.0 a
68.6 b
1.3 a
4.4 a
0.6 b
13.2 a
20
12.2 a
71.4 a
1.3 a
3.7 a
0.6 b
10.8 b
LSD
3.6
4.1
0.4
0.9
0.2
2.9
Linear Contrast
NSx
*
NS
NS
NS
*
Quadratic Contrast
NS
NS
NS
NS
NS
NS

z Abbreviations for individual anthocyanins: Dp-3-g = delphin-3glucoside; Cy-3-g = cyanidin-3-glucoside; Pt-3-g = petunidin-3-glucoside; Pn-3-g = peonidin-3-glucoside; Mv-3-g = malvidin-3-g = malvidin-3-glucoside; acylated Cy derivative = acylated cyanidin-glycoside.
y Mean separation by least significant difference (LSD) at P = 0.05 level.
x Orthogonal contrast at P = 0.05 or 0.01 level; *, **, NS = significant at 0.05 or 0.01 level, and not significant, respectively.


Cluster thinning had the opposite effect on the percentage of cyanidin-3-glucoside as for the percentage of acylated cyanidin derivative. The same relationship exists between the percentages of cyanidin-3-glucoside and malvidin-3-glucoside. The effect of cluster-thinning treatments on the percentage of individual anthocyanins does not appear to be only related to berry ripening. Roggero et al. (11) reports that anthocyanin composition in 'Syrah' is quickly set after veraison and remains nearly stable until the grapes mature except for the cyanidin derivative, which is the precursor of other pigments. Based on a calculation of ratios of individual anthocyanins presented by Cacho et al. (2), it seems that percentages of anthocyanins also remain stable after veraison. However, ethephon, a growth regulator that is well known to promote fruit ripening, altered the pigment makeup, characterized as percentages of individual anthocyanins, in 'Pinot noir' when applied at 500 ppm at veraison (9).

Cluster-thinning treatments seem to shift the balance among cyanidin-3-glucoside and acylated cyanidin derivative without significantly affecting the percentages of other anthocyanins. Cacho et al. (2) found that cyanidin-3-glucoside content naturally varies very little from ripening to maturity in 'Tempranillo,' 'Moristel,' and 'Garnacha.' The authors do not know whether cluster-thinning effects on the percentages of individual anthocyanins are related to the ethylene production in berries. It is still very interesting that both cluster thinning and ethephon could alter the percentages of anthocyanins in a grape cultivar. Measurements of ethelene production in berries after imposing cluster thinning, at different developmental stages, could help shed some light on the mechanism of anthocyanin metabolism.

The manipulation of fruit color in a grape cultivar can be achieved by alteration of anthocyanin composition alone without affecting total anthocyanin. Fruit-color improvement by cluster thinning could be attributed to increased total anthocyanin and change in anthocyanin composition. Increase in cyanidin-3-glucoside must have contributed to increased red color in 'Reliance.' However, it is very difficult to increase composition of an anthocyanin without affecting total anthocyanin content of field-grown grapes. How much of a change in anthocyanin composition is needed to significantly alter the color of a grape cultivar is still unknown.

Twenty clusters per vine were shown to be the best in improving fruit quality and color under single-curtain and high-cordon systems in northeastern Ohio. Grapevines thinned to 20 clusters per vine produced grapes of larger berry size, higher soluble solids, and better fruit color than those thinned to 40 or 60 clusters per vine. This study was an attempt to investigate the effects of cluster thinning on both total and individual anthocyanins of 'Reliance.' In the future, such crop levels, ranging from 20 to 80 clusters per vine, alone and in combination with other variables such as temperature, growth regulators, or fertilization, should be used to investigate change in individual anthocyanins from veraison to maturity by progressive sampling. More research is needed to understand anthocyanin metabolism in pigmented grapes and to achieve effective manipulation of fruit color.

Literature Cited

  1. Amerine, M. A. and C. S. Ough. 1980. Methods for analysis of musts and wines. John Wiley and Sons, New York, N.Y.
  2. Cacho, J., P. Fernandez, V. Ferreira, and J. E. Castellis. 1992. Evolution of five anthocyanidin-3-glucosides in the skin of the 'Tempranillo,' 'Moristel,' and 'Garnacha' grape varieties and influence of climatological variables. Amer. J. Enol. Vitic. 43:244-248.
  3. Gao, Y. and Garth A. Cahoon. 1995. High performance liquid chromatographic analysis of anthocyanins in the red seedless table grape 'Reliance.' Amer. J. Enol. Vitic. 46:339-345.
  4. Hepner, Y. and B. Bravdo. 1985. Effect of crop level and drip irrigation scheduling on the potassium status of 'Cabernet Sauvignon' and 'Carignane' vines and its influences on must and wine composition and quality. Amer. J. Enol. Vitic. 36:140-147.
  5. Hong, V. and R. E. Wrolstad. 1990. Use of HPLC/photodiode array detection for characterization of anthocyanins. J. Agric. Food. Chem. 38:708-715.
  6. Hrazdina, G. 1975. Anthocyanin composition of 'Concord' grapes. Lebensm. Wiss. Technol. 8:111-113.
  7. Hunter, J. J., O. T. Devilliers, and J. E. Watts. 1991. The effects of partial defoliation on quality characteristics of Vitis vinifera L. cv. Cabernet Sauvignon. Amer. J. Enol. Vitic. 41:13-18.
  8. Kliewer, W. M. and R. J. Weaver. 1971. Effect of crop level and leaf area on growth, composition, and coloration of 'Tokay' grapes. Amer. J. Enol. Vitic. 22:172-177.
  9. Powers, J. R., E. A. Shively, and C. W. Nagel. 1980. Effect of ethephon on color of 'Pinot Noir' fruit and wine. Amer. J. Enol. Vitic. 31:203-205.
  10. Reynolds, A. G. 1989. Impact of pruning, cluster thinning, and shoot removal on growth, yield, and fruit composition of low-vigor 'De Chaunac' vines. Can. J. Plant. Sci. 69:269-275.
  11. Roggero, J., P. S. Coen, and B. Ragonnet. 1986. High performance liquid chromatography survey in pigment content in ripening grapes of Syrah. An approach to anthocyanin metabolism. Amer. J. Enol. Vitic. 37:77-83.
  12. Roubelakis-Angelakis, K. A. and W. M. Kliewer. 1986. Effects of exogenous factors on anthocyanin and total phenolics in grape berries. Amer. J. Enol. Vitic. 37:275-280.
  13. Van Zyl, J. L. and H. W. Webber. 1977. Irrigation of 'Chenin blanc' in the Stellenbosch area within the framework of the climate-soil-water continuum. Int. Syp. Qual. Vintage 1977:331-350. (Abstr.). Amer. J. Enol.Vitic. 30:259.
  14. Weaver, R. J. and R. Montgomery. 1974. Effect of ethephon on coloration and maturation of wine grapes. Amer. J. Enol. Vitic. 25:39-41.
  15. Weaver, R. J., M. A. Amerine, and A. J. Winkler. 1957. Preliminary report on effect of level of crop on development of color in certain red wine grapes. Amer. J. Enol. Vitic. 9:157-166.
  16. Winkler, A. J., J. A. Cook, W. M. Kliewer, and L. A. Lider. 1974. General viticulture. 2nd Ed. University of California Press. Berkeley, Calif.
  17. Wulf, L. W., and C. W. Nagel. 1978. High pressure liquid chromatographic separation of anthocyanins of Vitis vinifera. Amer. J. Enol. Vitic. 24:42-49


Back | Forward | Table of Contents