Abstract
Background and goals Muscat Bailey A (MBA), a prominent red wine grape cultivar in Japan, is distinguished from traditional Vitis vinifera cultivars by its large berries and low sugar concentration. This study aimed to elucidate the relationship between berry size and three key quality parameters of MBA: total soluble solids (TSS), anthocyanins, and 4-hydroxy-2,5-dimethyl-3(2H)-furanone (furaneol). Furaneol is responsible for the characteristic aroma of MBA berries, reminiscent of cotton candy and strawberry.
Methods and key findings During the three growing seasons of 2018, 2020, and 2021, individual berries were classified based on their position within the bunch and their berry weight. Berry position within a bunch had no effect on berry weight or TSS. Across all three growing seasons, small berries consistently exhibited higher TSS, higher anthocyanins in the skin, and generally higher furaneol than larger berries, although annual variations were observed.
Conclusions and significance Smaller berry size in MBA was linked to higher TSS and anthocyanins, indicating potential quality enhancements. These findings have implications for viticultural practices aimed at reducing berry size and for breeding programs focused on developing small-berry phenotypes, ultimately enhancing the potential to produce higher-quality MBA wines with deeper color and higher alcohol concentration.
Introduction
The sugar content of winegrapes is a critical factor affecting wine alcohol content (Jordão et al. 2015), and certain cultivars naturally exhibit low sugar concentration in their berries. Muscat Bailey A (MBA; Vitis labruscana cv. Bailey × Vitis vinifera cv. Muscat Hamburg), developed in Japan in 1927 (McLeRoy and Renfro 2008), is currently the most commonly cultivated grape for red wine production in Japan; in 2013, MBA was listed in the International List of Vine Varieties and Their Synonyms (OIV 2013). The sugar concentration of MBA tends to be low overall, although it depends largely on the growing region. Even with optimal leaf removal, total soluble solids (TSS) typically reach ~19 Brix in Yamanashi Prefecture, a famous wine-producing region in Japan (Matsuyama et al. 2014). Unless sugar is added for fermentation, the inherently low TSS results in wines with low alcohol content. Alcohol affects the perception of aromatic compounds and the detection of volatile aroma compounds (Goldner et al. 2009). Currently, to produce standard MBA wines, the juice must be chaptalized to reach a sugar level of 22 to 24 Brix, using sucrose.
The low sugar content of MBA berries may be attributed to their large berry size. MBA berries weigh 5 to 6 g at harvest, larger than traditional European winegrapes (Matsuyama et al. 2014). Similarly, a study on Cabernet Sauvignon berries in Oakville, California demonstrated a negative relationship between berry size and TSS (Roby et al. 2004). The sugar concentration also varies within the bunch, with the lower position differing from the upper position. For example, within individual bunches of Cabernet franc, berries from the bottom third of the bunch were considerably higher in sugar concentration than berries sampled from the top third (Pagay and Cheng 2010). However, no scientific evidence yet shows a direct link between berry size and/or berry position within the bunch and the TSS of MBA.
Anthocyanins in grape berry skin are also dependent on berry size. Anthocyanins per berry weight decrease with increasing berry size (Roby et al. 2004). MBA, with its large berry size, exhibits a lower skin-to-juice ratio compared with V. vinifera cultivars. Leaf removal around MBA bunches can increase anthocyanins by activating anthocyanin biosynthesis genes (Matsuyama et al. 2014). However, the ongoing rise in global temperatures has led to poor color development in berry skins in current winegrape-producing regions, and has resulted in the location of viticultural areas shifting toward higher latitudes or elevations (Hannah et al. 2013). High temperatures during berry ripening significantly reduce anthocyanin accumulation (Mori et al. 2007). Abscisic acid, which regulates anthocyanin biosynthesis in berry skins (Pilati et al. 2017), is reduced under high temperatures from preveraison to harvest, thereby inhibiting anthocyanin accumulation (Koshita et al. 2007).
This study aimed to investigate the effect of natural berry weight variation on TSS and anthocyanins in MBA berries. This cultivar exhibits berry weight variation even within individual bunches (Figure 1A). The relationship was explored between natural berry weight variation and 4-hydroxy-2,5-dimethyl-3(2H)-furanone (furaneol), the compound responsible for the characteristic cotton candy and strawberry aromas of MBA berries (Kobayashi et al. 2013, Sasaki et al. 2015, Goto-Yamamoto 2019). Furaneol and its glucopyranoside primarily accumulate in the flesh of MBA berries (Sasaki et al. 2015). The modulation of furaneol concentration is contingent upon the intended sensory profile of the wine. While some winemakers strategically increase furaneol to accentuate fruity and confectionery-like aromas, others may minimize its presence to maintain a more traditional red wine flavor composition. In V. vinifera cultivars, berry size variation influences aroma composition (Xie et al. 2018). For example, terpenes were higher in medium (12 to 14 mm) Merlot and Cabernet Gernischt berries than in larger berries (>14 mm; Xie et al. 2018). Similarly, medium Riesling berries (12.5 to 14 mm) had higher terpenoids than large berries (14 to 16 mm; Friedel et al. 2016).
Japanese grape cultivar Muscat Bailey A (MBA). A) An MBA bunch showing variation in berry size. B) Overhead shelf system (pergola style). C) Comparison of berry size: MBA is larger than European Vitis vinifera cultivars Merlot (ME), Cabernet Sauvignon (CS), Syrah (SY), and Pinot noir (PN). Scale bar = 1 cm.
Materials and Methods
Plant materials and bunch sampling
MBA grapevines were cultivated in the experimental vineyard of the University of Yamanashi, Kofu, Yamanashi Prefecture, Japan (35°36′N; 138°34′E; 250 m asl). The grapevines were 9-yr-old, grafted onto Kober 5BB rootstock, and trained to an overhead shelf system (pergola style) (Figure 1B). Ten bunches were randomly selected from two grapevines (five bunches per grapevine) at harvest (15 Oct 2018, 5 Oct 2020, and 2 Oct 2021).
Berry sorting and categorization
Each berry was individually removed from the bunch and numbered. Berries were categorized into three groups (top, middle, and bottom) based on their position within the bunch. The weight of each berry was measured using an electronic balance (EK2000i; A & D Co.) and the mean berry weight for each bunch was calculated. The same berries were also categorized into three size groups: small (<1 g of the mean berry weight), medium (within 1 g of the mean berry weight), and large (>1 g of the mean berry weight).
Measuring berry TSS
Each berry was manually pressed to extract juice. The TSS of each juice sample was measured using a refractometer (PAL-BX/ACID2; Atago).
Quantifying berry skin anthocyanins
Seven berries were randomly selected from each berry weight group. The skin of each berry was carefully removed using tweezers, frozen in liquid nitrogen, and ground to a fine powder. Anthocyanin extraction from each berry skin was performed following the method described by Yokotsuka et al. (1999), with minor modifications. Briefly, 0.1 g of each powdered skin sample was macerated in 2 mL of hydrogen chloride-methanol [36:1 (v/v)] for 16 hr at 4°C in the dark. After thorough mixing, the absorbance at 520 nm (OD520) of each sample was measured using a spectrometer (UV-1800, Shimadzu). Total anthocyanins were calculated using a previously published formula (Bakker et al. 1986) and expressed as milligrams of malvidin 3-O-glucoside equivalent per gram of fresh skin weight.
Measuring furaneol and its glucopyranoside
Extraction and quantification of furaneol and its glucopyranoside from MBA berries were performed according to the method described by Sasaki et al. (2015), with slight modification. Briefly, six to eight berries were randomly selected from each berry weight group. The juice was extracted from each berry using a juicer, pressing each berry to 40% of its original weight. The supernatant was obtained by centrifuging each juice sample at 3000 × g for 8 min. The supernatant was then diluted with an equal volume of 0.1% formic acid and filtered through a 0.45-μm cellulose acetate filter (DISMIC-25CS; Advantec). The filtrate was analyzed using high-performance liquid chromatography-tandem mass spectrometry to quantify furaneol concentration, following the method by Sasaki et al. (2015).
Statistical analysis
To visualize the relationship between TSS and berry weight, a regression analysis was performed and the regression equation and p values were calculated using Excel (2019; Microsoft Corp.). Statistical analysis was conducted using analysis of variance and Tukey’s test in Excel statistics software 2012 (Social Survey Research Information). The number of replicates for each experiment is indicated in the figure captions.
Results
Berry position does not affect berry weight and TSS
The larger size of MBA berries in comparison to those of European winegrapes such as Merlot, Cabernet Sauvignon, Syrah, and Pinot noir is depicted (Figure 1C). With an average transverse diameter of ~2 cm, MBA berries exhibited size variation even within a single bunch, regardless of position (Figure 1A).
During the 2018 growing season, 814 berries were collected from 10 bunches: 358 berries from the top, 255 from the middle, and 201 from the bottom of each bunch. No differences in mean berry weight and TSS were observed among berries from different positions within the bunch (Figure 2A). Conversely, regression analysis of berry weight and TSS for individual berries within each position revealed a negative relationship between these two variables, regardless of berry position (Figure 2B).
Berry weight and total soluble solids (TSS) of Muscat Bailey A berries are independent of berry position within the bunch. A) Comparison of mean berry weight and TSS among berry positions: top, middle, and bottom of the bunch. Boxplots represent the distribution of the data: the horizontal line inside the boxes indicates the median, and the lower and upper edges of the boxes correspond to the first (Q1, 25th percentile) and third quartiles (Q3, 75th percentile), respectively. Whiskers represent the minimum and maximum values within a certain range, inner data points and outliers are indicated by open circles, and crosses (×) indicate the mean values for each berry position. B) Regression lines and their equations, calculated using Excel (2019; Microsoft Corp.), show the relationship between berry weight and TSS for berries from the top, middle, and bottom of the bunch. A negative relationship between berry weight and TSS for individual berries within each position was observed (p < 2.06 × 10−8 for top, p < 2.42 × 10−7 for middle, p < 2.98 × 10−8 for bottom).
Berry weight and TSS
Using the 814 berries collected from 10 bunches in the 2018 growing season, a regression analysis was first performed on all berries regardless of berry weight or position, followed by individual analyses for each cluster (Figure 3). The regression analysis using 814 berries revealed a negative relationship between smaller berry size and higher TSS (Figure 3A).
Negative regression between berry weight and total soluble solids (TSS) of Muscat Bailey A. A) Regression line between berry weight and TSS for 814 berries collected from 10 bunches in 2018. A negative relationship between smaller berry size and higher TSS was observed (p < 2.80 × 10−19). B) Regression line between berry weight and TSS for individual bunches collected in 2018. A significant relationship between berry weight TSS was observed in Bunch 1, 2, 3, 4, 8, and 9 (p < 0.0079, 0.014, 0.0023, 0.0059, 0.0015, and 0.029, respectively. Regression lines and equations were calculated using Excel (2019; Microsoft Corp.).
Berries were collected from 10 individual bunches, with the following counts: 110 berries from Bunch 1, 84 from Bunch 2, 76 from Bunch 3, 54 from Bunch 4, 79 from Bunch 5, 103 from Bunch 6, 74 from Bunch 7, 64 from Bunch 8, 98 from Bunch 9, and 72 from Bunch 10. In six of these bunches, a significant relationship between berry weight and TSS was observed (Figure 3B). These results suggest that berry weight may influence sugar accumulation.
To further examine the relationship between berry weight and TSS, 814, 813, and 800 berries collected from 10 bunches in 2018, 2020, and 2021, respectively, were categorized into three size groups (small, medium, and large), as described in the Materials and Methods. The number of berries in each group was as follows: 133 small, 529 medium, and 152 large in 2018; 263 small, 271 medium, and 279 large in 2020; and 255 small, 271 medium, and 274 large in 2021. Regardless of the growing season, small berries consistently exhibited high TSS and large berries had low TSS (Figure 4A). Overall, these results suggest an inverse relationship between berry weight and TSS in MBA.
Effect of berry size on berry characteristics of Muscat Bailey A. A) Total soluble solids (TSS) in 2018, 2020, and 2021; B) anthocyanins in 2020 and 2021; C) furaneol and furaneol glucopyranoside in 2018; and D) furaneol in 2020 and 2021. Berries were collected in 2018, 2020, and 2021 and categorized into three size groups: small, medium, and large. Boxplots represent the distribution of the data: the horizontal line inside the boxes indicates the median, and the lower and upper edges of the boxes correspond to the first (Q1, 25th percentile) and third quartiles (Q3, 75th percentile), respectively. Whiskers represent the minimum and maximum values within a certain range; inner data points and outliers are indicated by open circles; and crosses (×) indicate mean values of TSS (all berries), anthocyanins (seven berries), and furaneol and furaneol glycoside (six to eight berries). Different letters above the bars denote significant differences (p < 0.05, Tukey’s test).
Effect of berry size on skin anthocyanins
Differences in berry skin anthocyanins were confirmed based on skin weight in 2020 and 2021 (Figure 4B) (anthocyanin data were lost for the 2018 growing season due to a laboratory error). Small berries consistently exhibited high anthocyanins (~3 mg/g skin fresh weight), which was 150 to 200% greater than that of medium and large berries. There were no differences in anthocyanins between medium and large berries in either year. These results suggest that smaller MBA berries tend to accumulate higher levels of anthocyanins, compared with larger berries.
Berry size and furaneol
In the 2018 growing season, no differences in furaneol concentration were observed among different-sized berries, although smaller berries tended to have higher furaneol (Figure 4C). By contrast, no berry size-related differences were observed in furaneol glucopyranoside, the precursor of furaneol. Based on the 2018 results, furaneol glucopyranoside was not measured in 2020 and 2021. In the 2020 growing season, differences in berry furaneol were observed based on berry size: small berries had the highest furaneol and large berries had the lowest (Figure 4D). No differences in furaneol were observed among different-sized berries in 2021, although smaller berries tended to have higher furaneol, as in 2018.
Discussion
As demonstrated by Reshef et al. (2017), the intensity and direction of solar radiation can modulate sugar accumulation patterns in Cabernet Sauvignon berries. This suggests that berry orientation within a bunch can affect sugar accumulation. For instance, Cabernet Sauvignon berries from bunches on the east side of the vine row may exhibit sugar levels different from those on the west side (Reshef et al. 2019). Similarly, berry position within the bunch contributes to differences in sugar concentration among Sauvignon blanc berries, depending on row orientation and training (Trought et al. 2017). These effects of berry position on sugar accumulation may be attributable to the training system, particularly vertical shoot positioning. By contrast, MBA grapevines are typically trained to an overhead shelf system (Figure 1B), allowing for uniform sunlight penetration through the canopy to the bunches. Leaf removal around the bunches of MBA grapevines trained to an overhead shelf system can accelerate sugar accumulation by increasing solar radiation exposure to the bunches (Matsuyama et al. 2014). However, as leaf removal was not applied in this study, berry position within the bunch may not have had an effect on berry sugar concentration in MBA trained to an overhead shelf system.
Previous studies, along with the present findings, indicate a strong relationship between berry size and composition. In Cabernet Sauvignon, larger berries were associated with lower TSS and anthocyanins, whereas skin tannins remained stable (Roby et al. 2004). In Merlot, Cabernet Gernischt, and Riesling, terpenoid concentrations varied with berry size, with larger berries generally exhibiting lower terpenoids (Friedel et al. 2016, Xie et al. 2018). This study further revealed that smaller MBA berries accumulated higher TSS and anthocyanins than larger berries. Although berry size had a minimal effect on furaneol and no effect on furaneol glucopyranoside, furaneol, with a low odor threshold (10 μg/L; Re et al. 1973), continued to accumulate in MBA berries throughout the ripening period (Kobayashi et al. 2013). In the present study, furaneol in MBA berries exceeded the odor threshold at harvest, regardless of berry size or growing season. While the intensity of the cotton candy aroma can vary between wines, berry size seems to have minimal effect on the characteristic aroma of MBA. Overall, smaller berries may be more desirable for MBA wine production.
Multiple strategies are possible to reduce the size of MBA berries. One potential approach is water deficit, which has been shown to inhibit berry growth in Cabernet Sauvignon. For instance, delaying irrigation until midday leaf water potential reaches −1.5 MPa resulted in berries that were 14% smaller than those that received control irrigation (Roby and Matthews 2004). However, this approach is not possible for MBA in Japan because rainfall during the grapegrowing season is abundant and irrigation is uncommon. Another promising technique is the induction of lateral shoots, which has been developed to mitigate the effects of global warming on MBA ripening (Kishimoto et al. 2017). This technique reduced MBA berry size by 25%, compared with conventional methods, and increased TSS and anthocyanins (Kishimoto et al. 2017). Wines produced from berries on lateral shoots exhibited more than double the furaneol concentration compared with wines from conventionally managed grapevines. Interestingly, this technique also reduced the size of Merlot berries and increased the anthocyanins in their skins (Kishimoto et al. 2022). The present study does not account for the variability in berry size across different viticultural environments, which depend on regional climate conditions and vineyard training methods. Future investigations should focus on viticultural management strategies, including timing and intensity of bunch thinning, leaf removal and pruning, shoot density, and canopy structure for overhead shelf system, to better understand their effect on berry morphology.
Conclusion
Smaller berry size is often linked to improved wine quality parameters. MBA, which has berries larger than those from typical V. vinifera cultivars, exhibits lower TSS and skin anthocyanins, which could result in wines with reduced color intensity. This study demonstrated that smaller MBA berries have higher TSS and higher anthocyanins. Further research should investigate viticultural practices that can reduce berry size, and alternative breeding approaches such as radiation breeding, to develop small-berry phenotypes. These could expand the stylistic diversity of MBA wines.
Data Availability
All data underlying this study are included in the manuscript.
Footnotes
This research was funded by Japan Society for the Promotion of Science, grant number 24K08892
Aoki Y, Sasaki K, Kido E and Suzuki S. 2025. Effects of Muscat Bailey A berry size on total soluble solids, anthocyanins, and 4-hydroxy-2,5-dimethyl-3(2H)-furanone. Am J Enol Vitic 76:0760011. DOI: 10.5344/ajev.2025.24070
By downloading and/or receiving this article, you agree to the Disclaimer of Warranties and Liability. If you do not agree to the Disclaimers, do not download and/or accept this article.
- Received December 2024.
- Accepted February 2025.
- Published online May 2025
This is an open access article distributed under the CC BY 4.0 license.
References
Vol 76 Issue 1
Jump to section
Related Articles
Cited By...
More from this TOC section
Similar Articles
