OENO One The sugarless grape trait characterised by single berry phenotyping 1 UE INRAE de Pech Rouge, 11430, Gruissan, France 2 AGAP, University of Montpellier, CIRAD, INRAE, Institut Agro, 34060 Montpellier Cedex, France 3 GENOVIGNE, IFV, INRAE, Institut Agro, 2, place P. Viala, 34060 Montpellier Cedex, France 4 LEPSE, University of Montpellier, CIRAD, INRAE, Institut Agro, 34060 Montpellier Cedex, France Bigard Antoine 1 2 Romieu Charles 2 3 Torregrosa HernánOjeda1andLaurent 1 4 * * Key words: Fleshy fruit, fruit development, ripening, sugars, acidity, climate warming Introduction

With a hexose concentration (glucose + fructose, [Hex]) of higher than 1.1 mol/l at the ripe stage, grape is one of the richest fleshy fruits in terms of sugars. [Hex] is known to depend on the GxExM interaction (Suter et al., 2021), and until recently the adaptation of cultivars to local conditions was thought to be related to the thermal time needed to reach specific vegetative and reproductive phenological stages, such as budburst, flowering or fruit veraison (Parker et al., 2020). The selection of grapevine varieties is mainly based on the balance between sugars and acidity (Torregrosa et al., 2017; Ollat et al., 2018; Duchêne et al., 2020). In cold climates, early ripening varieties are preferred in order to ensure the accumulation of sugars and secondary metabolites before autumn. Conversely, in warm regions, late ripening varieties which shift fruit ripening to cooler days preserve organic acids (Rienth et al., 2016), anthocyanidin (Zhang et al., 2015) and aroma compounds (Alessandrini et al., 2018; Asproudi et al., 2016; Gutiérrez-Gamboa et al., 2018). In practice, however, a range of oenological processes are implemented to correct the sugars in the must and/or its acidity, demonstrating that the supposed adaptation of the varieties based on thermal time phenology is oversimplified. Over the last decades, temperature during the period of grapevine fruit ripening has been constantly increasing, which can lead to excessive sugars and insufficient acidity in hot vine growing areas, such as Mediterranean regions (Santillán et al., 2019, Bécart et al., 2022). Given the strength and the rate of ongoing climate changes, it is critical to precisely define the development and the metabolism of fruits to accurately identify their adaptation potential (Bigard et al., 2018 and 2020). There is thus increasing interest in breeding new cultivars which can ripen at lower sugar concentration while preserving fruit acidity.

In fact, warming does not only accelerate the whole ripening process (which would be easily solved by harvesting earlier), but it also desynchronises different aspects of ripening, thus accelerating malic acid breakdown (Rienth et al., 2016, Sweetmann et al., 2014) and inhibiting the accumulation of secondary metabolites, such as anthocyanins; this leads to making the decision of shifting the harvest date to when grapes have a higher [Hex] (Arrizabalaga et al., 2018). Moreover, comparing cultivars at the same developmental stage (the so-called ripe stage) raises both methodological and conceptual issues. In most comparative studies, authors predetermine an average harvest date, ripening duration, thermal time or even [Hex] threshold to compare accessions at a hopefully similar ripe stage, which contradicts the recognised impact of GxE on these variables (Liu et al., 2007; Dai et al., 2011; Duchêne et al., 2012). To circumvent this inconsistency and lack of consensus, the moment at which berry phloem unloading stops was recently proposed as a relevant definition of ripe stage from both a physiological and transcriptomic point of view (Bigard et al., 2018; Bigard et al., 2020; Shahood et al., 2015; Shahood et al., 2020; Savoi et al., 2021). This key transition which marks the arrest of the most intensive fruit biochemical pathways is associated with the transcriptional extinction of many genes, encoding, amongst others, sugar transporters, aquaporins and cell wall-related enzymes (Savoi et al., 2021). Unfortunately, this developmental stage cannot be directly inferred from [Hex] kinetics, which continue to evolve after the completion of sugar storage (or accumulation) due to subsequent berry shrivelling. Thus, additional information on berry growth is required to address the net rates of sugar accumulation and malic acid breakdown in berries, together with their respective timings. Very recently, single berry phenotyping approaches improved our understanding of berry growth and metabolism during ripening, avoiding any biases resulting from berry asynchronicity (Shahood et al., 2020; Savoi et al., 2021). Based on this new paradigm, this study aimed to decipher the genetic differences existing for [Hex] and fruit acidity in a set of genotypes encompassing traditional varieties and new hybrids exhibiting low [Hex] at harvest (Escudier et al., 2017; Ojeda et al., 2017).

Materials and method 1. Plant Material and sampling method

Experiments were performed outdoors at the INRAE Pech Rouge experimental unit (Gruissan, France, 43.14'/3.14''W) under a semi-arid Mediterranean climate in 2017 (temperature, rainfall and evapotranspiration data described in Alem et al., 2021). Experimental plots were managed through drip irrigation keeping the predawn leaf water potential (ΨPD) higher than -0.5 MPa (Giorgi & Lionello, 2008) to correspond to a moderate water stress. The set of the genotypes comprised 3 traditional varieties, Grenache N, Merlot N and Morrastel N (

https://plantgrape.plantnet-

project.org/fr/

), and 3 new disease-resistant varieties derived from 4 (3197-81B, 3197-373N) or 5 (3184-1-9N) backcrosses of Muscadinia rotundifolia with V. vinifera varieties (Bouquet al., 1980). These three latter genotypes are known to display a reduced [Hex] at harvest, enabling the production of wines at low ethanol levels, called VDQA, “ V ins D e Q ualité à teneur modérée en A lcool” (Ojeda et al. 2017). Hereafter, the names G5, G7 and G14 will be used for 3197-81B, 3197-373N and 3184-1-9N respectively. On a weekly and random basis, 60 berries were collected during a period of 2 weeks before to 2 weeks after the first signs of berry softening, and 30 berries were collected during the rest of the ripening period over a period of 1 week after fruit shriveling. Whole berries were sampled between 09:00 and 11:00 by cutting the fruit peduncle just below the calyx; they were kept in a plastic bag in a cool place and analysed on the same day.

 2. Berry firmness and composition

Firmness was monitored with a digital penetrometer called Pénélaup™ (Abbal et al., 1992; Robin et al., 1997) as described in Shahood et al. (2020). Immediately after the firmness measurement, berries were immersed in 4 times their weight of 0.25N HCl. Seeds were immediately removed and samples incubated for 48h. Samples were vigorously shaken and a first 100 L aliquot was diluted 11 times with 8.3 10-3 N acetic acid (internal control) + 16.4 10-3 N sulphuric acid and centrifuged for 5 min at 20°C at 18,500 g; supernatants were injected by HPLC to quantify glucose, fructose, malic and tartaric acids using a Biorad aminex-HPX87H column, which is also described in Shahood et al. (2020). A second 100 L aliquot was diluted 10 times in water and then centrifuged for 3 min at 12000 rpm (20°C). Next, 10 μl of clear supernatant was injected into the HPLC using a Waters® IC-Pak Cation M/D 3.9x150 mm column with the same parameters used in Bigard et al. (2020), in order to obtain Potassium ([K+]), Magnesium ([Mg2+]) and Calcium ([Ca2+]) concentrations. Titratable acidity (TA) was calculated as the sum of malic and tartaric acids minus K+ in mEq.L-1.

 3. Data normalisation and presentations

R® software (version 4.1.2) was used to build graphical representations and to analyse the data (R Core Team, 2017). Tests as shapiro.test (“stats”, for normality) and bartlett.test (for variance equality) were always performed before the selection of the appropriate statistical test. Software packages were “ggplot2” (Version 3.3.5), “car” (Version 3.0-12) and “rcompanion” (Version 2.4.6).

Results and discussion

The grapevine displays small fruits clustered in grape bunches showing a huge internal asynchrony (Gouthu et al., 2014; Doumouya et al., 2014; Bigard et al., 2019; Shahood et al., 2020), as illustrated in the following figures. In previous studies, extrapolating single fruit metabolic traits from the average values observed for the population of berries led to biased kinetic interpretations and chimeric metabolic concepts. In the grapevine, where the berry is the only truly relevant physiological unit, the most accurate method would be to non-destructively characterise each single fruit kinetics (Castellarin et al., 2015; Savoi et al., 2021), which is only possible for morphological attributes (fruit colour and volume) and firmness. Regarding solute accumulation, hypodermal sampling led to excessively elevated fluxes, possibly resulting from injury to the berries, thus the validity of this result was questioned (Coombe, 1992). In previous studies, destructive sampling of density sorted berries or large sets of individual fruits allowed to get accurate physiological insights on grapevine fruit development (Rolle et al., 2013; Bigard et al., 2019; Shahood et al., 2020). Here, we have phenotyped the ripening of 3 new sugarless genotypes (G5, G7 and G14) and 3 traditional varieties (Grenache N, Merlot N and Morrastel N) though the destructive chemical analysis of thousands of single berries, as in Shahood et al. (2020). Since it was not possible to record each individual flowering or softening date, data are interpreted as a function of berry sugar concentration, a proxy for the internal time of fruit ripening (Rienth et al., 2016).

1. Berry development and sugar accumulation

Figure1 shows the evolution of berry weight and tartaric acid concentration during the ripening of Morrastel N, which is representative of the panel of varieties described in this study. Observed trends are typical of V. vinifera genotypes, with nearly double the berry weight and a twofold dilution of tartaric acid during ripening. As described in Rienth et al. (2016) and in Bigard et al. (2020), the dilution of tartaric acid appears to be a relevant indicator of berry relative growth, as its quantity does not evolve during or after ripening (Ruffner, 1982; Lang and Thorpe, 1989; Terrier and Romieu, 2001; Rienth et al., 2014; Rösti et al., 2018; Burbidge et al., 2021). When the uploading of sugars and water in the berries stops (Coombe & McCarthy, 2000; Conde et al., 2007; Savoi et al., 2021), hexoses and tartaric acids just continue to concentrate due to evaporation. In theory, evaporation necessarily results in a straight line passing through the origin when each solute concentration is plotted against each other, as can be observed in Figure 1b. Tartaric acid concentration was obviously less scattered than fresh berry weight, thus facilitating the identification of the max berry weight stage. With this method, it was possible to determine the berry [Hex] at the maximum fruit weight for each variety; i.e., 920 mmol/L for G5, 900 mmol/L for G7, 800 mmol/L for G14, 1125 mmol/L for Grenache N, 1125mmol/L for Merlot N and 1000 mmol/L for Morrastel N. We used 10 berries which were the closest to each other for each genotype and both stages (end of green growth and max berry weight) to perform statistics as recorded in Table 1. For the end of the green growth period, the ten berries showing the highest malic acid concentration were selected.

Figure 1. Evolution of fresh weight (A) and tartaric acid concentration (B) of single berries of Morrastel during ripening. Each dot corresponds to a single hard (solid circle) or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem unloading arrest (indicated by the red arrow). In Figure 1B, a regression fit between the origin and all points beyond the arrow is displayed.

Figure 1. Evolution of fresh weight (A) and tartaric acid concentration (B) of single berries of Morrastel during ripening. Each dot corresponds to a single hard (solid circle) or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem unloading arrest (indicated by the red arrow). In Figure 1B, a regression fit between the origin and all points beyond the arrow is displayed.

Despite genotypic differences in berry weight at the green stage (Figure 2A), the 6 genotypes displayed classical fruit expansion kinetics during ripening (Coombe, 1976). The relative weight increment (i.e., Vripe/Vveraison) was obtained using all possible combinations of green- and ripe-selected berries, ranging from 1.8 +/- 0.5 to 3.1 +/- 0.7 for Merlot N and G5 respectively.

Table 1. Berry weight, firmness, sugar concentration and acidity of the 6 studied genotypes at the end of the green growth phase and at the physiological ripe stage. FW (Fresh weight), Text. (Texture), Hex (Hexoses), TA (Tartaric acid), MA (Malic acid). Statistical significance: * (p < 0.01), ** (p < 0.001), *** (p < 0.0001), na (p > 0.01).
titre du tableau
Genotype Stage FW (g.berry-1) Texture (g.mm-1) Hexoses (mmol.L-1) Acidity (meq.L-1)
Mean SD Mean SD Mean SD Mean SD
G14 Green 1.3 0.4 ab 1993 428 ab 66 11 a 576 23 ab
G5 0.9 0.1 cd 1911 331 ab 73 15 ab 712 54 c
G7 1.3 0.2 a 1683 605 ab 110 55 bcd 705 62 c
Grenache 1.1 0.3 abc 1861 446 ab 99 13 cd 562 38 a
Merlot 0.9 0.2 bcd 2088 431 b 78 17 abc 629 37 bd
Morrastel 0.7 0.2 d 1490 356 a 122 14 d 655 16 cd
Test & Post-Hoc Genotype effect Kruskal-Wallis & Dunn 4 10-5 *** Anova & Tukey 5 10-2 * Kruskal-Wallis & Dunn 2 10-5 *** Kruskal-Wallis & Dunn 1 10-8 ***
G14 Ripe 2.5 0.3 b 163 79 na 809 33 a 80 25 a
G5 2.5 0.4 bc 152 41 na 918 13 b 90 36 a
G7 3.1 0.7 c 196 59 na 905 17 ab 128 39 a
Grenache 2.2 0.5 ab 184 33 na 1130 23 c 116 37 a
Merlot 1.6 0.3 a 217 89 na 1123 9 cd 104 36 a
Morrastel 1.8 0.4 a 154 43 na 997 9 bd 116 47 a
Test & Post-Hoc Genotype effect Anova & Tukey 3 10-8 *** Anova & Tukey 1 10-1 Kruskal-Wallis & Dunn 2 10-10 *** Anova & Tukey 5 10-2 *

Regardless of the genotype, we observed considerable heterogeneity in berry size at a similar [Hex], as previously reported (Gouthu et al. 2014; Vondras et al., 2016; Bigard et al. 2019; Shahood et al., 2020). Both the maximum weight of the berries and [Hex] at this stage varied depending on the variety (Figure 2A). [Hex] increased during ripening, from values of around 100 mmol/L at the end of the green growth phase to 0.8 - 1.1 mol/L at the physiological ripe stage (Table 1). During berry ripening, at around 0.5 M each, glucose and fructose became the major osmoticums, as has been previously reported for V. vinifera (Hawker et al., 1976; Liu et al., 2006; Shiraishi et al., 2010). The [Hex] observed in this study at phloem arrest for Merlot N and Grenache N is lower than the usual concentration threshold of 1.2 - 1.5 mol/L [Hex], at which the industry considers the berries as technologically ripe. Kliewer (1967) reported the range of 1 mol/L- 1.5 mol/L [Hex] as being the technical ripeness threshold for usual V. vinifera varieties. This apparent discrepancy is due to the very common practice of pushing grapes towards over-ripeness to obtain redder, fuller and more aromatic and concentrated wines (Antalick et al., 2021). In the absence of supplementary physiological landmarks, the use of [Hex] for comparative studies is very hazardous, as this parameter steadily increases after phloem unloading arrest because of fruit shriveling (Friend et al., 2009; Shahood et al., 2020; Figure 2A).

Sucrose unloading in berries of all genotypes dramatically increased at softening, or at relaxation of turgor pressure (Figure S1). All genotypes displayed a glucose/fructose higher than 2.2 before fruit softening, then the ratio progressively converged to 1, as reported for other V. vinifera varieties(Varandas et al., 2004). No specific metabolic trends were observed for the G genotypes. It is known that the berry glucose/fructose balance, which depends on grapevine organs and the developmental stage, can be used as a metabolic indicator of fruit ripening (Kliewer et al., 1966). During green growth, the preferential use of fructose is obvious, leading to elevated G/F ratio. At softening, the import of sucrose dramatically accelerates, exceeding metabolic needs, insofar as malic acid replaces sugar as a respiratory substrate; consequently, the G/F ratio rapidly becomes 1 (Amerine and Thoukis, 1958; Liang et al., 2011; Houel et al., 2015).

Figure 2. Evolution of berry fresh mass (A) and firmness (B) during the fruit ripening of 6 grapevine varieties. Each dot represents the average of sets of 10 berries ranked according to their hexose concentration (bars on each dot represent the standard deviation).

Figure 2. Evolution of berry fresh mass (A) and firmness (B) during the fruit ripening of 6 grapevine varieties. Each dot represents the average of sets of 10 berries ranked according to their hexose concentration (bars on each dot represent the standard deviation).

For a range of extreme varieties and offsprings, Bigard et al. (2018) showed that [Hex] can vary from 750 to 1350 mmol/L when solute unloading stops just before berry shrivelling. In the present study, when phloem arrest was considered as the physiological ripe stage, the G genotypes displayed [Hex] levels of between 0.8 and 0.9 mol/L, while Morrastel N showed levels of 1 mol/L, and Merlot N and Grenache N levels over 1.1 mol/L (the highest).These data agree with previous results obtained at the whole berry population levels with similar genotypes (Ojeda et al., 2017; Bigard et al., 2019). Interestingly, the final quantity of sugar per fruit unit in sugarless genotypes is of the same magnitude as classical varieties; i.e/, 2.8 +/- 0.3 mmol (G7), 2.5 +/- 0.6 mmol (Gre), 2.3 +/- 0.4 mmol (G5), 2.0 +/- 0.3 mmol (G14), 1.8 +/- 0.3 mmol (Merlot) and 1.8 +/- 0.4 mmol (Morrastel) per berry.

Before softening, some differences were found in terms of berry firmness at the end of the green growth phase, with Morrastel N and Merlot N showing the least and the most firm fruits respectively (Table 1). From the green growth phase onwards, the mechanical properties of the berries evolved in the same way for all three genotypes (Figure 2B): the berries softened rapidly at the beginning of ripening to reach a low level of firmness until 500 mmol.L-1 of [Hex]; i.e., at more or less mid-ripening. This was followed by a slow and continuous decrease in firmness up to the physiological ripe stage and over-ripening, during which no differences in firmness were observed between the sugarless genotypes and traditional cultivars (Table 1, Figure S1). Therefore, although firmness is widely accepted as a sensitive and early indicator of the onset of ripening (Coombe et al. 1992; Abbal et al., 1992; Castellarin et al., 2015; Shahood et al., 2020; Bigard et al., 2020), this parameter cannot be used to tag the transition at the arrest of phloem unloading. Consequently, the only way to non-destructively determine the shift from fruit expansion to shrivelling remains the monitoring of berry growth. As discussed below, this can be done indirectly using tartrate concentration.

2. Evolution of the main determinants of fruit acidity 2.1. Tartaric acid

During ripening, tartaric acid concentration depended on the variety and the stage of berry development (Table 2). At the onset of ripening, tartaric acid concentration was 25 % lower (150 vs 200 mEq/L) in all G genotypes. Tartaric acid dilution (Figure S2) proceeded at a negligible rate until 220 mmol/L (G5, G14) to 300 mmol/L [Hex] (Grenache N) and then accelerated, confirming the delay between berry softening and growth resumption (Castellarin et al., 2015; Shahood et al., 2020 and other literature cited in this paper).

The change in tartaric acid from the start to the end of ripening (Figure 3) is consistent with an initial two- to three-fold dilution followed by an increase in concentration due to shrivelling, leading to a linear increase in sugar and tartaric acid passing the origin. At the arrest of phloem unloading, the concentration of tartaric acid reached a minimum, ranging from 75 +/- 10 mEq/L (G5) to 126 +/- 18 mEq/L (Merlot N). Grenache N displayed a 104 +/- 12 mEq/L concentration of tartaric acid at the ripe stage.

Table 2. Berry composition in major organic anions and inorganic cations of the 6 studied genotypes at the end of the green growth phase and at the physiological ripe stage. TA (Tartaric acid), MA (Malic acid), K (Potassium), Mg (Magnesium), Ca (Calcium). Statistical significance: * (p < 0.01), ** (p < 0.001), *** (p < 0.0001), na (p > 0.01).
titre du tableau
Genotype Stage TA (meq.L-1) MA (meq.L-1) K (mmol.L-1) Mg (mmol.L-1) Ca (mmol.L-1)
Mean SD Mean SD Mean SD Mean SD Mean SD
G14 Green 172 16 a 447 10 a 43 6 b 1.9 0.6 a 4.9 2.7 na
G5 209 33 abc 547 25 b 44 8 b 1.6 1.0 a 4.5 2.2 na
G7 194 38 ab 552 30 b 42 5 b 2.8 1.1 ab 5.3 1.8 na
Grenache 219 32 bc 375 27 c 31 5 a 2.7 0.9 ab 4.6 0.7 na
Merlot 254 34 c 430 17 ac 55 5 c 3.7 1.0 b 5.4 1.6 na
Morrastel 248 17 c 451 11 a 45 6 b 2.9 0.6 b 4.4 1.0 na
Test & Post-Hoc Genotype effect Kruskal-Wallis & Dunn 2 10-5 *** Anova & Tukey 9 10-10 *** Kruskal-Wallis & Dunn 1 10-9 *** Anova & Tukey 7 10-5 *** Kruskal-Wallis & Dunn 6 10-1
G14 Ripe 80 8 ab 57 26 ab 56 6 b 2.0 0.4 a 1.9 0.5 a
G5 75 10 a 72 30 ab 57 4 b 1.8 0.4 a 2.6 0.7 ab
G7 98 19 bc 87 27 b 57 4 b 2.9 0.3 b 2.2 0.6 a
Grenache 105 12 cd 59 31 ab 47 8 a 2.7 0.5 b 2.8 0.8 ab
Merlot 126 18 e 42 19 a 64 6 b 3.7 0.6 c 3.7 0.8 b
Morrastel 121 22 de 59 27 ab 64 8 b 2.9 0.7 b 3.0 1.9 ab
Test & Post-Hoc Genotype effect Anova & Tukey 4 10-10 *** Anova & Tukey 1 10-2 * Kruskal-Wallis & Dunn 5 10-7 *** Anova & Tukey 2 10-11 *** Kruskal-Wallis & Dunn 5 10-4 ***

Figure 3. Evolution of tartaric acid concentration during the berry ripening of Grenache (A) and G14 (B). Lines correspond to linear fitting during and after phloem unloading. Each dot corresponds to a single hard (solid circle) or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem loading arrest (3 latest dates). Hatched lines show the linear regression forced through the origin, as expected from evaporation alone, on all points after phloem arrest (red arrow).

Figure 3. Evolution of tartaric acid concentration during the berry ripening of Grenache (A) and G14 (B). Lines correspond to linear fitting during and after phloem unloading. Each dot corresponds to a single hard (solid circle) or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem loading arrest (3 latest dates). Hatched lines show the linear regression forced through the origin, as expected from evaporation alone, on all points after phloem arrest (red arrow).

Tartaric acid is the first organic acid to accumulate during young berry development and remains one of the main acids in ripe fruits of Vitis vinifera (Amerine et al., 1965; Kliewer, 1966). In the single berry analysis of this study, tartaric acid ranged from 170 to 250 mEq/L at the beginning of ripening to decrease to 70-120 mEq/L at phloem stop, as previously observed (Bigard et al., 2019). The statistical analyses confirmed that at the onset of ripening sugarless genotypes already displayed a lower tartaric acid concentration than traditional cultivars. This trend is amplified at the phloem loading arrest (or maximum weight) stage due to the higher expansion and resulting dilution during the ripening of these genotypes (Table 1). Morrastel N, also called Graciano in Spain, is a traditional cultivar that produces wines rich in polyphenols with moderate ethanol levels (Ramos and Martínez de Toda, 2021). Our study performed at single berry level confirms that Morrastel N can produce ripe fruit at lower [Hex] (i.e., below 1 mol/L) in comparison to other traditional varieties.

 2.2. Malic acid

Malic acid concentration peaked at 370-550mEq/L at the very end of the green growth period and then decreased to less than 90mEq/L at maximum berry volume regardless of the genotype (Table 1). At the onset of ripening, in contrast to tartaric acid, malic acid concentration was higher in sugarless genotypes than in traditional varieties. At the arrest of phloem unloading, the concentrations in malic acid ranged from 42 +/- 20 mEq/L (Grenache N) to 87 +/- 26 mEq/L (G7), with no obvious genotypic effects. After phloem arrest, in contrast to tartaric acid, malic acid concentration stayed stable or even slightly decreased (Figure 4), as observed in previous studies performed at berry population levels (Ojeda et al., 2017; Bigard et al., 2019).

Figure 4. Evolution of malic acid concentration during the berry ripening of Grenache (A) and G14 (B). Lines correspond to linear fitting during and after phloem unloading. Each dot corresponds to a single (solid circle) hard or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem loading arrest (the 3 last dates). Red arrows indicate the stage of phloem loading arrest.

Figure 4. Evolution of malic acid concentration during the berry ripening of Grenache (A) and G14 (B). Lines correspond to linear fitting during and after phloem unloading. Each dot corresponds to a single (solid circle) hard or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem loading arrest (the 3 last dates). Red arrows indicate the stage of phloem loading arrest.

Malic acid respiration provides a major fraction of the energic supply of the berry during early ripening (Famiani et al., 2014; Shahood, 2017), hence the decrease in concentration is faster than when only dependent on dilution due to berry growth. Here, despite different starting points (Table 2, Figure S3), the decrease in malic acid during the first phase of ripening (i.e., from 250 to 800 μmol/berry [Hex]) was characterised by an initial slope of -1 mEq per 2 hexoses; the slopes were noticeably similar for all 6 genotypes (Figure S3). During early ripening, the initial changes in the respective amounts of malic acid and sugar per fruit (concentration x weight; Figure 5) are fully consistent with the activation of a sucrose/H+ exchanger on the tonoplast of all V. vinifera cultivars investigated, including sugarless genotypes, agreeing with previous observations for Syrah and Pinot (Shahood et al., 2020). In single berries, genes linked to sucrose transport are strongly expressed until phloem arrest (Savoi et al., 2021). At the beginning of ripening, the sucrose/H+ exchange is electro-neutralised by the release of vacuolar malate, as detected here, while more and more H+ must be redirected to the vacuole as malic acid vanishes, as illustrated by the progressive activation of vacuolar ATPase and PPiase (Terrier et al., 2001).

Figure 5. Repeatability of the malic acid/sugar exchange expressed in quantity per fruit during berry ripening of 6 grapevine varieties. Each dot represents the average of sets of 10 berries ranked according to their hexose concentration (error bars are not displayed to improve readability).

Figure 5. Repeatability of the malic acid/sugar exchange expressed in quantity per fruit during berry ripening of 6 grapevine varieties. Each dot represents the average of sets of 10 berries ranked according to their hexose concentration (error bars are not displayed to improve readability).

When the cumulative evolution of tartaric + malic acids was monitored during early ripening (Figure S4A), no specific behaviours were observed in the sugarless genotypes in comparison to the other varieties. Considering that, depending on their concentrations, sugars and organic acids are the main contributors to the osmotic potential of the berry (Matthews et al., 1987), the present results totally exclude the possibility of the reduction in sugar concentration being compensated for by a greater accumulation of organic acids in the sugarless genotypes. As already mentioned, sugarless genotypes displayed lower tartaric and higher malic acid than traditional controls and consequently a higher malic acid/tartaric acid ratio (Figure S4B).

2.3. Potassium

K+, the main cation in the grapevine fruit, accumulates during both phases of growth (Cuellar et al., 2013). Here, concentrations before ripening ranged from 31 +/- 5 mEq/L for Grenache N to 55 +/- 5 mEq/L for Merlot N (Table 2). During ripening, [K+] increased moderately (Figure S5)with increments ranging from 16 % (Merlot N) to 50 % (Grenache N) and both genotypes displaying the lowest and the highest levels of [K+] at the ripe stage: 47 +/- 8 mEq/L for Grenache N and 64 +/- 6 mEq/L for Merlot N respectively. After the period of phloem loading arrest, [K+] steadily increased as a result of water loss associated with shrivelling, as was the case for tartaric acid (Figure 6).

Figure 6. Evolution of potassium concentration during the berry ripening of Grenache (A) and G14 (B). Lines correspond to linear fitting during and after phloem unloading. Each dot corresponds to a single (solid circle) hard or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem loading arrest (the 3 last dates). Red arrows indicate the stage of phloem loading arrest.

Figure 6. Evolution of potassium concentration during the berry ripening of Grenache (A) and G14 (B). Lines correspond to linear fitting during and after phloem unloading. Each dot corresponds to a single (solid circle) hard or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem loading arrest (the 3 last dates). Red arrows indicate the stage of phloem loading arrest.

After the main organic compounds, K+ is the fourth contributor to berry osmotic potential, which also neutralises a fraction of organic acids (Storey, 1987; Rogiers et al., 2017). During the early stages of development, low sugar accumulators (Morrastel N included) had a higher [K+] than high accumulators (Table 2). The fact that this element mainly accumulates in the skin may play a role in the difference in elasticity of the G genotypes. At the end of phloem unloading the average [K+] was around 50 mmol/L in the 6 genotypes, indicating strong homeostasis for this element. As observed at population level with other genotypes (Bigard et al., 2020), K+ concentration increased 20-40 times less than hexoses during the phloem unloading period from veraison to max berry weight, as observed at population level with other genotypes (Figures 5andS5, Table 2).

This obviously contradicts the so-called "massive" K+ import into the ripening berry (Villette et al., 2020). As recently discussed by Savoi et al. (2021), the belief that K+ transport will compensate for an intrinsic deficiency in the energisation of sugar imports is not supported by experimental data. In this respect, the simultaneous and parallel increases in [Hex] and [K+] observed after the arrest of phloem unloading is not indicative of a co-transport mechanism, but only a result of net water loss and berry shrivelling. Despite significant progress in the understanding of the import of potassium into grapevine berries (Rogiers et al., 2017; Villete et al., 2020; Savoi et al., 2021), the putative mechanistic links between potassium and sugar imports remain speculative.

 2.4. Evolution of fruit acidity

Green berries displayed high acidity (Figure 7, Table 1) ranging from 560 +/- 40 mEq/L (Grenache N) to 710 +/- 50 mEq/L (G5). Similar to tartaric acid dilution, malic acid respiration and dilution, and K+ accumulation, acidity was reduced to between 80 +/- 25 mEq/L (G14) and 130 +/- 40 mEq/L (G7) at the ripe stage, with no statistical differences between genotypes. Noticeably, total acidity tended to increase quite late during the ripening process for all genotypes: around 1250 mmol/L [Hex] in Grenache N, and 1000 mmol/L [Hex] in G14 (Figure 7).

Figure 7. Evolution of juice total acidity during the berry ripening of Grenache (A) and G14 (B). Lines correspond to linear fitting during and after phloem unloading. Each dot corresponds to a single (solid circle) hard or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem loading arrest (the 3 last dates). Red arrows indicate the stage of phloem loading arrest.

Figure 7. Evolution of juice total acidity during the berry ripening of Grenache (A) and G14 (B). Lines correspond to linear fitting during and after phloem unloading. Each dot corresponds to a single (solid circle) hard or soft (triangle) berry. Solid lines correspond to successive linear regressions for soft berries before and after phloem loading arrest (the 3 last dates). Red arrows indicate the stage of phloem loading arrest.

Obtaining optimal grape acidity for the production of quality wines is a major challenge (Champagnol, 1984; Sweetman et al., 2014; Ollat et al., 2018). The effect of temperature on grape acidity is well documented (Kliewer and Lider, 1970; Butrose et al., 1971; Seguin et al., 2004; Rienth et al., 2016). By virtue of the electroneutrality principle, titratable (or total) acidity represents the difference between acids (mainly tartaric and malic in grapevine) and cations (mainly K+ in plants). Bigard et al. (2018; 2020) and Duchène et al. (2020) have detailed genetic diversity for anions (i.e., organic acids) and cations and its consequence on grape acidity in a set of extreme V. vinifera varieties and offsprings. In the present study, we analysed the determinants of the acidity of 6 varieties, including 3 sugarless genotypes. As shown in previous sections, sugarless genotypes tended to display a higher malic acid/tartaric acid ratio than the 3 traditional cultivars, but they had similar contents in malic + tartaric acids and K+. As a result, the sugarless genotypes showed similar levels of acidity at the same physiological ripe stage as other varieties.

 3. Other cations (Mg2+, Ca2+)

Far less Magnesium (Mg2+) than K+ accumulated in all the genotypes. [Mg2+] displayed very few changes during ripening (Table 2, Figures S6 and S7). At the arrest of phloem unloading, [Mg2+] ranged from 1.6 +/- 1.0 (G5) to 3.7 +/- 1.0 (Merlot N) mEq/L. Ranging from 4.5 +/- 2.2 (G5) to 5.4 +/- 1.6 (Merlot N) mE/L, a higher accumulation of Calcium (Ca2+) was found in the green berries than of Mg2+ (Table 2); during ripening, [Ca2+] tended to decrease (Figures S8 and S9), until it reached concentrations of 1.9 +/- 0.5 (G14) to 3.7 +/- 0.8 (Merlot N) mEq/L, which is a relative decrease quite comparable to that of tartaric acid. Therefore, its total amount per berry remains constant during ripening, which is widely accepted in grapevine, and consistent with its almost exclusive transport by xylem (Glad et al., 1992; Creasy et al., 1993). Neither cation had a major impact on wine quality or varied much within the range of varieties.

Conclusion

Obtaining fruits with reduced [Hex] while preserving their acidity is an option for mitigating the effect of climate warming on grapevine fruit quality (Torregrosa et al., 2017). This objective cannot be fully achieved by applying viticultural practices (e.g., harvesting before complete sugar unloading or removing a fraction of the leaves to reduce C photoassimilation) without impacting the quality of the wines (Bobeica et al., 2015; Antalick et al., 2021). Some diversity in water, sugars and the determinants of acidity of the grape can be found in V. vinifera varieties, or it can be obtained by crossbreeding (Bigard et al., 2018, 2020). In this study, using advanced methods of berry phenotyping we characterised the fruit development and ripening of a set of new disease-tolerant varieties producing low alcoholic wines (Escudier et al., 2017). In previous studies, we have shown that combining single fruit phenotyping and precise physiological landmarks can significantly improve the understanding of berry development (Shahood et al., 2020; Savoi et al., 2021). Indeed, in these conditions, studying the relationships between the main solutes are no longer biased by averaging unsynchronised, hence developmentally and metabolically chimeric, samples. In the present study, to circumvent the imprecision of berry growth curves resulting from the heterogeneity of berry size, tartaric acid dilution was used to detect the timing of phloem unloading arrest. The sugarless genotypes were found to display a [Hex] reduced by 20-30 % at the ripe stage without impacting berry growth, organic acid or cation accumulation levels. No major differences were found in terms of fruit growth rates and quantity of sugars per berry in comparison to the control varieties, suggesting that the sugarless phenotypes undergo a greater cellular expansion at similar osmotic or turgor pressure. This property is not specific to genotypes derived from Muscadinia rotundifolia and table grape varieties, because Morrastel N also displayed a limited [Hex] in the ripe fruit (< 1 mol/L). Moreover, similar behaviours can be found in other traditional varieties, such as Aramon, Cornifesto and Mandilaria (Bigard et al., 2018) and Glera, a variety used for Prosecco wine production (

https://plantgrape.plantnet-project.org/fr/cepage/Glera

). Taken together, our results show that adaptive traits to climate changes can be pyramidised with QTLs of tolerance to diseases. By crossing G5 and G14, we generated microvine segregating progenies (Torregrosa et al., 2019) to further characterise the physiology of this trait and to investigate the genetic determinism of water, sugar and organic accumulations (Savoi et al., 2021).

Acknowledgements

We would like to thank Eleoneora Maoddi, Yannick Sire and Philippe Abbal for their excellent contribution to the experiments.

 Abbal P. , Boulet J. C. , Moutounet M. , 1992. Utilisation de paramètres physiques pour la caractérisation de la véraison des baies de raisin. Journal International des Sciences de la Vigne et du Vin, 26, 231-237. https://doi.org/10.20870/oeno-one.1992.26.4.1185. Alem H. , Ojeda H. , Rigou P. , Schneider R. , Torregrosa L. , 2021. The reduction of plant sink/source does not systematically improve the metabolic composition of the Vitis vinifera white fruit. Food Chemistry, 345, 128825. https://doi.org/10.1016/j.foodchem.2020.128825. Alessandrini M. , Gaiotti F. , Belfiore N. , Matarese F. , D'Onofrio C. , Tomasi D. , 2018. Influence of vineyard altitude on Glera grape ripening (Vitis vinifera L.): effects on aroma evolution and wine sensory profile. Journal of the Science of Food and Agriculture, 97, 2695-2705. https://doi.org/10.1002/jsfa.8093. Amerine M. A. , Thoukis G. , 1958. The glucose/fructose ratio of California grapes. Vitis, 1, 224-229.. Amerine M. A. , Roessler E. B. , Ough C. S. , 1965. Acids and the acid taste. I. The effect of pH and titratable acidity. American Journal of Enology and Viticulture, 16, 29-37.. Antalick G. , Šuklje K. , Blackman J. W. , Schmidtke L. M. , Deloire A. , 2021. Performing sequential harvests based on berry sugar accumulation (mg/berry) to obtain specific wine sensory profiles. OENO One, 55, 131–146. https://doi.org/10.20870/oeno-one.2021.55.2.4527.. Arrizabalaga M. , Morales F. , Oyarzun M. , Delrot S. , Gomès E. , Irigoyen J. J. , Hilbert G. , Pascual I. , 2018. Tempranillo clones differ in the response of berry sugar and anthocyanin accumulation to elevated temperature. Plant Science, 267, 74-83. https://doi.org/10.1016/j.plantsci.2017.11.009. Asproudi A. , Petrozziello M. , Cavalletto S. , Guidoni S. , 2016. Grape aroma precursors in cv. Nebbiolo is affected by vine microclimate. Food Chemistry, 211, 947-956. https://doi.org/10.1016/j.foodchem.2016.05.070. Bigard A. , Berhe D. T. , Maoddi E. , Sire Y. , Boursiquot J. M. , Ojeda H. , Péros J. P. , Doligez A. , Romieu C. , Torregrosa L. , 2018. Vitis vinifera L. fruit diversity to breed varieties anticipating climate changes. Frontiers in Plant Science. https://doi.org/10.3389/fpls.2018.00455. Bigard A. , Romieu C. , Sire Y. , Veyret M. , Ojeda H. , Torregrosa L. , 2019. Grape ripening revisited through berry density sorting. OENO One, 4, 719-724. https://doi.org/10.20870/oeno-one.2019.53.4.2224. Bigard A. , Romieu C. , Sire Y. , Torregrosa L. , 2020. Vitis vinifera L. diversity for cations and acidity is suitable for breeding fruits coping with climate warming. Frontiers in Plant Science. https://doi.org/10.3389/fpls.2020.01175. Bobeica N. , Poni S. , Hilbert G. , Renaud C. , Gomès E. , Delrot S. , Dai Z. , 2015. Differential responses of sugar, organic acids and anthocyanins tosource-sink modulation in Cabernet Sauvignon and Sangiovese grapevines. Frontiers in Plant Science, 382. ttps://doi.org/10.3389/fpls.2015.00382. Burbidge C. A. , Ford C. M. , Melino V .J. , Wong D .C. J. , Jia Y. , Jenkins C. L. D. , Soole K. L. , Castellarin S. D. , Darriet P. , Rienth M. , Bonghi C. , Walker R. P. , Famiani F. , Sweetman C. , 2021. Biosynthesis and cellular functions of tartaric acid in grapevines. Frontiers in Plant Science, 12. https://doi.org/10.3389/fpls.2021.643024. Butrose M. S. , Hale C. R. , Kliewer W. M. , 1971. Effect of temperature on the composition of Cabernet-Sauvignon berries. American Journal of Enology and Viticulture, 22, 71-75.. Castellarin S. D. , Gambetta G. A. , Wada H. , Krasnow M. N. , Cramer G. R. , Peterlunger E. , Shackel K. A. , Matthews M. A. , 2015. Characterization of major ripening events during softening in grape: turgor, sugar accumulation, abscisic acid metabolism, colour development, and their relationship with growth. Journal of Experimental Botany, 67, 709-722. https://doi.org/10.1093/jxb/erv483. Champagnol F. , 1984. Eléments de physiologie de la vigne et de viticulture générale, ed. Dehan (Montpellier, France: Dehan).. Conde C. , Silva P. , Fontes N. , Dias A. Tavares , R.M. Sousa , M. J. Agasse , A. Delrot , S. Gerós , H. , 2007. Biochemical Changes throughout Grape Berry Development and Fruit and Wine Quality. Food, 1: 1-22.. Coombe B. G. , 1976. The development of fleshy fruits. An. Rev. Plant Physiol., 27, 507-528.. Coombe B. G. , 1992. Research on development and ripening of the grape berry. American Journal of Enology and Viticulture, 43, 101-110.. Coombe B. G. , McCarthy M.G. , 2000. Dynamics of grape berry growth and physiology of ripening. Australian Journal of Grape and Wine Research, 6, 131-135. https://doi.org/10.1111/j.1755-0238.2000.tb00171.x. Cuellar T. , Azeem F. , Andrianteranagna M. , Pascaud F. , Verdeil J. L. , Sentenac H. , Zimmermann S. , Gaillard I. , 2013. Potassium transport in developing fleshy fruits: the grapevine inward K+ channel VvK1.2 is activated by CIPK–CBL complexes and induced in ripening berry flesh cells. Plant J.,73, 1006-1018. https://doi.org/10.1111/tpj.12092. Dai Z. W. , Ollat N. , Gomès E. , Decroocq S. , Tandonnet J. P. , Bordenave L. et al. , 2011. Ecophysiological, genetic, and molecular causes of variation in grape berry weight and composition: a review. American Journal of Enology and Viticulture, 62, 413-425. https://doi.org/10.5344/ajev.2011.10116. Doumouya S. , Lahaye M. , Maury C. , Siret R. , 2014. Physical and physiological heterogeneity within the grape bunch: impact on mechanical properties during maturation. American Journal of Enology and Viticulture, 65, 170-178. https://doi.org/10.5344/ajev.2014.13062. Duchêne E. , Dumas V. , Jaegli N. , Merdinoglu D. , 2012. Deciphering the ability of different grapevine genotypes to accumulate sugar in berries. Australian Journal of Grape and Wine Research, 18, 319–328. https://doi.org/10.1111/j.1755-0238.2012.00194.x. Duchêne E. , Dumas V. , Butterin G. , Jaegli N. , Rustenholtz C. , Chauveau A. , Berard A. , Le Paslier M. C. , Gaillard I. , Merdinoglu D. , 2020. Genetic variations of acidity in grape berries are controlled by the interplay between organic acids and potassium. Theor. App. Genet. https://doi.org/10.1007/s00122-019-03524-9. Escudier H. , Bigard A. , Ojeda H. , Samson A. , Caillé S. , Romieu C. , Torregrosa L. , 2017. De la vigne au vin : des créations variétales adaptées au changement climatique et résistant aux maladies cryptogamiques 1/2 : La résistance variétale. Revue des Oenologues, 44, 16-18.. Famiani F. , Farinelli D. , Palliotti A. , Moscatello S. , Battistelli A. , Walker R. B. , 2014. Is stored malate the quantitatively most important substrate utilised by respiration and ethanolic fermentation in grape berry pericarp during ripening? Plant Physiology and Biochemistry, 76, 52-57. https://doi.org/10.1016/j.plaphy.2013.12.017. Friend A.P. , Trought M.C.T. , Creasy G. L. , 2009. The influence of seed weight on the development and growth of berries and live green ovaries in Vitis vinifera L. cvs. Pinot noir and Cabernet-Sauvignon. Australian Journal of Grape and Wine Research, 15, 166-174. https://doi.org/10.1111/j.1755-0238.2009.00050.x. Gouthu S. , O'Neil S. T. , Di Y. , Ansarolia M. , Megraw M. , Deluc L.G. , 2014. A comparative study of ripening among berries of the grape cluster reveals an altered transcriptional programme and enhanced ripening rate in delayed berries. J. Exp. Bot., 65, 5889–5902. https://doi.org/10.1093/jxb/eru329. Gutiérrez-Gamboa G. , Garde-Cerdán T. , Carrasco-Quiroz M. , Pérez-Álvarez E.P. , Martínez-Gil A.M. , Del Alamo-Sanza M. , Moreno-Simunovic Y. , 2018. Volatile composition of Carignan noir wines from ungrafted and grafted onto País (Vitis vinifera L.) grapevines from ten wine-growing sites in Maule Valley, Chile. Journal of the Science of Food and Agriculture, 98, 4268-4278. https://doi.org/10.1002/jsfa.8949. Hawker J. S. , Ruffner H. P. , Walker R.R. , 1976. The sucrose content of some Australian grapes. American Journal of Enology and Viticulture, 27, 125-129.. Houel C. , Chatbanyong R. , Doligez A. Rienth , M. Foria , S. Luchaire , N. Roux , C. Adivèze , A. Lopez , G. Farnos , M. Pellegrino , A. This , P. Romieu , C. & Torregrosa L. , 2015. Identification of stable QTLs for vegetative and reproductive traits in the microvine (Vitis vinifera L.) using the 18K Infinium chip. BMC Plant Biol., 15,205. https://doi.org/10.1186/s12870-015-0588-0. Kliewer W. M. , 1966. Sugars and Organic Acids of Vitis vinifera. Plant Physiology, 41, 923-931. https://doi.org/10.1104/pp.41.6.923. Kliewer W. M. , 1967. The glucose-fructose ratio of Vitis vinifera grapes. American Journal of Enology and Viticulture, 18, 33-41. https://doi.org/10.1016/j.aca.2011.11.043. Kliewer W. M. , Lider L.A. , 1970. Effect of day temperature and light intensity on growth and composition of Vitis vinifera L. fruits. Journal of the American Society for Horticultural Science, 95, 766–769.. Lang A. , Thorpe M. R. , 1989. Xylem, phloem and transpiration flows in a grape: application of a technique for measuring the volume of attached fruits to high resolution using Archimedes' principle. Journal of Experimental Botany, 40, 1069–1078. https://doi.org/10.1093/jxb/40.10.1069. Liang Z. , Sang M. , Fan P. , Wu B. , Wang L. , Duan W. , Li S. , 2011. Changes of polyphenols, sugars, and organic acid in 5 Vitis genotypes during berry ripening. J. Food Sci., 76, 1231-1238. https://doi.org/10.1111/j.1750-3841.2011.02408.x. Liu H. , Wu B. , Fan P. , Xu H. , Li S. , 2006. Sugar and acid concentrations in 98 grape cultivars analyzed by principal component analysis. Journal of the Science of Food and Agriculture, 86, 1526-1536. https://doi.org/10.1002/jsfa.2541. Liu H. , Wu B. , Fan P. , Xu H. , Li S. , 2007. Inheritance of sugars and acids in berries of grape (Vitis vinifera L.). Euphytica,153, 99-107. https://doi.org/10.1007/s10681-006-9246-9. Matthews M. A. , Cheng G. , Weinbaum S. A. , 1987. Changes in water potential and dermal extensibility during grape berry development. J. Amer. Soc. Hort. Sci., 112, 314-319.. Ojeda H. , Bigard A. , Escudier J. L. , Samson A. , Caillé S. , Romieu C. , Torregrosa L. , 2017. De la vigne au vin : des créations variétales adaptées au changement climatique et résistant aux maladies cryptogamiques 2/2 : Approche viticole pour des vins de type VDQA. Revue des Œnologues, 44, 22-27.. Ollat N. , Marguerit E. , Lecourieux F. , Destrac-Irvine A. , Barrieu F. , Dai Z. , 2018. Grapevine adaptation to abiotic stresses. Acta Hortic., 1248. https://doi.org/10.17660/ActaHortic.2019.1248.68. Parker A. K. , de Cortázar-Atauri I. G. , Gény L. , Spring J. L. , Destrac A. , Schultz H., … , 2020. Temperature-based grapevine sugar ripeness modeling for a wide range of Vitis vinifera L. cultivars. Agric. For. Meteorol., 285, 107902. https://doi.org/10.1016/j.agrformet.2020.107902. Ramos M. C. , Martínez de Toda F. , 2021. Interannual and spatial variability of grape composition in the Rioja DOC show better resilience of cv. Graciano than cv. Tempranillo under a warming scenario. OENO One, 55, 85-100. https://doi.org/10.20870/oeno-one.2021.55.3.4695. R Core Team , 2017. R: A Language and Environment for Statistical Computing. Vienna: R Foundation for Statistical Computing.. Rienth M. , Torregrosa L. , Luchaire N. , Chatbanyong R. , Lecourieux D. , Kelly M. , … , 2014. Day and night heat stress trigger different transcriptomic responses in green and ripening grapevine (Vitis vinifera) fruit. BMC Plant Biol., 14,108. https://doi.org/10.1186/1471-2229-14-108. Rienth M. , Torregrosa L. , Gauthier S. , Ardisson M. , Brillouet J. L. & Romieu C. , 2016. Temperature desynchronizes sugar and organic acid metabolism in ripening grapevine fruits and remodels its transcriptome. BMC Plant Biology, 16, 164. https://doi.org/10.1186/s12870-016-0850-0. Robin J. P. , Abbal P. , Salmon J.M. , 1997. Firmness and grape berry maturation: definition of different rheological parameters during the ripening. Journal International des Sciences de la Vigne et du Vin, 31, 127-138. https://doi.org/10.20870/oeno-one.1997.31.3.1083. Rogiers S. Y. , Coetzee Z. A. , Walker R. R. , Deloire A. , Tyerman S.D. , 2017. Potassium in the grape (Vitis vinifera L.) berry: transport and function. Frontiers in Plant Science, 8, 1629. https://doi.org/10.3389/fpls.2017.01629. Rolle L. , Giacosa S. , Gerbi V. , Bertolino M. , Novello V. , 2013. Varietal comparison of the chemical, physical, and mechanical properties of five colored table grapes. International Journal of Food Properties, 16, 598-612. https://doi.org/10.1080/10942912.2011.558231. Rösti J. , Schumann M. , Cleroux M. , Lorenzini F. , Zufferey V. , Rienth M. , 2018. Effect of drying on tartaric acid and malic acid in Shiraz and Merlot berries. Australian Journal of Grape and Wine Research, 24, 421-429. https://doi.org/10.1111/ajgw.12344. Ruffner H. P. , 1982. Metabolism of tartaric and malic acids in Vitis: A review - Part A. Vitis, 21, 247–259, https://doi.org/ 10.5073/VITIS.1982.21.247-259.. Santillán D. , Iglesias A. , La Jeunesse I. , Garrote L. , Sotes V. , 2019. Vineyards in transition: A global assessment of the adaptation needs of grape producing regions under climate change. Science of The Total Environment, 657, 839-852. https://doi.org/10.1016/j.scitotenv.2018.12.079.. Savoi S. , Torregrosa L. , Romieu C. , 2021. Transcripts repressed at the stop of phloem unloading highlight the energy efficiency of sugar import in the ripening V. vinifera fruit. Horticulture Research, 8, 193. https://doi.org/10.1038/s41438-021-00628-6. Seguin B. , Stevez L. , Herbin C. , Rochard J. , 2004. Changements climatiques et perspectives pour la viticulture : conséquences potentielles d'une modification du climat. Revue des Oenologues, 111, 59-60.. Shahood R. , Rienth M. , Torregrosa L. , Romieu C. , 2015. Evolution of grapevine (Vitis vinifera L.) berry heterogeneity during ripening. 19th International Meeting of Viticulture GIESCO. Vol. 2, pp. 564–567.. Shahood R. , 2017. The berry within an asynchronous harvest: A new paradigm towards the quantitative interpretation of sugar and acid fluxes as major osmoticums and respiratory substrates during bimodal grape development. PhD of Montpellier SupAgro, http://www.theses.fr/s206075, 215 p.. Shahood R. , Torregrosa L. , Savoi S. , Romieu C. , 2020. Berry development hidden by its non-synchronous population. OENO One, 54, 1077-1092. https://doi.org/10.20870/oeno-one.2020.54.4.3787. Shiraishi M. , Fujishima H. , Chijiwa H. , 2010. Evaluation of table grape genetic resources for sugar, organic acid, and amino acid composition of berries. Euphytica, 174, 1-13. https://doi.org/10.1007/s10681-009-0084-4. Storey R. , 1987. Potassium localization in the grape berry pericarp by energy-dispersive x-ray microanalysis. American Journal of Enology and Viticulture, 38, 301-309.. Suter B. , Destrac-Irvine A. , Gowdy M. , Dai Z. , van Leeuwen C. , 2021. Adapting wine grape ripening to global change requires a multi-trait approach. Frontiers in Plant Science, 12, 624867. https://doi.org/10.3389/fpls.2021.624867. Sweetman C. , Sadras V. O. , Hancock R. D. , Soole K. L. , Ford C. M. , 2014. Metabolic effects of elevated temperature on organic acid degradation in ripening Vitis vinifera fruit. Journal of Experimental Botany, 65, 5975–5988. https://doi.org/10.1093/jxb/eru343. Terrier N. , Sauvage F. X. , Ageorges A. , Romieu C. , 2001. Changes in acidity and in proton transport at the tonoplast of grape berries during development. Planta, 213, 20-28. https://doi.org/10.1007/s004250000472. Torregrosa L. , Bigard A. , Doligez A. Lecourieux , D. Rienth M. , Luchaire N. , Pieri P. Chatbanyong , R. Shahood , … , 2017. Developmental, molecular and genetic studies on the grapevine response to temperature open breeding strategies for adaptation to warming. OENO One, 51, 155-165. https://doi.org/10.20870/oeno-one.2016.0.0.1587. Torregrosa L. , Rienth M. , Romieu C. , Pellegrino A. , 2019. The microvigne, a model for grapevine physiology studies and genetics. OENO One, 53. https://doi.org/10.20870/oeno-one.2019.53.3.2409. Varandas S. , Teixeira M. J. , Marques J. C. , Aguilar A. , Alves A. , Bastos M. , 2004. Glucose and fructose levels on grape skin: interference in Lobesia botrana behaviour. Analytica Chimica Acta, 513, 351-355. https://doi.org/10.1016/j.aca.2003.11.086. Villete J. , Cuélar T. , Verdeil J. L. , Delrot S. , Gaillard I. , 2020. Grapevine potassium nutrition and fruit quality in the context of climate change. Frontiers in Plant Science, 11, 123. https://doi.org/10.3389/fpls.2020.00123. Vondras A. M. , Gouthu S. , Schmidt J. A. , Petersen A. R. , Deluc L. G. , 2016. The contribution of flowering time and seed content to uneven ripening initiation among fruits within Vitis vinifera L. cv. Pinot noir clusters. Planta, 243, 1191-1202. https://doi.org/10.1007/s00425-016-2474-x. Zhang C. , Jia H. , Wu W. , Wang X. , Fang J. , Wang C. , 2015. Functional conservation analysis and expression modes of grape anthocyanin synthesis genes responsive to low temperature stress. Gene, 574, 168-177. https://doi.org/10.1016/j.gene.2015.08.003.