Abstract
The effects of grapevine red blotch disease (GRBD) on vine performance were studied in a Cabernet franc vineyard in the Okanagan Valley, British Columbia, over two years. Vines with GRBD had lower pruning mass, were less winter hardy than non-GRBD vines, and produced lower yields comprising fewer clusters and larger berries with more seeds. GRBD reduced photosynthesis and stomatal conductance for asymptomatic leaves through the growing season, and more so for symptomatic leaves. Leaf fall was delayed or incomplete for GRBD vines. GRBD reduced berry soluble solids, anthocyanins, yeast assimilable nitrogen, and tannins and increased berry pH and titratable acidity. Wine made from GRBD fruit had reduced anthocyanin and alcohol, lighter color, and more yellow hue. Effects of GRBD on wine sensory characteristics included less body and aftertaste, lower intensity of black and red fruit character, and increased acidity and intensity of vegetal character. Inclusion of low amounts, up to 20%, of GRBD fruit in wine reduced red fruit character in the first year and increased astringency and vegetal character in the second year. The deleterious effects of GRBD on yield, fruit composition for winemaking, and bud hardiness constitute a serious threat to grapegrowing regions in Canada and the United States. The degree of these impacts would be difficult to mitigate through management practices, which underscores the importance of planting virus-tested vines to prevent the introduction and spread of GRBD.
- anthocyanin
- bud hardiness
- grapevine red blotch virus
- photosynthesis
- Vitis vinifera
- wine sensory quality
- yield components
Grapevine red blotch disease (GRBD) is a recently recognized virus disease widely distributed in the United States and in Canada (Krenz et al. 2014, Sudarshana et al. 2015, Poojari et al. 2017, Xiao et al. 2018). Its symptoms were first reported in California in 2008 (Calvi 2011). The causative pathogen, grapevine red blotch virus (GRBV), was first characterized in New York in 2012 (Krenz et al. 2012). In California, GRBV was associated with GRBD in 2013 (Al Rwahnih et al. 2013).
GRBV has a single-stranded circular DNA genome and belongs to the genera Grablovirus, family Geminiviridae (Varsani et al. 2017). Some foliar symptoms of GRBD are similar to those associated with grapevine leaf roll disease, an unrelated widespread virus disease of grapes. Mature leaves of black-fruited Vitis vinifera cultivars with GRBD show late-season red blotches and marginal reddening (Golino et al. 2002, Calvi 2011, Al Rwahnih et al. 2013, Krenz et al. 2014, Poojari et al. 2017). In white-fruited cultivars, symptoms are less apparent but may include yellowing and marginal necrosis of mature leaves (Krenz et al. 2014, Sudarshana et al. 2015).
A large-scale survey conducted in British Columbia (BC) in 2014 and 2015 found a low incidence of GRBV infection in the Okanagan and Similkameen Valleys (Poojari et al. 2017). Most infected vineyard blocks found in this survey had been established between 2011 and 2014, with plants sourced from the United States, suggesting a recent introduction through infected plant material. While this finding underscores the need for a domestic clean plant propagation program in Canada, understanding the impact of GRBD on vine performance, especially impacts on fruit composition, wine characteristics, and bud hardiness, is critical for growers deciding how to manage the disease for the long term.
Reported effects of GRBD on leaf functional characteristics include reductions in leaf greenness (Reynard et al. 2018) and in leaf gas exchange parameters, including photosynthesis (Reynard et al. 2018, Martinez-Luscher et al. 2019), although differences between symptomatic and asymptomatic leaves have not been reported. In a study of leaf foliar metabolism before and after symptom development, GRBD leaves had increased levels of amino acids associated with host defences and increased fructose and glucose levels, revealing impacts of GRBV on the function of young leaves (Wallis and Sudarshana 2016). GRBD impacts on total vegetative growth, measured as pruning mass, have been reported as substantial—that is, ≥33% (Reynard et al. 2018) and ≥25% (Poojari et al. 2013), or inconsistent (Martinez-Luscher et al. 2019). Effects of GRBD on bud hardiness, a trait critical to vineyard performance in cold-winter regions, have not been reported.
Studies of GRBD effects on fruit characteristics have revealed several detrimental impacts on berry development and composition. All studies found reductions in berry soluble solids (Poojari et al. 2013, Reynard et al. 2018, Girardello et al. 2019, Levin and KC 2019, Martinez-Luscher et al. 2019, Girardello et al. 2020). Less consistent effects were decreased fruit yield (Poojari et al. 2013, Girardello et al. 2019, Martinez-Luscher et al. 2019), increased berry mass (Martinez-Luscher et al. 2019), lower berry anthocyanin concentrations in black-fruited cultivars (Reynard et al. 2018, Girardello et al. 2019, 2020, Levin and KC 2019, Martinez-Luscher et al. 2019), and increased berry pH and malic acid or titratable acidity (TA) (Reynard et al. 2018, Girardello et al. 2019, Martinez-Luscher et al. 2019). A study of GRBD impacts on transcriptional and hormonal regulation of grape ripening revealed a number of pathways abnormally induced or inhibited by GRBV infection, which were associated with retarded maturation processes (Blanco-Ulate et al. 2017). These impacts of GRBD on fruit development may limit the ability of growers to manage GRBD vines to achieve acceptable fruit composition. In an Oregon Pinot noir study, deficit irrigation enhanced berry anthocyanins and skin tannins in healthy vines but not in GRBD vines (Levin and KC 2019). Although GRBD effects on wine characteristics have not been evaluated from field-replicated studies, differences in the composition and sensory characteristics of Chardonnay wines made from uninfected and GRBD fruit were reported (Girardello et al. 2020). The annual economic cost of GRBD effects, considering a 5% infection level and a 25% price quality penalty, ranged from more than USD $2000/ha in eastern Washington State and Long Island, New York, to more than USD $4000/ha and $8000/ha in Sonoma and Napa Counties in California, respectively (Ricketts et al. 2017).
A Cabernet franc vineyard in the Okanagan Valley, BC, with GRBD vines randomly dispersed among mostly uninfected vines presented an opportunity to conduct a controlled study of GRBD impacts on vine, fruit, and wine characteristics. The study characterized GRBD impacts on vine growth, cold hardiness, and leaf characteristics, including pigments, nitrogen content, and photosynthesis of symptomatic and asymptomatic leaves. GRBD effects on fruit yield and composition, wine composition, and wine sensory characteristics were also characterized. Wines made with low proportions of GRBD fruit were evaluated to determine impacts on wine characteristics of low GRBD levels in vineyards, contributing up to 20% GRBD fruit in the harvest. The main objective of the study was to provide insights into the types and degree of GRBD impacts on vine performance that could assist growers in making informed decisions on how to manage vineyards with GRBD vines in the short and long terms.
Materials and Methods
Site conditions and management
The study was conducted in 2017 and 2018 in a 0.7-ha block of Cabernet franc (cl. 347 on 140 Ru) in a commercial vineyard 5.5 km southwest of Oliver, BC, in the Golden Mile Bench sub-geographical indication (sub-GI). The vineyard was planted in 2010 to 2011, with vines spaced 1.15 m apart within rows spaced 1.85 m apart and oriented north-south. The bilateral cordon-trained vines were shoot-thinned annually to retain ~20 shoots/vine (two shoots per spur). Retained shoots were positioned vertically and supported by three pairs of trellis catch wires. Clusters were thinned annually just after fruit set to one cluster per shoot. The canopy was hedged in midsummer each year to a height of 2.2 m. The vines were drip-irrigated and fertilized as required based on standard practices for the region.
Weather during both years of the study was typical for the region. Growing degree day (GDD) accumulation (base 10°C, from 1 April to 31 Oct) was 1653 in 2017 and 1638 in 2018, based on temperatures measured at a weather station <0.5 km from the study site. Rainfall totaled 293 mm in 2017 and 263 mm in 2018, and growing season (1 April to 31 Oct) rainfall was 171 mm in 2017 and 82 mm in 2018. Daily minimum temperatures from early November to mid-April prior to each study year were typical for the region (Figure 1). The minimum winter temperature was -18.7°C in Jan 2017 and -14.6°C in Feb 2018.
Red blotch diagnosis and experimental design
On 10 Oct 2017, vines in the block exhibiting foliar GRBD symptoms were tested to verify infection by GRBV. Testing was conducted by PCR using two sets of GRBV-specific primers, GVGF1/GVGR1 (Al Rwahnih et al. 2013) and GLRaV-4 For/GLRaV4Rev (Poojari et al. 2013), which amplify a part of the coat protein and replicase intergenic regions, respectively. Total DNA from petioles of individual vines (four petioles per vine) were extracted using DNeasy Plant Mini Kit (Qiagen Inc.) following the manufacturer’s instructions. Total DNA from uninfected grape leaf tissue was obtained from seedlings maintained under greenhouse conditions at the Summerland Research and Development Centre (SuRDC), Summerland, BC, and used as a negative control. Total DNA isolated from petioles of GRBV-infected grapevines maintained at a SuRDC greenhouse was used as a positive control. All the vines in the study were additionally tested for the presence of grapevine leafroll-associated virus (GLRaV)-1, GLRaV-2, GLRaV-3, and GLRaV-4, grapevine fanleaf virus, and grapevine Pinot gris virus by single-tube RT-PCR using virus-specific primers (Bertazzon and Angelini 2004, Osman and Rowhani 2006, Mekuria et al. 2008, Alabi et al. 2011, Cho et al. 2013, Poojari et al. 2016). The GRBV genome from each infected vine was characterized by Ilumina sequencing at the Cool Climate Oenology and Viticulture Institute, Brock University, St. Catharines, ON, as described by Poojari et al. (2017). The GRBV isolates were determined to belong to clade 2 by comparing the genome sequences against GRBV available in GenBank and conducting a phylogenetic analyses as described by Poojari et al. (2017). Eighteen infected vines, interspersed through the block, and non-infected vines flanking, in the row one vine away on each side of the infected vines, were selected for study. Each of the 18 sets of three vines was considered a replicate block for statistical analysis of GRBD effects on vine performance and included an extra non-GRBD vine in case one became infected with GRBV during the study. All selected vines were tested for GRBV before harvest each year to verify their infection status using the same methodology and primer sets as described above. All of the non-infected vines remained non-infected through the study. Unless indicated otherwise, measurements were taken on all three vines per block each year.
Leaf and canopy characteristics
Leaf gas exchange was measured for all vines on 6 and 12 July, 30 Aug, and 18 Sept 2018 using a portable leaf gas exchange system (6400XT, LI-COR) equipped with an artificial light source (6400-02B) that provided a photosynthetic photon flux of 1000 μmol/m2/sec. Chamber air flow was 500 μmol/sec, air temperature was 27 ± 2°C, relative humidity was 29 ± 1%, and CO2 concentration was maintained at 400 μL/L. Measurements were taken between 1000 hr and 1400 hr each day. For each vine, two mature exposed asymptomatic leaves at near midheight of the canopy were measured. On 6 and 12 July and 30 Aug 2018, the leaf canopy temperature was measured using a noncontact scanning infrared thermometer (InfraPro 3 model 35629-20, Oakton Instruments). The leaf canopy was scanned for 15 sec at ~45 cm from the surface, covering the sun-exposed side of the vine canopy and avoiding gap areas.
GRBD leaf symptoms became apparent at the end of June. On 12 July, 30 Aug, and 11 Sept 2018, 10 replicate blocks were randomly chosen to assess differences in leaf gas exchange of asymptomatic and symptomatic lamina areas of GRBD vines. For each GRBD vine, one asymptomatic and one symptomatic leaf, both mature and exposed, were selected for gas exchange measurements as described above. For the symptomatic leaf, asymptomatic (green) and symptomatic (red) lamina areas were measured separately.
On 5 July and 28 Sept 2018, leaf pigments and N content were measured in the 10 blocks selected for assessing leaf gas exchange. Three mature exposed asymptomatic leaves were sampled from the GRBD vine and from one of the non-GRBD vines, randomly chosen, per block. Before sampling on 28 Sept, gas exchange was measured on each leaf as described above. The leaves were bagged in plastic film and transported to the lab in an ice-filled cooler for analysis. Eight leaf discs, each 18 mm in diameter, were sampled from each side of the center vein of each leaf lamina, avoiding areas with large veins. Two samples of 24 discs were each made by combining eight discs from each of the three leaves sampled per vine. After weighing, one sample was dried for analysis of N concentration using a LECO N analyzer (model FP-528, LECO Corp.). The other sample was prepared for pigment analysis by cryomilling, then extracting a 0.3 g portion three times with 8 mL acetone. The extracts were centrifuged and the supernatants combined before acetone was added to make 25 mL. The combined extract was diluted by half with acetone and absorbance was measured at 470, 520, 645, 662, and 720 nm. Chlorophylls a and b, xanthophylls, and carotenes were determined following the calculation used for acetone extractions (Lichtenthaler and Buschman 2001).
Yield components and fruit composition
Basic composition of the berry juice was determined from a 20-berry sample (one to two berries per cluster) collected from each vine the day before harvest each year. The samples were weighed to determine average berry mass, then crushed with a mortar and pestle. The released juice was pressed through cheesecloth and analyzed for soluble solids by a digital refractometer (Pallette, Atago), pH by electrode (London Scientific), and TA by titrating 5 mL juice with 0.1 M NaOH to a pH endpoint of 8.1 using an automatic titrator (Metrohm Canada).
Fruit was harvested when the juice soluble solids of berries from non-infected vines averaged ~26 Brix on 24 Oct 2017 and 10 Oct 2018. Average cluster mass was calculated by dividing total yield/vine mass by the number of clusters per vine. A 120-berry sample was collected at random from the fruit harvested from each vine and stored at -20°C before analysis for phenolics. While frozen, the skins and seeds were removed, weighed, and lyophilized prior to phenolics extraction with acidified methanol (MeOH:HCl, 99:1, v/v). Anthocyanins were determined using the Glories method (Glories 1984). Tannins were determined using a vanillin assay (Burns 1971).
Bud hardiness and vine pruning mass
The cold hardiness of buds was measured using differential thermal analysis (DTA) for low temperature exotherm (LTE) detection, following the methods of Mills et al. (2006) and Bowen et al. (2016). Buds subjected to DTA were collected on 17 Nov 2017, and 8 Feb, 19 March, and 9 Nov 2018. On each date, one dormant cane per vine was sampled and transported to the lab for DTA the same day. The buds were excised from basal node positions 3 to 7. The DTA was performed inside a freezer (Tenney) programmed to ramp from 0 to -36°C at a rate of 4°C per hour. LTEs were detected by a Keithley multimeter data acquisition system (Keithley Instruments) as voltage outputs and recorded to a Microsoft program ExcelLINX (Emcore Corp.). The temperature at which 50% of the sampled buds were killed (LTE50) was determined from the LTE data. All pruned dormant canes, including those sampled for DTA, were weighed each year to determine vine pruning mass.
Winemaking
Four field blocks were delineated based on midafternoon foliage canopy temperature as an indicator of vine transpiration and water stress, measured using a drone-mounted thermal imager (FLIR) in late August each year. Each block had four GRBD vines and eight non-GRBD vines. The fruit was harvested from each vine, and after yield component data and berry samples for compositional analysis had been collected, five lots of ~2.2 kg of fruit were composed from each of the four replicate blocks to have different contents by mass of GRBD fruit: 0, 5, 10, 20, and 100%.
Wine was made from each fruit lot to produce four replicate wines, based on the four field blocks, for each for the five GRBD fruit-content treatments. To make the wines, the berries were removed from rachises and mixed before a 1.4 kg portion was lightly crushed by hand and placed into a 1.5-L glass coffee press (Bodum). A 50-mL juice sample was taken to measure soluble solids, pH, and TA by the methods described above, and yeast assimilable nitrogen (YAN) by the formol method (Gump et al. 2001). Potassium metabisulfite (KMS) (66 mg) was added to the must before cold soaking for three days at 10°C in a temperature-controlled room. Fermentation was initiated by increasing the temperature to 28°C and adding 0.5 g of yeast (Lalvin D254) with 0.5 g of GoFerm (both from Lallemand). The next day, 0.1 g of diammonium phosphate (DAP) was added, followed by 0.3 g of Fermaid K (Lallemand) and 0.137 to 0.496 g of DAP a day later to increase the YAN in all wines to 230 mg/L. During the cold soak and fermentation on skins, the musts were stirred daily and caps were kept submerged using the coffee press plunger. After fermenting for five days, the wines were pressed and transferred to 1-L Kimax glass jars (Kimble Chase) with air locks. Three days later all fermentations were complete (soluble solids ≤ -1.0 Brix). The room temperature was lowered to 20°C, and malolactic fermentation was initiated by adding 15 mg of Viniflora CH16 (Gusmer Enterprises Inc.) and filling the jar headspace with N2. After 39 days, the malic acid content of the wines, measured using an L-Malic Acid Unitab test kit (Unitech Scientific), was <0.2 g/L. The wines were racked 13, 32, and 49 days after the malolactic fermentation. After each racking, KMS was added to bring the SO2 content to 30 mg/L (first two times) or 20 mg/L (last time), and the jar headspace was filled with N2. The wines were cold stabilized at -1°C for four days, then racked, bottled, and stored at 10°C until the sensory evaluation was performed. In late May, ~19 weeks after bottling, wine samples were collected for color and compositional analysis just prior to the sensory evaluation.
Wine composition and sensory evaluation
Percent alcohol was determined using a multipurpose refractometer (Refracto 30PX, Mettler Toledo) that was calibrated (R2 = 0.983) with an ebulliometer (DuJardin-Salleron). Reducing sugar was determined using the Rebelein reducing sugar test (Amerine and Ough 1980). Wine color characteristics (lightness, hue, and chroma) were assessed using a Minolta color analysis system (Konica Minolta Sensing Americas Inc.).
The wine sensory evaluation each year followed procedures for descriptive analysis described by King et al. (2001). The wines were first tasted by five tasters experienced in red wine sensory evaluation to determine the main attributes that characterized and differentiated the wines. The tasters noted the aroma, flavor, and mouthfeel characteristics of each wine and then reached a consensus on the attributes and descriptors, which were the same each year (Table 1). The aroma and flavor attributes included red fruit, black fruit, cooked fruit, and vegetal. The other attributes were acidic flavor, aftertaste, body, and astringency. Each year, a panel of 10 tasters consisting of five male and five female winemakers from the Okanagan Valley evaluated the wines over a three hour period. All tasters had extensive experience in wine sensory evaluation, and most had previously participated in a defect and flavor recognition training program offered by the BC Wine Institute. The evaluations began with a training session that used a blended reference wine from the study to build consensus among the tasters on the magnitude or intensity of each sensory attribute. The rating form for the wines used a 10-cm unstructured line scale for rating the magnitude or intensity of each attribute, with anchors at 1 cm and 9 cm for low and high, respectively. The wines for evaluation were presented in a randomized block design as 35-mL samples at 20°C in ISO glasses, each labeled with a random-number code. There were four sets of five wines corresponding to the four replicate field blocks from which the five wines having different GRBD fruit content were made. The wines in each set were first evaluated to rate their aroma attributes and then tasted to rate their flavor, mouthfeel, and aftertaste attributes. The wine was expectorated after tasting. Bottled water was used for mouth rinsing between wines, followed by a serving of melba toast and further rinsing as required. The tasters took a short break (10 to 15 min) as desired after evaluating each set of wines.
Statistical analyses
The effects of GRBD on the individual vine and wine characteristics were determined by analysis of variance or repeated measures analysis appropriate for the randomized block designs, which had 18 blocks for most vine and fruit characteristics, 10 blocks for leaf chemistry and gas exchange measured on symptomatic leaves, and four blocks for wine characteristics. Trend analysis was used to detect trends in wine chemistry, color, and sensory characteristics in response to the GRBD fruit content in the wines. Principal component analysis (PCA) was performed on the mean sensory scores and wine color and chemistry characteristics found to be affected by GRBD for the 20 wines each year. All statistical analyses were conducted using SAS (SAS Institute, Inc.).
Results
Leaf gas exchange, composition, and abscission
For asymptomatic leaves, photosynthesis, stomatal conductance, and transpiration were reduced by GRBD on all measurement dates in July through September (Figure 2). On average, GRBD reduced photosynthesis by 27%, stomatal conductance by 40%, and transpiration by 34%, but the effects were greater in July (preveraison) than in September (postveraison). In early July, concentrations of the pigments, chlorophylls a and b, and carotenoids in leaf laminas were lower for asymptomatic leaves of GRBD vines than for leaves of uninfected vines (Table 2). In late September only chlorophyll b was lower in response to GRBD. Over the measurement period, leaves of non-GRBD vines had a greater decline in chlorophyll a and carotenoids than did asymptomatic GRBD leaves. Similarly, leaf N content was reduced by GRBD in early July and late September, but leaf N declined more in non-GRBD leaves than in GRBD leaves.
Within GRBD vines, photosynthesis was higher for asymptomatic than symptomatic leaves, especially in July when symptoms first appeared (Figure 3). Differences in photosynthesis between green and red lamina areas of symptomatic leaves were small (in September) or not significant (in July and August).
In mid-November, after leaf fall was complete for all non-GRBD vines (that is, they had no retained leaves), all GRBD vines were observed to retain some (>30) leaves (Figure 4).
Vine growth, yield components, and crop load
GRBD vines had less vegetative growth than did non-infected vines in both years, resulting in a 20% average reduction in pruning mass (Table 3). After shoot-thinning but prior to cluster thinning in late May 2018, the number of shoots per vine averaged 20.0 for both control and GRBD vines, but GRBD vines had 25% fewer clusters per shoot (data not shown and unavailable for 2017). In both years, fruit yield was substantially reduced, by 42% on average, by GRBD. Crop load, expressed as fruit per pruning mass, averaged 26% lower in GRBD vines. The effect of GRBD on yield per vine was due both to fewer clusters per vine (19%) and fewer berries per cluster (47%). GRBD also caused increases in berry mass, averaging 39%, and the number of seeds per berry (23%). The effects on berries per cluster and berry mass were visibly apparent (Figure 5).
Bud hardiness
At all measurement times—three during the 2017 to 2018 dormant period and one in fall 2018—bud hardiness was lower for GRBD vines. The increase in bud LTE50 ranged from 1.6 to 3.7°C, and averaged 2.5°C (Figure 6). No cold damage was observed during the bud LTE assessments. On each measurement date, bud LTE50 of the non-GRBD vines was similar to the LTE50 reported biweekly on the BC Wine Grape Council website (BCWGC.org) for Cabernet franc growing within 4 km of the study site (Figures 1 and 6).
Fruit composition
At harvest each year, fruit from GRBD vines was on average 4.10 lower in Brix, 2.26 g/L higher in TA, and 0.23 higher in pH (Table 4). Skin phenolics concentrations (dry mass basis) were also affected by GRBD (Table 5). Anthocyanin was reduced by GRBD in both years, by 22% on average. GRBD also reduced the concentrations of tartaric acid esters and flavonol in skins in 2017, reduced flavonol in seeds in 2018, and increased the concentration of condensed tannin in skins in 2018.
On a fresh berry mass basis, GRBD substantially reduced the anthocyanin concentration of berries, averaging 37% (Table 4). GRBD berries also had 18% less skin tannin concentration in 2017, and 7% less seed tannin and 7% less total tannin concentration in 2018.
Wine composition and sensory evaluation
GRBD fruit content in the wines, up to 100%, was negatively related to the concentrations of must YAN, and wine anthocyanin, flavonols, residual sugar, tartaric acid, and alcohol in both years, and condensed tannin in 2017 (Table 6). Wine lightness, chroma, and hue values were positively related to GRBD fruit content, indicating that GRBD fruit produced lighter, brighter, and less red (more yellow) wines. Sensory characteristics that declined in intensity with increasing GRBD fruit content were black fruit aroma and flavor, body, and aftertaste in both years; astringency in 2017; and red fruit flavor in 2018 (Table 7). Vegetal aroma and flavor, and acidity increased with GRBD fruit content in both years.
Inclusion of GRBD fruit content in wines, up to 20%, caused reductions in wine anthocyanin concentration both years, in red fruit aroma and flavor in 2017, and in must YAN and alcohol in 2018 (Tables 6 and 7). Vegetal aroma and flavor increased with GRBD fruit content up to 20% in 2018 (Table 7).
Results of the PCA for sensory and compositional characteristics of the wines were remarkably similar between years (Figures 7 and 8). In 2017 and 2018, respectively, the two main factors explained 69% and 76% of the total variation among the wines. In both years, the positive sensory variables of Factor 1 included black fruit flavor and aroma, body, and aftertaste, and were associated with anthocyanin and alcohol content of the wines. The negative sensory variables of Factor 1 included vegetal flavor and aroma, and were associated with lighter color, higher chroma, less red hue, and higher must TA and pH. All of the 100% GRBD wines associated negatively with Factor 1. Except for one wine made with 20% GRBD fruit in 2018, wines made from all or mostly non-GRBD fruit associated positively with Factor 1. The association was weakest for wines made with 20% GRBD fruit. GRBD fruit content in the wines had little association with Factor 2, whose main positive variable was astringency, and negative variables were inconsistent between years.
Discussion
This study found significant deleterious effects of GRBD on the growth, development, and fruiting performance of Cabernet franc grapevines in the Okanagan Valley, BC. Vegetative growth was impaired by GRBD in both years, consistent with results for Gamay grown in Switzerland (Reynard et al. 2018) and Merlot and Cabernet franc grown in the state of Washington (Poojari et al. 2013) but inconsistent with results for Cabernet Sauvignon in California, which showed no impact of GRBD on vine growth (Martinez-Luscher et al. 2019). The preveraison effect of GRBD in reducing photosynthesis of asymptomatic leaves was also reported for Gamay (Reynard et al. 2018). In our study, the effect on asymptomatic leaf photosynthesis continued after veraison but was diminished and associated with lower levels of chlorophylls a and b and leaf N content. A similar diminishing effect on photosynthesis was found for Cabernet Sauvignon in California (Martinez-Luscher et al. 2019). In our study there was likely a substantial effect of GRBD on whole vine photosynthesis after leaves became symptomatic, as a result of the lower vegetative growth of GRBD vines and highly reduced pho tosynthesis by symptomatic leaves. This effect would cause a substantial reduction in carbon availability within vines and may have contributed to a reduction in bud fruitfulness, which was indicated by the lower number of clusters per vine.
Effects of GRBD on yield components in our study, including the higher berry mass and lower numbers of clusters per vine and berries per cluster in both years, were similar to GRBD effects reported for Chardonnay in California (Girardello et al. 2019). The higher berry mass may have been compensatory for the lower berry number and yield per vine. These effects were not found in GRBD studies on Gamay in Switzerland or Cabernet Sauvignon in California, which found no effects on yield (Reynard et al. 2018, Martinez-Luscher et al. 2019) and a reduction in or no effect on berry mass (Martinez-Luscher et al. 2019). More research is needed to clarify whether the differences in GRBD effects on yield components result from differences among cultivars, study site conditions, or GRBV genetic variants. In our study, the GRBV infecting vines belonged to clade 2.
The reduction in bud hardiness by GRBD could affect the performance of vines growing in the cold-winter regions of Canada and the United States, where minimum temperatures are frequently near or below hardiness levels for V. vinifera. The hardiness reduction by GRBD averaged 2.5°C in this study, which could substantially reduce primary bud survival and fruit production after exposure to temperatures near bud hardiness levels. Grapevine bud hardiness is highly dependent on temperature acclimation but has been shown to be influenced by abscisic acid (ABA) applications to vine leaves in the previous growing season (Bowen et al. 2016). GRBD effects on leaf ABA metabolism have not yet been studied, but in berries, the metabolic pathways of ABA and other hormones were found to be compromised by GRBV infection (Blanco-Ulate et al. 2017). In the current study, leaf fall was delayed or incomplete for GRBD vines, whereas ABA applied postharvest to grapevine canopies has been found to advance leaf fall (Bowen et al. 2016), which further indicates an effect of GRBD on ABA metabolism. More research is needed to elucidate the role of ABA and other hormones in GRBD effects on vegetative development and physiological processes such as hardiness acclimation. GRBD effects on trunk, cane, and root hardiness could further impact yield and overwinter survival of GRBD vines in cold-winter regions.
In this study, GRBD vines had fewer clusters and berries per cluster, which could indicate primary bud damage due to cold exposure. However, no bud damage was observed during the hardiness assessments, and the minimum exposure temperatures were unlikely to have caused bud damage. During the dormant period (November to April) preceding the 2017 study year, daily minimum temperatures remained above the LTE50 reported for Cabernet franc growing nearby within the sub-GI (Figure 1). The lowest temperature, -18.7°C, was >5.7°C warmer than the LTE50 for local Cabernet franc at that time and therefore unlikely to damage buds of healthy Cabernet franc vines. For GRBD vines, the bud LTE50 may have been elevated by up to 3.7°C, based on the maximum effect of GRBD on LTE50, and would be at least 2°C lower than the minimum temperature exposure (-18.7°C). Bud damage would be unlikely under those conditions as the LTE range during winter is within 3°C (Mills et al. 2006). During the dormant period preceding the 2018 study year, minimum temperatures remained well above the bud LTE50 for Cabernet franc growing nearby within the sub-GI. The lowest temperature, -14.6°C, was 7.9°C warmer than the bud LTE50 for Cabernet franc at that time. If the bud LTE50 was 3.7°C higher in GRBD vines, it would be 4.2°C below the minimum exposure, making bud damage unlikely. More likely is that the effects of GRBD on yield components arose from decreased bud fruitfulness, perhaps due to reduced photosynthesis and carbohydrate availability during bud development, or from other physiological impairments caused by GRBD.
The detrimental effects of GRBD on fruit and wine characteristics were substantial and mostly consistent between years. The impaired development of GRBD berries was apparent from the reductions in soluble solids, anthocyanin, and YAN concentrations, and increases in pH and TA. The impacts on wine sensory characteristics included poor color, acidic taste, vegetative flavor and aroma, and a lack of body, aftertaste, and black fruit flavor and aroma. The effect on berry anthocyanin was substantial, which may have involved disrupted pathways for ABA metabolism that control and influence berry ripening and anthocyanin synthesis (Owen et al. 2009, Blanco-Ulate et al. 2017). The noted severity of the effects on fruit composition spurred our addition of wine-making treatments that included low amounts, up to 20%, of GRBD fruit to mimic fruit harvested from vineyards having low levels of GRBD, such as found in the Okanagan Valley (Poojari et al. 2017). Although some traits, including color, body, and black fruit character, were not individually impacted by including up to 20% GRBD fruit, adding low amounts of GRBD fruit reduced the concentration on anthocyanin in both years, increased the vegetal character and astringency in one year, and reduced red fruit character in the other. Wines with a combination of more body, aftertaste, and black fruit character, and less vegetative character, were made with low or no GRBD fruit content. These effects of GRBD on wine characteristics could substantially reduce the value of wines produced from mostly GRBD fruit.
Vines newly infected with GRBV were not found in the Cabernet franc vineyard block during the two-year study. The block was established with plants sourced from a California nursery within the period when GRBV was likely introduced into the Okanagan Valley (Poojari et al. 2017). Although the risk of spread appears to be low at the site, the high degree of impact on vine performance, especially fruit composition, would be difficult to mitigate through management practices, as found in another study (Levin and KC 2019). The severity of effects of GRBD on yield and fruit composition, and on bud hardiness, warranted removal of infected vines from the study block, which previously had produced superior-quality wines. The effects of GRBD constitute a serious threat to winegrape production, particularly in cold-winter regions in Canada and the U.S. Understanding the likely source of the GRBV infection and its potential impacts has emphasized the importance of planting virus-free transplants, and the value of virus testing to identify and manage diseased vines.
Conclusions
The deleterious effects of GRBD on yield, fruit composition, wine characteristics, and bud hardiness constitute a serious threat to grapegrowing regions in Canada and the United States where GRBD has become widespread. The degree of these impacts would be difficult to mitigate through management practices, which underscores the importance of planting virus-tested vines to prevent the introduction and spread of GRBD.
The observations that GRBD delayed or hindered leaf fall and reduced bud hardiness and berry anthocyanins all suggest the involvement of ABA in GRBD symptoms. More research is warranted to elucidate how GRBV affects the synthesis and catabolism of ABA in vine tissues and to determine whether ABA applications could alleviate GRBD symptoms in grapevines.
Although including up to 20% of GRBD fruit in winemaking consistently affected only anthocyanin concentration, the pronounced differences between wines made entirely with GRBD and non-GRBD fruit indicate that greater amounts of GRBD fruit in wine lots could significantly impact wine quality. The cost of removing and disposing of GRBD fruit from a wine lot needs to be measured against the loss of wine value due to GRBD.
Acknowledgments
This study was supported by the British Columbia Wine Grape Council, the Canadian Grapevine Certification Network, and Agriculture and Agri-Food Canada. Technical assistance was provided by Brad Estergaard, Steve Marsh, Emmanuelle Jean, John Drover, and Julie Boulé.
Footnotes
By downloading and/or receiving this article, you agree to the Disclaimer of Warranties and Liability. The full statement of the Disclaimers is available at http://www.ajevonline.org/content/proprietary-rights-notice-ajev-online. If you do not agree to the Disclaimers, do not download and/or accept this article.
- Received February 2020.
- Revision received June 2020.
- Accepted June 2020.
- Published online October 2020
- Copyright © Her Majesty the Queen in Right of Canada, as represented by the Minister of Agriculture and Agri-Food Canada, 2020.