Abstract
Field studies were conducted on Vitis vinifera L. cvs. Cabernet franc and Merlot to evaluate the effects of basal leaf removal timing and severity on 3-isobutyl-2-methoxypyrazine (IBMP) concentration in grape berries. Treatments consisted of removing either 50% or 100% of leaves from the fruiting zone at either 10 days after anthesis, 40 days after anthesis, or 60 days after anthesis. In the second year of the Cabernet franc study, a 15-day postveraison leaf removal treatment was also included. In both years of the Cabernet franc study, significant reductions in IBMP (range = 28 to 53%) were observed before veraison compared with the control in both 10 days after anthesis treatments (50% and 100% leaf removal). In 2007, all leaf removal treatments significantly reduced IBMP concentrations compared with the control (46 to 88%) in Cabernet franc berries at harvest, with the greatest reduction observed in the 100% leaf removal treatments at 10 days after anthesis and 40 days after anthesis. In 2008, the 100% leaf removal treatment at 10 days after anthesis and the 50 and 100% leaf removal treatments at 40 days after anthesis significantly reduced IBMP concentrations (34 to 60%) in mature Cabernet franc berries. In the Merlot trial, all leaf removal treatments significantly reduced IBMP concentrations (38 to 52%) at harvest. In summary, early season (10 to 40 day after anthesis) basal leaf removal reduced IBMP accumulation preveraison compared with the control in both studies, suggesting that early leaf removal is a more effective management strategy to reduce IBMP accumulation in grape berries than leaf removal later in the season.
The 3-alkyl-2-methoxypyrazines (MPs) are a class of odorants associated with “green,” herbaceous aromas of some Bordeaux winegrape (Vitis vinifera L.) cultivars. Quantitatively, 3-isobutyl-2-methoxypyrazine (IBMP) is the predominant MP in grapes and wine, typically an order of magnitude higher in concentration than 3-isopropyl-2-methoxypyrazine (IPMP) and 3-sec-butyl-2-methoxypyrazine (sBMP) (Alberts et al. 2009). The sensory detection threshold for IBMP is reported to range from 0.5 to 2 pg/g in water (Buttery et al. 1969, Kotseridis et al. 1998, Seifert et al. 1970) and 10 to 15 pg/g in red wine (de Boubee et al. 2000, Kotseridis et al. 1998). When present at concentrations near sensory threshold, MPs may contribute positively to wine quality by adding complexity and, in some cases, varietal character (Allen et al. 1991). At higher concentrations, MPs can result in excessive herbaceousness and suppressed fruitiness in wines (Allen and Lacey 1999, Hein et al. 2009, Pickering et al. 2005). MPs are efficiently extracted by conventional red wine practices, and their concentrations in wine are strongly correlated to their concentrations in grapes (Ryona et al. 2009). Several studies have evaluated the efficacy of vinification and cellaring practices in reducing MPs (Blake et al. 2009, de Boubee 2003, Marais 1998, Pickering et al. 2006) and have generally concluded that remediation of MPs is ineffective or else results in other nonselective changes to the wine. Viticultural management strategies that reduce MPs in the vineyard have thus been proposed to be the most effective way to control MP concentration in wine (Bogart and Bisson 2006).
In grape berries, IBMP begins to accumulate around 10 days after anthesis with a peak in concentration occurring approximately 0 to 14 days before veraison, followed by a rapid decline during maturation (de Boubee et al. 2000, Hashizume and Samuta 1999, Ryona et al. 2008, Sala et al. 2004). IBMP concentrations in mature berries are reported to be less than 10% of their preveraison peak concentrations. A strong correlation (R2 = 0.936) between IBMP concentrations in mature Cabernet franc berries and preveraison peak concentrations has been reported (Ryona et al. 2008), suggesting that final IBMP concentration is primarily determined preveraison. Thus, management practices that affect initial accumulation of MPs in grapes preveraison are expected to more dramatically impact final MP concentrations at harvest than interventions later in the season, assuming similar maturities.
Fruit-zone leaf removal is a common viticultural practice and has been demonstrated to yield improved fruit chemistry at harvest (Percival et al. 1994, Poni et al. 2006, Reynolds et al. 1996, Zoecklein et al. 1992, 1998) and to improve fungal control (Chellemi and Marois 1992, Percival et al. 1994, Wolf et al. 1986, Zoecklein et al. 1992). These effects are generally hypothesized to be mediated through an increase of sunlight reaching the fruiting zone. Several groups have observed that cluster light exposure results in lower MP concentrations in mature fruit (Allen et al. 1996, de Boubee et al. 2002, de Boubee 2003, Marais et al. 1999, Noble et al. 1995, Ryona et al. 2008). Recent work suggests that sun-exposed clusters accumulate less IBMP preveraison than shaded clusters within the same vine (Ryona et al. 2008) and that the proportional differences persist until harvest, although the physiological mechanisms behind these effects are not understood. Most of the aforementioned studies have observed differences between shaded and exposed fruit by using artificial shading or taking advantage of natural variation in light exposure within the canopy, but little work has been published on the effectiveness of specific vineyard practices (e.g., leaf removal) to reduce MP accumulation preveraison and subsequent levels at harvest. A 68% reduction in IBMP concentration of Cabernet Sauvignon at harvest resulted from removal of lateral shoots and basal leaves on the east side of the fruiting zone at fruit set compared to an unthinned control (de Boubee 2003). A similar treatment imposed postveraison resulted in only a 10% reduction in IBMP at harvest. However, that report did not consider more than one preveraison leaf removal timing, the period when the accumulation of MPs is greatest (Ryona et al. 2008), nor did it investigate the effects of the severity of leaf removal. We are unaware of any other literature that has quantified the impact of leaf removal on MPs in grape berries. The objective of this study was to investigate the impact of timing and severity of leaf removal on IBMP concentration in Cabernet franc in the Finger Lakes and Merlot on Long Island, New York.
Materials and Methods
Experimental design.
Two commercial vineyards located in Ovid, New York (42.67°N, 76.82°W; Finger Lakes AVA, Cayuga Lake) and Cutchogue, New York (40.99°N, 72.48°W; Long Island AVA, North Fork) were used in this study. The soil types were classified by the U.S.D.A. as Howard series with a gravelly loam structure, well drained, and a depth of >2 m and as Haven series with a loamy structure, well drained, and a depth of >2 m at Finger Lakes and Long Island, respectively. Vines at the Finger Lakes site were Vitis vinifera L. cv. Cabernet franc cl. 1 grafted on 3309C rootstock trained to a Scott Henry system with four canes. The upper canes were at 1.3 m height and shoots vertically positioned. The lower canes were at 1.0 m height and shoots downward positioned. Vines at the Long Island site were Merlot cl. 181 grafted on 3309C rootstock trained to a combination of low wire cordon and a flat cane system with either two cordons or two canes at 1.0 m height and shoots vertically positioned. Vine spacing for both sites was 2.0 m between vines and 2.5 m between rows. Vine management was performed according to the standard viticultural practices for vinifera in the Finger Lakes and Long Island regions. The experimental design was a randomized complete block with four replications. The experimental plot at each site consisted of four rows, and each experimental unit consisted of eight contiguous vines in each row.
Treatments consisted of a control (no leaf removal); removing the first, third, and fifth leaf from the base of each shoot at 10 days after anthesis (10 DAA 50%), 40 days after anthesis (40 DAA 50%), or 60 days after anthesis (60 DAA 50%); and removing the first five leaves beginning at the base of each shoot at 10 days after anthesis (10 DAA 100%), 40 days after anthesis (40 DAA 100%), or 60 days after anthesis (60 DAA 100%). Two additional treatments were added at the Cabernet franc site in the second year of the study: removing the first, third, and fifth leaf from the base of each shoot at 15 days after veraison (15 DAV 50%) or removing the first five leaves from the base of each shoot at 15 days after veraison (15 DAV 100%). All basal leaf removal treatments were applied by hand on all fruiting and nonfruiting shoots of each vine. The beginning of bloom was noted on 18 June 2007 and 19 June 2008 (Cabernet franc), and 22 June 2008 (Merlot). Time of anthesis was determined as the date on which 50% capfall was visually estimated. In 2007, the calendar dates for the treatments in Cabernet franc were anthesis (17 June), 40 days after anthesis (27 July), 60 days after anthesis (16 Aug), and harvest (21 Oct). In 2008, the calendar dates for the treatments in Cabernet franc and Merlot were anthesis (18 June and 21 June, respectively), 40 days after anthesis (28 July and 31 July), 60 days after anthesis (17 Aug and 20 Aug), and harvest (20 Oct and 16 Oct). The 15-day postveraison treatment was performed on Cabernet franc on 6 Sept 2008.
Sampling and harvest.
Five days after each basal leaf removal treatment was imposed in 2007 (15, 45, and 65 days after anthesis) and 5 to 15 days after each basal leaf removal treatment was imposed in 2008 (15, 50, 75, 85 days after anthesis) in Cabernet franc, 50-berry samples were collected at random from each experimental unit for IBMP quantification. At harvest, 150 berries were collected at random from each experimental unit in Cabernet franc and Merlot for IBMP quantification and chemical analysis. The berry samples were placed in plastic storage bags and immediately frozen followed by storage at −23°C for later analysis.
Yield components were assessed in the 2008 Cabernet franc and Merlot studies. At harvest, yield (measured with a hanging scale accurate to 0.01 kg; model SA3N340, Salter Brecknell, Fairmont, MN) and cluster counts were determined for each vine and an average was recorded for each replication. Crop weight and number of clusters were used to calculate average cluster weight. Yield data was not recorded in the 2007 Cabernet franc study as there was a significant “green harvest” of fruit by the grower several weeks before harvest. In the 2008 Cabernet franc study, cluster thinning at veraison was performed by the grower in all treatments to eliminate the least mature clusters.
Berry analysis for Brix, titratable acidity, and pH.
A subsample of 100 mature berries per experimental unit was removed from the −23°C freezer, placed in a 250-mL beaker, and heated to 65°C for one hour in a water bath to redissolve tartrates, pressed through cheesecloth with a pestle, and the juice was collected for analyses. Soluble solids (Brix) were measured using a digital refractometer (model 300017; SPER Scientific, Scottsdale, AZ) with temperature correction. Titratable acidity (TA) and pH were measured with an automatic titrater (Titrino model 798, Metrohm, Riverview, FL), and TA was measured with a 5.0-mL aliquot of juice by titration against 0.1 N NaOH to pH 8.2.
Berry analysis for IBMP.
3-Isobutyl-2-methoxypyrazine analysis was conducted using 50-berry samples. The extraction method was head-space solid-phase microextraction (HS–SPME) and quantification was performed by comprehensive two-dimensional gas chromatography time-of-flight mass spectrometry (GCxGC–TOF–MS) as described elsewhere (Ryona et al. 2009). In brief, HS–SPME was conducted using a LEAP CombiPAL Autosampler (Carrboro, NC) fitted with a three-phase fiber (DVB/CAR/PDMS). A 10-min online incubation at 650 rpm agitation rate and an incubation temperature of 80°C was applied before headspace-fiber insertion and equilibrium. Following fiber insertion, the vial was agitated at 100 rpm for 30 min at 80°C. Quantification was performed by GCxGC–TOF–MS (Pegasus IV, Leco Corp, St. Joseph, MI). SPME injections were splitless with a desorption temperature of 270°C. The first capillary column (30 m × 0.25 mm × 0.50 μm) was an RTX5 (Restek, Bellefonte, PA), and the second column (2.5 m × 0.10 mm × 0.10 μm) was a VF-WAXms (Varian, Palo Alto, CA). Helium was used as a carrier gas at a flow rate of 1 mL/min. The temperature program was as follows: initial hold for 5 min at 40°C, followed by a 5°C/min ramp to 120°C; then 2°C/min to 150°C, no hold; then 10°C/min to 250°C, 15 min hold. The GCxGC modulation time was 3 sec. The MS transfer line temperature was 230°C. The TOF–MS was operated in EI mode with an ionization energy of 70 eV. The electron multiplier was set to 1680 V. The TOF–MS data were stored at an effective acquisition rate of 120 Hz over a mass range of m/z 20 to 400. The qualifier ions were m/z = 124, 151, 166 for IBMP and m/z = 126, 153, 168 for [2H2]-IBMP. The quantifier ions were m/z = 124 and 126, respectively.
Statistical analysis.
Statistical analyses were conducted with SAS statistical software (SAS Institute, Cary, NC). Data was subjected to the Proc GLM procedure and means were separated using the Fisher’s protected least significant difference (LSD) at the 5% significance level. IBMP data for harvested Cabernet franc berries in 2007 and 2008 were not combined over years due to significant year by treatment interaction.
Results
Leaf removal in Cabernet franc.
Leaf removal timing and severity impacted the concentration of IBMP preveraison and at harvest in both 2007 and 2008. In 2007, at 15 days after anthesis, IBMP was present at quantifiable concentrations (data not shown), but no significant difference was observed in IBMP between the 10 DAA leaf removal treatment and the untreated vines. At 45 days after anthesis, both the 10 DAA 50% and 10 DAA 100% treatments significantly reduced IBMP concentrations by 52 and 53%, respectively, compared to the control (Figure 1A⇓). At 65 days after anthesis, the concentrations of IBMP in the 10 DAA 50% and 10 DAA 100% treatments were 55 and 65%, respectively, lower than the control (Figure 1B⇓). The period between veraison (65 days after anthesis) and harvest (125 days after anthesis) was marked by a decline in IBMP concentration. The IBMP concentration in mature fruit ranged from 0.5 to 4.3 pg/g (Figure 1C⇓) and averaged 1.1% of the observed maxima (65 days after anthesis). Although the only significant reduction in IBMP concentration at the three preharvest sample timings was observed for the 10 DAA 50%, 10 DAA 100%, and 40 DAA 100% treatments, all leaf removal treatments significantly reduced IBMP in mature berries with respect to the control (Figure 1C⇓). The range in Brix of the Cabernet franc berries at harvest in 2007 was 19.4 to 22.3 (Table 1⇓). The 10 DAA 50% and 10 DAA 100% treatments significantly increased Brix compared to the control by 5 and 10%, respectively. TA ranged from 6.4 to 8.6 g/L across treatments. All leaf removal treatments except the 60 DAA 50% treatment significantly reduced TA compared to the control.
In 2008, the 10 DAA 50% and 10 DAA 100% treatments significantly reduced the concentration of IBMP in Cabernet franc berries at 50 days after anthesis by 28% and 36% (Figure 1D⇑). At 75 days after anthesis the 10 DAA 100% and 40 DAA 100% treatments reduced IBMP concentrations by 25% and 48%, respectively (Figure 1E⇑). At 85 days after anthesis, there were no significant differences among treatments (data not shown). At harvest (124 days after anthesis), the range in IBMP concentration across all treatments was 1.2 to 3.5 pg/g (Figure 1F⇑) and averaged 1.3% of the observed preveraison (50 days after anthesis) maxima. Although the 10 DAA 50% and 100% treatments significantly reduced IBMP concentrations at the preveraison sample timing, the 10 DAA 100%, 40 DAA 50%, and 40 DAA 100% leaf removal treatments significantly reduced IBMP concentrations (range = 34 to 60%) at harvest. The range in Brix was 21.1 to 22.5, with no significant differences among treatments (Table 1⇑). TA ranged from 5.5 to 6.8 g/L among treatments. All treatments except 10 DAA 50% significantly reduced TA below the control. No differences in juice pH were observed among the leaf removal treatments. Yield, number of clusters, and average cluster weight per vine in 2008 ranged from 3.4 to 4.1 kg, 21.6 to 26.1, and 145.7 to 181.8 g, respectively, with no significant differences among treatments (data not shown).
Leaf removal in Merlot.
At harvest (117 days after anthesis), the range in IBMP concentration in Merlot berries across treatments was 3.2 to 6.7 pg/g (Figure 2⇓). Leaf removal at all timings and severities significantly reduced IBMP by a range of 37 to 52% compared to the control. Leaf removal timing and severity had no significant impact on Brix, TA, and pH (Table 1⇑). The 10 DAA 50% treatment had significantly lower yield (1.9 kg/vine) than the control and other treatments (range = 2.1 to 2.4 kg/vine). No significant differences were observed among treatments for number of clusters per vine (12.8 to 14.8) and average cluster weight (143.2 to 172.9 g).
Discussion
The highest concentrations of IBMP in Cabernet franc were observed at the preveraison sample timings (65 days after anthesis sampling in 2007, and at the 50 days after anthesis sampling in 2008 (Figure 1⇑). Differences in reported peaks between years are likely a function of different sample timings. In agreement with our results, previous research has demonstrated that IBMP reaches a maximum in the 2 to 3 weeks before veraison (de Boubee et al. 2000, Lacey et al. 1991, Ryona et al. 2008).
In both 2007 and 2008, significantly lower IBMP concentrations were observed in Cabernet franc berries in the 10 DAA 50% and 10 DAA 100% treatments compared to the control at the time points just before or just after veraison (65 days postanthesis in 2007, 50 days postanthesis in 2008). No significant effect of leaf removal was observed at these points with the 40 DAA 50% or the two 60 DAA treatments in either year, although the 40 DAA 100% treatment had lower IBMP than the control in 2007. These results are in concordance with a recent observation that cluster light exposure preveraison reduces IBMP accumulation (Ryona et al. 2008). Because basal leaf removal is widely shown to improve light penetration to the fruiting zone (Reynolds et al. 1996, 2006, Wolf et al. 1986, Zoecklein et al. 1992), the reductions in IBMP concentration that we observed are likely due to increased cluster light exposure. Generally, we did not observe a significant decrease in IBMP at the time point immediately following the treatment application. We did not observe significantly lower IBMP in the 40 DAA 50% treatment at 45 days after anthesis in 2007 or at 50 days after anthesis in 2008 nor did we observe a significant effect for the 40 DAA 100% treatment at 50 days after anthesis in 2008. We did, however, observe significantly lower IBMP in the 40 DAA 100% treatment at 75 days after anthesis in 2008, and at 65 days after anthesis in 2007. Similarly, no significant difference in IBMP was observed between the 10 DAA treatments and the control at 15 days after anthesis in both years nor was a difference observed at 75 days after anthesis between the 60 DAA treatments and the control. Thus, except for one case (40 DAA 100% in 2007), the impact of the leaf removal treatment was not observable until >15 days after the treatment was imposed.
Across all three studies, the largest and most consistent decreases for IBMP at harvest were observed in the early leaf removal treatments. In the 2007 Cabernet franc study, all treatments had significantly lower IBMP than the control at harvest, with the greatest reduction in the 10 DAA 100% and 40 DAA 100% treatments (Figure 1C⇑). In the 2008 Cabernet franc study, the 10 DAA 100%, 40 DAA 50%, and 40 DAA 100% treatments contained significantly lower IBMP at harvest compared to the control (Figure 1F⇑). In the 2008 Merlot study, all treatments resulted in lower IBMP than the control at harvest (Figure 2⇑). These results support the previous hypothesis that cluster light exposure preveraison inhibits accumulation preveraison, but has little effect postveraison, and that the relative differences in IBMP established before fruit maturation persist until harvest (Ryona et al. 2008).
Although preveraison leaf removal (10 or 40 days postanthesis) resulted in the largest decrease in IBMP levels at harvest compared to the control in the Cabernet franc studies, we also observed a smaller but still significant decrease in IBMP for the 60 DAA treatments in both the 2007 Cabernet franc and 2008 Merlot studies. Previous work (Allen et al. 1996, Marais et al. 1999, Ryona et al. 2008) indicates that cluster exposure reduces IBMP accumulation preveraison, but does not increase IBMP degradation postveraison on a percentage basis. A potential explanation is that IBMP synthesis and degradation are occurring simultaneously, at similar rates, in berries 40 to 60 days postanthesis. Thus, IBMP synthesis may still be occurring around veraison although the berry IBMP concentration is unchanged. In support of this hypothesis, we observed a sizeable decrease (41%) in 2008 Cabernet franc for the 40 DAA 100% treatment at Day 75 compared to the control, even though no significant decrease was observed at Day 50. However, IBMP synthesis likely does not persist late into the season. In the 15-day postveraison treatments (50% and 100%) in the 2008 Cabernet franc study, we observed no significant change in IBMP levels at harvest compared to the control (Figure 1F⇑). Similarly, postveraison cluster shading has been reported to have no impact on IBMP in Cabernet Sauvignon (Sala et al. 2004).
Several studies have reported that growing season temperature and MP content in mature berries are inversely correlated (Allen et al. 1991, 1994, Falcao et al. 2007). The total growing degree days (GDD, base 10°C) accumulated at the Finger Lakes site in 2007 and 2008 from 1 Jan to harvest (17 and 16 Oct) was 1552 and 1410, respectively (Figure 3⇓). Although there were 142 more total GDD accumulated in 2007, there was less than 1% difference between years in GDD accumulated from 10 days after anthesis to veraison. The period between veraison and harvest was much warmer in 2007 (492 GDD during ripening) than in 2008 (349 GDD). Thus far, the relative importance of preveraison versus postveraison growing season temperature in determining IBMP content in grapes has not been reported. Although we observed large differences in GDD between years, the average IBMP concentrations measured at harvest in Cabernet franc (2.0 pg/g in 2007 and 2.3 pg/g in 2008) were similar, suggesting that the postveraison GDD accumulation did not have a strong influence on final IBMP concentration. A strong correlation has been noted between IBMP concentrations at veraison and harvest (Ryona et al. 2008), suggesting that final concentration is dependent upon preveraison conditions.
Although the harvest concentrations of IBMP observed in this study are below reported sensory thresholds in red wine (de Boubee et al. 2000, Kotseridis et al. 1998), the leaf removal treatments in 2007 and 2008 reduced the final IBMP concentration in Cabernet franc by up to 88% and 60%, respectively, and in Merlot by up to 52% compared to the control. In Cabernet franc, IBMP accumulation was reduced by up to 65% (2007) and up to 36% (2008) by the 10 DAA 50% and 10 DAA 100% treatments at the observed maximum IBMP concentrations. Although we did not measure cluster light exposure for the various treatments in this study, our findings may be consistent with other groups that have evaluated the effects of preveraison cluster light exposure on IBMP concentration (Allen et al. 1996, de Boubee 2003, Marais et al. 1999, Ryona et al. 2008).
Conclusion
Preveraison basal leaf removal treatments reduced IBMP concentration in Cabernet franc and Merlot berries at harvest. In Cabernet franc, accumulation of IBMP in the preveraison period was reduced by leaf removal, likely due to improved light interception by the clusters. In a situation where IBMP is present in concentrations near sensory threshold, leaf removal during the growing season could be critical in reducing accumulation of IBMP. The earliest (10 days after anthesis and 40 days after anthesis) leaf removal treatments yielded the greatest benefit in reducing IBMP.
Footnotes
Acknowledgments: The authors thank Sheldrake Point Vineyard and Pellegrini Vineyards for cooperation on this project, Imelda Ryona for assistance with sample preparation, and Terry Bates for assistance in editing this manuscript.
- Received November 2009.
- Revision received March 2010.
- Accepted April 2010.
- Published online September 2010
- Copyright © 2010 by the American Society for Enology and Viticulture