Photosynthetic and Photoinhibition Behavior of Two Field-Grown Grapevine Cultivars under Multiple Summer Stresses

Experimental vineyard.

This study was carried out during summer 2003 and 2004 in a 7-year-old vineyard situated on a hillside with a southern exposure in the Marche region (central Italy; lat. 43°40′N; elevation 50 m asl). The vineyard covers ~2 ha, with 1 ha Sangiovese vines and 1 ha Montepulciano vines. Twenty vines of each cultivar on adjacent rows with a north-south orientation were used for all measurements. All vines were of clonal origin, grafted onto 420A rootstock, trained to cordons, spur-pruned, and spaced at 2.5 × 1 m. The main wire was 0.70 m above the soil surface, and shoots were maintained in a vertical position by three pair of movable wires, the highest of which was 1.80 m above the soil surface. All vines were uniformly pruned, with a bud charge of 12 ± 2 buds per vine. Vines were grown on a calcareous soil, of clayey texture, with 23.9% total calcium carbonate and 12.3% active lime. The soil was alkaline (pH 8.1), low in organic matter (0.7%) and phosphorous (11 g/kg), and high in potassium (308 mg/ kg) and magnesium (720 mg/kg). The climate is Mediterranean, with a mean annual rainfall of 780 mm, of which ~62% falls during the dormant period. Weather conditions during the study were monitored by an automated meteorological station located near the vineyard. In both study years, no irrigation was carried out and the manual canopy management practice included shoot-tip cutting (first week of July).

Chlorophyll fluorescence, gas exchange measurements, and leaf characteristics.

The data for the physiological, biochemical, and morphological traits were for leaves in the basal (4th–5th node from the base) and medial (12th–15th node from the base) portions of the Sangiovese and Montepulciano primary shoots. Each data set consisted of at least 10 replicates.

In 2003 and 2004, the fluorescence transients for chlorophyll a were measured in the morning (0600–0700 hr) and in the early afternoon (1300–1400 hr; denoted as “midday”), from just after berry set to the postveraison stage, at roughly two-week intervals, using a portable, continuous, excitation-type fluorimeter (Handy-PEA, Plant Efficiency Analyser, Hansatech Instruments, Norfolk, UK) and according to a published method (Strasser et al. 1995, 2000). Fluorescence emission was induced by homogeneous irradiation on a 4-mm diam area of the leaf sample (upper surface), using an array of six high-intensity, light-emitting diodes (peak 650 nm), and an irradiance >3,000 μmol photon m⁻² s⁻¹. The duration of the dark adaptation required to obtain relaxation of the rapid energy-dependent, nonphotochemical quenching was experimentally established as 20 min. An algorithm was used to determine the line of best fit through the initial 4–16 data points at the onset of illumination. This line of best fit was then extrapolated to time zero, to determine the ground fluorescence (Fo). The maximum fluorescence (Fm) was obtained from the same light intensity when the primary electron acceptors (Qa, plastoquinone) from photosystem II (PSII) became fully reduced. The variable fluorescence (Fv) was calculated by subtracting Fo from Fm. The Fv/Fm ratios, which represent the maximum quantum yield of PSII (Butler 1978, Russel et al. 1995), were calculated automatically by the Handy-PEA, as well as the area over the fluorescence curve between Fo and Fm (Area), which indicates the pool size of Qa on the reducing side of PSII. The integrated energy needed to close all of the reaction centers was also calculated, as S_m = Area / (Fm − Fo), which takes into account multiple turnover in the closure of the reaction centers (Strasser et al. 2000).

To quantify the irreversible and reversible photoinhibition for the data measurements for the hottest periods during summer 2003, the total photoinhibition assessed on basal and medial leaves from the primary shoot of the Montepulciano and Sangiovese vines was separated into chronic and dynamic components (Osmond 1994, Werner et al. 2002). The chronic photoinhibition represents the long-term sustainable decrease in Fv/Fm and was calculated as:

where (Fv/Fm)_max represents the maximum value recorded for mature leaves, early in the morning and before heat stress (mid-June). The (Fv/Fm)_predawn values were measured at 0700 hr.

The dynamic photoinhibition represents the diurnal decline in Fv/Fm that was fully reversible overnight and was calculated as:

where (Fv/Fm)_midday was measured at 1300 to 1400 hr.

In the 2003 and 2004 seasons, the physiological traits were recorded for the same leaves used for the chlorophyll fluorescence measurements. Net photosynthesis (Pn) and stomatal conductance (g_s) were measured by exposing 10 basal and 10 medial leaves of each cultivar to saturating light, as light with a photosynthetic photon flux density >1,800 μmol photon m⁻² s⁻¹. An LCA-4 portable, open-system, infrared gas analyzer (Analytical Development, Hoddesdon, Herts, UK) and a Parkinson-type leaf chamber were used. The intrinsic water-use efficiency (WUEi, μmol CO₂ mol⁻¹ H₂O) was calculated as the Pn/g_s ratio.

For both years, the leaves used for the gas-exchange measurements were collected to determine their relative water content (RWC). The RWC was measured on 10 leaf discs, as (FW-DW)/(TW-DW) × 100, where FW is the fresh weight, DW the dry weight after three days of drying at 80°C, and TW the turgid weight obtained after 24 hr in distilled water at 4°C in the dark. The specific leaf weight (mg DW/cm²) was also calculated, together with the water content/dry mass ratio (g/g).

In 2003, the leaves used for gas-exchange measurements were also used to determine the chlorophyll a and b and carotenoid contents, with eight leaf discs (1.76 cm² each) per treatment, according to a published method (Lichtenthaler and Wellburn 1983). Leaf reflectance, transmittance, and absorbance were measured on 20 leaves per treatment, between 1300 and 1400 hr, using a proper tripod and an LI-190S Quantum Sensor (LI-COR, Lincoln, NE) (Schultz 1996b). The stomata density and sizes were measured using impressions taken by painting clear nail polish on the abaxial surface of 10 basal Sangiovese and Montepulciano leaves collected on 19 Aug. After drying at room temperature, the film was removed from the leaf surface and mounted on a glass microscope slide. The stomata were counted under a binocular microscope on stomata impressions taken from each leaf. The mean stomatal size was calculated from stomata length × width × 0.785 (the standard formula for an ellipse).

Leaf orientation was calculated by measuring the angles between the main vein of the leaves relative to the horizontal (0°) using a compass protractor. The measurements were taken between 1300 and 1400 hr on 200 leaves for each leaf type and cultivar selected on both sides of the canopy.

During the third week of August in both 2003 and 2004, the percentages of basal (from the base of the shoot to 10th node) and medial (from the 11th node to the apex of the shoot) leaves affected by chlorosis and necrosis were determined.

Growth and replenishing of carbohydrate reserves.

At the end of February in 2003 and 2004, canes from 20 vines for each cultivar were pruned and weighed to estimate the annual vine growth. In March 2004, three weeks before budbreak, the soluble sugars and starch concentrations in stems, permanent cordons, and roots (fine and medium brown roots, 1.5 ± 0.2 and 3.4 ± 0.3 mm mean diam, respectively) were determined on six replicates. Soluble sugars were extracted from 500 mg dry samples with ethanol:water (80:20). The soluble solids in the extract were determined by the colorimetric method using the anthrone reagent (Merck, Darmstadt, Germany), at 620 nm on a LKB-Ultrospec III spectrophotometer (Pharmacia, Uppsala, Sweden), according to Morris (1948), and as modified by Loewus (1952). To break down the starch, the remaining pellets were rehydrated with water, boiled for 15 min, and incubated for 24 hr at 27°C with α-amylase and amylo-glucosidase in a sodium acetate buffer at pH 5.0. The resulting soluble carbohydrates were determined with the anthrone reagent, as above. A calibration curve was constructed immediately prior to the measurements, using 0.0, 0.25, 0.50, 0.75, 1.00, 1.25, and 1.50 mg/mL glucose solutions (1 mL).

Statistical analysis.

Three-way analysis of variance (ANOVA) was used to examine year, cultivar, and leaf position effects on chlorophyll fluorescence and gas-exchange parameters. One-way ANOVA was applied to the data for the morphostructural characteristics of sun-exposed leaves from the basal and medial portions of primary shoots and for the replenishing of carbohydrate reserves. The Student-Newman-Keuls test was used to compare adjacent means at a level of 0.05, using SigmaStat 3.5 (Systat Software, San Jose, CA).

Environmental conditions.

Unlike 2004, the 2003 season was very hot and dry, with a rainfall of only 92 mm from budbreak to grape harvest, compared with 255 mm during the same period in 2004 and with the 42-year average of 353 mm from April to September (2003 rainfall was thus 64% and 74% lower, respectively) (Figure 1⇓). In addition, the 2003 average temperatures were higher compared with 2004 and the 42-year average, especially from May to August. In August 2003, the maximum air temperature at midday (1300–1400 hr) was 40.4°C, compared with 36.2°C in August 2004.

• Download figure
• Open in new tab

Figure 1

Monthly total rainfall and mean air temperature at the experimental site during 2003 and 2004 and mean values over 42 years (1959–2001).

Chlorophyll fluorescence and gas exchange.

In general, in June and July the three variables of year, cultivar, and leaf position had no significant effects on the Fv/Fm ratio (photoinhibition, as the maximum quantum yield of PSII), net photosynthesis (Pn), and stomatal conductance (g_s) (Table 1⇓). In July, there were significant effects of leaf position on Pn and g_s (with the medial leaves, which in general were more efficient compared to the basal leaves). However, on 20 Aug, after two weeks of drought conditions, year, cultivar, and leaf position and all relative interactions showed significant effects for the variability of Fv/Fm, Pn, and g_s.

During the summers of 2003 and 2004, although with the exception of 20 Aug in 2003, for both cultivars and leaf type, there were no indications of permanent photoinhibition, as seen by the midday Fv/Fm values that were always close to 0.8 (Figure 2⇓). During the hottest hours of 20 Aug 2003, which were characterized by clear skies (1,900 μmol photon m⁻² s⁻¹, VPD of 3.2 kPa at midday, and total radiation of 22.8 MJ m² d⁻¹), there was a significant down-regulation of PSII activity in the Sangiovese basal leaves, which showed an ~33% decrease in the Fv/Fm ratio (Figure 2A⇓). There was only a moderate decrease in the Fv/Fm ratio for the Montepulciano basal leaves on 20 Aug 2003. The medial leaves of both cultivars performed similarly, showing only slight reductions in the Fv/Fm ratio compared with the prior measurements (Figure 2C⇓), although the environmental conditions were particularly favorable for the onset of photoinhibition.

• Download figure
• Open in new tab

Figure 2

Seasonal evolution of maximum quantum yield (Fv/Fm) measured at midday (1300–1400 hr) in light-exposed leaves located in basal (A and B) (4th–5th nodes from base) and medial (C and D) (13th–15th nodes from base) portions of Montepulciano and Sangiovese shoots in 2003 and 2004. For 20 Aug 2003 and 18 Aug 2004, morning Fv/Fm (measured between 0600 and 0700 hr, solar time) and leaf relative water contents (values in parentheses) are given. Data are mean ± SE (n = 10).

As expected, midday RWC measured on 20 Aug 2003 was reduced in comparison to the morning RWC of both cultivars, irrespective of leaf position along the shoot (Figure 2A, C⇑). The extent of the RWC drop from the morning to midday was ~8% in basal leaves of both cultivars. Of note, when compared with the RWC of Montepulciano basal leaves at midday (86.1%), there was only a minor reduction in Sangiovese leaves (<3.5%). Thus, even for the basal leaves that show considerable photoinhibition, leaf water status was not the primary cause of the reduction in PSII activity, as has been reported previously for the other genotypes of Vitis vinifera L. (Schultz 1996a) and Vitis californica Benth (Gamon and Pearcy 1990). Also, for both cultivars and years, the RWC of the basal and medial leaves measured at midday during June, July, and the first week in August was consistently greater than 88% (data not shown), indicating good water availability in the soil and the absence of any manifest environmental stress.

The Fv/Fm decrease in Sangiovese basal leaves as compared with Montepulciano leaves was due to an increase in Fo (+16%) and a decrease in Fm (−22%) (Table 2⇓). There were also significant reductions in Fv (−41%) and Area (−57%) for Sangiovese basal leaves. Furthermore, a high value was determined for the integrated energy needed to close all of the reaction centers, the S_m parameter, for Montepulciano basal leaves (+38% over Sangiovese), indicating the high amount of energy required (Table 2⇓).

Leaves located in the medial portions of Sangiovese shoots did not differ from the homologous Montepulciano leaves in regard to the fluorescence parameters (Table 2⇑). In comparison with basal leaves, these young sun-exposed leaves of both cultivars had high Fv/Fm values during the hottest hours of the sunny days (Figure 2C, D⇑) and were never affected by chronic photoinhibition.

In contrast, Sangiovese basal leaves clearly showed higher levels of chronic and dynamic photoinhibition than Montepulciano leaves (Figure 3⇓). Sangiovese was more prone to a reduction in transpiration loss through leaf abscission, since chronic photoinhibition is associated with the permanent loss of function in the PSII reaction centers and leads to the irreversible destruction of the photosynthetic pigment in the antenna system, followed by chlorosis and necrosis (Long et al. 1994). Contrary to the 2004 season and the Montepulciano leaves, during the third week of August 2003, ~57% of basal leaves and only 5% of medial leaves on the primary shoots of Sangiovese vines were chlorotic (discolored and yellowed) and necrotic (Table 3⇓). This occurred predominantly on leaves situated on the western side of the row, which was exposed to direct solar radiation during the afternoon, suggesting that direct exposure of these leaves to sunlight triggered leaf chlorosis followed by necrosis, with irreversible damage to the photosynthetic apparatus. Such conditions were not seen for Montepulciano grapevines. The Sangiovese leaves bordering on irreversible photoinhibition were the older leaves, located in the cluster zone and temporarily exposed to direct sunlight.

• Download figure
• Open in new tab

Figure 3

Chronic and dynamic photoinhibition in basal (A) and medial (B) light-exposed leaves of Montepulciano and Sangiovese cultivars measured on a clear, hot day in August 2003. Chronic photoinhibition was calculated as the reduction (%) in predawn Fv/Fm relative to the seasonal maximum Fv/Fm, whereas dynamic photoinhibition represents the additional diurnal reduction of Fv/Fm (fully reversible overnight, therefore a flexible adjustment).

As recorded in 2004 (data not shown), in the 2003 season, the Pn and g_s measured at midday during June, July, and the first week of August were similar for both cultivars (Figure 4⇓). During the hottest hours of 20 Aug 2003, Sangiovese had significantly higher Pn (+56% and +173% for basal and medial leaves, respectively) and g_s (+26% and +56% for basal and medial leaves, respectively) than the Montepulciano (Figure 4⇓). At the same time, the Sangiovese medial leaves showed significantly higher values of WUEi (+75%) compared with the homologous Montepulciano leaves (Figure 5⇓). In June, July and the first weeks of August, no significant differences were seen in WUEi, neither between Sangiovese and Montepulciano, nor between basal and medial leaves (Figure 5⇓).

• Download figure
• Open in new tab

Figure 4

Seasonal evolution of midday net photosynthesis (Pn) and stomatal conductance (g_s) in light-exposed leaves located in basal (A and B) (4th– 5th nodes from base) and medial (C and D) (13th–15th nodes from base) portions of Montepulciano and Sangiovese shoots in 2003. For 20 Aug 2003, morning (measured between 0800 and 0900 hr, solar time) Pn and g_s values are also given. Data are mean ± SE (n = 10).

• Download figure
• Open in new tab

Figure 5

Seasonal evolution of midday intrinsic water use efficiency (WUEi, calculated as the Pn/g_s ratio) in light-exposed leaves located in the basal (A) (4th–5th nodes nodes from base) and medial (B) (13th–15th nodes from base) portions of Montepulciano and Sangiovese shoots. Data are mean ± SE (n = 10).

Leaf characteristics.

In general, medial leaves had significantly higher photosynthetic pigment content than basal leaves, whereas chlorophyll a/b and chlorophyll/carotenoid ratios did not change (Table 3⇑, pooled means). Irrespective of their position, Montepulciano leaves had significantly higher concentrations of chlorophyll a and b and of carotenoid than Sangiovese leaves (Table 3⇑). No significant differences in chlorophyll a/b and chlorophyll/carotenoid ratios were seen between the cultivars. A high level of photosynthetic pigment was detected in Montepulciano leaves, compared with data published on other grapevine cultivars (Palliotti et al. 2000, Smithyman et al. 2001). Large levels of photosynthetic pigment in Montepulciano leaves might, on the one hand, enhance light absorption, and on the other, increase energy dissipation by the chlorophyll fluorescence processes.

Again irrespective of their position, Montepulciano leaves had significantly higher specific weights than Sangiovese leaves (Table 3⇑), perhaps due to a greater lamina thickness and/or to greater solute accumulation. The higher water content/dry weight ratio of Sangiovese leaves (+23% and +18% in basal and medial leaves, respectively) might indicate an increased osmotic adjustment. Several studies have, indeed, reported good correlations between the water content/dry weight ratio and changes in osmotic adjustment (Sobrado and Turner 1983, Turner et al. 1987).

The two cultivars showed different optical characteristics for the leaves (Table 2⇑), with the exception of reflectance. Montepulciano leaves showed higher absorptance (+6.2% and +5.1% for basal and medial leaves, respectively) and a lower transmittance (−4.5% for both leaf types). The leaf absorptance and reflectance values for both cultivars were comparable to those previously reported for grapevines (Smart 1987, Schultz 1996b). The absorption of incident radiant energy in the wavelength range 400 to 700 nm was enhanced by an increase in the total chlorophyll concentrations of the leaves. As the total chlorophyll concentrations increased, leaf transmittance decreased significantly (r² = 0.91), with an exponential trend (Figure 6⇓).

• Download figure
• Open in new tab

Figure 6

Leaf transmittance in leaves of Vitis vinifera L. cvs. Montepulciano and Sangiovese as a function of total chlorophyll concentration. The nonlinear regression of leaf transmittance versus total chlorophyll concentration is represented by: y = 0.60 + 53.31 e^{(−0.0441 x)}; R² = 0.931.

Sangiovese basal leaves had a lower stomatal density (−16%) and size (−14%) than Montepulciano leaves (Table 3⇑).

Contrary to 2004, during the hottest hours of the sunny days in August 2003, a significantly more vertical position with increased lamina inclination was observed for both basal and medial leaves on Sangiovese vines, with an average angle between the leaf axis and the horizon of ~84° to 89°, as opposed to the average range of 64° to 71° for Montepulciano leaves (Table 3⇑). This behavior gives a more closely woven canopy (erectophile vegetation) and reduces the amount of direct irradiation on the adaxial surface of Sangiovese leaves. To decrease light absorption, leaf lamina orientation may change during drought (Gamon and Pearcy 1989, Flexas et al. 1998). This movement has been attributed to leaf wilting and not to pulvin tissue (Flexas et al. 1998), as found in other species (Koller 1990).

Growth and replenishing of carbohydrate reserves.

As determined by the weight of one-year-old wood removed by winter pruning, in both years vine vigor was similar for Montepulciano and Sangiovese (809 and 827 g/vine in 2003, respectively, and 565 and 611 g/vine in 2004, respectively). In 2004, three weeks before budbreak, starch concentration measured in aboveground vine organs and in roots was greater in Sangiovese vines than in Montepulciano vines (Table 4⇓). Compared with Montepulciano vines, Sangiovese vines had starch concentrations that were 18% higher in the canes, 41% higher in the permanent cordons, and 42% higher in the fine brown roots. Sangiovese fine brown roots had a 49% higher soluble sugar concentration than Montepulciano roots.