The analysis of stable carbon isotopes in plants is a powerful tool that allows for obtaining information on ecophysiological processes (Gessler et al., 2014). It has been proven that by analysing the isotopic ratios between ¹²C and ¹³C in grape must, it is possible to obtain a continuous integrator of the plant water status throughout the ripening period (Gaudillère et al., 2002). This technique is based on the discrimination produced during the CO₂ fixation process (the Calvin–Benson cycle) by the enzyme ribulose 1,5-diphosphate. This enzyme has a higher affinity for ¹²C, the most abundant stable carbon isotope in the atmosphere (Brillante et al., 2018). When the stomata close and gas exchange with the outside of the leaf and the substomatal chamber ceases, discrimination in ¹³C fixation with respect to ¹²C decreases. This is because the ¹³C/¹²C ratio (δ¹³C) increases within the substomatal chamber and photosynthetic products become enriched in ¹³C (Farquhar et al., 1980). Previous studies have concluded that water deficit is the main factor affecting this isotopic ratio (Brillante et al., 2018). Currently, in Vitis vinifera L., there are numerous studies on the δ¹³C mainly carried out on berry, must and wine samples with various objectives. Among them, the evaluation of the effects of irrigation and plant water status, the identification of the isotopic signal as a function of the content of sugars and organic acids or isotopic characterisations of the cultivar are commonly found in the literature (Brillante et al., 2018; Costinel et al., 2011; Damiano et al., 2019; Gaudillère et al., 2002; Gómez‐Alonso and García‐Romero, 2009; Santesteban et al., 2010; Tardaguila et al., 1997). The studies on the aforementioned sample types have the limitation that they are bounded to the year in which the berry or wine was developed.

The study of stable carbon isotope ratios in growth rings is widely used to understand the incidence of environmental conditions on the physiology and development of woody plants (Harlow et al., 2006; Schleser et al., 1999). For this reason, the analysis of δ¹³C contained in growth rings (dendrochemistry) is recognised as a valuable environmental proxy, reflecting climatic signals on a local, regional, and even hemispheric scales (Verheyden et al., 2005; Andreu et al., 2008). Therefore, unlike analysis in berries, must or wine, the isotopic analysis of δ¹³C contained in growth rings would provide information on the impact of climate on plant development and growth throughout its life. In the case of climbing plants, Santesteban et al. (2010), Tyminski (2013), De Micco et al. (2018), and Roig-Puscama et al. (2021) have demonstrated the potential of dendrochronology to analyse various aspects of plant physiology, including average productivity, berry sugar production, vine health, rootstock interaction and relations with the edaphic environment, among others. Therefore, δ¹³C measurements can be expected to provide metric and historical data on ecophysiological aspects of vineyards, which can be useful for generating management strategies and predictive models of wine yield and quality in the context of climate variability and change processes.

Isotopic analysis in wood is nowadays predominantly based on measurements of α-cellulose material, a component that is dominant in the cell wall. The use of α-cellulose is based on the fact that certain non-structural components, such as resins, organic acids, sugars or gums, can affect the isotopic value of wood (Harlow et al., 2006; Taylor et al., 2008; Wilson and Grinsted, 1977). Therefore, α-cellulose extraction is proposed as a standardised way to perform isotopic analysis on wood (Loader et al., 2003). However, due to the long time required for the α-cellulose extraction and the cost involved, this methodological step is often a limiting factor when many samples must be analysed (D'Alessandro et al., 2004; Loader et al., 2003; Verheyden et al., 2005). In addition, α-cellulose extraction can be more difficult when samples are small (< 1 g), which can occur ingrapevine due to their small annual radial growth (Tyminski, 2013).

Studies on different woody plant species have shown that there is a strong positive correlation between the isotopic δ¹³C signal of bulk wood and α-cellulose (Borella et al., 1998; Guerrieri et al., 2017; Harlow et al., 2006; Loader et al., 2003; Taylor et al., 2008), suggesting that isolation of α-cellulose may not be necessary for δ¹³C analysis. These relationships should always be corroborated for each plant species (Ma et al., 2011).

At present, no indicators have been developed on the isotopic (δ¹³C) ratio between bulk wood and α-cellulose material in Vitis vinifera. In the scarce antecedents that currently exist on isotopic analyses in grapevine wood (Brook et al., 2020), the studies were carried out directly on whole wood without having made a previous comparison with α-cellulose. This could lead to errors in the calculation of derived physiological indicators since the presence of gums or resins could produce an important bias. Therefore, it is necessary to consider whether the isotopic variability between bulk wood and cellulose remains highly correlated and constant throughout the life of the plant. This information would facilitate more consistent ecophysiological interpretations and the understanding of how the climate influences the growth and development of the vine, crucial aspects in the management of these plants in the face of the challenges posed by climate change. The main goal of this contribution is to compare the values of δ¹³C both in bulk wood and α-cellulose of Vitis vinifera.

Samples of 3 cultivars from two different vineyards were used. One of the vineyards, made up of 10-year-old Malbec and Bonarda cultivars, is located in Mendoza, Argentina (33°46’20.29’’ S; 69°9’14.62’’ W; 1100 m altitude). The other vineyard is located in the region of Burgundy, France (47°16’44.45’’ N; 4°58’52.44’’ E; 295 m altitude), composed of 6-year-old Chardonnay cultivar plants. During the 2019 dormant season, wood samples were obtained from the main trunks at 0.4 m height: in the Argentine vineyard by using a 6” Pressler increment borer of 5.15 mm in diameter, and in the French vineyard, by cross-cutting sections of the main stems. From each plant/individual, 3 replicates were extracted and pooled into a single sample.

Once the annual growth rings were anatomically identified, they were separated with a cutting knife under a binocular microscope. The rings of the different samples were then pooled according to their corresponding calendar age, resulting in a set of composite samples, each one including the rings formed in the same year for all plants (Ma et al., 2011). For the French vineyard samples, the growth rings corresponding to the 2018 growing season were sectioned, while for the Argentine vineyards, those rings corresponding to the 2017–2018 austral growing season were sectioned. Subsequently, the wood samples were dried for 24 hours at 60 °C and ground with a mixer mill (MM 400 Retsch®, Germany).

For each ring sample, a portion of the wood material was kept as bulk wood, while the rest was used for α-cellulose extraction. To extract α-cellulose, the methodology of Brendel et al. (2000) for small samples (≥100 mg) was followed. The procedure consisted of weighing 100 mg of ground wood in 10 mL Pyrex™ tubes. Subsequently, under a fume hood, acid digestion of each sample was performed with acetic and nitric acid in a thermostatic bath at 120 °C for 20 min. Once cooled, ethanol was added, and the samples were centrifuged. The supernatant was then removed, and the pellets were washed sequentially with ethanol, deionised water, and acetone.

For isotopic analyses, 1 mg of bulk wood and 1 mg α-cellulose of each sample were weighted. Three replications of each sample were used. Carbon isotopic analyses were performed on a Vario Micro Cube elemental analyser coupled in a continuous flow mode to an isotope ratio mass spectrometer (IsoPrime, Elementar). USGS40 L-Glutamic acid (δ¹³C_VPDB = -26.39 ± 0.04 ‰) and IAEA-600 Caffeine (δ¹³C_VPDB = -27.77 ± 0.04 ‰) were used as internal standards. The δ¹³C values were reported in parts per thousand (‰) relative to the Vienna Pee Dee Belemnite (VPDB) international reference.

In grapevine wood, α-cellulose represents 44.3 % (± 0.012 %) with respect to bulk wood (w/w). The δ¹³C values of bulk wood and α-cellulose are shown in Table 1. The mean δ¹³C_{bulk wood} value was -27.35 ± 0.11 ‰, while the δ¹³C_α-cellulose value was slightly higher, reaching -26.49 ± 0.11 ‰. The δ¹³C_α-cellulose values differed significantly with respect to the δ¹³C_{bulk wood} values. The mean δ¹³C value of the difference between the two types of samples was +0.86 ‰. Because the analysed samples came from a pool of different cultivars and from different geographical locations, the isotopic values and the variation patterns were different. However, the δ¹³C_{bulk wood} and δ¹³C_α-cellulose values showed a significant linear relationship (δ¹³C_α-cellulose = 0.78 (± 0.10) * δ¹³C_{bulk wood} -5.24 (± 2.73); r² = 0.86; Figure 1), where Student’s t-test indicated that the slope of the LLSR is not significantly different from one (p = 0.05244). The regression confidence interval was 0.56 (lower limit; 95 %) and 1.00 (upper limit; 95 %).

Table

titre du tableau
	δ13C (‰)
	Average	Standard deviation	Max	Min
Bulk wood	-27.35	0.41	-26.33	-27.85
α-cellulose	-26.49	0.35	-25.74	-26.93

Table

Each values represent a single measurement. The dashed line corresponds to the least square regression (δ¹³C_α-cellulose = 0.78 (± 0.10) * δ¹³C_{bulk wood} - 5.24 (± 2.73); p = 0.05244; n = 12; r² = 0.86) and the solid line represent the one-to-one line.

This research was possible thanks to the “Bourse de Doctorat Chaire Unesco « Culture et Traditions du Vin » de l’Université de Bourgogne, and supported by Biogeosciences Laboratory funding and Bodega Familia Zuccardi (La Agrícola S.A.). The authors wish to express their gratitude for the technical assistance to Anne-Lise Santoni (GISMO platform, Biogéosciences, University Bourgogne Franche-Comté), Laura Principiano, Martín Di Stefano, Sebastian Zuccardi (La Agrícola S.A.) and Marcelo Alberto (Facultad de Ciencias Agrarias UNCuyo, Mendoza, Argentina).