Identification and Regionalization of Cold Resistance of Wine Grape Germplasms (V. vinifera)

With the extreme changes of the global climate, winter freezing injury has become an important limiting factor for the development of the global grape industry. Therefore, there is a significant need for the screening of cold-resistant wine grape germplasms and cold regionalization for cold-resistant breeding and the development of grapevine cultivation in cold regions. In this study, the low-temperature half-lethal temperature (LT50) values were determined for the annual dormant branches of 124 wine grape germplasms (V. vinifera) to evaluate their cold resistance. The LT50 values of the 124 tested germplasms ranged from −22.01 °C to −13.18 °C, with six cold-resistant germplasms below −20 °C. Based on the LT50 values, the 124 germplasms were clustered into four types, with cold resistance from strong to weak in the order of type Ⅱ > type Ⅰ > type Ⅳ > type Ⅲ, corresponding to the four cold hardiness zones. Zones 1, 2, 3, and 4 included 6, 22, 68, and 28 germplasms, respectively, with decreasing cold resistance. The number of germplasms in different hardiness zones followed a normal distribution, with the most in zone 3. In Type Ⅱ, the fruit skin color of germplasms was positively correlated with cold hardiness, while the temperature of origin was negatively correlated with cold hardiness. The average LT50 of germplasms in different origin regions ranged from −17.44 °C to −16.26 °C, with differences among some regions. The cold regionalization analysis resulted in the distribution of 124 germplasms in four temperature regions in China with six germplasms in region A (−22 °C ≤ LT50 ≤ −20 °C), 30 germplasms in region B (−20°C ≤ LT50 ≤ −18°C), 71 germplasms in region C (−18 °C ≤ LT50 ≤ −15 °C), and 17 germplasms in region D (−15 °C ≤ LT50 ≤ −13 °C). Strong cold-resistant wine grape germplasms (V. vinifera) were identified, and these could be used as parental material for cold-resistant breeding. In some areas in China, soil-burial over-wintering strategies are used, but our results suggest that some wine grapes could be cultivated without requiring winter burial during overwintering. The results of this study should provide guidance for the selection of promising strains for cold-resistant breeding for expanded cultivation of improved varieties for wine grape production in China.


Introduction
Mean climate change and increased extreme events (i.e., days with Temperature max > 30-35 • C, or days with Temperature min < 0 • C) results in colder winters, especially in

Low Temperature Treatment and LT 50 Determination
The collected branches were rinsed with tap water and then deionized water, three times for each, and then the water was adsorbed using filter paper. Each sample was divided into six equal parts and placed in a heat test box with controlled temperature and humidity (Model: YSGJS-408, Shanghai lanhao instrument & equipment Co., Ltd., Shanghai, China) for low temperature treatment. The groups were as follows: 4 • C (control group, no low temperature treatment), −10 • C, −14 • C, −18 • C, −22 • C, and −26 • C, for a total of six temperature gradient treatments. For each condition, the temperature was decreased to the treatment temperature at a rate of 4 • C/h for 12 h, and then heated to 4 • C at a rate of 4 • C/h. At the conclusion of the treatment, the samples were removed and placed at room temperature for 4 h prior to testing.
Since the buds are greatly affected by the full state and the position of the internodes, we choose the more stable internodes of the branches as the identification material for cold resistance. Remove the epidermis of the processed branches, avoiding the bud eyes, select the stems and cut them into 1~2 mm slices and mix evenly. Samples weighing 2 g were transferred into a 25 mL test tube with a stopper, and then 20 mL deionized water was added and shaken well. This was done three times for each treatment. After shaking for 12 h in a shaker, a conductivity meter (DDS-11C, Shanghai optical instrument factory, Shanghai, China) was used to determine the initial conductivity value. Each test tube was then boiled for 40 min, allowed to stand for 2 h, and then the final conductivity value was determined. The relative conductivity was calculated as follows. Relative conductivity (%) = (initial conductivity value/final conductivity value) × 100 (1) The semi-lethal temperature (LT 50 ) value was calculated using the logistic regression equation: Y is the relative conductivity, x is the processing temperature, and K is the maximum leakage (K = 100).
In practical application: that is, the relative conductivity Y is converted into Y , and the relationship between it and the processing temperature is expressed by linearity. The parameter a and b of the equation were obtained by linear regression. The inflection point temperature is the LT 50 .

Clustering of Cold Resistance of Wine Grape Germplasms (V. vinifera)
According to the measured LT 50 values, the cold resistance of the 124 germplasms was subjected to cluster analysis. The climate temperature of the areas where the germplasms originated were then correlated with the LT 50 values of germplasms in different types. Correlation of LT 50 values with the fruit skin color of different germplasms (yellow-green, light-red, crimson, and black) was also performed.

Zoning of Cold Resistance of Wine Grape Germplasms (V. vinifera)
According to the results of cluster analysis, germplasms were divided into cold hardiness zones. The LT 50 values were organized from low to high, and the temperature range and the distribution of different germplasms were determined for each zone. Average LT 50 values were calculated for germplasms from different origins.

Cold Regionalization of Wine Grape Germplasms (V. vinifera) in China
The daily minimum temperatures of surface meteorological elements for 30 consecutive years (1982-2011) provided by the National Climate Center at 2294 weather stations across the country ( Figure 1) were used in this study. The temperature range of each cold hardiness zone was rounded to the nearest tenth degree to determine the standards for the cold regionalization of wine grape germplasms (V. vinifera) in China. Based on the climatic regionalization, cold regionalization of wine grape was next carried out using the daily minimum temperature of surface meteorological elements in each region and the critical low temperature occurring at least three times in 30 years as the applied index. The specific distribution of 124 germplasms in cold regionalization were determined according to LT 50 .

Data Analysis
Origin 9.0 (OriginLab, Northampton, MA, USA) software was used to fit the logistic equation, and LT 50 values were obtained. Descriptive statistics were analyzed via SPSS 22.0 (Statistical Product and Service Solutions, Inc., Chicago, IL, USA). Values were presented as the mean ± standard deviation using triplicate measurements. Cluster analysis and graphing were carried out using R language software and complete-linkage clustering. ArcGIS 10.2 (Environmental Systems Research Institute, Inc., California, CA, USA) software spatial analysis module was used to carry out spatial interpolation processing and to construct the cold regionalization map.

Data Analysis
Origin 9.0 (OriginLab, Northampton, MA, USA) software was used to fit the logistic equation, and LT50 values were obtained. Descriptive statistics were analyzed via SPSS 22.0 (Statistical Product and Service Solutions, Inc., Chicago, IL, USA). Values were presented as the mean ± standard deviation using triplicate measurements. Cluster analysis and graphing were carried out using R language software and complete-linkage clustering. ArcGIS 10.2 (Environmental Systems Research Institute, Inc., California, CA, USA) software spatial analysis module was used to carry out spatial interpolation processing and to construct the cold regionalization map.

LT50 Values
LT50 values were obtained by fitting the relative electrical conductivity of grapevine branches under different temperature stresses with the corresponding stress temperatures. As shown in Table 2, the LT50 values of the 124 tested germplasms ranged from

LT 50 Values
LT 50 values were obtained by fitting the relative electrical conductivity of grapevine branches under different temperature stresses with the corresponding stress temperatures. As shown in Table 2

ClusterAanalysis of LT 50 Values
Cluster analysis was conducted on the determined LT 50 values, and the results are shown in Figure 2. The germplasms were divided into four types. Type-I included 22 germplasms, with the lowest LT 50 −19.95 • C and the highest LT 50 −18.46 • C. Of these germplasms, 9 germplasms had grapes with black skin, 10 germplasms had grapes with yellow-green skin, 1 germplasm had grapes with light-red skin, and 2 germplasms had grapes with crimson skin. Among these germplasms, 2 germplasms originated from Germany, 12 germplasms originated from the Soviet Union, 4 germplasms originated from France, 1 germplasm originated from Spain, 1 germplasm originated from China, and 3 germplasms originated from Italy. Type II included six germplasms, ranging in LT 50 value from −22.01 • C to −20.46 • C. Among these germplasms, 5 germplasms had grapes with yellow-green skin and 1 germplasm had grapes with black skin. A total of four originated from the Soviet Union, one germplasm originated from Spain, and one germplasm originated from China. Type III included 28 germplasms, ranging in LT 50 value from −15.51 • C to −13.18 • C. Of these, 11 germplasms had grapes with black skin, 16 germplasms had grapes with yellow-green skin, and 1 germplasm had light-red skin color. A total of 3 of these germplasms originated from Germany, 10 germplasms originated from the Soviet Union, 10 germplasms originated from France, 2 germplasms originated from Spain, 1 germplasm originated from China, and 2 germplasms originated from Italy. Type IV included 68 germplasms, with LT 50 values ranging from −18.29 • C to −15.53 • C. Among these germplasms, 26 germplasms had grapes with black skin, 31 germplasms had grapes with yellow-green skin, five germplasms had grapes with light red skin, and six germplasms had grapes with crimson skin. For these germplasms, 8 germplasms originated from Germany, 29 germplasms originated from the Soviet Union, 23 germplasms originated from France, 1 germplasm originated from Spain, 2 germplasms originated from China, 3 germplasms originated from Italy, 1 germplasm originated from Romania, and 1 germplasm originated from Switzerland. The cold resistance based on LT 50 value was from strong to weak as follows: type II > type I > type IV > type III, with temperature ranges of type II: The correlation between LT 50 and temperature of germplasm origin and fruit skin color was further explored, and the results are shown in Table 3. The temperature of germplasm origin was positively correlated with LT 50 for the different types, with correlation coefficients of 0.580, 0.224, 0.280, and 0.343, respectively, for Type I, II, III, and Type IV; this correlation was significant for Type IV. The fruit skin color was positively correlated with LT 50 value in Type I and Type IV, with correlation coefficients of 0.249 and 0.336, respectively, reaching an extremely significant level for Type IV. The fruit skin color was negatively correlated with LT 50 in Type II and Type III, with correlation coefficients of −0.591 and −0.099, respectively.  Table 3 were tested by Student's test, * p < 0.05 and ** p < 0.01 represent significant differences between treatments.
Agriculture 2021, 11, 1117 9 of 16 Agriculture 2021, 11, x FOR PEER REVIEW 9 of 17 Figure 2. Clustering results of cold hardiness identification in 124 germplasms. Layer a corresponds to germplasm origin information, layer b corresponds to germplasm fruit skin color information, and layer c corresponds to LT50 value. In the information of fruit skin color, color is qualitative rather than quantitative to some extent. Wine grape fruit skin color is mainly defined according to the type of wine. According to the color of the wine can be roughly divided into yellow-green, light-red, crimson, and black, corresponding to white wine, rose wine, red wine, and toning or purplish wine.
The correlation between LT50 and temperature of germplasm origin and fruit skin color was further explored, and the results are shown in Table 3. The temperature of germplasm origin was positively correlated with LT50 for the different types, with correlation coefficients of 0.580, 0.224, 0.280, and 0.343, respectively, for Type I, II, III, and Type IV; this correlation was significant for Type IV. The fruit skin color was positively correlated with LT50 value in Type I and Type IV, with correlation coefficients of 0.249 and 0.336, respectively, reaching an extremely significant level for Type IV. The fruit skin color was negatively correlated with LT50 in Type II and Type Ⅲ, with correlation coefficients of −0.591 and −0.099, respectively. Clustering results of cold hardiness identification in 124 germplasms. Layer a corresponds to germplasm origin information, layer b corresponds to germplasm fruit skin color information, and layer c corresponds to LT 50 value. In the information of fruit skin color, color is qualitative rather than quantitative to some extent. Wine grape fruit skin color is mainly defined according to the type of wine. According to the color of the wine can be roughly divided into yellow-green, light-red, crimson, and black, corresponding to white wine, rose wine, red wine, and toning or purplish wine.

Hardiness Zones of Wine Grape Germplasms for Different Origin Areas
Based on the cluster analysis combined with the cold resistance, the 124 germplasms were divided into four hardiness zones. From zone 1 to zone 4, the cold resistance decreased successively: hardiness zone 1: −22.01-20.46 • C; hardiness zone 2: −19.95-18.46 • C; hardiness zone 3: −18.29-15.53 • C; and hardiness zone 4: −15.51-13.18 • C. These correspond to type II, type I, type IV, and type III in the cluster analysis. Table 4 shows the distribution of 124 germplasms in the four hardiness zones. The mean LT 50 values of germplasms from different origin regions ranged from −17.44 • C to −16.26 • C, with no difference between the mean LT 50 values of germplasms from France and Italy or those from Germany, the Soviet Union, Spain, and China. Histograms of the distribution of germplasms from different regions in different hardiness zones were generated, as shown in Table 4. The results showed that the cold-tolerance distributions of the tested germplasms originating in Germany, France, the Soviet Union, and Italy all showed a normal distribution, with most germplasms from different regions clustered in zone 3, and discretely distributed in zones 1, 2, and zone 4. With fewer germplasm samples from Spain, China, Romania, and Switzerland, frequency histograms were not drawn. germplasms from different regions in different hardiness zones were generated, as shown in Table 4. The results showed that the cold-tolerance distributions of the tested germplasms originating in Germany, France, the Soviet Union, and Italy all showed a normal distribution, with most germplasms from different regions clustered in zone 3, and discretely distributed in zones 1, 2, and zone 4. With fewer germplasm samples from Spain, China, Romania, and Switzerland, frequency histograms were not drawn.  from Germany, the Soviet Union, Spain, and China. Histograms of the distribution of germplasms from different regions in different hardiness zones were generated, as shown in Table 4. The results showed that the cold-tolerance distributions of the tested germplasms originating in Germany, France, the Soviet Union, and Italy all showed a normal distribution, with most germplasms from different regions clustered in zone 3, and discretely distributed in zones 1, 2, and zone 4. With fewer germplasm samples from Spain, China, Romania, and Switzerland, frequency histograms were not drawn.  from Germany, the Soviet Union, Spain, and China. Histograms of the distribution of germplasms from different regions in different hardiness zones were generated, as shown in Table 4. The results showed that the cold-tolerance distributions of the tested germplasms originating in Germany, France, the Soviet Union, and Italy all showed a normal distribution, with most germplasms from different regions clustered in zone 3, and discretely distributed in zones 1, 2, and zone 4. With fewer germplasm samples from Spain, China, Romania, and Switzerland, frequency histograms were not drawn.   Zone 3 1 / The data based on three replicates are represented as mean ± standard deviation. Data were tested by one-way ANOVA, and means followed by the same letter do not differ. Significance analysis and frequency histogram were not performed for quantities ≤5. The data based on three replicates are represented as mean ± standard deviation. Data were tested by one-way ANOVA, and means followed by the same letter do not differ. Significance analysis and frequency histogram were not performed for quantities ≤5.

Cold Regionalization of Wine Grape Germplasms (V. vinifera) in China
Based on the climatic regionalization of wine grape in China and the clustering results, the critical low temperatures for cold regionalization of V. vinifera were selected as −13 • C, −15 • C, −18 • C, −20 • C, and −22 • C. Using the daily minimum temperature measured at 2294 meteorological stations for 30 consecutive years  in China, the critical low temperature occurring at least three times in 30 years was used as the dividing line to regionalize the cold resistance of V. vinifera, as shown in Figure 3.  Finally, based on the LT50 values of the 124 germplasms, the distribution of each variety in the region was determined. The results are shown in Table 5

Evaluation of Cold Resistance of Wine Grape Germplasms (V. vinifera)
The LT 50 value is used as an indicator of plant stress injury, and has been widely applied in grape cold resistance identification [63,64]. The lower the LT 50 value, the stronger the cold resistance [65]. With the most economically valuable cultivars in the world, V. vinifera varieties exhibit good drought resistance, but poor resistance to disease and cold [6,11]. Although there is variation in cold resistance among different germplasms of V. vinifera, reported differences are not considered significant [17]. However, there are some varieties with strong cold resistance [47,66]. For example, Ecolly, a V. vinifera variety, allows burial-free cultivation overwintering in some areas in China where soil-burial practices are required for over-wintering of grapevines [47]. In this work, 124 wine grape germplasms (V. vinifera) were tested for cold resistance by conductivity method. The LT 50 values of Gordan, Sateni, Crimean Cornish ♀, Ecolly, Spitak, and Petroximegne were all lower than −20 • C, indicating that these germplasms exhibit strong cold resistance and could be used as parent materials for cold resistance breeding.
Cold resistance is an important characteristic of fruit trees and is acquired by genetic variation and natural selection during long-term adaptation of different plants to low temperature and cold environment [67]. The cold resistance of grapevine is controlled by the genetic characteristics of the variety [68], grapevine conditions such as age, growth, development stage, and nutrient accumulation [69]; the morphological structure, thickness, and maturity of branches [70]; and environmental conditions such as external light and temperature [66,71]. In this study, the fruit skin color of five of the six germplasms in type II (hardiness zone 1) were yellow-green, while in the other three types, black, yellow-green, light-red and crimson showed discrete distributions. Therefore, there may be a relationship between fruit skin color and cold resistance. Correlation analysis of fruit skin color and LT 50 of germplasms showed that type II had the highest correlation coefficient between LT 50 and fruit skin color and this correlation was negative, where the lighter the fruit skin color, the stronger the cold resistance.
In addition to fruit skin color, grapevine cold resistance is closely related to its origin and geographical distribution area [72]. Previous studies have shown that the cold resistance of germplasm (V. amurensis) is related to the climate temperature in the original distribution area [70]. The lower the temperature in the original area, the stronger the cold resistance of germplasm in this area. Our results showed that the average LT 50 values of germplasms from different origin areas ranged from −17.44 • C to −16.26 • C, though greater variation in LT 50 values might be observed if a large number of germplasms were screened. The analysis revealed that the cold resistance of germplasms from different origins was normally distributed, mainly concentrated in hardiness zone 3 and distributed discretely in zones 1, 2, and 4. Correlation analysis showed that geographical origin exhibited a positive correlation with LT 50 for different types. In Type II, LT 50 and origin of germplasm were most highly correlated, indicating that in germplasms with high cold resistance, the lower the temperature of origin, the stronger the cold resistance might be. Therefore, it may be a feasible strategy to select wine grape germplasms that originated in colder areas for cultivation in cooler parts of China.

Cold Regionalization of Wine Grape Germplasms (V. vinifera) in China
Starting with the previous climatic regionalization of wine grape in China [73] and utilizing the clustering results from this work, the daily minimum temperature data were applied to delineate four regions of cold regionalization for more accurate regionalization. There has been significant work on grape regionalization for wine making [57,58,73]. Wang et al. used the low temperatures of −15 • C and −18 • C that occurred more than three times in a certain area within 30 consecutive years to determine the soil-burial over-wintering line for V. vinifera and V. labrusca grapevines [55]. The critical low temperature line for that cold regionalization strategy is quite consistent with the regionalization strategy used here. Overall, the regionalization results are consistent with the currently used overwintering and cold protection regions of grapevines in China, suggesting this is an effective strategy for the regional mapping of cold resistance.
In some marginal areas in China, soil-burial over-wintering is required, as well as other modifications of cultivation conditions [47,66,74]. In this study, six germplasms correlating to region A (−22 • C ≤ LT 50 ≤ −20 • C) showed strong cold resistance, which should enable these strains to be able to adapt to burial-free cultivation in the Loess Plateau, Jin-Jin-Ji, and Shandong regions of China. The 30 germplasms for region B (−20 • C ≤ LT 50 ≤−18 • C) showed medium cold resistance, which should enable their adaptation to burial-free cultivation in the Loess Plateau and Shandong. Overall, the cold regionalization information can be used to provide reference and cultivation guidance for areas of China that currently utilize soil-burial over-wintering of grape vines.
Annual extreme minimum temperature is the most important determinant of the safe overwintering of grapevines, but other factors can affect grapevine survival. For example, dryness in winter and low temperature frost in late spring can significantly affect the safe overwintering of grapevines. The cold regionalization strategy applied here is solely based on the extreme low temperature, and there are limitations of this approach. In future work, other factors affecting the safe overwintering of grapevines can be considered, and a more comprehensive approach to overwintering regionalization can further guide the optimal selection of new varieties for cultivation in different regions.

Conclusions
In this study, 124 V. vinifera germplasms were screened for cold resistance, allowing their classification into four hardiness zones by cluster analysis. Of particular interest, Gordan, Sateni, Crimean Cornish ♀, and Spitak from Soviet Union, Ecolly from China, and Petroximegne from Spain have high cold resistance are distributed in hardiness zone 1, suggesting these varieties are promising parent materials for cold-resistant breeding. The skin color of grapes and the climate temperature of the areas where the germplasms originated were also related to cold resistance. Based on the cold regionalization of V. vinifera in China, the wine grape germplasms were classified into four regions. This distribution of different germplasms provides reference value for cultivation and introduction of vari-eties into each region. Importantly, germplasms were identified that may be suitable for burial-free cultivation in colder regions that typically require the burial of grapevines for safe overwintering.