WANG Yan-xiu, HU Ya, CHEN Bai-hong, ZHU Yan-fang, Mohammed Mujitaba Dawuda,, Sofkova Svetla

1 College of Horticulture, Gansu Agricultural University, Lanzhou 730070, P.R.China

2 Deparment of Horticulture, Faculty of Agriculture, University for Development Studies, Tamale P.O. BoxTL 1882, Ghanas

3 Institute of Agriculture and Environment, Massey University, Palmerston North 4442, New Zealand

1. Introduction

Rootstock is one of the most important factors influencing tree vigor, yield and fruit size in tree crops (Mariniet al.2006). In fruit tree industries, dwarfing rootstocks have been widely applied for several decades all over the world, to intensive orchards to restrict tree volume and promote earlier flowering (Lockard and Schneider 1981; Webster 2004). In the western countries, the use of apple dwarfing rootstocks has accounted for more than 80% of the total orchard area and resulted in great achievements in terms of yield, quality and mechanization management. However, there are many problems with the application of dwarfing rootstocks in the apple production in China. Some main factors limiting the development of dwarfing cultivation in China are: utilization of mixed rootstocks, poor adaptability of the rootstocks and lack of systematic study on regionalization of dwarfing rootstocks(Shao 2015). As environment and climate diversity in China,it is becoming increasingly difficult to grow the dwarfing rootstocks adaptable to a specific local climate in other areas with different environmental and climatic conditions (Denget al.2012). Among the various abiotic stresses such as drought, heat, ultraviolet light and air pollution, cold stress is one of the most crucial limitations to growth, affecting many genetic, physiological, and biochemical responses in plants (Soydamet al.2013). Cold is a typical environmental stress factor that limits the geographical distribution and growth of various plants, and affects crop quality and productivity (Chenet al.2014). Under low temperature stress, changes in the relative gene expression activated the antioxidant system and accumulation of osmotic adjustment substances (Cyrilet al.2002). The freezing tolerance was associated with acclimation processes, such as cellular osmotic stabilization, antioxidant production, modifications in hormone metabolism (Lucau-Danilaet al.2012). Numerous studies have demonstrated that the destruction of the cell membrane system is the major mechanism of plant cold injury(Pomeroyet al.1985; Shaoet al.2013). In the apple growing districts in northern China, development of intensive apple orchards with dense tree planting is greatly limited by frequent occurrence of freezing injuries which result from the improper application of dwarfing rootstocks with poor cold resistance.Therefore, it is of a great significance to determine suitable cold resistance indices and establish an evaluation system for cold resistance selection of proper apple rootstocks with good cold resistance as well as providing support for their application in the regional apple cultivation.

It has been reported that the relative electrical conductivity(REC), malondialdehyde (MDA), proline and starch contents can be used as the indices of cold resistance identification(Wanget al.2013; Zhanget al.2015). In many published reports, semi-lethal temperature (LT50) was applied to determine the cold resistance of plants (Armstronget al.2015; Peixotobet al.2015). Meiet al.(2008) concluded that the REC, MDA content and superoxide dismutase (SOD)activity could serve as three indices to comprehensively evaluate the early cold resistance of rootstocks under laboratory conditions. Yinet al.(2011) supposed that the soluble protein content and POD activity would increase under low temperature stress, regulating the plant cold hardiness. Zhuet al.(1986) used the Logistic function to obtain the inflection point temperature to estimate the LT50of plant tissues, which could act as the quantity index of cold resistance. Different tree species and varieties vary in their low temperature adaptability, therefore it is difficult to use a single physiological or morphological indicator to evaluate the level of the cold hardiness. The multivariate statistical methods are commonly used in multi-index comprehensive evaluation (Zhanget al.2016). Therefore,by using principal component analysis to comprehensively analyze the various indicators related to cold hardiness,the differences in cold resistance among varieties can be compared. It is significant to establish a comprehensive evaluation system of cold hardiness and to guide regional application for apple dwarfing rootstocks.

In this study, the principal component analysis was used to comprehensively evaluate the REC and corresponding physiological indices after low temperature treatments. On the basis of introducing and collecting the main dwarfing apple rootstocks in northern China, this study aimed to establish a set of suitable methods for evaluating cold resistance of dwarfing apple rootstocks and select the rootstocks characterized by high level cold resistance.

2. Materials and methods

2.1. Plant materials

During mid-December 2015, one-year branches with uniform size were collected from two-year healthily growing apple trees in an open field after defoliation. The information of materials is presented in Table 1 (Liet al.1994; Gao and Bai 2010).

Table 1 Characteristics of the ten apple rootstocks employed in the study

2.2. Low temperature treatment

The low temperature treatment of apple stem segment was applied as described by Meiet al.(2008). The one-year-old rootstocks were first cut into 20-cm long cuttings from the basal fourth bud and the wounds at both ends closed with paraffin and then the sticks were washed thoroughly with distilled water. Each rootstock was divided into 6 groups of 12 cuttings in each group. The cuttings were wrapped with clean gauze, kept in plastic bags and placed in the ultra low temperature refrigerator to carry out artificial freezing treatment. The temperatures were set at: –15, –20, –25,–30, –35°C, and 0°C as control. After the temperature reached the set temperature and maintained for 12 h, one group was taken out from each rootstock. Each group after being taken out of the refrigerator was stood for 8 h at 0°C to thaw and placed at room temperature for 12 h for indices determination. The determinations were repeated for 3 times.

2.3. Measurement of indices

REC was determined by DDS-307 conductivity (LEICI,China). REC was measured by placing 1.5-g branch pieces in a flask containing 30 mL of deionized water. The initial electrical conductivity (EC0) was the electrical conductivity of deionized water. Samples were immersed in deionized water for 3.5 h at room temperature for determination of the second round of electrical conductivity (EC1). Samples were then boiled in water-bath for 20 min and cooled to room temperature to measure the final electrical conductivity(EC2). Then REC was determined as follows:

Soluble protein content was analyzed by the Coomassie brilliant method according to Zhaoet al.(2002).

Soluble sugar content was quantified by the method of Zhaoet al.(2002). A total of 0.5-g samples were ground in mortars with 4 mL of 80% ethanol, and the mixture was centrifuged. The supernatant was supplemented by 0.5 g of acticarbon, and kept in a water-bath at 80°C for 30 min and then filtered. Then, 1 mL aliquots of the extract were supplemented by 5 mL of anthrone. The mixture was boiled for 10 min and cooled. Absorbance was monitored at 630 nm.

Proline content was determined by the method of Meloniet al.(2004). Sulfosalicylic acid and acidic ninhydrin were used to dissolve proline and stain, respectively. Absorbance was determined at 520 nm.

MDA content was extracted and detected using the thiobarbituric acid assay according to Zhaoet al.(2002).Absorbance of supernatant was monitored at 600, 532 and 450 nm.

Anthocyanin content was measured by methanol hydrochloric acid colorimetry method of Liuet al.(2008).

Soluble starch content was quantified by anthrone sulfuric acid method according to Zhaoet al.(2002).

SOD activity was assayed spectrophotometrically based on the classical method of Giannopolitis and Ries (1977)by measuring the inhibition of the photochemical reduction of nitroblue tetrazolium.

POD activity was assayed at 470 nm using the reaction substrates of hydrogen peroxide and guaiacol (Gao, 2006).The OD470was recorded every minute. One unit of enzyme activity is equal to OD470increase 0.01 per min.

2.4. Statistical analyses

Zhuet al.(1986) proposed that the treatment temperature and REC were fit with Logistic function to obtain the inflection point temperature of the curve which can be used as the LT50of tissue. REC fitting logistic regression function is:Y=K/(1+ae–bX), LT50=–lna/b, where,Yrepresents the injury rate of the cells (in the calculation, the equidistant three points were selected in the measured value);Kmeans the asymptote of the curve or the maximum rate of electrolyte leakage,Xis the processing temperature, and a and b are the function parameters.

All values reported are means of three independent replicates. Mean analysis was assessed by one-way ANOVA with SPSS Software (SPSS version 22.0 Inc., USA).Mean differences were established by Duncan’s test.

3. Results

3.1. Changes in REC and LT50

Fig. 1 Relative electric conductivities in 10 apple rootstocks under cold stress conditions from 0 to –35°C. Data are presented as means±SE (n=3).

Changes in RECChanges in REC of different rootstocks exposed to different low temperatures are shown in Fig. 1:The REC of all rootstocks continuously increased with decreasing temperature. At –15 and 0°C, REC rapidly increased in all rootstocks, particularly in M9, JM7, and T337. At –15–(–25)°C, the increasing trend of REC in most of the rootstocks slowed down. However, JM7, T337, M7,and M26 increased sharply at the same temperature. At–25 and –30°C, all the increases in REC were relatively similar. At –25 and –30°C, the REC of GM256 increased rapidly and SC1 and SH38 flattened.

The logistic function and LT50The data on the Logistic functions and LT50s of the 10 apple rootstocks are shown in Table 2. The results showed that all functions’ fitting degrees were greater than 0.9 and reached an extremely significant level. The three species which attained LT50s above –25°C were M9, T337 and JM7. LT50s of SH1, SC1,M7 and M26 occurred between –25 and –30°C; the species which attained LT50at –30 to –40°C were SH6 and SH38.LT50of GM256 was attained at a much lower temperature of –41.6°C compared with the other rootstocks .

3.2. Changes in osmotic substances

Changes in anthocyanin contentAs shown in Fig. 2-A,when the temperature was above –20°C, the anthocyanin content of all rootstocks increased. When the temperature reached –20°C, the anthocyanin content decreased withthe decreases in temperature. The anthocyanin content of GM256 was consistently higher than those of the other rootstocks, while that of M9 was the least. GM256 had the most quickly rising trend which increased to 112.63%. M9 had the least increase rate of only 26.28%. Increase rates of the others were between 36.01 and 103.38%.

Table 2 Logistic function and semi-lethal temperature (LT50)

Fig. 2 Anthocyanin (A), protein (B), soluble sugar (C), soluble starch (D), proline (E) and malondialdehyde (F) contents in 10 apple rootstocks under cold stress conditions from 0 to –35°C. FW, fresh weight. Data are presented as means±SE (n=3).

Changes in protein contentWith decreasing temperature,the protein content in the different rootstocks changed differently (Fig. 2-B). Based on the changes in protein content, these rootstocks can be divided into three categories. The protein content of GM256 gradually increased. The protein contents of SH38 and SH6 reached their peaks of 2.76 and 2.54 μg g–1, respectively, at –25°C.The other rootstocks attained the highest protein content at–20°C and decreased gradually at –20–(–35)°C.

Changes in soluble sugar contentFrom Fig. 2–C,the soluble sugar content increased with the decreasing temperature. The sharp increases of soluble sugar in SH1, SC1, and M26 occurred at 0–(–25)°C and the upward tendency leveled off at below –25°C. Although similar to SH1, SC1, and M26 rootstocks, the soluble sugar content of T337 increased slowly. The soluble sugar contents of GM256 and SH6 continuously increased. In the cooling process, the soluble sugar contents of M9 and SH38 decreased at –30 and –35°C, while that of JM7 and M7 reduced at –25 and –30°C.

Changes in soluble starch contentFrom Fig. 2-D, the starch content in the rootstocks increased slightly at 0 and–15°C and sharply at –15 and –20°C, but reached the peak at –20°C. The starch contents decreased rapidly at –20 and–25°C and changed slightly at –25–(–35)°C. The increases in starch content of all the rootstocks were quite different.The starch contents in GM256, SH38, SH6 increased by 427, 313, and 301%, respectively. Those in SH1, SC1, M7,M26, JM7, and T337 changed between 224 and 284%. The least increase was 195% which occurred in M9.

Changes in proline contentIn Fig. 2-E, the proline contents of JM7, M7, M9, M26 and T337 showed a similar trend. Initially, proline increased and then decreased at–30°C. The proline contents in SH38, SH6, SH1 and SC1 increased slightly and then flattened sharply at –30 and–35°C. In GM256, proline consistently increased with in temperature decreasing.

Changes in MDA contentIn Fig. 2-F, the MDA contents of the 10 rootstocks reached the highest value at –20°C and then decreased. In SC1 and M9, MDA changed slightly at–30 and –35°C. Both SH1 and SH38 showed a plateau at 0 and –15°C, and SH6 showed a similar trend at –25 and–30°C. The MDA content of GM256 increased slightly at 0 and –15°C, soared into 41.83 μmol g–1at –15 and –20°C,dropped sharply at –25 and –30°C and then declined. The trends in JM7 and T337 were consistent similar to that of GM256. The MDA content of M9 was higher than other rootstocks at –20°C.

3.3. Changes in antioxidant enzymes activity

Changes in SOD activityWith decreasing temperature,SOD activities of 10 rootstocks increased and then declined(Fig. 3-A). The peaks occurred at –30°C, suggesting that the elevation of SOD activity may be a response to cold stress. GM256 exhibited higher SOD activity than the other rootstocks. In contrast, the SOD activity of M9 was relatively poor and only reached 63.23 U g–1at –30°C.

Fig. 3 Superoxide dismutase (A) and peroxidase (B) activities in 10 apple rootstocks under cold stress conditions from 0 to –35°C.Data are presented as means±SE (n=3).

Changes in POD activityPOD activities of all the rootstocks exhibited increasing trends under different low temperatures stress (Fig. 3-B). In the rise and fall trend,the POD activity of GM256 reached the maximum (230.74 U g–1min–1) at –35°C. The M9 showed a different trend compared with the other rootstocks. It decreased at –30°C and its highest POD activity only reached 130.73 U g–1min–1.The increases in POD activities of T337, JM7, M26 and M7 were in a rise and fall pattern.

3.4. The correlation analysis between LT50 and physiological indices

With the exception of POD, other indices were correlated with LT50(Table 3). The results showed that LT50was positively correlated with MDA and negatively correlated with other indices. LT50had a significant positive correlation with MDA at 0 and –15°C and had significantly negative correlation with soluble sugar at –20°C, with protein at–25–(–35)°C, with proline at –20, –30, and –35°C, with anthocyanin at –15°C and –25–(–35)°C, and with SOD at–15 and –35°C. Moreover, LT50and MDA at –20°C were extremely positive, whiles it correlated extremely negative with soluble sugar at –25–(–35)°C, starch at –15°C, protein at –20°C, anthocyanin at –20°C, proline at –25°C and SOD at –25°C.

3.5. Principal component analysis

In order to determine the relationship between LT50on one hand and the physiological indices and cold resistance on the other, the principal component analysis was used to filter out the major factor group. Based on the principal component analysis, when the cumulative variance contribution rate was greater than 85%, the information could be used to reflect the variation of the system. As can be seen from Table 4, the first, second, and third principal component variance contribution rates reached 62.204,16.269, and 12.082%, respectively. Notably, the cumulative variance contribution rate was 90.555% (more than 85%)without missing variables. Therefore, the first three principal components can reflect completely the different information of the cold resistance system and most of the data had already been included in the three principal components.

The information in the first principal component mainly describing the cold resistance indices accounted for 62.204% of all the information. All the indices contributing to the first principal component had a larger absolute value of the feature vector and showed much positive load. The second principal component mainly included proline, soluble sugar and anthocyanin which had a larger absolute value of the feature vector, and the contribution rate was 16.269%.Especially, proline had a larger negative load than the other indices. As for the third main component, both protein and proline had greater eigenvectors up to 0.52 and 0.49,respectively (Table 5).

Table 6 showed that the principal component scores which directly evaluated the cold resistance. The final scores were derived from the three principal components together. The rootstock with a higher score had stronger cold resistance.Therefore, the order of the evaluating scores from high to low was: GM256>SH6>SH38>SH1>SC1>M26>M7>JM7>T337>M9.

4. Discussion

In this study, the principal component analysis was used to extract three principal components which total contribution rate was 90.555%. Analysis of the values of the three principal component factors showed that the top three rootstocks of the first principal component were GM256,SH6 and SH38 which had low LT50. This indicated thatthese rootstocks may be more sensitive to oxidative stressviaantioxidant enzyme activity regulation. In the second principal component, the first three rootstocks with the highest score were GM256, SH6 and SC1. It suggested that the three rootstocks performed well in terms of the accumulation of osmotic adjustment substances under adverse stress. In the third principal component, the three rootstocks with the highest score were SH6, M26 and M7, which performed well in the production of protein and proline. It was possible to speculate that SH6, M26 and M7 would generate more stress proteins in response to low temperature stress.

Table 3 Correlation between semi-lethal temperature (LT50) and physiological indices1)

Table 4 Total variance explained

Antioxidant enzymes can protect plants from damage by removing reactive oxygen species (ROS). Under low temperature stress, the increase of ROS can cause membrane lipid peroxidation if protective enzymes couldn’t eliminate more ROS. Accumulation of MDA, a membrane lipid peroxidation product, can lead to the increase of membrane permeability, electrolyte leakage which results in higher conductivity value and damage of cell integrity. The increase of cell membrane permeability causes effusion of intracellular soluble materials. The contents of soluble protein, soluble sugar, soluble starch and free amino acid increase the cell concentration, enhance water holding capacity of the cells, decrease the freezing point and improve the protection ability of protoplast.

Under cold stress, the plasma membrane permeability increases leading to electrolyte leakage. The cell plasma membrane permeability of species with stronger cold resistance changes less (Niyazet al.2014). REC is a physiological symptom that could measure whether the insoluble substance diffuses outside or the cytoplasmic membrane is damaged (Chenet al.2014). The determination of REC is an effective method to identify plant cold resistance.LT50has been widely used to reflect the suboptimal temperature limit of plants response to cold (Armstronget al.2015; Peixotoet al.2015). In this study, our results for LT50values of GM256, M26, SH38 and SH6 were very similar to those of Yinet al.(2011), Gaoet al.(2000) and Yanget al.(2011) with the difference between our and their LT50estimations of no more than 4°C. However, LT50value of M9 was very different. Two factors that may have increased our LT50compared to other research results may be environmental differences and/or the genotypes. The LT50s of the other rootstocks have not yet been reported.

Increase of the anthocyanin content in shoots, under coldstress, is one of the adaptive responses of the cultivated apple tree. Generally, the anthocyanin accumulation in apple tree shoots and the degree of cold resistance are related.From our results, as the stress duration extended the content of anthocyanin increased firstly and then decreased,which were similar to the changes in starch content. Zhanget al.(2013) suggested that excess carbohydrate may be the proximate trigger for the induction of anthocyanin biosynthesis. This interpretation is correspond with the results of our experiment. Where the soluble sugar content of all the rootstocks changed at different degrees, they all exhibited a tendency to increase. Soluble sugar is an important osmolyte to plant cells, which can increase the intracellular solute content. The accumulation of soluble sugar in low temperature stress can improve cell osmotic pressure, so it can enhance the ability of water retention.Soluble sugar also has a cryoprotectant effect, which can reduce the freezing point of the cell solution, buffering the cytoplasm excessive dehydration to prevent protein from freezing and thawing, thus reducing the chilling temperature to the cells (Wang 1987). Penget al.(2015) found that at suboptimal temperature, starch cloud hydrolyse into sugar. This suggests that the early accumulation of polysaccharides can be used to improve the soluble sugar content in late winter. The decrease of starch content may be hydrolyzed into soluble sugar to increase the content of osmotic substances in the cell.

Table 5 Tested feature vector analysis of resistant rootstocks

Table 6 Factor scores of different apple rootstocks1)

Soluble protein with high water absorption capacity can be helpful in increasing the bound water in the cell, lowering the freezing point, and reducing the chance of the freezing injury in protoplast (Gaoet al.2007). Wang (1987) indicated that the soluble protein was hydrophilic substance that could significantly enhance the protection of cells. Walliset al.(1997) found a significant positive correlation between the content of soluble protein and cold resistance. The soluble protein contents of apple seedlings in this study exhibited an upward trend or it remained relatively stable after the increase, which was again conducive to cold stress.

Proline is an ideal organic osmotic adjustment substance as a result of its high solubility, non-toxic and high water potential of aqueous solution (Delauneyet al.1993). The protective effect of proline on the protein may be due to the formation of hydrophilic colloid in aqueous solution, which can produce a hydrophobic framework interacting with the protein (Chadalavadaet al.1994). Proline can stimulate the activityies of catalase (CAT), SOD and polyphenol oxidase (PPO) in plants (Matysiket al.2002). Role of proline in improving cold resistance of plants is reflected in the above aspects. Free proline content increases in many plants under low temperature, drought, and high salinity stress (Hareet al.1998). Our results showed that the proline contents of all the rootstocks increased in different degrees during the cold temperature treatment, which was beneficial to alleviating the damage caused by suboptimal temperature.

One of the reasons for the destruction of plants subjected to low temperature stress is lipid peroxidation of unsaturated fatty acids in membrane induced by free radicals in cells (Ndony 1997). MDA is the final product of membrane lipid peroxidation, which reflects the degree of plant damage (Velikovaet al.2002). Rootstock M9 had relatively more MDA accumulation amount, so its cold resistance was interpreted as poor while GM256 had less MDA accumulation amount, and thus its cold resistance was interpreted as strong.

Antioxidant enzymes including SOD, POD, and CAT remove excessive free radicals, maintaining the normal physiological activities of plants, to a certain extent (Gechevet al.2003). SOD can eliminate superoxide anion radical and catalyze the reaction between superoxide radical and hydrogen ions, producing H2O2and O2, while the increase of POD activity can effectively remove the H2O2(Wang 1982). Becket al.(1998) believed that there are two kinds of changing patterns in the activity of SOD enzyme. One is that SOD activity gradually decreases with the increase of temperature, and the second is that the enzyme activity increases until it reached the critical temperature, and decreases rapidly afterwards. In this experiment, SOD enzyme activity of apple rootstocks increased firstly and then decreased which fell into the second pattern. The increase trend of POD activity appeared at –20°C, but SOD activity increased at the beginning of the cold stress treatment. The reason for the phenomenon might be that the H2O2produced by antioxygenation of SOD causes POD activity to increase.

5. Conclusion

The principal component analysis was successfully used in the study of cold resistance of 10 elite apple rootstocks.And the rootstocks’ cold resistant were identified and the order from high to low was: GM256, SH6, SH38, SH1, SC1,M26, M7, JM7, T337, and M9.

Nine indices, including LT50(X1), soluble sugar (X2),soluble starch (X3), protein (X4), proline (X5), anthocyanin(X6), MDA (X7), POD (X8), and SOD (X9), were used to establish a mathematical model of apple cold resistant evaluation:

The LT50and comprehensive evaluation results can provide the basis for the selection of apple rootstocks in different climatic regions of China. For the regions where the minimum temperature was higher than –20°C, all rootstocks could be used for apple production. At this point, other indices affecting fruit quality rather than cold resistance should be considered. For instance, in the areas where the climate is colder, it would be better to select GM256, SH38 and SH6 as dwarfing rootstocks.

Acknowledgements

This work was supported by the Fundamental Research Funds for the Universities of Gansu Province, China (035-041051) and the Natural Science Foundation of Gansu Province, China (145RJZA167).

Armstrong J J, Takebayashi N, Sformo T, Wolf D E. 2015. Cold tolerance inArabidopsis kamchatica.American Journal of Botany, 102, 439–448.

Beck S L, Dunlop R, Van S J. 1998. Rejuvenation and micropropagation of adultAcacia mearnsiiusing coppice material.Plant Growth Regulation, 26, 149–153.

Bing H E, Chen Q B, Pan Y Z, Liu Y G. 2004. Research on physiology characteristic performance of severalEuphorbia pulcherrimawilld cultivars under low temperature stress.Journal of Sichuan Agricultural University, 22, 331–332.(in Chinese)

Chen L J, Xiang H Z, Miao Y, Zhang L, Guo Z F, Zhao X H, Lin J W, Li T L. 2014. An overview of cold resistance in plants.Journal of Agronomy and Crop Science, 200, 237–245.

Cyril J, Powell G L, Duncan R R, Baird W V. 2002. Changes in membrane polar lipid fatty acids of seashore paspalum in response to low temperature exposure.Crop Science,42, 2031–2037.

Delauney A J, Verma D P S. 1993. Proline biosynthesis and osmoregulation in plants.The Plant Journal, 4, 215–223.

Gao A, Jiang S, Zhao X, Deng J, Sha S, Liu Z, Zhang M.2000. Study on the hardiness of apple cultivars.Journal of Fruitence, 17, 17–21. (in Chinese)

Gao D, Guo J, Wei Z, Fan Q. 2012. Evaluation of productivity and light quality in two high density dwarf rootstock apple orchards in central China.Scientia Agricultura Sinica, 13,1848–1853. (in Chinese)

Gao J F. 2006. Experimental guidance for plant physiology.Higher Education Press, Beijing. pp. 217–219. (in Chinese)

Gao Y, Bai H X. 2010. Characteristics and application of apple dwarf rootstock M9 lines.Northwest Horticulture, 6, 34–35.(in Chinese)

Gao Y, Qi X H, Yang J H, Zhang M F. 2007. The response mechanism of cold stress in higher plants.Northern Horticulture, 10, 58–61. (in Chinese)

Gechev T, Willekens H, Van Montagu M, Inzé D, Van Camp W I M, Toneva V, Minkov I. 2003. Different responses of tobacco antioxidant enzymes to light and chilling stress.Journal of Plant Physiology, 160, 509–515.

Giannopolitis C N, Ries S K.1977. Superoxide dismutases.I. Occurrence in higher plants.The Plant Physiology, 59,309–314.

Hare P D, Cress W A, Van Staden J. 1998. Dissecting the roles of osmolyte accumulation during stress.Plant Cell and Environment, 21, 535–553.

Li D K, Shao K J, Zhang Z R. 1994. Main characters and utilization techniques of SH lines of apple dwarf rootstocks.Northern Fruits, 1, 8–10. (in Chinese)

Liu R D, Zhang M, Li X X. 2008. Comparisons of extraction solvents and quantitative-methods for analysis of anthocyanins in strawberry and blueberry fruits.Acta Horticulturae Sinica, 35, 655–660. (in Chinese)

Lochard R G, Schneider G W. 2011. Stock and scion growth relationships and the dwarfing mechanism in apple.Horticultural Review, 3, 315–375.

Lucau-Danila A, Toitot C, Goulas E, Blervacq A S, Hot D,Bahrman N, Sellier H, Lejeune H I, Delbreil B. 2012.Transcriptome analysis in pea allows to distinguish chilling and acclimation mechanisms.Plant Physiology and Biochemistry, 58, 236–244.

Marini R P, Anderson J L, Autio W R, Barritt B H, Cline J, Wpjr C, Crassweller R C, Garner R M, Gauss A, Godin R. 2006.Performance of ‘Gala’ apple trees on 18 dwarfing rootstocks:Ten-year summary of the 1994 NC-140 rootstock trial.Journal of the American Pomological Society, 60, 69–83.

Matysik J, Alia Bhalu B, Mohanty P. 2002. Molecular mechanisms of quenching of reactive oxygen species by proline under stress in plants.Current Science, 82,525–532.

Mei L X, Jiang B, Zhao Z Y, Liang D Q. 2008. Comparison of Cold hardiness indices of apple dwarf rootstocks determined by different methods.Acta Agriculturae Boreali-occidentalis Sinica, 17, 103–106. (in Chinese)

Meloni D A, Gulotta M R, Martínez C A, Oliva M A. 2004. The effects of salt stress on growth, nitrate reduction and proline and glycinebetaine accumulation inProsopis alba.Brazilian Journal of Plant Physiology, 16, 39–46.

Ndong C, Ouellet F, Houde M, Sarhan F. 1997. Gene expression during cold acclimation in strawberry.Plant and Cell Physiology, 38, 863–870.

Niyaz A, Ablat P, Dong S L, Du R Q, Zhang S K, Xu L, Li W H. 2014. Comparative study on the cold resistance of four cherry cultivars under chilling stress.Agricultural Science and Technology, 15, 1919–1922.

Peixoto M M, Friesen P C, Sage R F. 2015. Winter coldtolerance thresholds in field-grownMiscanthushybrid rhizomes.Journal of Experimental Botany, 66, 4415–4425.

Peng X, Teng L, Yan X, Zhao M, Shen S. 2014. The cold responsive mechanism of the paper mulberry: Decreased photosynthesis capacity and increased starch accumulation.BMC Genomics, 16, 1–20.

Pomeroy M K, Andrews C J. 1985. Effect of low temperature and calcium on survival and membrane properties of isolated winter wheat cells.The Plant Physiology, 78, 484–488.

Rajendrakumar C S V, Reddy B V B, Reddy A R. 1994. Prolineprotein interactions: Protection of structural and functional integrity of M4 lactate dehydrogenase.Biochemical and Biophysical Research Communications, 201, 957–963.

Shao L Q. 2015. Techno-economic evaluation on high-density dwarfing cultivation pattern of apple in China. Ph D thesis,Northwest A&F University, China. (in Chinese)

Shao Y R, Xu J X, Xue L, Zhang R, Wu C Q. 2013. Effects of low temperature stress on physiological-biochemical indexes and photosynthetic characteristics of seedlings of four plant species.Acta Ecologica Sinica, 33, 4237–4247. (in Chinese)Soydam A S, Büyük I, Aras S. 2012. Relationships among lipid peroxidation, SOD enzyme activity, and SOD gene expression profile inLycopersicum esculentumL. exposed to cold stress.Genetics & Molecular Research, 12,3220–3229.

Velikova V, Yordanov I, Edreva A. 2000. Oxidative stress and some antioxidant systems in acid rain-treated bean plants:Protective role of exogenous polyamines.Plant Science,151, 59–66.

Wallis J G, Wang H, Guerra D J. 1997. Expression of a synthetic antifreeze protein in potato reduces electrolyte release at freezing temperatures.Plant Molecular Biology,35, 323–330.

Wang C. 1982. Physiological and biochemical responses of plants to chilling stress.Hortscience, 17, 173–186.

Wang G., Wang J, Xue X, Chao L U, Nie P. 2013. Research progress and identification method of apple stress resistance.Agricultural Science and Technology, 18,1413–1416.

Wang R F. 1987. The kinds of plant hardiness criteria and their application.Plant Physiology Communications, 3, 49–55.(in Chinese)

Webster A D. 2004. Vigour mechanisms in dwarfing rootstocks for temperate fruit trees.Acta Horticulturae, 658, 29–41.

Yang F Q, Chen D M, Zhao Y B, Zhang X S, Xu J T. 2011.Study on cold hardiness of apple dwarf rootstocks.Journal of Hebei Agricultural Sciences, 15, 8–9. (in Chinese)

Yin L L, Li X Y, Liu X L, Zhu J, Dai H Y. 2011. Study on cold resistance of several apple rootstocks.Journal of Qingdao Agricultural University(Natural Science), 28, 198–200. (in Chinese)

Zhang B Q, Yang L T, Li Y R, Zhang B Q. 2015. Physiological and biochemical characteristics related to cold resistance in sugarcane.Sugar Technolgy, 17, 49–58.

Zhang G, Bai J, Xi M, Zhao Q, Lu Q, Jia J. 2016. Soil quality assessment of coastal wetlands in the Yellow River Delta of China based on the minimum data set.Ecological Indicators,66, 458–466.

Zhang K M, Li Z, Li Y, Li Y H, Kong D Z, Wu R H. 2013.Carbohydrate accumulation may be the proximate trigger of anthocyanin biosynthesis under autumn conditions inBegonia semperflorens.Plant Biology, 15, 991–1000.

Zhao S J, Shi G A, Dong X C. 2002.Techniques of Plant Physiological Experiment.China Agriculture Science and Technology Press, Beijing. pp. 84–85. (in Chinese)

Zhu G, Liu Z, Zhu P. 1986. A study on determination of lethal temperature with logistic function.Journal of Nanjing Agricultural University, 3, 11–16. (in Chinese)