Influence of Simulated Acid Snow Stress on Leaf Tissue of Wintering Herbaceous Plants

; Acid snow might be an environmental stress factor for wintering plants since acid precipitates are locally concentrated in snow and the period in which ice crystals are in contact with shoots might be longer than that of acid precipitates in rain. In this study, ‘equilibrium’ and ‘prolonged’ freezing tests with sulfuric acid, which simulate situations of temperature depression and chronic freezing at a subzero temperature with acid precipitate as acid snow stress, respectively, were carried out using leaf segments of cold-acclimated winter wheat. When leaf segments were frozen in the presence of sulfuric acid solution (pH 4.0, 3.0 or 2.0) by equilibrium freezing with ice seeding, the survival rate of leaf samples treated with sulfuric acid solution of pH 2.0 decreased markedly. Leaf samples after supercooling to –4 and –8°C in the presence of sulfuric acid solution (pH 2.0) without ice seeding were less damaged. When leaf samples were subjected to prolonged freezing at –4 and –8°C for 7 d with sulfuric acid (pH 2.0), the survival rates of leaf samples exposed to sulfuric acid decreased more than those of leaf samples treated with water. On the other hand, leaf samples were less damaged by prolonged supercooling at –4 and –8°C for 7 d with sulfuric acid (pH 2.0). The results suggest that an acid condition (pH 2.0) in the process of extracellular freezing and/or thawing promotes freezing injury of wheat leaves.


Introduction
Acid precipitation is an environmental problem that has affected ecosystems, human health and agriculture (Evans 1989, Likens andBormann 1995), especially in Europe, North America and Eastern Asia.Many studies on the effects of acid precipitates on plants have been carried out using ecological and physiological approaches.
Several experimental studies have revealed morphological and physiological changes in herbaceous and woody plants treated with simulated acid rain (SAR) (Evans 1989, Haines andCarlson 1989) or mist (Igawa et al. 1997, Sheppard et al. 1998) containing sulfuric acid, nitric acid or hydrochloric acid as typical acid pollutants.Sensitivity to SAR differs depending on the plant species.Based on visible effects on foliage of woody plants, the sensitivity of woody dicots to SAR is higher than that of conifers (Evans 1989).Periodic treatment with SAR of pH ranging from 3.0 to 4.0 for several months had only a weak effect or no effect on some coniferous trees (Haines and Carlson 1989, Izuta 1998, Izuta et al. 1998).Treatment with SAR of pH 3.0 for 8 weeks had no significant effects on dry matter production in Japanese cedar seedlings, but treatment with SAR of pH 2.0 for 8 weeks induced visible injuries of needles and reduction in dry mass of seedlings (Izuta et al. 1990).SAR has also been postulated as a predisposing factor with respect to frost damage of red spruce (Fowler et al. 1989).An electron microscopic study showed structural changes in chloroplasts and protoplasts in Norway spruce needles treated with SAR during winter (Bäck et al. 1993).
Herbaceous dicot plants seem to be more sensitive than woody plants to SAR of pH 4.0 or lower (Evans 1989).Decreases in yields of various crops have been documented in controlled environment and field experiments at pH levels of 4.0 or lower for a few months.Treatment with SAR of pH 4.5 or lower for 5 weeks caused a reduction in growth and yield of wheat (Singh and Agrawal 1996).Anatomical changes in leaves of beans (Stoyanova 1998), ultrastructural changes in organelles of beans (Stoyanova and Vellikova 1998) and tomatoes (Gabara et al. 2003), and inhibition of photosynthetic activities of cucumbers (Yu et al. 2002) and beans (Vellikova et al. 1999) were observed by single treatment with SAR of pH <3.0.
Almost 30% of the total land surface of the earth is covered by snow in winter.It has been found over the past decade that the pH of meltwater of snow has become more acidic than that of pure water in which saturated carbon dioxide is dissolved (pH 5.6) (Fushimi et al. 2001, Noguchi et al. 2001), indicating precipitation of acid snow during winter.It is generally thought that the advantage of snow cover is to protect wintering plants from winter-related abiotic stresses such as severe freezing temperatures and desiccation (Neuner et al. 1999, Decker et al. 2003).
However, air pollutants, including acid substances, in snow may become winter stress factors.In the melting-refreezing processes of granular snow crystals in snow layers, acid substances may be locally concentrated, especially in the outer and inner parts of granular snow (Bales et al. 1989, Fushimi 1994, Suzuki 2000).Under field conditions, the lowest pH value of acid snow is usually >4.0 (Suzuki 2000, Noguchi et al. 2001).At the early and/or late stages of the snowmelt season, meltwater containing 2-5 times higher concentrations of acid substances than the average concentrations of snow layers may be released from snow layers into soils, streams and lakes (Johannessen andHenriksen 1978, Fushimi et al. 2001).
Shoots of wintering herbaceous plants under the snow cover would be continuously exposed to acid pollutants concentrated in the outer parts of snow crystals in winter.Furthermore, repeated freeze-thaws and being immersed in the acid meltwater in early spring and/or early winter will affect shoots more than roots in agricultural field.Therefore, wintering herbaceous plants covered with acid snow may be subjected to severe stresses compared with wintering herbaceous plants covered with snow in the absence of acid substances.However, there have been few studies on the responses of wintering plants to stresses caused by acid snow.
In this study, acid snow stress was conveniently simulated as freezing stress under acid conditions by two in vitro assay systems, equilibrium freezing under acid conditions (EQFA) and prolonged freezing under acid conditions (PRFA).In this study, we carried out the EQFA that simulate a situation of a temperature depression below subzero temperature with acid snow or acid meltwater, and the PRFA that simulate a situation of chronic freezing at a constant subzero temperature for a long time with acid snow crystals.The effects of acid snow stress by EQFA and PRFA on the viability of leaves of winter wheat (Triticum aestivum L. cv.Chihokukomugi) were investigated.

Simulation of acid snow stress during a temperature depression by EQFA
In order to simulate acid snow stress in vitro, leaf segments were frozen with sulfuric acid solution, instead of water, in two freezing tests, equilibrium freezing (EQF) and prolonged freezing (PRF) (Fig. 1).In the EQFA test, sulfuric acid added before ice seeding remained in the subsequent processes (i.e.freezing, thawing and amino acid leakage) at various concentrations in each step.For example, the pH value of the aqueous solution in the test tube changed from 2.0 just before ice seeding to <4.0 after thawing at 4°C overnight and to about 5.0 after leakage of amino acids in the EQFA test.However, these pH changes after thawing might cause little or no damage to wheat leaves in EQFA (Inada et al. 2005).Also, the ninhydrin reaction for quantification of amino acid contents is suitable for calculation of survival rates of leaf samples even in the presence of sulfuric acid (Inada et al. 2005).Therefore, we used both assay systems to assess the influence of freezing under acid conditions on plant tissues as a simulation of acid snow stress.
In order to estimate the influences of freezing stress caused by acid snow (acid snow stress) on winter wheat leaves during a temperature depression, EQFA tests were carried out at various pH values.Survival rates of leaf samples after EQFA with sulfuric acid solutions of pH 4.0 and 3.0 were similar to the survival rate in pure water (pH 5.6) as a control (Fig. 2A), but the survival rate after EQFA with sulfuric acid solution of pH 2.0 markedly decreased.Differences between the survival rates of leaf samples frozen at -4, -6 and -10°C with water and sulfuric acid (pH 2.0) were about 10, 15 and 40%, respectively.Damage after EQFA (pH 2.0) was enhanced by lowering the treatment temperature.
To determine the effects of sulfuric acid on the viability of leaf samples during the process of amino acid leakage, a modified EQFA test was carried out.A 300 µl aliquot of sulfuric acid solution (pH 2.0) was added to the leaf samples after freeze-thawing with water in the EQF test to the same final concentration of sulfuric acid as that used in a standard EQFA test, and the survival rates were compared with those of leaf samples subjected to the standard EQFA test.The survival rates were hardly affected by the existence of sulfuric acid during the process of amino acid leakage (Fig. 2B).More severe acidification during the process of amino acid leakage after freezethawing in water was examined.The survival rates of leaf samples were slightly decreased during the process of amino acid leakage at pH 1.3 after EQF at -4 and -8°C in water but were hardly affected by amino acid leakage at pH 2.0 under this experimental condition (data not shown).The viability of wheat leaves was not significantly influenced by the existence of sulfuric acid in the process of leakage of intracellular amino acids after EQF.
When leaf samples were supercooled under an acid condition (pH 2.0) without ice seeding, there was little decrease in survival rates after the treatment (Fig. 3).Survival rates of leaf samples supercooled at -12°C in the presence and absence of sulfuric acid were >80 and 90%, respectively, although survival rates of leaf samples frozen at -12°C in the presence and absence of sulfuric acid were <10 and 50%, respectively (Fig. 3).These results indicated that leaf samples of winter wheat  were mainly damaged by acidification rather than by exposure to a subzero temperature in the process of extracellular freezing and/or thawing.
In order to study the effects of sulfate ions and low pH on the viability of leaf samples in the EQFA test, some sulfate salts (Fig. 4A) and acids (Fig. 4B) were added to the samples instead of sulfuric acid before ice seeding.There was little decrease in the survival rates of leaf samples by EQF in the presence of potassium sulfate or sodium sulfate compared with the survival rates of leaf samples in the presence of sulfuric acid (Fig. 4A).On the other hand, the survival rates were mark-edly decreased by EQFA in the presence of nitric acid or hydrochloric acid, even to -4°C, as well as sulfuric acid (Fig. 4B).Therefore, leaf samples of winter wheat were mainly damaged by acidification in the process of freeze-thawing rather than by concentration of sulfate anions.

Simulation of acid snow stress at a constant subzero temperature by PRFA
To study the influence of acid snow stress at a constant subzero temperature on the survival of winter wheat, PRFA tests were carried out using sulfuric acid solution instead of  pure water for prolonged freezing (PRF) (Fig. 1).In these tests, the samples were cooled in the same manner as that used in the EQFA tests and were kept at the desired freezing temperature for 4 weeks.PRFA tests were carried out at -4 and -8°C for 4 weeks under acid conditions of pH 4.0, 3.0 or 2.0 (Fig. 5).The survival rates of leaf samples frozen at -4°C under both water (control) and acid conditions were reduced as a function of the freezing period.Damage was slightly enhanced at pH 4.0 and 3.0 and was significantly enhanced at pH 2.0 (Fig. 5A).In the PRFA test at -8°C, the survival rate under both control and acid conditions markedly decreased within 3 d (Fig. 5B).The survival rate of leaf samples at pH 4.0 was similar to that under the control condition, but the survival rates of leaf samples treated for 7 d at pH 3.0 and 2.0 decreased by 20 and 35%, respectively, compared with the survival rates under control (20%) and acid (pH 4.0) conditions.Decreases in survival rates under the control condition and conditions of acidification except pH 2.0 virtually reached plateaus at about 40% within 4 weeks.The decrease in survival rate under the acid condition of pH 2.0 reached about 10% within 7 d.When the viability of wheat leaves after PRFA tests for 7 d at various freezing temperatures was studied, it was found that the survival rate of wheat leaves at pH 2.0 decreased over the course of the freezing period.The survival rates also decreased as the freezing temperature was lowered (Inada et al. 2005).
Fig. 6 Injury of leaves of cold-acclimated winter wheat caused by extracellular freezing and supercooling under acid conditions.In the PRF test, sulfuric acid solution of pH 2.0 (PRFA) or pure water of pH 5.6 (PRF) was added to the leaf samples before ice seeding.After EQF to -4°C (A) or -8°C (B), samples were kept at the freezing temperatures for 7 d.In prolonged cooling under acid conditions, sulfuric acid solution of pH 2.0 (SCA) or pure water of pH 5.6 (SC) was added to the leaf samples, and prolonged cooling was started without ice seeding to maintain a supercooling state.After cooling to -4°C (A) or -8°C (B), samples were kept at those temperatures for 7 d.Survival rates of leaf samples after treatments were estimated by measuring amino acid leakage.Fig. 7 Influence of the presence of sulfuric acid added just before amino acid leakage on the survival rates of leaves of winter wheat after PRF.After EQFA to -4°C (A) or -8°C (B), samples were kept at the freezing temperatures for 7 d.In the PRFA test, pure water of pH 5.6 or sulfuric acid solution of pH 2.0 was added to the leaf samples before ice seeding, or sulfuric acid solution of pH 2.0 was added just before amino acid leakage (pH 5.6 + pH 2.0).Survival rates of the leaf samples were estimated by measuring amino acid leakage.When tests of prolonged supercooling at pH 2.0 were carried out by the same method as that used for the PRFA tests without ice seeding, the survival rates after prolonged supercooling under acid conditions at -4 and -8°C for 7 d were >90 and 65%, respectively.These survival rates were still higher than survival rates after PRFA (Fig. 6A, B).
In order to determine the effects of sulfuric acid on the viability of leaf samples during the process of amino acid leakage in a PRFA test (pH 2.0) for 7 d, the survival rates were compared with those after a modified PRFA test in which sulfuric acid solution (pH 2.0) was added to the leaf samples after freeze-thawing with water.Damage to leaf samples in PRF tests in the presence of at -4 and -8°C for 7 d was hardly enhanced by addition of sulfuric acid to the leaf samples just before amino acid leakage (Fig. 7A, B).The rates of leaf samples after PRFA were not significantly affected by the existence of sulfuric acid during the process of amino acid leakage as was the case in the EQFA test.

Discussion
There have been many studies on the responses of plants to acid rain or mist, but there have been few studies on the responses of plants to acid snow.Although it seems that direct effects of acid precipitation on plant growth are not serious under natural conditions, unusually severe acidification or combination with other environmental stimuli, such as ozone, showed inhibitory effects on plant growth (Matsumura 2001).Since acid snow is regarded as acid precipitation at subzero temperatures, acid snow consists of two stress components, acidification and freeze-thawing, for wintering plants.Acid substances in snow crystals may be locally concentrated during the processes of recrystallization or freeze-thawing of snow and ice crystals in winter.It is possible that the local pH around acid snow crystals may be much lower than the pH of the meltwater.
In field conditions, wintering plants may be subjected to various patterns of freeze-thaw cycles in winter such as cooling below a severe subzero temperature, exposure to a constant subzero temperature for a long time or repeated freeze-thaws.Several assay systems with different experimental conditions are needed to estimate the effects of acid snow stress on tissues of wintering crops.Thus, we tried to clarify the responses of winter wheat to acid snow stress using in vitro assay systems, EQFA and PRFA, that simulate the situation of acid snow stress by freeze-thawing of leaf tissues with sulfuric acid solution.
It is known that plant cells are dehydrated and deformed by the growth of extracellular ice in an extracellular freezing process (Steponkus 1984, Fujikawa et al. 1999).Severe freezing stress induced irreversible structural changes in cell membranes, especially the plasma membrane, that resulted from interaction with endomembranes in a state of extreme approach toward each other due to freezing-induced deformation of the cells (Fujikawa and Miura 1986, Fujikawa et al. 1999, Uemura and Steponkus 1999).In EQFA and PRFA, therefore, it is possible that harmful effects caused by an acid condition (pH <3.0) may magnify the irreversible structural changes of cell membranes due to interaction between endomembranes in a process of EQF of plant cells or promote the injury of membranes and/ or cell structure in a way that is different from freeze-induced structural changes.
In PRF and PRFA, the survival rates of leaves of winter wheat gradually decreased over the duration of freezing treatment in both the presence and absence of sulfuric acid at given temperatures.Freezing injury was enhanced as the temperature of PRF was lowered under acid conditions.It has been reported that prolonged exposure to subzero temperatures induced additional injury of Arabidopsis leaf cells with different features of injury, mainly due to the solution effect, which is the effect of the concentration of intracellular and/or extracellular ions caused by extracellular freezing (Mazur 1966, M. Nagao et al. unpublished observation).A recent study using freeze-fracture electron microscopy has suggested that structural change of the plasma membrane induced by PRF at a constant subzero temperature occurred with a mode of injury different from that by EQF in Arabidopsis leaves (M.Nagao et al. unpublished observation).According to this hypothesis, acidification due to freeze-induced dehydration and chronic acid conditions at subzero temperatures may be toxic for leaf samples in PRFA.Considering the situation of wintering plants covered with acid snow, the existence of acid substances in snow may lead to non-negligible damage to the shoots.The survival rates of leaf samples after PRFA tests (pH 3.0 or 4.0) at -4 and -8°C for >7 d were slightly lower than those of leaf samples in the absence of sulfuric acid.Also, the leaf samples were barely injured by supercooling under an acid condition (pH 2.0) that would not have a solution effect.These results support the hypothesis that the solution effect of sulfuric acid solution is the main cause of the additional injury caused by PRFA.In addition to the solution effect, it is possible that ice encasement or anaerobic conditions of leaf samples may have had an effect on the samples.
Furthermore, it is reasonable to assume that acid treatment has unfavorable effects on plant cells.The degree of damage may depend on experimental and physiological conditions such as pH and duration of treatment, composition of acid solution, treatment methods and the sensitivity of plants to acidification (Evans 1989).Extracellular and/or intracellular acidification due to strong acid treatment at a subzero temperature might cause damage due to conformational changes in macromolecules, phase transition of some membrane lipids (Cevc 1987), decrease in various metabolic activities and generation of reactive oxygen species (Yu et al. 2002) in some herbaceous plants sensitive to acid treatment.
In an EQF test, the survival rates of leaves of winter wheat gradually decreased as the freezing temperature was lowered.In the EQFA and PRFA tests, addition of sulfuric acid solution (pH 2.0) before ice seeding caused extensive damage to the leaf samples compared with the damage to leaf samples in sulfuric acid (pH 4.0 and 3.0) and water.Limiting factors for the sensitivity to acid snow stress in plants are still unknown.The sensitivity may be dependent on the buffering capacity of intracellular pH of wheat cells that is related to the proton pump activities in the plasma membrane and vacuolar membrane, cytosolic buffering action with ions, and membrane intactness.Since little damage was caused by supercooling under an acid condition (pH 2.0), by the addition of sulfuric acid just before leakage of amino acids or by freeze-thawing in the presence of sulfate salts, it is thought that a low pH condition due to proton concentration in the process of extracellular freezing is harmful to leaf samples of winter wheat.
Extracellular freezing of tissues with an acid solution results in the concentration of acid substances in extracellular unfrozen water.Therefore, the pH of extracellular unfrozen water at a subzero temperature may be drastically lowered.As a result, plant tissues would be exposed to ice crystals with local acidification on their surfaces and to extracellular unfrozen water containing a high concentration of acid substances in the process of EQFA.This may be the reason for the differences in the response of leaf tissues in extracellular freezing and supercooling states.The pH of unfrozen water at a subzero temperature may be similar in sulfuric acid solutions with various concentrations at a non-freezing temperature since the freezing point of the solution is dependent on the concentration of solutes.Therefore, the volume of unfrozen water will be dependent on the initial volume, concentration of acid solution and freezing temperature.Since the initial volumes of sulfuric acid solution in the EQFA and PRFA tests were the same, the volumes of unfrozen water in sulfuric acid solutions of pH 2.0, 3.0 and 4.0 at the same subzero temperature might be different.This may be one of the causes of the damage to wheat leaves in EQFA and PRFA tests.Therefore, an increase in the initial volume of sulfuric acid solution will enhance the damage to wheat leaves.
EQFA and PRFA in the present study were different from those in many previous studies on the effects of SAR treatment of plants grown under normal conditions or field conditions since the effects of light irradiation in both the EQFA and PRFA tests were minimized to protect wheat leaves from secondary damage due to production of reactive oxygen species from photosynthetic organs induced by light.In addition, atmospheric chemical studies have shown that freezing of a dilute ionic solution can promote some chemical and photochemical reactions that occur on the surface of growing ice or snowpacks via production of hydroxyl radicals, etc. (Finnegan et al. 1991, Takenaka et al. 1992, Chu and Anastasio 2003).Light irradiation during or after acid snow stress with realistic acidity of the sulfuric acid solution may enhance the damage to wheat leaves.If acid snow stress disturbs some physiological process of shoots or damages wintering plants, the inhibitory effects may be easily enhanced by other additional abiotic and biotic stimuli such as light irradiation, ozone and pathogen attacks in field conditions, although direct evidence for this hypothesis has not yet been obtained.
In summary, the assay systems used in this study, EQFA and PRFA tests, enabled determination of the influence of simulated acid snow stress on leaf tissues of winter wheat.The results suggested that acid snow is a potential environmental stress factor for wintering plants because of the combined effects of acid substances concentrated in snow crystals and freeze-thaws in an acid solution concentrated by freezing.Since damage caused by acid snow stress was enhanced by depression of subzero temperatures, extension of the stress period and decrease in pH, it is possible that unfavorable effects of acid snow on vegetation under natural conditions, regardless of whether the effects are direct or indirect, may arise in areas of the extreme cold environmental conditions such as high latitudes or polar regions.For further characterization of the responses of wintering plants to acid snow stress, studies on the mechanism of injury caused by acid snow stress are in progress.

Plant materials
Seeds of winter wheat (T.aestivum L. cv.Chihokukomugi) were germinated for 2 d on wet paper at 18°C in the dark.Seedlings were planted in soil and grown at 18°C (12 h light)/16°C (12 h dark) for 1 week in a growth chamber with irradiance of about 120 µmol m -2 s -1 from daylight-type white fluorescent tubes at soil level.One-week-old seedlings were cold acclimated at 4°C (12 h light)/2°C (12 h dark) for 4 weeks in a growth chamber with irradiance of about 120 µmol m -2 s -1 at seedling level.

Freezing tests under acid conditions
Leaves of cold-acclimated winter wheat were excised, washed with water and cut into small pieces (5 mm in width) with a razor.Leaf segments (50 ± 2 mg fresh weight) were put into each test tube.A 300 µl aliquot of pure water or sulfuric acid solution of which the pH was adjusted to 4.0, 3.0 or 2.0 was added to each test tube.
After the test tubes had been set in a programmable freezer (EP40-MV, Julabo, Germany) kept at 0°C, they were immediately cooled to -1°C and kept for 5 min.Freezing was initiated by ice seeding and the temperature was kept at -1°C for 1 h.Then two procedures for programmed freezing, EQF and PRF, were carried out.In EQF, the tubes were cooled at 2.4°C/h to various subzero temperatures.When the desired temperature had been reached, the tubes were withdrawn from the freezer and kept on ice in the dark.In PRF, the tubes were also cooled at 2.4°C h -1 to various subzero temperatures.When the desired temperature had been reached, the tubes were kept at that subzero temperature for 4 weeks.After PRF for the desired period, the test tubes were taken out of the freezer and kept on ice in the dark.In both freezing tests, values for 100 and 0% injury were obtained from fresh leaf samples treated with liquid nitrogen for 10 min and kept at 4°C, respectively.After all of the test tubes were thawed at 4°C overnight in the dark, survival rates of the samples were estimated by amino acid leakage measurement as described below.

Fig. 1
Fig. 1 Experimental diagrams of the two freezing tests, EQF (A) and PRF (B), used in this study.See the text for details.

Fig. 2
Fig. 2 Freezing injury of leaves of cold-acclimated winter wheat under acid conditions.Leaves of cold-acclimated winter wheat were examined by an EQFA test.(A) Sulfuric acid solutions of pH 4.0 (filled circles), 3.0 (filled squares) and 2.0 (filled triangles) or pure water of pH 5.6 (open circles) were added to leaf samples before ice seeding.(B) Pure water of pH 5.6 (open circles) or sulfuric acid solution of pH 2.0 (filled triangles) was added to leaf samples before ice seeding, or sulfuric acid solution of pH 2.0 was added just before amino acid leakage (filled squares: pH 5.6 + pH 2.0).Survival rates of the leaf samples after thawing were estimated by measuring amino acid leakage.

Fig. 3
Fig.3Injury of leaves of cold-acclimated winter wheat caused by extracellular freezing and supercooling under acid conditions.In the EQFA test, sulfuric acid solution of pH 2.0 (EQFA) or pure water of pH 5.6 (EQF) was added to the leaf samples before ice seeding.In equilibrium cooling under acid conditions, sulfuric acid solution of pH 2.0 (SCA) or pure water of pH 5.6 (SC) was added to the leaf samples, and programmed cooling was started without ice seeding to maintain a supercooling state.Survival rates of leaf samples after treatments were estimated by measuring amino acid leakage.

Fig. 4
Fig.4Influence of sulfate ions and pH in EQF on the survival of leaves of winter wheat.In the EQF test, potassium sulfate or sodium sulfate (A) and nitric acid or hydrochloric acid (B) were added to leaf samples instead of sulfuric acid before freezing.Survival rates of the leaf samples after thawing were estimated by measuring amino acid leakage.

Fig. 5
Fig. 5 Influence of PRF with different acidities on the survival of leaves of winter wheat.In the PRFA test, sulfuric acid solutions of pH 4.0 (filled circles), 3.0 (filled squares) and 2.0 (filled triangles) or pure water of pH 5.6 (open circles) were added to the leaf samples before ice seeding.After EQFA to -4°C (A) or -8°C (B), samples were kept at the freezing temperature for the desired periods.Survival rates of leaf samples were estimated by measuring amino acid leakage.