In Vitro Cellular & Developmental Biology - Plant

, Volume 45, Issue 4, pp 483–490

Analysis of ultrastructure and reactive oxygen species of hyperhydric garlic (Allium sativum L.) shoots


    • College of HorticultureNanjing Agricultural University
  • L. J. Chen
    • College of HorticultureNanjing Agricultural University
  • Y. J. Long
    • College of HorticultureNanjing Agricultural University

DOI: 10.1007/s11627-008-9180-8

Cite this article as:
Wu, Z., Chen, L.J. & Long, Y.J. In Vitro Cell.Dev.Biol.-Plant (2009) 45: 483. doi:10.1007/s11627-008-9180-8


Hyperhydricity is a physiological disorder frequently affecting shoots propagated in vitro. Since it negatively affects shoot multiplication vigor, and impedes the successful transfer of micropropagated plants to in vivo conditions, hyperhydricity is a major problem in plant tissue culture. In commercial plant micropropagation, there are reports of up to 60% of cultured shoots or plantlets which demonstrate hyperhydricity, which reflects the pervasiveness of this problem. The phenomenon has been correlated to water availability, microelements, and/or hormonal imbalance in the tissue culture. In this study, the ultrastructure and the characteristics of reactive oxygen species between hyperhydric and normal shoots of garlic were studied. We observed that in some cells of hyperhydric tissues, the intranuclear inclusion was separated, the mitochondrion was swollen and its intracristae had splits, the organelles were compressed against the cell wall, and the chloroplasts and intergranal thylakoids were also compressed. Additionally, the content of chlorophyll and soluble protein in hyperhydric shoots decreased significantly. For instance, chlorophyll a decreased 43.61%, chlorophyll b decreased 49.29%, chlorophyll a+b decreased 48.10%, and soluble protein dropped 47.36%. In contrast, the O2 generation rate and H2O2 level increased 45.36% and 63.98%, respectively, obviously higher than the normal shoots. Lipoxygenase activity and malondialdehyde content in the hyperhydric shoots increased significantly, while the electrolyte leakage rose, indicating a serious membrane lipid peroxidatic reaction. Superoxide dismutase, peroxidase, catalase, glutathione peroxidase, and ascorbate peroxidase activities in hyperhydric tissue were all significantly higher than in normal leaf tissue. The antioxidant metabolism demostrated a close connection between hyperhydricity and reactivated oxygen species.


Antioxidant enzymesGarlicHyperhydricityOxidative stressLipid peroxidation


Hyperhydricity is a specific physiological disorder frequently caused by environmental stress during in vitro micropropagation. Hyperhydric tissues undergo changes in metabolism, leading to an abnormal morphology and anatomy. The visual symptoms of hyperhydricity were water-soaked, thick, and translucent leaves, and, shoots with shorter internodes. The affected plants appear glassy and have reduced or retarded growth, have a bushy habit, thickened and malformed stems and leaves with hypertrophy of cortical and pith parenchyma cells (Debergh et al. 1981; Kevers et al. 1984; Olmos and Hellin 1998; Kevers et al. 2004). High relative humidity, poor gaseous exchange between the internal atmosphere of the culture vessel and its surrounding environment, and the accumulation of ethylene may induce hyperhydricity in plant micropropagation. Someone believes it is a response to the changes of physiology induced by oxidative stress (Enrique and Abel 1997).

Oxidative stress is caused by the imbalance between peroxidation and antoxidation in cells, resulting in the increase of peroxidation, thus injuring cells (Sies 1991). Explant tissues being prepared for initiation and subculturing in vitro cultures are wounded, which is known to cause oxidative stress. Hypochlorite (Wiseman and Halliwell 1996) and mercuric salts (Patra et al. 1997), are used to surface sterilize the primary explants and cause oxidative stress. Many factors associated with aberrations in plant tissue culture such as mineral deficiency (Elstner 1991), high salt content (McKersie and Leshem 1994), water stress (Navari-Izzo et al. 1996), excess metal ions (Caro and Puntarulo 1996), and possible overexposure to PGRs in the media (Droog 1997). As a defense mechanism, plants compete against these oxidative stresses by the synchronous action of various enzymatic and non-enzymatic antioxidants. Enzymatic antioxidants like superoxide dismutase (SOD), peroxidase (POD), catalase (CAT), glutathione peroxidase (GPX), ascorbate peroxidase (APX), are the principal reactivated oxygen species (ROS) scavenging systems, which quench molecular oxygen, superoxide radicals, singlet oxygen, hydroxyl radicals, and hydrogen peroxide like \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) and H2O2 under stressful conditions (Ahmed et al. 2002; Sairam et al. 2002).

Franck et al. (1995) found that oxidative stress occurrence and ROS accumulation are important factors causing the hyperhydricity. Because SOD activity in hyperhydric tissues was higher than that in normal shoots, Franck et al. believed that the primary cause of hyperhydricity was the higher SOD, inducing H2O2 accumulation. Piqueras et al. (1998, 2002) found higher peroxides activity, lower lignifications, and higher malondialdehyde (MDA) content in hyperhydric leaves than non-hyperhydric leaves in micropropagated carnation plants. This indicates that hyperhydricity is caused by oxidative injury. However, it remains difficult to determine whether hyperhydric shoots are stressed or not. A better understanding of the physiological bases of hyperhydricity in relation to its stress response would be of great interest in order to prevent this abnormality. Here we study several aspects of peroxides activity, ROS metabolism, physiology, and biochemistry in normal and hyperhydric shoots of micropropagated garlic plants, examine whether oxidative stress is involved in hyperhydricity or not, focusing on the connection of ROS and hyperhydricity.

Materials and Methods

Plant materials.

Hyperhydric and normal shoots of garlic (Allium sativum L.) ‘Ershuizao’ (provided by the Plant Tissue Culture Center of Horticulture College, Nanjing Agriculture University, Nanjing, China) were obtained from in vitro garlic plants cultured following the method of Ayabe and Sumi (1998, 2001). We observed the morphology and ultrastructure, and evaluated the severity of the hyperhydric shoots according to their vitrification (thicker and translucent stems and leaves; wrinkled and curled leaves).


Leaf central sections from the moderately hyperhydric shoots and normal shoots were pre-fixed under mild vacuum in a phosphate buffer (pH 7.2) and the 3% solution of glutaraldehyde pre-fixed for 12 h. These samples were then rinsed 3 times in a phosphate buffer (pH 7.2), post-fixed in 1% osmic acid for 2 h. The fixed samples were dehydrated in a graded series of acetone. Afterwards, the samples were infiltrated with Epon-812 epoxy resin. Ultra-thin sections were picked up on nickel grids. Grids were subsequently stained with uranyl acetate and lead citrate. Specimens were observed with a JGM-100CX II type electron microscope.

Measurements for contents of water, protein, and chlorophyll.

The fresh weight and dry weight of the samples were measured. To determine the fresh weight (fw), hyperhydric and normal shoots were separated and individual tissues were weighed. For dry weight (dw), samples (500 mg fw) of hyperhydric and normal shoots were placed in 105°C–110°C for 15 min, and then were dried at 70°C for 48 h and weighed. Water content (WC) as percentage of fresh weight was calculated using the formula: RWC % = ( fw–dw) ×100/fw. Soluble protein content was measured by Coomassie brilliant blue G-250 staining; the standard curve was made by bovine serum albumin Bradford (1976). Chlorophyll content was determined according to Wintermans and De Mots (1965) after extraction in 96% (v/v) ethanol.

Measurements for Solute leakage and \(O^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) generation rate, H2O2 content.

Electrolyte leakage was measured by RLO60C conductivity meter (Orion, Beverly, Massachusetts). First, electrical conductivity R1 of the bathing solution at room temperature was measured. Second, the bathing solution was heated at 100°C for 10 min, and then electrical conductivity R2 was determined. Electrolyte leakage (E) was calculated according to the following equation: E = R1/R2 ×100%.

\({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) assay was performed as described by Wang and Luo (1990). We detected \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) by vacuum injecting leaf discs with 0.1 mgmL-1 p-nitroblue tetrazolium (NBT) in 0.2 M sodium phosphate buffer (pH 7.8) for 15 min. The discs were incubated at 25°C in the dark for 2 h. Controls in the presence of SOD and MnCl2 (O2 scavengers) were also performed simultaneously by adding 100 U of SOD and 10 mM of MnCl2 into the infiltration buffer.

H2O2 content was determined as described by Uchida et al. (2002). Leaf tissue (500 mg) was homogenized in ice bath with 5 mL of a cold 0.1 % (m/v) trichloroacetic acid (TCA). The homogenate was centrifuged (12,000 × g, 15 min, 4°C) and 0.5 mL of the supernatant was added to 0.5 mL of 100 mM potassium phosphate buffer (pH 7.0) and 1 ml of KI reagent (1 M KI w/v in distilled H2O). The control probe consists of 0.1% TCA in the absence of leaf extract. The reaction was allowed 1 h in darkness and the absorbance was measured at 390 nm. The amount of H2O2 was calculated using a standard graph of known concentrations.

Measurements for Lipid peroxidation and LOX activity.

Lipid peroxidation rates were estimated by measuring the levels of MDA using the TBA reaction as described by Dhindsa et al. (1981). Samples of 500 mg (fw) were homogenized in 3 mL of 5% TCA (w/v). The homogenate was centrifuged at 3 000 rpm for 10 min at 4°C and 2 ml of 0.6 % TBA was added to 2 ml aliquots of the supernatant. The mixture was heated at 100°C for 15 min, and the reaction was stopped by quickly cooling to room temperature with ice. The cooled mixture was centrifuged at 10,000 rpm for 15 min. Absorbances were read at 450, 532 and 600 nm, respectively. After subtracting the non-specific absorbance at 600 nm, the amount of MDA-TBA complex (red pigment) was calculated using an absorbance coefficient 155 mM−1 cm−1.

Lipoxygenase (LOX; EC activity was measured according to the methods of Surrey (1963). The plant tissues were ground in 50 mM phosphate buffer (pH 7.0) containing 5% (w/v) insoluble polyvinylpolypyrrolidone (PVPP) and 0.25% Triton X-100. Crude extract initiated the reaction mixture containing 50 mM phosphate buffer (pH 7.0) and 0.4 mM linoleic acid. The absorbance was recorded at 234 nm. The molar extinction coefficient was 25 mM−1cm−1.

Measurements for some antioxidant enzymes activities.

Samples (500 mg fw) were homogenized at (4°C for 20 min at 12,000 × g) in 1.6 mL of 50 mmolL−1 potassium phosphate (pH 7.8), which included 1% PVP and 0.1 mmolL−1 EDTA. The supernatant was used to determine enzyme activity.

SOD (EC activity was assayed by measuring its inhibition of the photochemical reduction of NBT (Giannoplitis and Ries 1977). The 3 mL reaction mixture contained 50 mM phosphate buffer (pH 7.8), 13 mM methionine, 750 μmolL−1 NBT, 20 μmolL−1 riboflavin, 0.1 mM EDTA, and 0.1 mL of enzyme extract. The reaction was initiated by adding 20 μmolL−1 riboflavin and placing the tubes under 15W fluorescent lamps for 15 min. A complete reaction mixture without enzymes served as control. The reaction was terminated after 10 min by removing the reaction tubes from the light source, and then the tubes were covered with a black cloth. Reaction products were measured at 560 nm. SOD activity was expressed as units per minute per milligram protein, while 1U was defined as the amount of enzyme required to result in a 50% inhibition of the rate of nitro blue tetrazolium reduction.

POD (EC activity was assayed by following the change in absorption at 470 nm due to guaiacol oxidation according to Putter (1974). The reaction solution was composed of 2.9 mL 0.05 molL−1 phosphate buffer (pH 5.5), 1.0 ml 0.05 molL−1 guaiacol, 1.0 ml 2% H2O2, and 0.1 mL tissue extract; change in absorbance after 5 min was measured at 470 nm. The activity of the enzyme was expressed as Units mg−1protein, while one unit was defined as DA470 of 0.01 between the control and the sample per minute of reaction time.

CAT (EC activity was measured as the decline in absorbance at 240 nm due to the decline of H2O2 extinction according to the method of Alir and Mayber (1981). The specific activity was expressed as Units mg−1protein, while 1U of CAT converts 1 μmol H2O2 per min. The reaction mixture contained 20 mM sodium phosphate buffer (pH 7.8), 10 mM H2O2, and 0.1 mL enzyme extract. The reaction was initiated by adding H2O2. The consumption of H2O2 by CAT was detected by measuring the decrease in absorbance at 240 nm after 10 min at 30°C. Enzyme solution containing hydrogen peroxide-free phosphate buffer was used as control.

APX (EC activity was measured according to Hossain and Asada (1984). The 2 ml reaction mixture contained 25 mM sodium phosphate buffer (pH 7.0), 0.5 mM ascorbate, 0.1 mM H2O2, 0.1 mM EDTA, and 0.1 mL enzyme extract. The reaction was started by adding H2O2. Both minus tissue-extract and minus H2O2 blanks were run and the changes in absorbance every 15 s were read at 290 nm. The activity of APX was calculated in terms of l mol ascorbate oxidized per minute per milligram protein.

GPX (EC activity was determined in a reaction mixture composed of 50 mM potassium phosphate buffer (pH 7.0), 9 mM guaiacol, 10 mM H2O2, and enzyme extract (Leopold and Wolfang, 1984). The reaction was initiated by adding H2O2. We measured the increase of absorbance at 470 nm due to guaiacol oxidation. The molar extinction coefficient was 26.6 mM−1cm−1.

All absorbance was measured on spectrophotometer UV-2450 (Shimadzu Kyoto, Japan). Measurements for each species were compared by using analysis of variance (ANOVA) in SPSS13.0. Differences among the species were evaluated using the least significance difference (LSD) method with significance level of p < 0.05 or p < 0.01.


Visual and ultra structure observation.

The non-hyperhydric garlic in vitro shoots had normal growth and morphology (Fig. 1a). The mildly hyperhydric shoots grew normally, but had semitransparent leaves (Fig. 1b). The moderately hyperhydric shoots had stiff and abnormal leaves. The shoots could grow normally, but were difficult to multiply (Fig. 1c). The heavily hyperhydric shoots were primarily transparent and water-logged. The leaf tips turned yellow, even white. The plants experienced poor growth and many failed to survive (Fig. 1d).
Figure 1.

Hyperhydric shoots and normal shoots of garlic. (a) Normal shoots; (b) Mildly hyperhydric shoots; (c) Moderately hyperhydric shoots; (d) Severely hyperhydric shoots.

For the ultrastructure, we observed that in some hyperhydric cells, the intranuclear inclusion was separate (Fig. 2b), the mitochondrion was swollen, and its intracristae had splits (Fig. 2d), the vacuole pushed the organelle to the cell wall edge (Fig. 2d), the chloroplast became slender (Fig. 2f), and the intergranal thylakoid were compressed (Fig. 2h).
Figure 2.

Ultrastructure of leaves in hyperhydric and normal shoots of garlic. (a) Nuclei in normal shoot × 4800; (b) Nuclei in hyperhydric shoot × 3600; (c) Mitochondria in normal shoot × 7200; (d) Mitochondria in hyperhydric shoot × 10000; (e) Chloroplasts in normal shoot × 58000; (f) Chloroplasts in hyperhydric shoot × 58000; (g) Mitochondria and chloroplasts in normal shoot × 14000; (h) Chloroplasts in hyperhydric shoot × 14000.

Content of water, protein and chlorophyll.

We found that the water content in hyperhydric leaves of garlic (Allium sativum L.) ‘Ershuizao’ was over 92%, with a significantly higher value than that in normal ones (p < 0.05). The protein content was 32.09 mgg−1FW in the normal tissues, while it was 16.89 mgg−1FW in the hyperhydric shoots, showing a significant decrease of 47.36% (p < 0.01) (Fig. 3). The content of chlorophyll a, b and a+b in the hyperhydric shoots were much lower than that in the normal ones, reduced by 43.61%, 49.29% and 48.10%, respectively (p < 0.01). On the contrary, the ratio of chlorophyll a/b had increased. It showed that the decrease of chlorophyll b was more than that of chlorophyll a in the hyperhydric shoots (Table 1).
Figure 3.

Change of water content, soluble protein content, and solute leakage in hyperhydric and normal shoots of garlic. NS-normal shoots; HS-hyperhydric shoots.

Table 1.

Change of chlorophyll a, b, a/b, total content in normal and hyperhydric shoots of garlic

Shoot type

chl a (mg g−1FW)

chl b (mg g−1FW)

chl a+b (mg g−1FW)

chl a/b











Capital letters indicating significant at p = 0.01.

Solute leakage and \(O^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) generation rate, H2O2 and MDA content, LOX activity.

The electrolyte osmosis rate in the hyperhydric shoots was remarkably higher than that of normal ones, up to 52.62% (p < 0.01) (Fig. 3). Lipid peroxidation, protein denaturation, and other oxidative damage to cells are often caused by \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) and H2O2 when they accumulate (Smirnoff 1993). The results showed that the \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) generation rate and H2O2 content had significant differences between hyperhydric shoots and normal shoots, increasing 45.36% and 63.98% in hyperhydric shoots, respectively (p < 0.05). The MDA content in normal shoots was only 0.0179 μmolmg−1protein, while it was up to 0.0511 μmolmg−1protein in the hyperhydric ones, showing significant difference (p < 0.01)(Fig. 4). The LOX activity in normal shoots was 0.0224 μmolmg−1protein, while was 0.0479 μmolmg−1protein in the hyperhydric shoots, (p < 0.01).
Figure 4.

Generation rate of \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \), H2O2, MDA content, and LOX activity in hyperhydric and normal shoots of garlic. NS-normal shoots; HS-hyperhydric shoots.

Some antioxidant enzymes activities.

The results in Fig. 5 shows that the activities of antioxidant enzymes all increased significantly in the hyperhydric shoots. The SOD and POD activities were remarkably higher than those in normal shoots, increasing 36.19% and 100.02%, respectively (p < 0.01). The CAT activity increased 32.49% (p < 0.05). The GPX activity was 0.1650 nmolmin−1mg−1protein in the hyperhydric shoots, significantly higher than that in the normal ones with GPX activity, 0.1148 nmolmin−1mg−1protein (p < 0.05); while the APX activity in the hyperhydric shoots was 0.0379 nmolmin−1mg−1protein, extremely significantly higher than which was 0.0046 in the normal shoots (p < 0.01)(Fig. 5).
Figure 5.

Change of activities of SOD, POD, CAT, GPX, and APX in hyperhydric and normal shoots of garlic. NS-normal shoots; HS-hyperhydric shoots.


Environmental stresses, such as salinity or low oxygen, can limit photosynthesis, which then increases ROS generation and induces cellular damage (Mittler 2002). In our experiment we found chlorophyll was dramatically reduced in the hyperhydric shoots. Electron microscopy of hyperhydric tissues also indicated that the normal function of chloroplast was affected, reducing photosynthetic efficiency. Few reports have been made on hyperhydric tissue ultrastructure, but our results are similar to the established literature (Shi and Li 1990; Macpherson et al. 1993; Zhao and Xu 2001).

Increased water content in hyperhydric leaves resulted in low oxygen, protein, and chlorophyll. At the same time, plants are compelled to accept the stress and accumulate more ROS, which can enhance proteolysis and result in significant reduction of protein content in the hyperhydric shoots. These and other biochemical events indicate that physiological metabolic disorder and oxidative injury occurred. Most of plants in stress promote ROS production, which then causes cell injury.

We also found that the \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) generation rate and the H2O2 content in the hyperhydric shoots increased when compared to the normal. The toxic effects of \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) and H2O2 can initiate a chain reaction resulting in hydroxyl radicals and lipid peroxides. Higher H2O2 concentration is harmful to cells, resulting in localized oxidative damage, lipid peroxidation, disruption of metabolic function, and loss of cell integrity. It may also trigger an antioxidative response (Foyer et al. 1997; Velikova et al. 2000). Therefore, we suggest that the increased level of H2O2 we observed was due to oxidative damages. This indicates that the concentration of ROS in the hyperhydric shoots improved membrane lipid peroxidatic reaction.

Lipid peroxidation is used as an indicator of stress-induced oxidative damage. MDA is a decomposition product of polyunsaturated fatty acids hydroperoxides, and is used as a suitable biomarker for lipid peroxidation. In the ROS metabolic process, MDA is an important reporter for evaluating the status of membrane redox in plants, and LOX is considered to be partly responsible for the formation of \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) and singlet oxygen (Smirnoff 1993). In our research, MDA content and LOX activity in the hyperhydric shoots were increased significantly, thus, the electrolyte leakage increased, demonstrating a serious membrane lipid peroxidatic reaction. Higher LOX and increased lipid peroxidation in hyperhydric shoots of garlic could reflect a similar process of oxidative stress. Meanwhile, the chloroplasts became slender (Fig. 2f), and the intergranal thylakoid were compressed.This indicates that \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) and H2O2 are being generated in hyperhydric samples, caused by the leakage of electrons to oxygen from electron-transport chains in chloroplasts and mitochondria during photosynthesis and respiration. Normal metabolic activities could be disrupted by the generation of toxic levels of H2O2 (Asada and Takahashi 1987).

Antioxidative enzymes are known to protect cell structures against ROS generated by stress. SOD, POD, CAT, GPX, and APX are involved in overcoming of oxidative stress (Reddy et al. 2004). and hyperhydricity. The up regulation of these enzymes would help reduce ROS buildup. Previous reports probed the relationship between antioxidant enzyme activities and hyperhydricity in other plants, such as Gypsophila elegans (Wang et al. 1997), pyrus pyrifoliacult (Zhao and Yang 1998), Dianthus caryophyllus (Franck et al. 1995; Piqueras et al. 2002), Nicotiana tabacum L (Piqueras et al. 1998), Prunus avium (Thierry et al. 2004) et al.

In our research, we find that SOD, POD, CAT, GPX, and APX activity was significantly higher in hyperhydric tissue compared to non-hyperhydric leaf tissue. Other authors also reported an increase in SOD activity in plants under oxidative stress (Gupta et al. 1993; Kang and Saltveit 2002). SOD is usually considered the first line of defense against oxidative stress. It catalyzes the dismutation of \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) into H2O2 and O2. Dismutation of \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) by SOD results in H2O2, which can be effectively scavenged by POD and CAT. CAT is located predominantly in peroxisomes (and also in glyoxysomes) where its function is chiefly to remove the H2O2, while POD reduces H2O2 to H2O using several reductants available to the cells (Mittler 2002; del Rio et al. 2003). In plants, APX located throughout the cell is a specific peroxidase catalyzing the breakdown of H2O2 at the expense of oxidizing ascorbate to MDA. CAT and APX are involved in the scavenging of H2O2 (Jagtap and Bhargava 1995; Shigeoka et al. 2002). They reduce H2O2 into water (H2O), and thus diminish the adverse effects of oxidative stress. In accordance with other authors, the results reported similar patterns of CAT and APX activities in different stress situations, such as acid rain stress (Velikova et al. 2000), arsenic toxicity (Stoeva et al. 2003), and drought (Zlatev et al. 2006). Therefore, the increase in CAT and APX activity in hyperhydric leaves is necessary to scavenge H2O2 in peroxisomes and cytosol, where the H2O2 might have diffused from chloroplasts as a result of SOD activity.

In conclusion, our study showed that there was a close connection between ROS and hyperhydricity. Compared with the normal shoots, the hyperhydric shoots had abnormal morphological and and ultrastructure, reduced chlorophyll and soluble protein content, and increased levels of \({\text{O}}^{{{\mathop {^{ \cdot } }\limits^ - }}}_{2} \) and H2O2. In addition, enzymatic antioxidants such as SOD, POD, CAT, GPX, and APX all increased as well. This indicates an antioxidative defense system was used by garlic cells to survive under hyperhydric conditions, and the hyperhydricity might be an outcome of the oxidative stress.

Copyright information

© The Society for In Vitro Biology 2009