Introduction

Lack of oxygen or hypoxia and its complete absence or anoxia is a common environmental challenge, which plants have to face throughout their life. Seed imbibitions, flood irrigation, floods, and excess of rainfall are examples of natural conditions leading to root hypoxia or anoxia. A decrease in adenylate energy charge (ATP/ATP + ADP + AMP), cytoplasmic acidification, anaerobic fermentation, elevation in cytosolic Ca2+ concentration, changes in the redox state [NAD(P)H/NAD(P)], and a decrease in the membrane barrier function are the main features caused by lack of oxygen (Richard et al. 1994; Crawford and Braendle 1996; Drew 1997; Tadege et al. 1999; Sairam et al. 2008). Regulation of anoxic/hypoxic metabolism is complex and not all the features are well elucidated.

Excessive generation of reactive oxygen species (ROS), or oxidative stress is an integral part of many stress situations, including hypoxia. Hydrogen peroxide accumulation under hypoxic conditions has been shown in the roots and leaves of Hordeum vulgare (Kalashnikov Ju et al. 1994) and in wheat roots (Biemelt et al. 2000). The presence of H2O2 in the apoplast and in association with the plasma membrane has been visualized by transmission electron microscopy under hypoxic conditions in four plant species (Blokhina et al. 2001). Recently Kumutha et al. (2009) and Sairam et al. (2009b) have shown that hypoxia induced ROS in pigeon pea roots are due to induction of membrane linked NADPH oxidase. To control the levels of ROS and to protect cells under stress conditions, plant tissues contain several ROS scavenging enzymes viz., superoxide dismutase (SOD: EC1.15.1.1), ascorbate peroxidase (APX: EC1.11.1.11), glutathione reductase (GR: EC 1.6.4.2), and catalase (CAT: EC 1.11.1.6), and a network of low molecular mass antioxidants (ascorbate, glutathione, phenolic compounds, and α-tocopherol). An increase in the activity of antioxidant enzymes in response to environmental stresses have been reported in various studies (Elstner 1986; Bowler et al. 1992; Sairam et al. 2000, 2001, 2002). A 14-fold increase in SOD activity under waterlogged condition has been reported in Iris pseudacorus (Monk et al. 1989). The degree of increase in total SOD activity in wheat roots under anoxia was reported to be positively correlated with duration of anoxia (Van Toai and Bolles 1991). Arbona et al. (2008) reported that water-logging induced positive antioxidant response in citrus cultivars Carrizo citrange and Citrumelo CPB 4475 might be responsible for a higher tolerance to flooding stress, whereas the susceptibility of Cleopatra mandarin could be due to less activity of antioxidant enzymes. Wang and Jiang (2007) reported that SOD and APX were mainly involved in water-logging induced antioxidant responses, and the partial water-logging could also significantly affect root antioxidant activities, particularly in water-logging-sensitive cultivar. Similarly, Kumutha et al. (2009) have reported higher activity of SOD, APX, and GR in the water-logging tolerant pigeon pea genotype ICP 301 than susceptible Pusa 207. Increase in the activity of enzymes involved in the ascorbate–glutathione cycle (APX, MDHAR, DHAR, and GR) has also been shown for anaerobically germinated rice seedlings and roots of wheat seedlings (Ushimaru et al. 1997, Albrecht and Wiedenroth 1994).

Mung bean, an important summer–rainy season pulse crop, is exposed to water-logging condition during germination and early vegetative growth phases. Very little physiological and molecular studies have been done on the expression of antioxidant defense genes under water-logging. The objective of this study was to characterize the extent of ROS production during and after water-logging, to validate the role of antioxidant enzymes in water-logging tolerance, and to clone the partial cDNA/genes of SOD and APX for their subsequent use in developing full length genes.

Materials and methods

Plant material and growth conditions

An experiment was conducted with two cultivated mung bean (V. radiata (L.) Wilczek) cvs. T 44 (tolerant) and Pusa Baisakhi (susceptible), and highly tolerant wild genotype Vigna luteola under pot-culture condition to study their response to water-logging stress. V. luteola, T 44 and, Pusa Baisakhi were obtained from National Bureau of Plant Genetic resources, New Delhi, India, Directorate of Pulse Research, Kanpur, Uttar Pradesh, India and Division of Genetics, Indian Agricultural Research Institute, New Delhi, India, respectively. Sowing was done in 30 × 30 cm (h × dia) earthen pots filled with clay-loam soil and compost in 3:1 ratio during the summer–rainy season. Pots were supplied with basal dose of 25 kg ha−1 nitrogen and 60 kg ha−1 each of phosphorus and potassium. Before sowing, seeds were treated with the required Rhizobium culture. Initially four plants were sown in each pot, which were thinned to two plants per pot after 20 days. Water-logging treatment was given by placing pots with 25 days old plants in plastic troughs measuring 100 × 70 × 35 cm (l × b × h), and filled with water to a height just 1–2 cm below the soil level in pots. Treatments consisted of control, 2, 4, 6 and 8 days of water-logging, and recovery after 4 days of termination of treatment for 6-day waterlogged plants, as 8 days of water-logging in Pusa Baisakhi (susceptible genotype) resulted in more than 75% mortality during recovery. At each stage samples were collected in quadruplicate from four pots. Observations were recorded on superoxide radical, hydrogen peroxide contents, and activity of superoxide dismutase (SOD), ascorbate peroxidase (APX) and glutathione reductase (GR), and expression and cloning of Cu/Zn-SOD and APX. The values are means of four replicates assayed twice (n = 8). The design of the experiment was complete randomized (CRD) and data was analyzed by factorial CRD.

Assay of oxidative stress

Superoxide radical content was quantified by its capacity to reduce nitroblue tetrazolium chloride (NBT) and the absorption of end product measured at 540 nm. One gram root tissue was homogenized in 10 mL of pre-cooled phosphate buffer (0.2 M, pH 7.2) containing 1 mM diethyl dithio carbamate (DEDC), to inhibit SOD activity. The homogenate was centrifuged in Sigma refrigerated centrifuge (model 3 K 30, Osterode, Germany) at 10,000 g for 10 min and the supernatant was used immediately for the estimation of superoxide radical (Chaitanya and Naithani 1994). The reaction mixture contained 0.25 mL tissue extract, 0.075 mM NBT, 25 mM Na2CO3, 0.1 mM EDTA, 13.33 mM L-methionine and water to make the volume 3 mL. Reaction mixtures were incubated at 30°C for 10 min and absorbance was recorded at 540 nm. Superoxide radical content was calculated according to its extinction coefficient \( \varepsilon \) = 12.8 mM−1 cm−1.

Hydrogen peroxide was estimated by forming titanium–hydro peroxide complex (Rao et al. 1997). One gram root material was ground with liquid nitrogen and the fine powdered material was mixed with 10 mL cooled acetone in a cold room (10°C). Mixture was filtered with Whatman No.1 filter paper followed by the addition of 4 mL titanium reagent and 5 mL ammonium solution to precipitate the titanium–hydro peroxide complex. Reaction mixture was centrifuged at 10,000 g for 10 min in the Sigma refrigerated centrifuge (model 3 K 30, Osterode, Germany). Precipitate was dissolved in 10 mL of 2 M H2SO4 and then recentrifuged. Supernatant was read at 415 nm against blank in UV–visible spectrophotometer (model Specord Bio-200, AnalytikJena, Germany).

Preparation of enzyme extract

Enzyme extract for SOD, APX, and GR was prepared by freezing 1 g root samples in liquid nitrogen, followed by grinding with 5 mL extraction buffer consisting of 0.1 M phosphate buffer, pH 7.5, containing 0.5 mM EDTA in case of SOD and GR or 0.1 M phosphate buffer, pH 7.5 containing 0.5 mM EDTA and 1 mM ascorbic acid in case of APX. Brie was passed through four layers of cheesecloth and filtrate was centrifuged for 20 min at 4°C and 21,000 g, and the supernatant was used as enzyme extract.

Antioxidant enzymes estimation

Superoxide dismutase activity was assayed by following the enzyme mediated decrease in absorbance of formazone formed by superoxide anion with NBT (Dhindsa et al. 1981). The reaction mixture contained 13.33 mM methionine, 25 mM NBT, 0.1 mM EDTA, 50 mM potassium phosphate buffer (pH 7.8), 50 mM sodium carbonate and 0.1 mL enzyme. Reaction was started by adding 2 μM riboflavin and placing the tubes under two 15 W fluorescent lamps for 15 min. To distinguish SOD isoforms viz., Cu/Zn-SOD, Fe-SOD and Mn-SOD, the sensitivity of Cu/Zn-SOD to cyanide (3 mM), and Cu/Zn-SOD and Fe-SOD to hydrogen peroxide (5 mM) were used, whereas Mn-SOD is unaffected (Yu and Rengel 1999). A complete reaction mixture without enzyme, which gave the maximal color, served as a blank. Reaction was stopped by switching off the light and putting the tubes into dark. A non-irradiated complete reaction mixture served as a reference. The absorbance was recorded at 560 nm, and 1 unit of enzyme activity was taken as that amount of enzyme, which reduced the absorbance reading to 50% in comparison with tubes lacking enzyme (blank).

Ascorbate peroxidase activity was assayed by recording the decrease in absorbance due to enzyme induced oxidation of ascorbic acid at 290 nm (Nakano and Asada 1981). The 3 mL reaction mixture contained 50 mM potassium phosphate buffer (pH 7.0), 0.5 mM ascorbic acid, 0.1 mM EDTA, 1.5 mM H2O2 and 0.1 mL enzyme. The reaction was started with the addition of hydrogen peroxide. Absorbance was measured at 290 nm in a UV–visible spectrophotometer (model Specord Bio-200, AnalytikJena, Germany).

Glutathione reductase activity was assayed as per the method of Smith et al. (1988). The reaction mixture contained, 66.67 mM potassium phosphate buffer (pH 7.5) and 0.33 mM EDTA, 0.5 mM DTNB (in 0.01 M potassium phosphate buffer, pH 7.5), 66.67 μM NADPH, 666.67 μM oxidized glutathione (GSSG), 0.1 mL enzyme extract. Reaction was started by adding GSSG. The increase in absorbance at 412 nm was recorded spectrophotometrically.

Gene expression by RT-PCR

Nucleotide sequences for candidate genes were obtained from National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/). The Basic Local Alignment Search Tool (Altschul et al. 1997) (http://www.ncbi.nlm.nih.gov/BLAST/) was used to identify the homologs of candidate genes. For RT-PCR expression analysis and cloning of cDNAs, the following oligonucleotide primers were designed manually, and oligo quality (to avoid primer dimmer, self dimer etc.), GC % and Tm were analyzed by using Oligoanalyzer 3.0 tool (http://www.idtdna.com/analyzer/Applications/OligoAnalyzer/, Intergrated DNA Technologies, Coralville, IA 52241, USA).

Primer Name

Sequence

GC content (%)

Tm (°C)

Amplicon size (bp)

Tubulin-F

CTTGACTGCATCTGCTATGTTCAG

45.8

55.5

422

Tubulin-R

CCAGCTAATGCTCGGCATACTG

54.5

58.4

APX-F

CTCAGAGGYTTMATCGCTGAGAAG

45.7

64.6

689

APX-R

CAGCAAACCCAAGCTCRGARAGC

49.7

66.3

Cu/Zn-SOD-F

ATGGTKAAGGCTGTGGCAGTTC

52.3

58.9

408

Cu/Zn-SOD-R

ACCTTTCCCAAGATCATCAGGAGGATC

45.8

56.5

For gene induction studies 25 days old plants were subjected to water-logging treatment for 24 h, as 48 h water-logging treatment resulted in decline in gene expression of Cu/Zn-SOD and APX in all the genotypes. Root samples were harvested from control and treated plants. Total RNA from the root tissue was extracted using Trizol reagent (GibcoBRL) as per the recommendations of the manufacturer. DNA contamination was removed from the RNA samples using DNase I (Qiagen Science, Maryland, USA). One microgram of total RNA was reverse transcribed using gene specific degenerate primers and Qiagen one step RT-PCR kit. PCR conditions were standardized using gene-specific primers for tubulin. Linear amplification for semi-quantitative RT-PCR was obtained with 35 cycles. Reactions were conducted using My Genie 32 Thermal Block PCR (Bioneer, Korea) under the following conditions: initial PCR activation step, 15 min at 95°C; reverse transcription, 30 min at 50°C; denaturation, 1 min at 94°C; annealing, 1 min at 60°C; extension, 1 min at 72°C; final extension, 10 min at 72°C. The amplification products were electrophoresed on 1.2% agarose gel at 120 V in TBE buffer (0.4 M Tris–borate, 0.001 M EDTA, pH 8.0) using known concentration DNA ladders. Gels were stained with ethidium bromide and visualized on Uvi Pro Gel Documentation system (Uvitec, England).

Gene sequencing

RT-PCR amplified cDNAs were fractionated on agarose gel and purified. The purified cDNAs for each gene were cloned into pTz57R/T vector and transformed into E. coli (strain DH5α) cells. DH5α cells transformed with recombinant plasmids were selected based on antibiotic resistance as well as α-complementation method. Ampicillin resistant putative recombinants were selected for further analysis. Plasmid were isolated from the confirmed colonies and restriction analysis was carried out by using KpnI and HindIII enzymes flanking the cloning site of the vector pTz57R/T, to confirm the presence of cloned insert cDNA. Cloned insert cDNA in the pTz57R/T vector was sequenced by dideoxy chain termination method (Sanger et al. 1977) using T7 and SP6 primers.

Results

ROS contents

Two days after water-logging total superoxide radical (O2 ·−) content declined over control plants in all the genotypes (Fig. 1a). However, continuous water-logging up to 8 days resulted in significant increase in total O2 ·− production over 2-day waterlogged plants in the three genotypes, though remaining less than the control plants. The increase was greater in V. luteola and T 44 than in Pusa Baisakhi. Recovery recorded after 4 days of termination of treatment resulted in a drastic increase in O2 ·− content over that observed in waterlogged and control plants of the three genotypes, and the increase in O2 ·− production was highest in Pusa Baisakhi compared to V. luteola and T 44.

Fig. 1
figure 1

Effect of water-logging on total superoxide radicals (a) and hydrogen peroxide content (b) in root tissues in V. luteola and V. radiata genotypes T 44 (tolerant) and Pusa Baisakhi (susceptible). LSD significant (P ≤ 0.05). Vertical bars show ±SE of mean

Full size image

Hydrogen peroxide content showed pattern similar to O2 ·− content. In control and recovered plants H2O2 content was higher in Pusa Baisakhi than in V. luteola and T 44 (Fig. 1b). Two days of water-logging resulted in significant decline in hydrogen peroxide content in the three genotypes compared to that observed in control plants. However, further water-logging up to 8 days resulted in significant increase in H2O2 content compared to 2-day waterlogged, but remained lower than control plants in all the genotypes. During the water-logging phase V. luteola and T 44 showed higher H2O2 content than observed in Pusa Baisakhi. Recovery resulted in a drastic increase in H2O2 content over that observed in control and waterlogged plants in the three genotypes, and Pusa Baisakhi showed significantly higher H2O2 content than V. luteola and T 44 (Fig. 1b).

Enzyme activities

Water-logging caused significant increase in the activity of SOD and its isozymes in V. luteola and T 44 up to 8th day of stress, while in Pusa Baisakhi a slight increase was observed only on 2nd day of water-logging and thereafter, the activity decreased with duration of stress (Fig. 2a). During recovery the SOD activity slightly decreased in V. luteola and T 44 and increased in Pusa Baisakhi, however, the activity in V. luteola and T 44 was 445 and 381% higher than observed in Pusa Baisakhi. Water-logging induced increase in total SOD activity in V. luteola and T 44 which was mainly due to the increase in Cu/Zn-SOD, which contributed to about 51, 54, 58, 59% and 57, 56, 58, 59% at 2, 4, 6 and 8-day water-logging treatment, respectively. Mn-SOD accounted for about 29, 27, 25, 25% and 23, 25, 25, 25% of total SOD activity at 2, 4, 6 and 8-day of water-logging treatment in V. luteola and T 44, respectively. Fe-SOD had lesser contribution to total SOD activity. All the isoforms of SOD were greatly reduced at 4, 6 and 8 days of water-logging stress in Pusa Baisakhi (Fig. 2b–d).

Fig. 2
figure 2

Effect of water-logging on the activity of total superoxide dismutase (SOD) (a), Cu/Zn-SOD (b), Mn-SOD (c) and Fe-SOD (d) in root tissues in V. luteola and V. radiata genotypes T 44 (tolerant) and Pusa Baisakhi (susceptible). LSD significant (P ≤ 0.05). Vertical bars show ±SE of mean

Full size image

There was significant increase in the activities of APX and GR at 2 d of water-logging in all the genotypes (Fig. 3a, b). Further water-logging of 4, 6, and 8 days caused continuous increase in APX and GR activity in V. luteola and T 44. However, in case of Pusa Baisakhi, 4-day water-logging showed only slight increase in APX and GR activity, while further water-logging of 6 and 8 days resulted in significant decline in all the enzymes. During recovery, Pusa Baisakhi showed slight increase in the activity of APX, and decline in GR activity, while V. luteola and T 44 showed decline in the activity of both the enzymes, however, the activities were many times higher than the control and recovered plants of Pusa Baisakhi.

Fig. 3
figure 3

Effect of water-logging on the activity of ascorbate peroxidase (a) and glutathione reductase (b) in root tissues in V. luteola and V. radiata genotypes T 44 (tolerant) and Pusa Baisakhi (susceptible). LSD significant (P ≤ 0.05). Vertical bars show ±SE of mean

Full size image

Gene expression of Cu/Zn-SOD and APX

Gene expression studies for Cu/Zn-SOD and APX by RT-PCR elicited bands of about 410 and 690 bp. There was very little constitutive expression of cyt.-Cu/Zn-SOD gene (mRNA) in V. luteola, T 44, and Pusa Baisakhi. However, 24-h water-logging treatment induced a very significant increase in cyt.-Cu/Zn-SOD gene expression only in V. luteola and T 44, with only a little increase in Pusa Baisakhi (Fig. 4a). Cyt-APX gene showed significant expression in V. luteola and T 44 under 24 h water-logging. In case of Pusa Baisakhi very little APX expression was observed both in control and water-logging conditions (Fig. 4b). The tubulin expression was almost constant in both the genotypes, and did not change under control and waterlogged conditions (Fig. 4c).

Fig. 4
figure 4

Expression analysis of a cytosolic Cu/Zn-superoxide dismutase (cyt-Cu/Zn-SOD), b cytosolic ascorbate peroxidase (cyt.-APX) and c tubulin genes under 24 h water-logging stress and control conditions. Gene expression was determined by RT-PCR utilizing gene specific primer sets for each gene. β-Tubulin gene expression was used as an internal control. (1-1 Kb ladder, 2-Control V. luteola, 3-Treated V. luteola, 4-Control T 44, 5-Treated T 44, 6-Control Pusa Baisakhi, 7-Treated Pusa Baisakhi)

Full size image

Sequencing of Cu/Zn-SOD and APX

Cyt.-Cu/Zn-SOD gene specific primers yielded partial gene coding sequences of 399, 406, and 392 bp in V. luteola, T 44 and Pusa Baisakhi, respectively. Partial nucleotide sequences for cyt.-Cu/Zn-SOD of the three genotypes are given below:

Vigna luteola

AGTAACGAAGTCTCGGGTACTATTAACTTCAGTCAGGAGGGAAATGGTCCAACCACTGTAACTGGAACTCTTGCTGGTCTTAAGCCTGGCCTCCACGGCTTCCATATCCATGCCTTGGGAGACACCACAAACGGTTGCATTTCAACTGGACCACATTTCAATCCTAATGGGAAGGAACATGGTGCCCCTGAGGATGAGACTAGACATGCTGGTGATTTAGGAAATATCAATGTTGGTGATGATGGAACTGTAAGCTTCACCATTACTGACAACCATATCCCTCTCACTGGAACAAACTCCATCATAGGAAGGGCTGTTGTTGTCCATGCCGATCCTGATGATCTTGGGAAAGGTGGTCACGAGCTTAGCAAAACTACTGGAAATGCTGGTGGCAGAGTA

T 44

GTTTAGTAACAGTAACGAAGTCTCGGGTACTATTAACTTCAGTCAGGAGGGAAATGGTCCAACCACTGTAACTGGAACTCTTGCTGGTCTTAAGCCTGGCCTCCACGGCTTCCATATCCATGCCTTGGGAGACACCACAAACGGTTGCATTTCAACTGGACCACATTTCAATCCTAATGGGAAGGAACATGGTGCCCCTGAGGATGAGACTAGACATGCTGGTGATTTAGGAAATATCAATGTTGGTGATGATGGAACTGTAAGCTTCACCATTACTGACAACCATATCCCTCTCACTGGAACAAACTCCATCATAGGAAGGGCTGTTGTTGTCCATGCCGATCCTGATGATCTTGGGAAAGGTGGTCACGAGCTTAGCAAAACTACTGGAAATGCTGGTGGCAGA.

Pusa Baisakhi

AGTTAGTAACAGTAACGAAGTCTCGGGTACTATTAACTTCAGTCAGGAGGGAAATGGTCCAACCACTGTAACTGGAACTCTTGCTGGTCTTAAGCCTGGCCTCCACGGCTTCCATATCCATGCCTTGGGAGACACCACAAACGGTTGCATTTCAACTGGACCACATTTCAATCCTAATGGGAAGGAACATGGTGCCCCTGAGGATGAGACTAGACATGCTGGTGATTTAGGAAATATCAATGTTGGTGATGATGGAACTGTAAGCCGCACCATTACTGGCAACCATATCCCTCTCACTGGAACAAACTCCATCATAGGAAGGGCTGTTGTTGTCCATGCCGATCCTGATGATCTTGGGAAAGGTGGTCACGAGCTTAGCAAAACTACTGGAA.

V. luteola, T 44, and Pusa Baisakhi showed 100, 99, 98% similarity with cyt.-Cu/Zn-SOD of Pisum sativum (GenBank Acc. No. ABQ189165). V. luteola showed 99 and 96% similarity with T 44 and Pusa Baisakhi, respectively, while the similarity between T 44 and Pusa Baisakhi was 98%.

Cyt-APX gene specific primers yielded partial gene coding sequences of 662, 630, and 622 bp in V. luteola, T 44, and Pusa Baisakhi, respectively. Partial nucleotide sequences for cyt.-APX of the three genotypes are given below:

Vigna luteola

TTCATCGCTGAGAAGAGATGCGCTCCTCTGATGCTCCGTTTGGCATGGCACTCTGCTGGTACCTACGACGTTAGCTCGAAGACCGGTGGTCCCTTCGGAACCATGAAGCACCCCGCTGAACTCGCTCACGGTGCCAACAACGGTCTCGATATCGCTGTTAGGCTTTTGGAACCAATTAAAGCAGAGTTTCCTATCTTGAGCTACGCAGATTTCTACCAGTTGGCTGGCGTTGTCGCAGTTGAGGTCACTGGTGGACCTGAAGTTCCTTTCCACCCGGGAAGAGAGGACAAGCCAGAACCACCTCCAGAGGGTCGCTTGCCCGATGCAACCAAGGGGTCTGATCATCTTAGGGATGTGTTCGGCAAAGCTATGGGCCTTAGTGATCAAGATATTGTTGCTCTATCTGGTGGCCACACCATTGGAGCGGCACACAAGGAGCGTTCAGGATTTGAAGGCCCGTGGACCTCAAACCCTCTTATTTTCGACAACTCATACTTTAAGGAGTTGTTGAGTGGTGAAAAGGAAGGCCTCCTTCAGCTGCCTTCTGACAAGGCACTTTTGTCTGATCCTGTATTCCGCCCTCTTGTTGAAAAATATGCAGCGGACGAAGATGCCTTCTTCGCTGATTACGCAGTTGCTCACCAAAAGCTCTCCGAGCTGGG.

T 44

GCATGGCACTCTGCTGGTACCTTTGACGTTAGCACGAAGACCGGTGGTCCCTTCGGAACCATCAAGCACCCTGCGGAACTCGCTCACGGTGCCAACAACGGTCTCGACATAGCTGTTAGGCTTTTGGAGCCAATCAAAGCAGAGTTTCCTATCTTGAGCTACGCAGATTTCTACCAGCTGGCTGGCGTTGTTGCAGTTGAGATAACTGGTGGACCCGAAGTACCTTTTCACCCGGGCAGAGAGGACAAACCAGAACCACCTCCAGAGGGTCGCTTGCCTGATGCAACCAAGGGGTCTGATCATCTTAGGGATGTGTTCGGCAAGGCTATGGGGCTTAGTGATCAGGATATTGTTGCTCTATCTGGTGGTCACACCATCGGAGCGGCACACAAGGAGCGTTCAGGATTTGAGGGTCCGTGGACCTCAGACCCTCTTATATTTGACAACTCACACTTTAAGGAGTTGTTGAGTGGTGAAAAGGAAGGCCTCCTTCAGCTGCCTTCTGACAAGGCACTTCTGTCAGATCCCGTATTCCGCCCTCTTGTTGAAAAATATGCAGCGGACGAAGATGCCTTCTTTGCTGATTATGCAGTTGCTCACCAAATGCTCTCCGAGCTAGGGTTTTGACTG.

Pusa Baisakhi

GATTTCTACCAGCTGGCTGGCGTTGTTGCAGTTGAGATAACTGGTGGACCCGAAGTACCTTTTCACCCGGGCAGAGAGACTCTGCTGGTACCTTTGACGTTAGCACGAAGACCGGTGGTCCCTTCGGAACCATCAAGCACCCTGCGGAACTCGCTCACGGTGCCAACAACGGTCTCGACATAGCTGTTAGGCTTTTGGAGCCAATCAAAGCAGAGTTTCCTATCTTGAGCTACGCAGACAAACCAGAACCACCTCCAGAGGGTCGCTTGCCTGATGCAACCAAGGGGTCTGATCATCTTAGGGATGTGTTCGGCAAGGCTATGGGGCTTAGTGATCAGGATATTGTTGCTCTATCTGGTGGTCACACCATCGGAGCGGCACACAAGGAGCGTTCAGGATTTGAGGGTCCGTGGACCTCAGACCCTCTTATATTTGACAACTCACACTTTAAGGAGTTGTTGAGTGGTGAAAAGGAAGGCCTCCTTCAGCTGCCTTCTGACAAGGCACTTCTGTCAGATCCCGTATTCCGCCCTCTTGTTGAAAAATATGCAGCGGACGAAGATGCCGTCTTTGCTGATTATGCAATTGCTCACCAAAGCTCTCCGAGCTAGGGTTTTGCTGA.

V. luteola, T 44, and Pusa Baisakhi showed 95, 92, and 93% similarity with the APX complete CDS of V. unguiculata (GenBank Acc. No. U61379). V. luteola showed 92% similarity with T 44 and Pusa Baisakhi, while there was 98% similarity between T 44 and Pusa Baisakhi.

Discussion

We have earlier reported that part of the basis of water-logging tolerance of wild species V. luteola and mung bean (V. radiata) cv. T 44 could be comparatively greater accumulation of total and reducing sugars, mRNA expression and activity of sucrose synthase and alcohol dehydrogenase in roots, which was manifested in higher relative water content and chlorophyll contents in leaves, and membrane stability index (lower membrane injury) both in roots and leaves than susceptible cv. Pusa Baisakhi (Sairam et al. 2009a, b).

A very striking observation under water-logging is the production of various ROS, especially superoxide radical and hydrogen peroxide (Yan et al. 1996). Though 2 day waterlogged plants showed a decline in O2 ·− and H2O2 contents as compared to control or untreated plants, from 4th to 8th day there was significant increase in the contents of O2 ·− and H2O2. Further comparatively tolerant V. luteola and T 44 showed relatively higher O2 ·− and H2O2 levels compared to Pusa Baisakhi (susceptible cv.) during the water-logging phase as against higher ROS levels in Pusa Baisakhi in control and revived plants. The decline in O2 ·− and H2O2 production on 2nd day of water-logging could be attributed to a shift from aerobic respiration to fermentation and consequently a blockage in the mitochondrial site of ROS production. Further, the rise in ROS from 4th to 8th day of water-logging treatment could be due to the induction of membrane linked NADPH oxidase. Water-logging induced increase in the activity of diphenylene iodonium chloride-sensitive NADPH oxidase dependent superoxide production as well as increase in the expression of NADPH oxidase has been reported in pigeon pea (Kumutha et al. 2009 and Sairam et al. 2009b). Blokhina et al. (2001) reported H2O2 formation during anoxia as evidenced by electron dense insoluble precipitate of cerium perhydroxide, visualized through electron microscope. Greater ROS production in tolerant genotypes V. luteola and T 44 under water-logging could also have a role in signaling anaerobiosis tolerance mechanism, as H2O2 signaling has been reported in the induction of ADH gene (Fukao and Bailey-Serres 2004) and SOD and APX (Agarwal et al. 2005).

Results observed on various antioxidative enzymes like SOD, APX and GR under waterlogged condition revealed a continuous increase in all the enzymes up to 8 days of water-logging in tolerant genotypes V. luteola and T 44, while in case of susceptible genotype Pusa Baisakhi the increase in antioxidative enzymes activity was observed only under 2–4 days of water-logging, and at subsequent stages there was decline in activity of all the enzymes as compared to control and 2/4 days waterlogged plants. The increase in SOD activity was mainly due to the Cu/Zn-SOD, followed by Mn-SOD, while very little increase was observed in Fe-SOD. The increase in the activity of SOD and APX under water-logging was observed to be associated with increase in mRNA expression of Cu/Zn-SOD and APX genes, which were significantly more in tolerant genotypes V. luteola and T 44 than susceptible cv. Pusa Baisakhi. Mittler and Zilinskas (1992) reported increase in APXI transcript levels in response to drought, heat, and application of ethephon, abscisic acid, and paraquat. It thus seems plausible that water-logging induced increase in ROS could have a role in inducing APX and SOD genes. Pasentsis et al. (2007) reported anoxia and/or hypoxia regulated 11 genes in Citrus flavedo. RNA blot analysis confirmed that the induction of O2 regulated genes was related to antioxidative mechanism. Hypoxia induced increase in the transcripts of Cu/Zn-SOD and APX have also been reported in pigeon pea (Kumutha et al. 2009; Sairam et al. 2009b).

RT-PCR amplified Cu/Zn-SOD cDNAs of V. luteola, T 44, and Pusa Baisakhi were cloned and sequenced. Partial coding sequences of about 400 bp were obtained for Cu/Zn-SOD and compared with Pisum sativum Cu/Zn-SOD (GenBank Acc. No. ABQ189165) using BLAST tool and CLUSTAL W (1.83) multiple alignment. In this comparison, V. luteola, T 44, and Pusa Baisakhi showed 100, 99, and 98% homology, respectively, with Pisum sativum. This confirmed the cloned product is Cu/Zn-SOD. Conserved domains were identified using ‘PROSITE’, and the partial amino acid sequence of Cu/Zn-SOD showed two conserved domains: (1) SOD_Cu_Zn_1, Cu/Zn SOD metal binding site signature (PS 00087) (amino acid residues 33–43 in V. luteola; 36–46 in T 44 and Pusa Baisakhi), where H (Histidine) residues are the copper ligands. (2) SOD_Cu_Zn_2, Cu/Zn SOD metal binding site signature (PS 00332) (amino acid residues 73–92 in V. luteola; 76–95 in T 44 and Pusa Baisakhi) (Fig. 5).

Fig. 5
figure 5

Clustal W (1.83) multiple sequence alignment and comparison of deduced amino acid sequences of cyt.-Cu/Zn-SOD in V. luteola and V. radiata genotypes T 44 (tolerant) and Pusa Baisakhi (susceptible) with Pisum sativum (GenBank Acc. No. ABQ189165) (*shows conserved nucleotides; dark/bold letters show nucleotide polymorphisms)

Full size image

Similarly, partial coding sequences of about 650 bp were obtained for cytosolic APX gene and compared with Vigna unguiculata cyt.-APX (GenBank Acc. No. U 61379) using BLAST tool and CLUSTAL W (1.83) multiple alignment. The three genotypes showed more than 90% similarity with V. unguiculata APX gene sequence. The deduced amino acid sequence of the three genotypes (Fig. 6) belonged to cl00196: plant peroxidase super-family, which catalyzes a multistep oxidative reaction involving hydrogen peroxide as the electron acceptor. Along with animal peroxidases, these enzymes belong to a group of heme-dependent peroxidases containing a heme prosthetic group (ferriprotoporphyrin IX) All the three sequences contained the active site of plant peroxidase and substrate binding site.

Fig. 6
figure 6

Clustal W (1.83) multiple sequence alignment and comparison of deduced amino acid sequences of cyt.-APX in V. luteola and V. radiata genotypes T 44 (tolerant) and Pusa Baisakhi (susceptible) with Vigna unguiculata (Acc. No. U 61379) (*shows conserved nucleotides; dark/bold letters show nucleotide polymorphisms)

Full size image

The expression of β-tubulin did not vary under control or waterlogged condition in all the three genotypes. β-Tubulin, is component of a heterodimeric protein composed of two closely related 55 kDa proteins called α and β tubulin. The sequences of these genes are highly conserved throughout the eukaryotic kingdom. The expression of tubulin is not affected by environmental conditions, and therefore, β-tubulin was used as an internal control.

It has been suggested by various studies that increase in antioxidative enzymes activity during water-logging is primarily to scavenge post hypoxia ROS generation. Monk et al. (1987) were one of the first workers to report a continuous increase in SOD activity in rhizomes of Iris pseudocorus under water-logging stress. The results obtained from this study suggest that the increase in antioxidative enzymes during water-logging may also be to scavenge the build up of oxidative stress during water-logging. However, plants suffer heightened oxidative stress (ROS production) when they were shifted to aerobic condition and this explains overall higher antioxidant enzymes activity in comparatively tolerant genotypes V. luteola and T 44 during water-logging and recovery as compared to susceptible genotype Pusa Baisakhi. The increases in the activity of various antioxidative enzymes under water-logging/flooding have also been reported by other studies, viz., SOD (Biemelt et al. 2000), APX (Biemelt et al. 1998), GR and CAT (Ushimaru et al. 1997).

From the results it is obvious that in spite of apparent oxygen deficiency during water-logging, wild species V. luteola and V. radiata genotype T 44 experienced increase in ROS production under hypoxic condition, though less than the control and recovered plants. Role of hypoxia induced O2 ·−/H2O2 as a positive signaling molecule responsible for the induction of defense genes is confirmed by the observation that productions of O2 ·−/H2O2 were significantly more in tolerant genotypes (V. luteola and T 44), which also showed higher expression of Cu/Zn-SOD and APX, as well as activity of antioxidant enzymes than susceptible genotype Pusa Baisakhi.