Detection of S-Nitrosothiol and Nitrosylated Proteins in Arachis hypogaea Functional Nodule: Response of the Nitrogen Fixing Symbiont

To detect the presence of NO, ROS and RNS in nodules of crack entry legumes, we used Arachis hypogaea functional nodule. The response of two cognate partner rhizobia was compared towards NO and GSNO using S. meliloti and Bradyrhizobium sp NC921001. ROS, NO, nitrosothiol and bacteroids were detected by fluorescence microscopy. Redox enzymes and thiol pools were detected biochemically. Nitrosothiols were found to be present but ROS and NO were absent in A. hypogaea nodule. A number of S-nitrosylated proteins were also detected. The total thiol pool and most of the redox enzymes were low in nodule cytosolic extract but these were found to be high in the partner microorganisms indicating partner rhizobia could protect the nodule environment against the nitrosothiols. Both S. meliloti and Bradyrhizobium sp NC921001 were found to contain GSNO reductase. Interestingly, there was a marked difference in growth pattern between S. meliloti and Bradyrhizobium sp in presence of sodium nitroprusside (SNP) and S-nitrosoglutathione (GSNO). Bradyrhizobium sp was found to be much more tolerant to NO donor compounds than the S. meliloti. In contrast, S. meliloti showed resistance to GSNO but was sensitive to SNP. Together our data indicate that nodule environment of crack entry legumes is different than the nodules of infection mode entry in terms of NO, ROS and RNS. Based on our biochemical characterization, we propose that exchange of redox molecules and reactive chemical species is possible between the bacteroid and nodule compartment.


Introduction
Nitric oxide (NO) is a small lipophilic molecule with immense physiological and pathological roles in biological system [1]. NO is involved in a number of physiological processes in plants [2,3]. Presence of nitric oxide synthase activity was shown in roots and nodules of Lupinus albus [4]. Despite the completion of a plethora of plant genome sequences, genes encoding the NO synthesizing enzymes known as nitric oxide synthases (EC 1.14.13.39) have not been identified in plants. NO can be produced by other sources, such as xanthine oxidoreductase, peroxidase and cytochrome P450 [5]. Although descriptions of NO-mediated processes are accumulating, the plant signaling pathways governed by NO are still largely unknown [6,7]. NO regulates the activity of numerous proteins by S-or metal nitrosylation and tyrosine nitration. In plant systems, S-nitrosylation of proteins are found to contribute to gene regulation [8][9][10][11], modulates phytohormonal signaling [12] and cell death [13][14][15]. Recently, glutamine synthetase has been found to be a molecular target of nitric oxide in root nodules of Medicago truncatula where it is regulated by tyrosine nitration [16]. However, the biological significance of protein nitration is still not clearly understood.
Arachis hypogaea (peanut or groundnut), a well-known member of the Leguminosae family, subfamily Papilionoideae, tribe Aeschynomeneae is an important oil seed crop. The interaction between legume plants and some soil bacteria, collectively called rhizobia, leads to the establishment of symbiotic relationships, characterized by the formation of new differentiated root organs called nodules which provide a niche for bacterial nitrogen fixation. Arachis genus consists of 22 species among which 9 species are reported to be nodulated [17,18]. The distinctive features of aeschynomenoid nodules have been studied in depth [19][20][21][22]. Arachis roots are distinctly different from other root systems in lacking root hairs. However, tufted clusters or rosettes of multicellular hairs are found in young roots in the junction of the taproot and lateral root of Arachis which are thought to be important for bacterial invasion [23][24][25]. The penetration occurs through epidermal breaches i.e. crack at sites where lateral roots protrude in A. hypogaea [26]. Nod factor has been identified from Arachis-Bradyrhizobium sp interaction [27] but it is not yet known whether Arachis sp like the other members of the aeschynomenoid/dalbergioid tribe, can undergo successful nodulation in the absence of nod factor [28]. Another mode of infection of legumes by rhizobia may occur through root hair invasion [29] as for e.g. in M. trancatula infected by Sinorhizobium meliloti. Transient presence of NO was observed in Lotus japonicus [30] and Medicago sativa [31] roots within a few hours after infection by their cognate symbiont. NO is found to be produced in infection threads and nodule primordia of M. trancatula [32]. NO bound to leghemoglobin (Lb) as nitrosyl hemoglobin (LbNO) was detected in nodules of soybean infected by Bradyrhizobium japonicum [33]. Moreover, Baudouin and colleagues clearly showed the presence of NO in functional nodules of M. trancatula infected by S. meliloti [34]. Pii and coworkers confirmed this observation and extended it to M. sativa [35]. Recently, it has been shown that NO is involved in the regeneration of energy in M. truncatula functioning nodules [36]. In contrast to all that has been carried out in nodules of legumes where infections occur through root hair invasion, the response of nitric oxide in nodules evolved through crack mode entry has been poorly documented. In addition, how rhizobia cope with the presence of NO in nodules and what role is played by the bacterial NO response in the interaction with the host plants are not clearly known.
To detect the presence of NO, ROS and RNS in nodules of crack entry legumes, we used A. hypogaea functional nodules. The response of rhizobia towards RNS was studied in Bradyrhizobium sp NC921001 and S. meliloti. In this study we report the presence of Snitrosothiols, not nitric oxide in the functional A. hypogaea nodules. Bradyrhizobium sp was found to be much more tolerant to NO donor compounds than the S. meliloti. Our biochemical studies showed that the partner symbiont might have some contributions to restore the cellular redox environment against S-nitrosothiols. We also report the presence of S-nitrosylated proteins in functional A. hypogaea nodules suggesting importance of S-nitrosylation in NO dependent signal transduction in nodule environment.

Absence of Free Nitric Oxide and ROS in A. hypogaea Nodules
The presence of nitric oxide has been reported in 10, 20 and 30 day old nodules of M. trancatula and 40 day old functional nodules of M. sativa using DAF-2DA by two different laboratories [34,35]. To detect the presence of nitric oxide in A. hypogaea functional nodules of 20, 40 and 80 day, nodule sections were stained with cell permeable NO specific probe 4, 5-diaminofluorescein diacetate (DAF-2DA) [37]. M. sativa nodules of 143 day collected from agricultural firm were used as a positive control. A. hypogaea nodule sections did not show any NO specific fluorescence ( Figure 1 Panel A, C and E). However, nodule sections of M. sativa showed significant positive NO specific fluorescence (Figure 1 Panel G) indicating that nitric oxide is formed in M. sativa nodules of different ages (40 day [35] and 143 day). This result was confirmed by treating nodule slices with the NO scavenger 2-(4carboxyphenyl) -4, 4, 5, 5-tetramethyl imidazoline-1-oxyl-3-oxide (cPTIO) ( Figure S1). There was no ROS specific fluorescence in nodule sections of A. hypogaea (20, 40 and 80 day) using Dichlorofluorescein diacetate (DCF-DA). Nodule sections of M. sativa (143 day) also showed similar observations ( Figure S2).

Presence of Nitrosothiols and Bacteroids in A. hypogaea Nodules
Alexafluor 488 is Hg-link phenylmercury compound, which forms stable thiolates with free sulfhydryls as well as other thiol moieties including nitrosylated thiols [38]. Figure 2 and Figure 3 represent the presence of nitrosothiols in nodule sections of A. hypogaea and M. sativa stained with alexafluor 488. Presence of nitrosothiol was confirmed by using N-ethylmaleimide (L-NEM) as a thiol blocking agent. There was ,23% decrease in fluorescence intensity in A. hypogaea 20 day old nodule sections following L-NEM addition ( Figure 4) indicating free thiol content was low in it. The decrease in fluorescence intensity was found to be negligible in 40 day (,2.8%) and 80 day old nodule sections (,5.7%) of A. hypogaea following L-NEM treatment. A further decrease in fluorescence intensity was observed in 20 day (,38%), 40 day (,22%) and 80 day (,53%) old nodule sections of A. hypogaea when they were reduced with ascorbate followed by blocking with L-NEM indicating the presence of significant amount of nitrosothiols. On the other hand, there was a ,48% decrease in fluorescence intensity following the addition of L-NEM in nodule sections of M. sativa indicating high free thiol content present in it. The mean intensity of green color fluorescence was plotted to reproduce the intensity curve using Image J software ( Figure 4). This figure clearly shows the differential fluorescence intensity profile of alexafluor 488 for thiols and nitrosothiols for both the plants' nodules.
Saville assay is specific for quantification of nitrosothiols in terms of p mole nitrite mg 21 protein. Figure 5 represents nitrosothiol profile of 22, 40, 50, 65 and 78 day old nodule extracts of A. hypogaea. Nitrosothiol content was changed significantly in different crude nodule extracts of different time points ranging from 10.860.3 to 16.4761.61 p mole nitrite mg 21 proteins. We used a fluorescent nucleic acid dye of the Syto family, which has been previously reported as being efficient for bacteroid detection in nodules [39]. In regard to the spectral characteristics of the different Syto molecules, we chose the Syto83 dye, the fluorescent emission of which can be fully discriminated from the alexafluor 488 fluorescent emission. Nodule slices were cotreated with Syto83 and alexafluor 488 dyes and were subsequently observed by confocal microscopy. As shown in Figure 6, a strong correlation between the cells that contained bacteroids and the cells that were producing nitrosothiol was observed. Thus it can be concluded that bacteroids are exposed to nitrosothiols within the nodule environment.

Presence of Nitrosylated Proteins in Functional Nodules of A. hypogaea
S-nitrosylation is emerging as a key redox based post translational modification during plant immune function. The modulation of cellular SNO level as well as S-nitrosylation was found to regulate the accumulation of the plant immune activator, salicylic acid (SA) and expression of SA-dependent genes by TGA1, the transcriptional regulator [40]. Although there is increasing evidence of finding nitrosothiols in several plant samples [41,42], the presence of nitrosothiols in A. hypogaea nodules is still novel information in legumes. Figure 7 represents S-nitrosylated protein profile of A. hypogaea nodule extracts. Lane 1, 2 and 3 represent reduced and biotinylated samples of 20, 40 and 80 day old nodule extracts. Lane 4 represents the marker. Lane 5, 6 and 7 represent the endogenously biotinylated protein profiles corresponding to lanes 1, 2 and 3. In the bacteroid free nodule extract, mixture of proteins with free cysteine thiols and proteins with cysteine Snitrosylation could be present. So first free cysteine thiols were blocked by S-methylthiolation with MMTS (methyl methanethiosulfonate); then S-nitrosylated cysteines or SNOs were converted to thiols via transnitrosation with ascorbate; then the nascent thiols were labeled in situ by S-biotinylation with N-[6-(Biotinamido)hexyl]-39-(29-pyridyldithio)propionamide (biotin-HPDP), a reactive mixed disulfide of biotin. So test of endogenously biotinylated protein samples with the probe is essential to detect the actual profile of S-nitrosylated proteins in the sample. Endogenously biotinylated proteins probably vary with the age of the plant. That is why only 40 day old sample represents hybridization with the probe. In other samples the level of endogenously biotinylated proteins could not be detected under our experimental conditions indicating the level of endogenously biotinylated proteins were low. We also checked for protein tyrosine nitration in nodule extracts. We did not observe any significant level of nitrated proteins in nodules of A. hypogaea ( Figure S3).

Presence of Low Thiols and Redox Active Enzymes in Functioning Nodules of A. hypogaea
In order to gain a precise view of the redox environment of A. hypogaea functional nodules, we determined the total thiol content in A. hypogaea nodule. The total thiol content was very low in comparison to the total thiol contents in nodules reported in other plants which encourage infection mode of entry [43,44]. We also measured the glutathione reductase, catalase, ascorbate peroxidase (APx) (EC 1.11.1.11) and superoxide dismutase (SOD) activity in 20, 40 and 80 day old A. hypogaea nodule extracts (Table 1). In our system we could only measure the GR activity in 20 and 40 day old nodule extracts. GR activity was found to have increased significantly in 40 day old nodules (8.360.49 n mole min 21 mg 21 ) compared with the 20 day old nodules (160.15 n mole min 21 mg 21 ). In 80 day old nodule extracts, the expression level of GR was probably very low which could not be detected by our assay system. APx activity did not change much in 20, 40 and 80 day old nodules ranging from 0.3860.007 to 0.560.03 m mole min 21 mg 21 proteins. Interestingly, the specific activity of SOD was found to be high in all three time points ranging from 90.5365.38 to 133.2369.66 m mole 21 min 21 mg 21 compared to the reported SOD activity (49 m mole min 21 mg 21 ) of soybean nodules [45]. On the other hand, the catalase activity was found to be lower in 20, 40 and 80 day old nodules (sp. activity ranging from 12.5860.18 to 17.3561.36 m mole min 21 mg 21 ) compared to the specific activity of the crude soybean nodule extracts (172 m mole min 21 mg 21 ) of 30 day after their appearance in the roots [45]. Next we measured the Glutathione peroxidase (GPx) (EC 1.11.1.9) and S-nitrosoglutathione (GSNO) reductase (EC 1.1.1.284) enzyme activities in nodule extracts. But GPx and GSNO reductase activities could not be detected in A. hypogaea nodule extracts. GSNO reductase activity was reported only in Arabidopsis thaliana [46] and pea leaves [47].

Bacterial Redox Active Enzymes and GSNO Reductase Protect them Against Nitrosative Stress
Our microscopic data clearly showed that bacteroids were exposed to nitrosothiols. So we wanted to check the response of the symbiont against the reactive nitrogen species. For this we used sodium nitroprusside (SNP) and S-nitrosoglutathione (GSNO) as the NO donating compound to create nitrosative stress on these bacteria. We tested the capacity of GSNO and SNP to inhibit the growth of these two organisms in aerobic condition. Figure 8 represents the growth curves of S. meliloti 1021 and Bradyrhizobium sp in the presence of different concentrations of GSNO and SNP. There was no significant growth inhibition up to 1 mM GSNO in S. meliloti 1021. The growth of S. meliloti 1021 was clearly affected at 0.5 mM, 1 mM and 3 mM SNP. The growth of Bradyrhizobium sp was not significantly affected in the presence of SNP. However, Bradyrhizobium cells were sensitive to GSNO. There was a clear lag period in the growth curve in presence of 3 mM GSNO concentrations. However, cells were capable to reach an absorbance value close to the control set after six hours. In eukaryotes and prokaryotes, the tripeptide glutathione (GSH) in its reduced state acts as an important constituent of cellular first line of defense against oxidative stress and helps to maintain intracellular redox balance. So we investigated the effect of nitrosative stress on cellular glutathione level both in Bradyrhizobium sp and S. meliloti (Table 2). We measured total glutathione (GSH + GSSG), oxidized glutathione (GSSG) and calculated reduced glutathione (GSH) as markers of cellular glutathione status following 1 mM and 3 mM SNP stress for 30 min. There was a , 3.3 fold increase in total thiol in S. meliloti 1021 following 3 mM SNP treatment compared to the control set. However, in Bradyrhizobium sp fold increase in total thiol was much less i.e. ,1.76 following 1 mM SNP exposure under similar experimental conditions. The basal level total thiol GSH was higher (, 5.84 fold) in S. meliloti than Bradyrhizobium sp. We could not detect the GSSG and GSH level in control and 1 mM SNP treated Bradyrhizobium sp under our experimental conditions [48]. However, detectable amount of GSSG and GSH were found following 3 mM SNP exposure for 30 min in Bradyrhizobium sp. There was a two fold increase in GSSG content following 3 mM SNP treatment in S. meliloti. But it did not affect the cellular reducing environment as there was a 3.36 fold increase in GSH level in S. meliloti following 3 mM SNP treatment for 30 min which resulted in 1.76 fold increase in GSH/GSSG ratio.
To investigate the cellular capacity to restore the GSH content after nitrosative stress, GR and GPx activities were assessed using 1 mM and 3 mM SNP treatment under similar experimental conditions. Interestingly, basal level GR was much higher (, 3 fold) in Bradyrhizobium sp compared to S. meliloti ( Table 3) which indicated that cells are well equipped for recycling of GSSG, generated under oxidative or nitrosative stress. There was a little change in GR in Bradyrhizobium sp following 1 mM and 3 mM SNP exposure.
The basal level catalase of Bradyrhizobium sp was 13.5 fold higher than S. meliloti. However, there was no significant change in catalase level in Bradyrhizobium sp following SNP treatment. Similar result was obtained in S. meliloti. Interestingly GPx level was also high in Bradyrhizobium sp compared to S. meliloti (Table 3). Both Bradyrhizobium sp and S. meliloti possess GSNO reductase activity. The basal level GSNO reductase activity was almost same in both the organisms but it was induced (, 24% increase) in presence of 3 mM SNP in both the organisms.

Discussion
Symbiotic interactions between legumes and rhizobia represent an original physiological context that results from a combination of developmental and infectious events. The importance of NO during the interaction of plants with pathogens brought us to characterize the occurrence of this molecule during the infection of the model legume A. hypogaea by its bacterial partner Bradyrhizobium sp. There is little information available on the crack entry legumes regarding the nitric oxide production and its effect on the partner microorganisms [49]. In this study, we used a biochemical approach to characterize the nodule environment of A. hypogaea in terms of reactive nitrogen species and the effect of nitrosative stress on the symbiont. Furthermore, we first report the presence of nitrosothiols and nitrosylated proteins in nodules of A. hypogaea. It is evident from our fluorimetric as well as biochemical study that RSNOs are present in a substantial amount in A. hypogaea nodules. It has been reported that RSNOs formed from low molecular weight biological thiols (L-cysteine and glutathione) are degraded in the presence of superoxide [50] producing nitrite and nitrate in the reaction mixture. ROS detoxifying enzymes like SOD, APx and catalase were found to become high in A. hypogaea nodule environment. That is why we did not observe any superoxide in the nodule environment. Absence of ROS and nitric oxide as determined by fluorescent probes distinguished A. hypogaea nodule as a separate group of plant where the signaling molecules related to the symbiotic association are different. There are multiple pathways of NO production thought to exist in plant [7,51]. These route(s) may control NO production but the bioavailability of NO depends on its turnover. NO can be metabolized to nitrate by nonsymbiotic hemoglobins. NO also reacts with GSH to form GSNO, which can release NO or functions as a transnitrosylating agent and can be degraded by GSNO reductase. NO can also react with superoxide to form peroxynitrite. So a network of RNS transformation and turnover balance the bioavailability of these signal compounds.
Flavohemoglobins (Hmp) are proteins involved in NO detoxification in many bacterial species. In S. meliloti, the role of Hmp protein has been demonstrated by the team through the production of hmp-null mutants and hmp-overexpressing (hmp++) strains [49]. RSNOs can be decomposed by various pathways that may include thiol [52], glutathione peroxidase [53] and GSNO reductase [54,55]. GSNO is equilibrated with other RSNO through transnitrosation. As a result GSNO reductase can  indirectly regulate the concentration of RSNOs in vivo. However, GSNO reductase is absent in A. hypogaea nodule extract. In our growth curve study we found that both S. meliloti and Bradyrhizobium sp are affected in presence of the graded concentrations of GSNO. Interestingly, both Bradyrhizobium sp and S. meliloti were found to contain substantial level of GSNO reductase activity. So the presence of GSNO reductase activity in bacteria might reduce the actual concentrations of GSNO in which they were exposed. So the bacterial GSNO reductase might act as potential defense enzyme along with other redox reactive enzymes like GR, GPx, and catalase against RSNOs in A. hypogaea nodules.
In the biological system, S-nitrosylated proteins can be formed through a number of mechanisms that include transnitrosylation, metalloprotein catalyzed reactions, or reaction with electrophilic nitrogen oxide species. But the actual mechanisms of S-nitrosylation as well as their functions in A. hypogaea nodules are not known. Together our data indicate that nodule environment of crack entry legumes is different than the nodules of infection mode entry in terms of NO, ROS and RNS. Based on our biochemical characterization, we propose a scheme (Figure 9) for the exchange of redox molecules and reactive chemical species between the bacteroid and nodule compartment. In future, the identification of nitrosylated proteins could provide new insights into the redox based machinery that controls the nodule environment as well as effective symbiosis in A. hypogaea nodule.

Chemicals
All the reagents are of highest purity and purchased from Sigma Chemical Co., (St. Louis, MO USA) unless otherwise stated.

Plant Material
A. hypogaea JL 24 (ICG 7827) was a generous gift from ICRISAT, Patancheru 502 324, Andhra Pradesh and were cultivated in the village fields of Gopalnagar, P.O. -Malighati, Paschim Medinipur, West Bengal, state of India as seasonal crop. No specific permits were required for the described field studies. The location is not protected in any way and it is confirmed that the field studies did not involve endangered or protected species.
Nodules were collected from the roots of 22, 40 and 80 day old plantations. M. sativa nodules of 143 day were collected from field.

Detection of Thiols and/or Nitrosothiols by Fluorescence Microscopy
Thiols and/or nitrosothiols were detected using the fluorescent reagent Alexa fluor 488 Hg-link phenylmercury [42]. A. hypogaea as well as M. sativa nodule sections were incubated with 10 mM Alexa fluor 488 Hg-link phenylmercury (Molecular Probes) at 25uC for 2 h in darkness washed thrice and images were captured with

Blocking of Nitrosothiol
Three sets were prepared for the experimental study. In one, the nodule cross sections were treated with 50 mM N-ethyl-maleimide (NEM) prepared in ethanol and was incubated at 25uC for 2 h in darkness. In the other, the nodule cross sections were incubated with 20 mM ascorbate prepared in HEN buffer for one hour at room temperature followed by similar NEM treatment. The last set prepared as control was only treated with buffer without ascorbate and NEM. Then all the sections were stained with 10 mM Alexa fluor 488 Hg-link phenylmercury (Molecular Probes) and images were captured as stated previously.

Bacteroid Detection
Bacteroids were detected together with RSNO in fresh nodule slices. The nucleic acid dye Syto83 (Invitrogen) was added with alexafluor 488 both at a final concentration of 10 mM. The sections were incubated for 1 hour followed by three washings. Syto83 emission was recorded using an upper 560 band-pass filter upon excitation at 543 nm with a helium laser. The bacteroids of M. sativa were stained only with Alexa fluor 488. The emission of Alexa fluor 488 Hg-link phenylmercury was recorded using 505 to 570 band-pass filter upon excitation at 495 nm.

Preparation of Bacteroid Free Nodule Cytosolic Extract
Bacteroid free nodule extracts were prepared as stated by Oehrle et al. (2008) [56] with slight modifications. Nodules were removed from the roots and were immediately placed on ice. 1 g of nodules were homogenized in a pre-chilled mortar and pestle with 2 ml of pre-chilled extraction buffer that consists of 100 mM Tris-HCl of pH 8.8 with 17% (w/v) sucrose, 100 mM DTT, 1 mM PMSF, 10 mM EDTA, 20 mg ml 21 pepstatin A, 20 mg ml 21 leupeptin and 20 mg ml 21 aprotenin. To precipitate the plant phenolic compounds, 0.33 g of insoluble Poly(N-vinyl -2pyrrolidone) (PVP) was used for per g of nodule. Homogenized nodules were filtered by passing through four layers of cheeseclothes those were pre-moistened with extraction buffer. Next the filtrates were centrifuged at 400 g for 10 minute. The supernatants containing the soluble plant proteins and bacteroids were centrifuged at 8,000 g for 20 min to pellet the bacteroids and the clear supernatant was used for enzymatic assay and Snitrosylation determination.

Preparation of Bacterial Cell Free Extract
Bradyrhizobium sp and S. meliloti 1021 were grown up to mid log phase at 30uC. Sodium nitroprusside (SNP) was added to mid log phase cell in the final concentration of 1.0 and 3.0 mM and was incubated at 30uC for 30 minute. Next cells were pelleted down by centrifugation at 3,400 g for 5 minute at 4uC. Cell pellets were washed thrice with pre chilled phosphate buffer saline (PBS) (NaCl Table 2. Quantification of thiol pool in S. meliloti 1021 and Bradyrhizobium sp after 30 minute stress with SNP in media.

S. meliloti
Bradyrhizobium sp S. meliloti cells were grown to mid log phase at 30uC in TY media whereas Bradyrhizobium sp cells were grown similarly in YEM media. Sodium nitroprusside (SNP) was added to mid log phase cell in the final concentration of 1.0 and 3.0 mM and incubated at 30uC for 30 minute. Cells were harvested, lysed and cell free crude extracts were subjected to quantification of total thiol, oxidized glutathione, reduced glutathione and ratio for reduced to oxidized glutathione. Results are expressed as mean 6 SD, for n = 3 experiments. P#0.01, using one-way ANOVA. doi:10.1371/journal.pone.0045526.t002 Table 3. Assessment of Glutathione reductase (GR), Catalase, Glutathione peroxidase (GPx) and GSNO reductase activity in S. meliloti 1021 and Bradyrhizobium sp after 30 minute stress with SNP in media. Lysozyme solution was added to the slurry at a final concentration of 1 mg ml 21 and again was kept in ice for 15 minute. Then the cells were sonicated using 6 pulses each for 10 seconds at each 10 second intervals at highest amplitude percentage of in Hilscher UP50H sonicator. Total procedure was performed on ice to prevent the temperature increase and was centrifuged at 13,000 g for 15 minute at 4uC. The supernatants were collected and were used for enzyme assay.

Biochemical Measurement of S-Nitrosothiol Content with the Saville Assay
The S-nitrosothiol content was determined by the Saville method [57]. For the analysis of the S-nitrosothiol content, three sets of solutions were prepared. Solution A consisted of 1% sulfanilamide dissolved in 0.5 M HCl, solution B contained 1% sulfanilamide and 0.2% HgCl 2 in 0.5 M HCl and solution C was prepared by dissolving 0.02% N-(1-naphthyl) -ethylendiaminedihydrochloride in 0.5 M HCl. Two protein samples were prepared in 100 mM phosphate buffer (500 ml). Solution A (250 ml) was then added to one of the samples, and the same amount of solution B was added to the other. After 5 minute, when the formation of the diazonium salt was complete, 250 ml of solution C were added to each of the two samples. After 5 minute, color formation of the azo dye was complete and the absorbance of 540 nm was recorded spectrophotometrically. The absorbance at 540 nm of the reaction with solution A was for the presence of free nitrite in plant samples and the absorbance at the same wavelength with solution B was for the presence of free nitrite and nitrosothiol in plant samples. So the actual nitrosothiol content in plant sample was the difference in absorbance between solution A and solution B. Finally the nitrosothiol content was determined in terms of nitrite from the standard curve of nitrite.

Preparation of GSNO
Preparation of GSNO solution was performed according to Hart (1985) [58] with slight modifications. 1 M solution of NaNO 2 was prepared in distilled water and 1 M solution of GSH was prepared in 1 N HCl. These two solutions were mixed on ice (1:1, v/v) to obtain GSNO solution and GSNO concentration was determined using e 335 nm = 586 M 21 cm 21 .

Bacterial Growth Study in Presence of SNP and GSNO
Mid log phase cultures of bacterial cells were used as inoculum following dilution in fresh media to O.D. 590 = 0.13-0.15 in the presence of graded concentrations of SNP or GSNO. Cell growth was monitored by measuring the cell density spectrophotometrically at 590 nm. SNP and GSNO were kept in 0.1 M Tris buffer, pH-7.4 at room temperature for three days in presence of light for decomposition. In all the above experiments, control sets were performed using decomposed NO donors.

Assessment of Enzymatic Status
Glutathione reductase and catalase. Glutathione reductase (GR) and catalase activities were measured in the cytosolic fractions of plant nodules, Bradyrhizobium sp and S. meliloti. According to Carlberg et al. (1975) [59] NADPH oxidation was followed spectrophotometrically at 340 nm over 2 min to assay Glutathione Reductase. Catalase activity was measured according to Aebi (1984) [60] following the H 2 O 2 degradation spectrophotometrically at 240 nm over 2 minute.
GSNO reductase. GSNO reductase activity of crude cell-free extracts were determined according to Liu et al. (2001) [54] by measuring the decrease in absorbance at 340 nm due to utilization of NADH. The specific activity was calculated using the extinction coefficient of NADH (6.22 mM 21 cm 21 ) and was expressed in n mol min 21 mg 21 protein.
Glutathione peroxidase and ascorbate peroxidase. The Glutathione peroxidase (GPx) activity of crude cell-free extracts were determined according to Wendel (1981) [61] by measuring the change in absorbance at 340 nm due to utilization of NADPH. Ascorbate peroxidase (APx) assay was performed according to Miyake et al. (1992) [62]. The H 2 O 2 dependent oxidation of ascorbate was followed by monitoring the decrease in absorbance at 240 nm and the amount of consumed ascorbate was determined by using absorption coefficient 2.8 mM 21 cm 21 . One unit of APx was defined as the enzyme that oxidized 1 m mol of ascorbate per minute at 25uC.
Superoxide dismutase. The superoxide dismutase (SOD) assay was performed using xanthine oxidase (XO)/NBT system according to Ukeda et al., (1997) [63] with slight modification. The reaction was performed in 0.5 ml of 50 mM potassium phosphate buffer that consisted of 0.125 mM xanthine, 0.125 mM EDTA, 0.03 mM NBT and SOD solution. The reaction was initiated by the addition of XO. The change of absorbance was monitored at 560 nm at 25uC for 2 minute. One unit of SOD was defined as the amount of enzyme that caused a 50% decrease in SOD -inhibited NBT reduction.

Total Glutathione (GSH + GSSG), Reduced Glutathione (GSH) and Oxidized Glutathione (GSSG) Content Measurement
The measurements of these three parameters were performed according to the method described by Akerboom and Sies, (1981) [48]. The crude cell free extracts were first deproteinised and were centrifuged at 5,000 g for 5 minute. Then cell free extracts were neutralized with KOH containing 0.3 M HEPES. These neutralized samples were then centrifuged at 5,000 g for 5 minute and were utilized for determination of the above mentioned parameters. Total glutathione (GSH+2GSSG) concentrations were estimated using glutathione reductase dependent DTNB reduction method. The reactions were performed in 100 mM potassium phosphate buffer of pH 7.0 containing 1 mM EDTA, 0.12 U GR, 0.2 mM NADPH, 0.063 mM DTNB with a total volume of 500 ml. The same neutralized samples were used for GSSG estimation after addition of 2-vinyl pyridine (Lancaster Chemicals, England) (50:1, v/v) and were incubated for 1 hour. The contents of reduced glutathione were determined from the difference between total and oxidized glutathione contents of the samples and both GSH and GSSG concentration were expressed in n mol mg 21 protein.

Determination of Protein S-nitrosylation
The detection of protein S-nitrosylation was based upon the modified Biotin switch technique described by Jaffrey et al. (2001) [64]. Each 400 mg proteins from bacteroid free nodule extracts of 20, 40 and 80 day were taken and were processed according to the method described in the S-nitrosylated protein detection assay kit of Cayman Chemicals, Ann Arbor, Michigan. 5 mg proteins of each processed sample were mixed with Laemmli buffer and were incubated in a boiling water bath for 5 min followed by 5 min on ice prior to loading to covalently bind the biotin label. Next the samples were separated in 10% SDS-polyacrylamide gel electrophoresis. Then the proteins were blotted on PVDF membrane. The membrane was blocked at 4uC for overnight. Next day the blot was washed with TBS for twice and incubated with 1:200 dilution of S-nitrosylation Detection Reagent I (HRP) in TBS for 1 hour at room temperature. Then the membrane was washed extensively with TBS and immunopositive bands were visualized by using chemilluminiscent reagent (Pierce ECL Western Blotting Substrate) as directed by the manufacturer.

Detection of Bacteroid Free Nodule Protein Nitration
The proteins of nodule cytosolic extracts were separated with 10% SDS PAGE. Next the proteins were blotted on PVDF membrane and blocked overnight at 4uC. PVDF membrane containing proteins was then incubated for 60 min at room temperature with anti nitrotyrosine monoclonal antibody (Upstate, USA) at 1:2000 dilutions. Next the membrane was washed and was incubated with a HRP conjugated goat antimouse IgG antibody (Santa Cruz biotechnology) at 1:500 dilutions. After that membrane was washed extensively and the immunopositive spots were visualized by using chemilluminiscent reagent (Perkin Elmer, Boston, MA) as directed by the manufacturer.