Introduction

Sulfur (S) is an essential nutrient in all living organisms as it is involved in a variety of vital biological functions (Ulrich and Jakob 2019; Rouault 2019). For instance, oxidative phosphorylation (OXPHOS), a process central in ATP synthesis during aerobic respiration cannot operate in absence of S. This moiety is a component of most of the enzymes participating in the electron transport chain (ETC). Here, the Fe–S clusters mediate the transfer of electrons (e−) located within the reducing factors (NADH and FADH2) to O2 (Middaugh et al. 2005; Mailloux et al. 2007; Bignucolo et al. 2013; Thomas et al. 2016). The tricarboxylic acid (TCA) cycle that supplies the reducing factors for OXPHOS is ineffective if enzymes like aconitase, fumarase and α-ketoglutarate dehydrogenase (KGDH) are devoid of their S-containing reactive constituent. The latter contains lipoic acid, a moiety that requires S while the former two enzymes are dependent on Fe–S clusters (Jordan et al. 1999; Tseng et al. 2001; Cheng et al. 2006). The synthesis of amino acids such as cysteine and methionine necessitates the participation of S. Cysteine is also pivotal in maintaining the proper redox potential in a cell due to its involvement in the biosynthesis of glutathione, a key antioxidant. The homeostasis of the oxidized and reduced forms of this tripeptide is known to orchestrate the fine-reducing balance all cellular systems need in order to survive in an aerobic environment (Forman et al. 2009).

Numerous redox sensing networks that living systems rely on to gauge intracellular oxygen tension are dependent on cysteine and Fe–S containing biomolecules. Upon oxidation of the sulfhydryl (SH) group into disulfide S–S and other oxidized S moieties, proteins are able to relay this information into actionable physiological responses (Groitl and Jakob 2014). For instance, the transcription factor SOXR upon reacting with superoxide (O ·−2 ) is activated following the modification of SH in a cysteine residue (Kobayashi 2017). This conformational change enables this protein to interact with the promoter region in the genetic machinery responsible for triggering a set of proteins involved in the detoxification of reactive oxygen species (ROS), thus diminishing oxidative tension. As part of our investigation to decipher the metabolic networks aimed at the survival of living organisms subjected to abiotic stress (Mailloux et al. 2007; Alhasawi et al. 2015a, b; Auger et al. 2013), we have examined how the soil microbe Pseudomonas fluorescens adapts in an environment deficient in S. The microorganism was grown in a mineral medium with no added S in the presence of glutamine as the source of carbon and nitrogen.

Here we report that the oxidative stress evoked by S stress compels P. fluorescens to reprogram its metabolic pathways in order to limit the production of the pro-oxidant NADH and enhance the synthesis of NADPH, an antioxidant. The modulation of isocitrate dehydrogenase (ICDH-NADP and NAD dependent) coupled with the concomitant activation of NAD-kinase (NADK) is central in combatting the oxidative stress induced by S deficiency as it essentially results in the conversion of NADH into NADPH. This strategy also promotes a sharp decline in the TCA cycle and oxidation phosphorylation, two metabolic processes known for their ROS-generating attributes.

Materials and methods

Bacterial cultures and biomass measurement

Pseudomonas fluorescens was obtained from the American Type Culture Collection (ATCC 13525). It was grown in a defined mineral medium with glutamine (19 mM) as the sole source of carbon and nitrogen. The salt and trace metals were added as described in (Anderson et al. 1992; Aldarini et al. 2017). The control medium contained 0.8 mM of MgSO4 while the stress medium had no added sulfate. Instead MgCl2 (0.8 mM) was utilized. Media were dispensed in 200 mL aliquots in 500 mL Erlenmeyer flasks and autoclaved for 20 min at 121 °C. The cultures were inoculated with 1 mL of P. fluorescens preculture as previously described (Anderson et al. 1992) and grown on a Gyrotory® Water Bath Shaker Model G76 (New Brunswick Scientific) at ambient temperature. To measure cellular biomass, 10 mL of microbial culture was spun down at 12,000 × g for 20 min. The pellet was resuspended in 1 mL 0.5 M NaOH. Biomass was determined by measuring the amount of soluble proteins with a Bradford assay (Bradford 1976).

Cell fractionation and metabolic profiling

Bacteria cultures were spun down at 12,000 × g for 20 min. The pellets were washed with ~ 100 mL of 0.85% m/v NaCl and the samples were centrifuged again. Following suspension of the cellular pellet in 1 mL of Cell Storage Buffer (CSB) (50 mM Tris–HCl, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, pH 7.6), the bacteria was broken down with ultrasonic waves using an ultrasonic processor (Johns Scientific). Resuspended cells were sonicated for 15 s, five times, with a three minutes interval between each sonication in an ice-bath. Centrifugation at 110,000 × g for 3 h at 4 °C resulted in a soluble fraction (supernatant) and a membrane fraction (pellet). The membrane fraction was resuspended in 1 mL of CSB. The cell free extracts were stored at 4 °C for a maximum of 7 days and were used for analytical studies including High Performance Liquid Chromatography (HPLC) analysis, Blue Native Polyacrylamide Gel Electrophoresis (BN-PAGE) and enzymatic assays.

HPLC analysis was performed on a Waters 2695 Separation Module HPLC system with a Synergi 4 μm Hydro-RP 80A column (Phenomenex) with dimensions of 250 × 4.6 mm at a flow rate of 0.2 mL/min. A Waters 2478 Dual λ absorbance detector, set at wavelengths of 210 nm to identify organic acids and 254 nm to monitor nucleotides. The mobile phase consisted of 5% v/v acetonitrile, 20 mM KH2PO4 diluted in Milli-Q water, pH 2.9 at room temperature. The mobile phase was vacuum filtered using a Stericup® Vacuum Driven Disposable Filtration System. The funnel used for the filtration had a 0.22 μm Millipore Express® PLUS Membrane filter. To monitor the metabolites and appropriate standards, samples were prepared and run in the HPLC as previously described (Singh et al. 2009). Metabolite amounts were determined from the area under the curve of the corresponding peaks using the Empower software (Waters Corporation).

Blue native polyacrylamide gel electrophoresis (BN-PAGE) and enzyme activity

The activity of enzymes of the TCA cycle as well as enzymes involved in NADH/NADPH production was probed by BN-PAGE. In order to obtain optimal protein separation, gels containing a 4–16% linear gradient of polyacrylamide were cast with the MiniProteanTM2 gel system (Bio-Rad Laboratories) with 1.0 mm spacers. The final concentration of the protein samples prepared was 4 mg/L. For membrane proteins, a concentration of 1% (m/V) of η-dodecyl β-d-maltoside was added in order to solubilize the proteins. A mass of 60 μg of proteins was loaded into each well and the proteins were migrated under native conditions (50 mm є-aminocaproic acid, 15 mM Bis–Tris, pH 7.0 at 4 °C). The electrophoresis was carried out at 75 V and 15 mA with blue cathode buffer (50 mM Tricine, 15 mM Bis–Tris, 0.02% (m/v) Coomassie G-250, pH 7.0 at 4 °C) until the migration front reached the resolving gel, then the voltage and current were changed to 150 V and 25 mA respectively. When the migration was halfway through the gel, the blue cathode buffer was swapped for a colourless cathode buffer (50 mM Tricine, 15 mM Bis–Tris, pH 7.0 at 4 °C) to allow for better visualization.

Detection of enzymatic activity was performed using formazan precipitation. Gels were incubated in a reaction mixture containing substrates and cofactors utilized by the enzyme of interest as well as 0.4 mg/mL or iodonitrotetrazolium (INT) and 0.2 mg/mL of either phenazine methosulfate (PMS) or dichlorophenolindophenol (DCPIP). The reaction buffer contained 25 mM Tris–HCl and 5 mM MgCl2, pH 7.4 at room temperature. To prevent complete utilization of substrates in the experiments with membrane preparations, KCN (5 mM) was included (Han et al. 2012). Complex I activity was tested on the membrane fraction using 0.5 mM NADH, 0.4 mg/mL INT while Complex IV activity was visualized with 5 mg/mL diaminobenzidine, 13.3 mM of sucrose and 10 mg/mL of cytochrome C. Malic enzyme (ME) and malate dehydrogenase (MDH) were probed using 5 mM malate, 0.5 mM NADP, 0.2 mg/mL PMS and 0.4 mg/mL INT in the soluble and membrane fractions respectively. Fumarase activity was identified in the membrane fraction using 5 mM fumarate, 0.5 mM NAD, 1 unit of malate dehydrogenase, 0.2 mg/mL PMS and 0.4 mg/mL INT. Isocitrate dehydrogenase (ICDH)-NAD dependant was assessed in the membrane fraction while ICDH-NADP dependant was monitored in the soluble fraction using 5 mM isocitrate, 0.5 mM of NADP, 0.2 mg/mL PMS and 0.4 mg/mL INT respectively. The reverse ICDH was assessed in the membrane fraction with 5 mM α-ketoglutarate, 5 mM sodium bicarbonate, 0.5 mM of NADH, 0.2 mg/mL DCPIP and 0.4 mg/mL INT while in the soluble faction the reverse ICDH was probed with NADPH (0.5 mM) as the co-factor. Glutamate dehydrogenase was probed as described in (Aldarini et al. 2017). To further confirm the nature of the enzymes, the activity bands were excised and incubated in the reaction buffer with the appropriate substrates. This reaction mixture was then analysed by HPLC for the formation of the products and the utilization of the substrates (Alhasawi et al. 2015a, b, 2016; Auger et al. 2011; Lemire et al. 2008).

Enzyme activity analysis by spectrophotometry

Spectrophotometric analysis was used to quantify the activity of some selected enzymes by monitoring the presence or absence of NADH/NADPH at 340 nm. For enzymes of interest, 200 µg of protein was added to 2 mM of substrate and 0.5 mM of NAD/NADP or NADH/NADPH based on the enzyme in a volume of 1000 μL of the Reaction Buffer (25 mM Tris–HCl and 5 mM MgCl2 in water, pH 7.4 at ambient temperature). Proteins were added last and the spectrophotometer was blanked before adding the proteins. After the addition of proteins, the mixture was allowed to react and the change in absorbance was recorded after 60 s to measure the reaction rate. For KGDH, 2 mM of α-ketoglutarate and 0.5 mM of NAD was used. For ICDH-NAD, 2 mM of isocitrate and 0.5 mM of NAD were used. For GDH-NAD, 2 mM of glutamate was used with 0.5 mM of NAD+. For Reverse-ICDH-NADH, 2 mM of α-ketoglutarate, 2 mM of sodium bicarbonate and 0.5 mM of NADH was used.

Superoxide dismutase and catalase assays

The activity of catalase (CAT) was determined with the aid of the reagent p-anisidine, and the absorbance at 458 nm was monitored. 200 µg of protein equivalent from control or sulfur deprived proteins were incubated with 15 mM hydrogen peroxide (H2O2). 10 mM p-anisidine was added immediately in a final volume of 1.0 mL and the absorbance was measured after 60 min. Blanks were prepared similarly lacking the peroxide component. The activity of the control enzyme was normalized to 100% and the activity of the enzyme in the absence of sulfur was quantified as fold change compared to control (Munoz et al. 1997).

The activity of SOD was obtained by using iodonitrotetrazolium violet (INT), with an oxidized absorbance of 485 nm € = 11 mM−1 cm−1 (Beyer and Fridovich 1987). Two hundred micrograms of control or sulfur deprived protein were incubated with 5 mM menadione, 15 µL of INT (4 mg/mL stock) was added for a final volume of 1.0 mL and the absorbance was measured after 60 min. Blanks were prepared similarly lacking the presence of menadione. The activity of the control enzyme was normalized to 100% and the activity of the enzyme in the absence of sulfur was quantified as fold change compared to control.

Free sulfhydryl measurement

The amount of free sulfhydryl groups in the cell free extracts of control and stressed cells was calculated with the aid of the Ellman’s reagent (Boyne and Ellman 1972). A 100 μg protein equivalent of cell free extract from control cells and cells grown in a media without added sulfur was incubated with 500 μM of 5,5′-dithio-bis-2-nitrobenzoic acid (DTNB) and was incubated for 30 min at room temperature. All dilutions were done using an aqueous buffer containing 0.4 M Tris–HCl and 0.1 M EDTA. Free sulfhydryl concentration was monitored by measuring the absorbance at 420 nm after the incubation period. A standard curve using reduced glutathione concentration ranging from 0.1 to 1.0 mM was made to calculate the concentration of free sulfhydryl groups in solution. The data is represented as relative concentration, with the control being assigned a value of 1.

ICDH-NADP and NAD kinase activity analysis by HPLC

The activity of ICDH-NADP and NADK was monitored by HPLC. For ICDH-NADP, 200 μg of protein was added to 2 mM isocitrate and 0.5 mM of NAD. For NADK, 200 μg of protein was added to 2 mM ATP and 2 mM of NAD+ and 0.5 mM of NAD. The reactions were completed to a final volume of 1000 μL with Reaction Buffer (25 mM Tris–HCl and 5 mM MgCl2 in water, pH 7.4 at ambient temperature) in an Eppendorf 1.5 mL microtube. The mixture was reacted for 30 min., then heat inactivated by boiling the samples in a water bath for 10 min. A volume of 100 μL was added to 900 μL of HPLC-grade water in a 2.0 mL scintillation vial and 10 uL was injected unto a Synergi 4 μm Hydro-RP 80A column (Phenomenex) with dimensions of 250 × 4.6 mm at a flow rate of 0.7 mL/min. A Waters 2478 Dual λ absorbance detector, set at wavelengths of 210 nm to identify the amount of α-ketoglutarate produced from ICDH and 254 nm to identify the amount of NADP produced by NADK, was utilized. The mobile phase consisted of a filtered 20 mM KH2PO4 solution diluted in Milli-Q water, pH 2.9 at room temperature similarly to what was used for metabolic profiling. NADP and α-ketoglutarate standards were run in parallel to identify their retention time in the column.

RNA extraction and cDNA synthesis

Total RNA was extracted using the nucleic acid extraction kit from Qiagen (Cat #80234) according to manufacturer’s instructions. Purified RNA with 260/280 absorbance (ND-1000 spectrophotometer) ratios below 1.8 were excluded from analysis. Contaminating genomic DNA was removed from the purified RNA samples using the DNAse kit (Sigma) according to manufacturer’s instructions. The DNAse treated RNA samples were reverse transcribed using random primers (Sigma), oligo dT (VWR), and M-MLV reverse transcriptase (Promega) according to manufacturer’s instructions to obtain complementary DNA (cDNA).

Real time quantitative polymerase chain reaction (RT-qPCR)

RT-qPCR reactions were performed using QuantStudio5 (ThermoFisher) to obtain threshold cycle (Ct) values. Each reaction was performed in 15 μL volumes containing 1 × POWERUP SYBR Master Mix (ThermoFisher), 600 nM forward/reverse primers and 10 ng cDNA. The cycling conditions were as follows: (a) 60 °C for 2 min, (b) 95 °C for 2 min, (c) 95 °C for 10 s (cDNA denaturation), (d) 56 °C for 10 s (primer annealing), (e) 72 °C for 20 s (template extension), (f) plate read and data collection, and (g) steps c to f repeated for 40 cycles. DNA melt curve analysis was performed at the end of each qPCR run to confirm specificity.

Forward and reverse primer pair sequences for genes of interest were designed using Primer-BLAST (NCBI). The complete list of primer sequences can be found in Table 1. All primer pairs were subjected to stringent validation tests as demonstrated previously (Alhasawi et al. 2019). All samples were normalized to two independent control housekeeping genes (16S rRNA and cpn60). The relative mRNA transcript level of each gene was reported according to the ΔΔCT method as mRNA fold increase (Livak and Schmittgen 2001; Alhasawi et al. 2019).

Table 1 RT-qPCR primer sequences
Full size table

Statistical analysis

All experiments were executed in duplicate or triplicate with at least three biological replicates. Statistical analysis was performed using Student’s T Test (paired; two-tail) on Microsoft Excel.

Results and discussion

Pseudomonas fluorescens multiplied in a mineral medium without added S. Glutamine, the sole of carbon and nitrogen was consumed more rapidly in the control cultures compared to those deficient in S. At stationary phase of growth, the biomass was relatively higher in the control cultures (Fig. 1a). The soluble cell-free extracts (CFE) obtained at stationary phase of growth revealed a disparate profile of nicotinamide adenosine nucleotides. The cells grown in the S-deficient medium was characterized with a higher amount of NADPH and lower concentration of NADH compared to the cultures with added S (Fig. 1b). As lack of S is known to trigger oxidative stress, enzymes involved in the detoxification of ROS were monitored. Superoxide dismutase (SOD), an enzyme that nullifies O ·−2 and catalase, an enzyme that neutralizes H2O2 were monitored. An increase in activity of SOD was observed in the cultures deficient in S (Fig. 1c). Free sulfhydryl measurements were performed in order to assess the effect of S starvation in SH groups availability. The cell free extracts from cells grown in a media without added sulfur did show free sulfhydryl, but in a lesser amount than the cells grown in control media (Fig. 1c). The presence of S in the other mineral salts utilized may account for this observation. Regardless, its diminished level is known to pose an oxidative challenge (Sekowska et al. 2000). Since oxidative tension was higher in the stressed cultures, its impact on select enzymes in the TCA cycle and electron transport chain (ETC) was assessed. The activities of such enzymes as ICDH-NAD, fumarase and Complex I, were significantly down-regulated in cells harvested from the media with no added S while there did not appear to be a significant change in the activity of MDH (Fig. 1d).

Fig. 1
figure 1

Sulfur starvation induces oxidative stress in P. fluorescens. a Growth curve of P. fluorescens cultures in glutamine media with (black) and without no added S (grey). b Nicotinamide nucleotide profiles in P. fluorescens obtained from control (left) and S-deficient cultures (middle) at stationary growth phase. NAD standard (right) is shown as an example. c SOD (left panel) and Catalase (middle panel) activity assay and free sulfhydryl measurement (right). Catalase activity was monitored at 485 nm after 60 min. Superoxide dismutase activity was monitored at 458 nm after 60 min. Activity is presented as fold change relative activity, where the control activity was assigned a value of 1. Free sulfhydryl was quantified at 420 nm after 30 min. D. Activity analysis of ICDH NAD (full gel), Complex I (CI), Fumarase (FUM) and malate dehydrogenase (MDH) by Blue Native polyacrylamide gel electrophoresis (C = control, S = stress, MW = molecular mass markers). Results are shown as mean ± standard deviation. Results are representative of three independent experiments. Statistical significance was determined at p < 0.05 (*) using student’s t test

Full size image

As there was a marked shift in oxidative metabolism in the S-deficient cultures, it became important to examine the fate of nicotinamide adenosine dinucleotides in these conditions as there was a significant difference in the amounts of these moieties in stressed cells compared to the controls (Fig. 1b). These observations prompted the investigation of enzymes involved in the synthesis of NADH and NADPH. We have previously reported that under oxidative stress induced by H2O2, P. fluorescens generates glutamate from glutamine in an energy efficient manner (Aldarini et al. 2017). Glutamate can subsequently be metabolized to α-KG, a reaction mediated by GDH-NAD/NADP dependent. The activity of GDH-NAD dependent was more pronounced in the control cells than in the stressed cells. On the other hand, the activity of GDH-NADP dependent was elevated in the stressed cells compared to the control (Fig. 2a). The α-KG formed can either follow the reductive or oxidative TCA cycle. The enzyme KGDH, a pivotal participant in the formation of NADH was virtually absent in the S-deficient cultures. No discernible activity band was observed even after 4 h of incubation. On the other hand, malic enzyme (ME) known to decarboxylate malate with the concomitant formation of NADPH was elevated in the stressed with a lack of S (Fig. 2b). The reductive TCA cycle propelled by the enzyme ICDH was markedly elevated in the stressed cultures. This reaction converts α-KG into isocitrate in the presence of HCO3− and NADH/NADPH. NADH was a preferred co-factor for the enzyme located in the membrane component of the stressed cells. The reductive decarboxylation mediated by ICDH that results in the formation of α-KG, NADH and CO2 was more pronounced in control cells. Hence, it became quite evident that the membrane fractions isolated from cells grown in the S-deficient cultures possessed isocitrate-producing enzyme that consumed NADH more rapidly than their control counterparts. The NADH-generating attribute of these fractions was more elevated in the controls. However, in the soluble CFE where the enzyme of the ICDH-NADP is located, the activity was sharply increased in the stressed cells (Fig. 2c). The activity was evident even when the co-factor (NADP) was decreased tenfold in the reaction mixture (data not shown). The reaction of the excised activity band was readily followed by HPLC where the utilization of isocitrate and the formation of α-KG were recorded (Fig. 2d). Hence, the dearth of S has compelled P. fluorescens to produce more NADPH and limit the synthesis of NADH.

Fig. 2
figure 2

NADH and NADPH homeostasis in control and stressed cultures. a BN-PAGE activity analysis of GDH NAD/NADP dependent. b Activity analysis of ME and KGDH. c BN-PAGE Activity analysis of ICDH-NAD/NADP dependent. C = control cell free extract, S = cell-free extract from sulfur starved culture. d HPLC analysis of ICDH-NADP dependent activity. 200 µg of protein was incubated with 2 mM isocitrate and 0.2 mM of NADP and incubated for 1 h. Samples were analyzed by a C-18 reverse-phase HPLC column at a flow rate of 0.2 mL/min and monitored at 210 nm. The production KG and consumption of isocitrate (IC) were followed. e NADK activity analysis by HPLC. 200 μg of protein was incubated with 2 mM ATP and 2 mM NAD and incubated for 60 min. Samples were then ran into a C-18 reverse phase HPLC column and detected using a UV lamp set at 254 nm. NADP formation was monitored via the characteristic NADP peak (showed by arrow). NADK activity is increased under sulfur starvation

Full size image

These findings were further confirmed by the ability of the soluble CFE from the stressed cells to readily generate α-KG upon incubation with isocitrate and NADP. It was numerous folds higher compared to the controls as evidenced by the production of α-KG. This trend to reconfigure the metabolic pathways dedicated to NADPH formation coupled to a decrease in the production of NADH was further tested by examining ME, an enzyme that liberates NADPH following the decarboxylation of malate and KGDH, an enzyme that generates NADH via the decarboxylation of α-KG. While the later was decreased, a marked increase was observed with the former (Fig. 2d) When the soluble CFE from the stressed cultures was incubated with NAD and ATP, NADP with the concomitant formation of ADP was readily obtained’ thus revealing the presence of enhanced NADK. This reaction was sharply reduced in the controls (Fig. 2e). These BN-PAGE studies were further confirmed by the spectrophotometric analyses of these enzymes. While KGDH was barely evident in the stressed cells, the reversible ICDH-NADH was tenfold higher (Table 2). The drastic down-regulation of KGDH has also been observed in the enhanced production of KG in P. fluorescens cultured in a glycerol in the presence of manganese (Alhasawi and Appanna 2016; Alhasawi et al. 2019). RT-qPCR analyses pointed to a significant increase in the mRNA transcripts of ICDH-NADP and ICDH kinase, two enzymes that are known to contribute to NADPH synthesis (Fig. 3).

Table 2 Monitoring the activity of select TCA cycle enzymes by spectrophotometric analysis
Full size table
Fig. 3
figure 3

mRNA expression analysis of metabolic enzymes in stressed and control cultures by RT-qPCR analysis. mRNA transcripts of isocitrate dehydrogenase NADP (ICDH-NADP) and isocitrate dehydrogenase kinase (ICDH Kinase) were determined relative to 16S rRNA and chaperonin 60 reference genes. Stress cultures demonstrated approximately 2.5 and 3.0 fold increased expression of ICDH-NADP and ICDH kinase respectively. Results are representative of 3 independent experiments. Statistical significance was determined at p < 0.05 (*) using student’s t test

Full size image

The aforementioned data demonstrate the ability of P. fluorescens to reconfigure its metabolism in order to fend the oxidative stress triggered by a S-deficient environment. Sulfur that helps maintain a reductive milieu is essential for the survival of all aerobic organisms. In this instance, the microbe opts for a metabolism geared towards the reduction in the production of the pro-oxidant NADH and an increase in the synthesis of the NADPH, an antioxidant. The increase in activities of the enzyme SOD in the S-deficient culture pinpoint to an oxidative stress encountered by the microorganism when subjected to this challenge (Mailloux et al. 2007; Singh et al. 2005). To quell this situation, P. fluorescens impedes the TCA cycle and OXPHOS, two metabolic networks known to be the major generator of ROS. Furthermore, the consumption of NADH is enhanced by the ability of the reductive TCA cycle to consume NADH through its interaction with α-KG and HCO3−. This not only decreases the intracellular levels of NADH but also enables the ICDH localized in the soluble CFE to generate NADPH with the subsequent formation of KG. The latter can then again fuel the fixation of NADH and the production of isocitrate, a moiety whose decarboxylation is coupled to the synthesis of NADPH by the ICDH residing in the soluble CFE. This metabolic cycle indeed provides an effective route to convert NADH into NADPH by activating enzymes located in different cellular fractions and in this instance fending off an oxidative assault induced by a dearth of S. A metabolic network involving malate, oxaloacetate and pyruvate and the enzymes malate dehydrogenase, malic enzyme and pyruvate carboxylase has been shown to convert NADH into NADPH (Hamel et al. 1999; Hamel and Appanna 2003; Lemire et al. 2008).

This fine-balancing act in modulating the homeostasis of NADH and NADPH is further aided by NADK, an enzyme that police the availability of NAD and NADP in the cells. An upregulation of the activity of the enzyme will tilt the balance towards NADP while a reduction in activity will favour the synthesis of NAD. Such an adjustment of the NAD/NADP ratio will have major implication of the downstream enzymes that require these cofactors for their activities. The lack of NAD is known to arrest the TCA cycle, a metabolic module that incorporates numerous enzymes dependent of this nicotinamide moiety. Hence, in this instance where the aim is to limit ROS production due to a scarcity of S, the enhanced activity of NADK restricts the availability of NAD and subsequently turns the TCA cycle and OXPHOS, two metabolic machines known to release ROS either by design or accidentally. The S-deficient microbe has to impede this if it is to survive this toxic situation. The enhanced formation of NADP will undoubtedly promote enzymes committed to the synthesis of NADPH, a well-known biological antioxidant. Indeed, an augmentation in activities of numerous NADPH-generating enzymes was observed in this study.

In conclusion, this report illustrates the plasticity of cellular metabolism and how it can be moulded to enable the adaptation of living organisms confronted by nutritional challenges. The dearth of S that precipitates an oxidative stress is averted by a metabolic reconfiguration mediated by ICDH, an enzyme that can modulate the two effectors of oxidative tension NADH and NADPH. This cyclic metabolic module converts NADH into NADPH with the aid of the substrates isocitrate and KG. The crosstalk involving NADK helps support this system by modulating the concentration of NADP and NAD. NADK resembles a molecular key that turns on and off the activity of NAD and NADP dependent. Hence, the metabolic adaptation orchestrated by molecular partners ICDH, NAD, NADP, isocitrate and KG helps provide an effective stratagem to counter the S-triggered oxidative stress (Fig. 4).

Fig. 4
figure 4

Schematic representation of the conversion of NADH into NADPH in a sulfur-deficient environment. The formation of isocitrate consumes NADH while its degradation produces NADPH; NADK funnels NAD to NADP-requiring enzymes. ICDH isocitrate dehydrogenase, NADK NAD kinase, KGDH α-ketoglutarate dehydrogenase, FUM fumarase, MDH malate dehydrogenase, OXPHOS oxidative phosphorylation

Full size image