Introduction

Ashbya gossypii is a natural overproducer of riboflavin and the first filamentous hemiascomycete fungus used in industrial large-scale riboflavin production [1]. Whether riboflavin overproduction is beneficial or toxic to A. gossypii has been an intriguing question. Riboflavin production was increased by nutrient limitation (nutrient stress) and decreased by cAMP supplementation (a negative stress signal) [2, 3]. In our previous study, we showed an increase in riboflavin production on supplementation of antioxidant vitamin E (VE) and oxidant menadione [4]. A recent study depicted Yap1-mediated induction of riboflavin when supplemented with H2O2 [5]. A. gossypii is a plant pathogen and has to deal with the plant defense mechanisms for survival. One important plant defense is oxidative burst which leads to production of reactive oxygen species (ROS), particularly H2O2 [5]. Therefore, it is now understood that riboflavin overproduction in A. gossypii constitutes a scavenging mechanism against the free radicals produced by plant defense [5].

Apart from overproduction of riboflavin, the fungus may host other antioxidant defense molecules to protect itself. This includes stress enzymes like catalase (CAT), superoxide dismutase (SOD), and glutathione peroxidase (GPX), which are known to increase during stress, as in Saccharomyces cerevisiae [6]. Exogenous supplementation of H2O2 and menadione leads to increased CAT, SOD, and GPX induction in Schizosaccharomyces pombe and Aspergillus niger [7, 8].

However, in A. gossypii, the study of crucial stress defense enzymes like CAT, SOD, and GPX under stress conditions is fragmentary. The present study aims at finding the role of these crucial enzymes, apart from riboflavin overproduction, in protection against H2O2 and menadione stress.

The reduced glutathione (GSH) and GSH metabolic enzymes are also involved in protection against ROS apart from SOD and CAT in fungal species such as Penicillium chrysogenum [9]. The study of GSH metabolism is very important in A. gossypii organism, due to the following reasons: Under stress conditions, GSH is utilized to scavenge the ROS directly (Fig. 1) by donating a reducing equivalent (H++ e−) from its thiol group of cysteine to other unstable reactive oxygen species [10]. Riboflavin is also an antioxidant and free radical scavenger by itself [11]. Hence, glutathione and riboflavin may reciprocate in this riboflavin overproducer, and any significant findings of the study could be exploited for improving riboflavinogenesis. Secondly, riboflavin is the precursor for the coenzymes flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD), which participate in various oxidation and reduction reactions in the cells [12]. Glutathione reductase requires FAD as a cofactor along with NADPH for the reduction of oxidized glutathione to GSH (Fig. 1) [13]. Thus, riboflavin overproduction by A. gossypii could alter the GSH metabolism differently when compared to other filamentous fungi. In view of the above, the present study explored the oxidative stress response and GSH metabolism at the biochemical and molecular levels in the presence of peroxide and superoxide.

Fig. 1
figure 1

Possible influence of hydrogen peroxide, menadione, or riboflavin on the enzymatic and nonenzymatic components of GSH metabolism (compiled from Herrero et al. [6] and Masella et al. [13])

Full size image

Materials and Methods

Menadione, tetramethoxy propane (TMP), thiobarbituric acid (TBA), reduced and oxidized glutathione, glutathione reductase, dithio-bis-nitrobenzoic acid (DTNB), agarose, and ethidium bromide were procured from Sigma-Aldrich, IL, USA. 1-Chloro-2,4-dinitrobenzene (CDNB) and other chemicals used for the study were from Sisco Research Laboratories (SRL), Mumbai, India. The messenger RNA (mRNA) isolation kit, M-MulV RT-PCR Kit (Moloney murine leukemia virus), and DNA gel extraction kit were procured from Bangalore Genei, Bangalore, India. The complementary DNA (cDNA) sequencing was done from Bangalore Genei, Bangalore, India.

Organism, Growth Conditions, and Exposure to Stressors

A. gossypii culture, NRRL Y-1056, was obtained from NCAUR, IL, USA. It was maintained on yeast-malt extract agar (YMA) slants of the following composition (g/l): 5 g peptone, 3 g yeast extract, 3 g malt extract, 10 g glucose, and 20 g agar. Mycelium was grown in Ashbya full medium (AFM) containing (g/l) 10 g casein, 10 g yeast extract, 20 g glucose, and 1 g myo-inositol [14] at 180 rpm at 30 °C for all experiments. A preinoculum of 0.5 % grown for 48 h in the above medium was used to inoculate 50 ml of AFM in a 250 ml brown Erlenmeyer flask for all experiments. H2O2 (MW 34.01) was added fresh from a stock bottle of 30 % to obtain final concentrations of 10, 25, and 50 mM; 218 μl was added from a stock of 0.1 mg/ml of menadione (MW 376.23) to obtain a concentration of 2.5 μM. GSH (MW 307.3) was dissolved at concentrations of 154 mg/ml (stock 1) and 1.54 mg/ml (stock 2) in sterile water. From stock 1, 10 μl, 100 μl, and 1.0 ml were added to the AFM broth to get the final concentrations of 100 μM, 1 mΜ, and 10 mM, respectively. From stock 2, 40 and 100 μl were added to 50 ml of AFM broth to obtain 4 and 10 μM, respectively.

Biomass was determined after drying the cells, harvested by centrifugation, to constant weight at 98 °C [4]. Total and extracellular riboflavin was estimated spectrofluorimetrically using the ISI standard procedure as described elsewhere [4]. A constant amount of frozen cells was used for cell-free extract (CFE) preparation using a mortar and pestle as mentioned previously [4].

Analysis of Nonenzymatic Stress Parameters

GSH was measured, in the protein free cell extract, by the formation of 5-thio-2-nitrobenzoic acid with DTNB spectrophotometrically at 412 nm as described earlier [4]. Protein was precipitated with 10 % TCA immediately after cell-free extract preparation and GSH measurements were performed. The total glutathione (GSH + oxidized glutathione (GSSG)) was measured after the GSSG present in the sample was converted to GSH by the highly specific glutathione reductase and NADPH. The total GSH was estimated similar to that of reduced GSH [4]. Lipid peroxidation (LPX) and membrane lipid peroxidation (MLPX) were assayed by their reaction with thiobarbituric acid (TBA) to form a colored adduct with a maximum absorbance at 532 nm. The quantification of the adduct formed was done by RP-HPLC using a RP-C18 Hibar column as described earlier [4].

Analysis of Enzymatic Stress Parameters

The activity of SOD was measured as the inhibition of the rate of reduction of cytochrome c by the superoxide radical, formed by the xanthine-xanthine oxidase system. The reduction was observed at 550 nm by continuous spectrophotometric rate determinations. One unit inhibits the rate of reduction of cytochrome c by 50 % at pH 7.8; 0.1 ml of CFE (containing 1.0–2.0 mg of protein/ml) was used for the assay [15].

The CAT activity was measured by the decrease in absorbance of H2O2 at 240 nm as described earlier [4]. The amount of reduced glutathione consumed for decomposition of hydrogen peroxide was measured for assaying the GPX activity as described [4]. Glutathione reductase (GR) was assayed by measuring the formation of NADP, which is accompanied by a decrease in absorbance at 340 nm. One unit of the enzyme is the oxidation of 1 μmol of NADPH/min at 25 °C at pH 7.0. CFE (0.1 ml containing 1.0–2.0 mg of protein/ml) was used for the assay [16]. The glutathione S transferase (GST)-catalyzed formation of GS-DNB (a dinitrophenyl thioether) was detected by a spectrophotometer at 340 nm. One unit of GST activity is defined as the amount of enzyme producing 1 μmol of GS-DNB conjugate/min; 0.3 ml of CFE (containing 1.0–2.5 mg/ml of protein) was used for the assay [17]. The protein content in the CFE was determined by the method of Lowry et al. using bovine serum albumin as standard [18].

mRNA Expression of the Stress Transcription Factors and Stress Genes

The primers for genes were designed from the Ashbya Genome Database using “PRIMER-BLAST” and “NET PRIMER” softwares (Table 1). The cells grown in the presence of H2O2 and menadione were harvested on days 2 and 3, and the total RNA was isolated followed by mRNA purification using the mRNA purification kit. The mRNA (90–100 ng) was reverse transcribed to cDNA followed by amplification using gene-specific primers. Multiplex PCR was carried out for amplification of cDNA of SOD1, GSH1, YAP1, and TEF in one tube and RIB3, CTT1, and MSN2 in another tube. All procedures were carried out after quantification and normalization of the RNA and DNA at each step. The PCR reactions were done at 30 cycles after ensuring that saturation is not attained at 30 cycles. The PCR products were electrophoresed by loading 15 μl of sample (100–200 ng/μl), 2 μl of 6× gel loading buffer, and 2 μl of ethidium bromide (0.5 μg/ml) in each lane. The bands were photographed using gel doc and intensity was quantified using Quantity One software. The PCR products were sequenced to confirm whether the primers used amplified the desired gene sequences. The experiments were repeated in three sets to confirm the results obtained.

Table 1 Primer sequences designed for amplifying the gene of interest
Full size table

Analysis of Stress Response Elements and Yap Response Elements on the Promoter Region of Stress Genes and RIB Genes

The upstream 1,000 bp of the genes along with the open reading frame (ORF) was analyzed for the presence of sequences of stress response elements (STRE) (5′-AGGGG-3′, 5′-CCCCT-3′) and Yap response elements (YRE) (5′-TTA(C/G)TAA-3′) using the Needleman-Wunsch global sequence alignment tool (http://blast.ncbi.nlm.nih.gov/). The other YRE motifs analyzed were 5′-TGACTAA-3′ and 5′-TGACTCA-3′. In the present study, the presence of a single sequence with 100 % identity to the STRE and YRE in the upstream 1,000 bp was considered as a response element.

Statistical Analysis

ANOVA and paired t test were used for statistical significance analysis. p values less than 0.05 were considered significant. The values represented are means of three independent experiments.

Results

A time course of A. gossypii showed maximum biomass on days 2 and 3 followed by autolysis on day 4 (Fig. 2, control). The total riboflavin production increased from day 2 and reached maximum on day 3, after which it remained constant. The extracellular riboflavin increased progressively from day 2 and kept increasing throughout the time course. However, the increase in extracellular secretion on days 4 and 5 could be attributed to the autolysis of the cells. Due to the above reasons, most of the further experiments were performed only on days 2 and 3 of growth.

Fig. 2
figure 2

Effect of the addition of different concentrations of hydrogen peroxide on biomass and riboflavin production

Full size image

Effect of Different Concentrations of H2O2 on Biomass, Total Riboflavin, and Extracellular Riboflavin

The abovementioned parameters were studied with different levels of H2O2 to find the concentration at which H2O2 influenced riboflavin production without compromising the biomass. Hydrogen peroxide was used in three different concentrations (10, 25, and 50 mM) to explore the effect on biomass and riboflavin production as a time course (Fig. 2). The biomass was not affected with 10 and 25 mm H2O2 and was decreased with 50 mM. H2O2 at 25 mM increased riboflavin production by 1.6-fold on day 2 and 1.7-fold on day 3 and was chosen for further experiments on stress response and GSH metabolic studies due to the following reasons: Since 25 mM was not toxic, it is understood that the cells could mount an effective stress response at this concentration. Apart from this, 25 mM had increased riboflavin production and thus is appropriate for GSH metabolism studies. In an earlier study, menadione (2.5 μM), a superoxide generating agent, also increased riboflavin production (207 ± 7.1 mg/l) and secretion (76.2 ± 1.7 mg/l) compared to controls without affecting biomass (Fig. 3) [4]. Menadione (2.5 μM) was used in the present study to compare its effects with hydrogen peroxide.

Fig. 3
figure 3

Effect of different concentrations of menadione on biomass and riboflavin production [4]

Full size image

Effect of H2O2 and Menadione on Oxidative Stress Parameters

  1. (a)

    Catalase and Superoxide Dismutase Enzyme Activities

    The activities of CAT and SOD were measured as a time course for 5 days in controls and on supplementation of H2O2 (25 mM) and menadione (2.5 μM) (Table 2). Though analysis was done on all days, the results of days 2 and 3 are discussed, since autolysis ensues from day 4. H2O2 and menadione significantly increased CAT and SOD on days 2 and 3 compared to controls. Menadione predominantly increased SOD, with the highest value on day 2 (6.2 ± 0.16 U/mg). There was a preferential increase in CAT activity on H2O2 supplementation and the highest value was recorded on day 2 (428 ± 28.8 U/mg) (Table 2).

    Table 2 Time course of superoxide dismutase and catalase enzyme activities in hydrogen peroxide- and menadione-treated cells
    Full size table
  2. (b)

    Lipid Peroxide and Membrane Lipid Peroxide Formation

    Lipid peroxidation indicates damage to lipids caused by oxidative stress. In the present study, lipid peroxidation measurements were used as a marker to confirm oxidative stress in the cells. LPX and MLPX were significantly increased on days 2 and 3 by H2O2 and menadione (Table 3), indicating oxidative damage, which could not be prevented, by an increase in SOD and CAT (Table 2).

    Table 3 Lipid peroxide, membrane lipid peroxide, GSH, and GSSG-related enzymes in hydrogen peroxide- and menadione-treated cells
    Full size table

Effect of H2O2 and Menadione on GSH and GSH Metabolism

The GSH levels decreased compared to controls when supplemented with H2O2 and menadione (Table 3). The decreased GSH is reflected in an increased GSSG and lower GSH/GSSG ratio (Table 3). The results indicate that GSH has been oxidized to GSSG in an effort to scavenge the ROS. The GSH utilized to scavenge ROS is replenished by increased GR activity [9, 19]. In the present study, the GR levels increased in H2O2 (2-fold)-supplemented cells, and the GSH/GSSG ratio was restored on day 3 due to this increase. In contrast, the GR levels were not increased on both days in menadione-treated cells and could be the reason for its low GSH/GSSG ratio (Table 3).

H2O2 and menadione supplementation showed an overall increase in GPX and GST (Table 3). GPX activity was drastically increased by 4-fold in H2O2-supplemented cells on day 3, which was much higher than the 1.7-fold increase in menadione-supplemented cells. In general, an increase in GSSG, GPX, GR, and GST and a decrease in GSH and GSH/GSSG ratio are indicators of oxidative stress as far as GSH metabolism is concerned [9, 19], indicating that H2O2- and menadione-treated cells suffer from oxidative stress. However, a few exceptions, such as restored GSH/GSSG ratio in H2O2-treated cells and decreased GR in menadione-treated cells, were observed.

Effect of Exogenous GSH on Biomass and Riboflavin Production on Days 2 and 3

Different concentrations of GSH (4 μM–1 mM) were tested for its effects on biomass and extracellular and total riboflavin levels on days 2 and 3 (Table 4). On day 3, all concentrations of GSH tested increased biomass significantly compared to controls with a maximum at 10 μM of exogenous GSH (9.9 g/l), indicating GSH is protective to cells. The total riboflavin production did not increase with any of the GSH concentrations tested, which was expected. The results show that available exogenous GSH could diminish the need for in vivo production of riboflavin; 1 mM GSH appeared slightly toxic (Table 4).

Table 4 Effect of the addition of different concentrations of GSH on biomass and riboflavin production
Full size table

Effect of H2O2 and Menadione on the Levels of Expression of Stress Genes, Riboflavin Biosynthetic Genes, and Stress Transcription Factors

  1. (a)

    Expression of Stress Genes (SOD1, CTT1, and GSH1)

    The gene expression levels of stress genes SOD1, encoding cytoplasmic superoxide dismutase, and CTT1, encoding peroxisomal catalase, were chosen for analyses (Fig. 4, Table 5).The SOD1 and CTT1 mRNA levels increased in response to H2O2 and menadione on days 2 and 3 (Fig. 4, Table 5), indicating that these molecules are inducible by oxidative stress. The genetic expressions were in agreement with the enzyme activities (Table 2), with H2O2 increasing CTT1 and menadione increasing SOD1 predominantly (Fig. 4, Table 5). In H2O2- and menadione-treated cells, the expression levels of GSH1 were very high on days 2 and 3 compared to controls in an effort to increase GSH production (Fig. 4, Table 5).

    Fig. 4
    figure 4

    Expression levels of genes of SOD1, GSH1, YAP1, TEF, RIB3, CTT1, and MSN2 on days 2 and 3 in cells treated with hydrogen peroxide or menadione. C control, H H2O2, M menadione

    Full size image
    Table 5 The ratio of peak intensity of stress transcription factors and genes with TEF on days 2 and 3 in control and H2O2- and menadione-treated cells
    Full size table
  2. (b)

    Expression of RIB3

    The genes encoding the riboflavin biosynthesizing enzymes are RIB1 (GTP cyclohydrolase), RIB2 (DRAP deaminase), RIB3 (DHBP synthase), RIB4 (lumazine synthase), RIB5 (riboflavin synthase), and RIB7 (diaminohydroxyphoshoribosylaminopyrimidine deaminase) distributed on chromosomes IV, V, and VII in A. gossypii [20]. RIB3 (DHBP synthase) was chosen for the study, out of the abovementioned enzymes for two reasons. DHBP synthase is the rate-limiting enzyme of the riboflavin biosynthetic pathway. The other reason is that it is needed twice in the pathway, while the other enzymes are involved only once [21].

    RIB3 mRNA levels increased in wild-type A. gossypii on oxidative challenge with H2O2 (25 mM) and menadione (2.5 μM) on day 2 (Fig. 4, Table 5). An increased expression of RIB3 by H2O2 and menadione was reflected in a net increase in riboflavin production (Figs. 2 and 3).

  3. (c)

    Expression of Stress Transcription Factors YAP1 and MSN2

    Apart from the above, the gene expression levels of genes encoding the stress transcription factors YAP1 and MSN2 were also measured. YAP1 is a stress transcription factor directly regulated by the redox status of the cells. Yap1 binds to the YRE consensus sequence in the nucleus 5′-TTA(C/G)TAA-3′ [22] and leads to H2O2-mediated induction of SOD1, CTT1, GSH1, and GSH2 [23]. MSN2 is a transcription factor involved in general stress response and mediates its effects via STRE sequences 5′-AGGGG-3′ [23].

    The mRNA expression of YAP1 was induced to a higher level in H2O2- and menadione-supplemented cells. However, MSN2 did not show much difference in its expression levels (Fig. 4, Table 5).

Analysis of YRE and STRE Upstream of Stress Genes and RIB Genes in A. gossypii

A preliminary study for the presence of the STRE and YRE showed their presence in SOD1 and GSH1. GPX7 (encoding glutathione peroxidase), GLR1 (encoding glutathione reductase), and GST1 (encoding glutathione S transferase) do not host YRE, but they have STRE (Table 6). Surprisingly, CTT1 did not have either STRE/YRE.

Table 6 Presence of YRE and STRE on the RIB genes and stress genes analyzed by BLAST
Full size table

All the riboflavin biosynthesizing genes except RIB7 hosted either YRE/STRE (Table 6). RIB3 had a YRE-like element with a single base change from 5′-TTAGTAA-3′ to 5′-TTGGTAA-3′. Riboflavin synthase gene (RIB5) had a YRE-like sequence 5′-TTAGTCA-3′ at −128 bp upstream. A similar YRE sequence was present in the promoter of GSH1 in S. cerevisiae [22]. RIB4 had a YRE element at −165 bp upstream. RIB2 and Lumazine synthase gene (RIB4) hosted STRE (Table 6).

Discussion

Riboflavin Metabolism and Oxidative Stress Defense in the Presence of H2O2 and Menadione in A. gossypii

Riboflavin is a pseudosecondary metabolite overproduced by A. gossypii in thr presence of stressors (Figs. 2 and 3). Recent literature depicts a link between stress response and secondary metabolite production. Many of these metabolites are increased in response to nutritional stress in fungi [1]. Butylated hydroxyl toluene (BHT), an antioxidant, on supplementation increased the production of carotenoids by filamentous fungus Blakslea trispora [24]. Paraquat, a superoxide generator, increased riboflavin production in Pichia guilliermondii [25].

Our previous study showed that VE and menadione increased riboflavin production via oxidative stress in A. gossypii [4]. Hence, oxidative stress could be exploited to increase secondary metabolite production industrially. Though oxidative stress could be exploited to increase secondary metabolite production, the concentration at which the oxidants increase production with minimum toxicity should be explored. Surprisingly, A. gossypii was able to grow at 25 mM of H2O2 supplementation with no toxic effects and increased riboflavin production (Fig. 2).

In the present study, increased SOD and CAT could be an important reason for the survival of the organism in the presence of H2O2 and menadione (Table 2). A. gossypii was able to grow at 1 M concentration of H2O2 though not to the levels of controls in a recent study [5]. Similar tolerance to high levels of H2O2 (0.35–0.70 M) and tert-BOOH (0.5-2.0 mM) was observed in P. chrysogenum. This increased tolerance was attributed to the increase in CAT and GPX levels [9]. P. chrysogenum was also resistant to menadione supplementation (250–500 μM) which was attributed to an increase in SOD and involvement of GSH metabolism [19]. From the above, it is understood that a well-established stress defense via CAT and SOD could enable the cell to survive in the presence of oxidants.

An interesting observation in the present study is that the cellular response to H2O2 and menadione was different with H2O2 predominantly increasing CAT and menadione predominantly increasing SOD (Table 2). Previous studies in S. pombe showed that H2O2 treatment mainly increased CAT compared to SOD activity in agreement with the present study [7]. Escherichia coli and S. cerevisiae exhibited diverse stress response to different kinds of oxidative stress. In E. coli, H2O2 triggered the oxyR regulon, and excess superoxide triggered the soxRS regulon [26]. Similarly, in S. cerevisiae, H2O2 increased CAT and excess superoxide increased SOD activity predominantly [27].

Though the cells were protected due to an increase in the stress enzymes (SOD and CAT), damage to the lipid molecules could not be prevented completely (Table 3). Similarly, in a previous study, the induction of the antioxidant defense such as SOD and CAT in filamentous fungi could not prevent protein carbonylation, but facilitated cell survival [28]. Increased lipid peroxidation and membrane lipid peroxidation can affect membrane fluidity [29] and impair crucial membrane functions such as transport and permeability [30]. Three mechanistic models have been proposed for riboflavin efflux in A. gossypii. The efflux is caused by increased membrane permeability or due to functional inversion of the uptake system or excretion via a specific riboflavin export carrier [31]. In the present study, we observed an increase in riboflavin secretion with H2O2 and menadione (Figs. 2 and 3), which could be attributed to the increased lipid peroxidation (Table 3). Therefore, supplementation with mild quantities of these oxidants could be a strategy for product recovery.

Lipid peroxidation is a process that could be propagated and requires protective mechanisms other than catalase and SOD. In P. chrysogenum, the GSH-dependent enzymes are involved in the elimination of toxic agents released during lipid peroxidation and in the restoration of the altered GSH/GSSG balance of the cells [9].

Involvement of GSH Metabolism in Oxidative Stress Defense and Its Possible Link with Riboflavin Metabolism in A. gossypii

GSH and riboflavin are potent natural antioxidants due to their free radical scavenging activity. Since A. gossypii is a natural overproducer of riboflavin, and riboflavinogenesis is further enhanced by oxidants, we hypothesized that riboflavin could have a sparing effect on GSH and vice versa. GSH is directly involved in scavenging ROS generated by exogenous H2O2 and menadione [9, 19] and could be the reason for its decrease on supplementation of menadione and H2O2 in the present study (Table 3). Apart from a decrease in GSH, there was a simultaneous increase in riboflavin production on H2O2 and menadione supplementation (Figs. 2 and 3). The decreased GSH could have triggered increased riboflavinogenesis in an effort to supplement the scavenging activity of GSH and to spare the rapidly utilized GSH.

Interestingly, in P. guilliermondii, blocking of GSH biosynthesis, by creating ∆gsh1 (encoding γ-glutamyl cysteine synthetase) and ∆gsh2 (encoding glutathione synthetase) mutants, increased riboflavin production by 365 and 148 times, respectively [32]. This supports our hypothesis that riboflavin could have a sparing action on GSH. In the future, GSH levels could be brought down by mutations or by oxidants to further improve riboflavinogenesis in A. gossypii.

In the present study, the very high increase in riboflavin by H2O2 on day 2 (2-fold) (Fig. 2) could be the reason for a very high GR activity on day 2 (Table 3). Riboflavin is essential for the synthesis of FMN and FAD, which is involved in the glutathione reductase activity. Thus, apart from the GSH levels, the activity of the GSH metabolic enzymes is also influenced by riboflavin overproduction.

The reason for lower GR activity in menadione-treated cells could be due to GSSG levels (Table 3). The levels of GSSG in menadione-supplemented cells did not increase as high as in H2O2-supplemented cells (Table 3). In pine shoots, exogenous GSSG increased GR levels [33]. The reason for a lower GSSG compared to H2O2 could be that GSH detoxified menadione by conjugation rather by donating electrons. GSH is not only involved in direct scavenging of ROS produced by menadione but also in excretion of menadione as GSH conjugates [13].

The decrease in GSH could also be attributed to the demand by increased GPX and GST [13] (Fig. 1). In the present study, the significant increase in GPX and GST under both H2O2 and menadione stress contributed to cell survival besides an increase in CAT and SOD activity. The GSTs also play an important role in removing lipid peroxide end products like endogenous unsaturated aldehydes, quinones, epoxides, and hydroperoxides as GSH conjugates [13] and are known to protect membrane damage [9].

Surprisingly, in the present study, the activity of the GSH metabolic enzymes in response to H2O2 and menadione was slightly different from the other filamentous fungus P. chrysogenum. In P. chrysogenum, both menadione and H2O2 supplementation had increased GR activity which differs from the present results with A. gossypii [19]. While both H2O2 and menadione increased GPX activity in A. gossypii, only H2O2 increased GPX activity in P. chrysogenum [9, 19]. In the present study, GST activity was increased by H2O2 and menadione. However, in P. chrysogenum, menadione increased GST, while H2O2 did not increase GST [9, 19]. Thus, this is one of the few studies which shows the involvement of GSH and GSH metabolism in oxidative stress protection in filamentous fungi and is the first report on A. gossypii.

Influence of Exogenous GSH on Riboflavin Metabolism and Biomass

Exogenous GSH could be transported into the cells by high affinity oligopeptide transporter Hgt1p in S. cerevisiae and increase the intracellular GSH levels [34]. Interestingly, a lower intracellular GSH levels, due to oxidative stress (H2O2 and menadione), increased riboflavin production in the present study. Hence, exogenous GSH was expected to decrease riboflavin production, but surprisingly, there was no decrease in riboflavin production, but the levels were similar to that of controls (Table 4). In a similar study, 10 μM of GSH supplementation in the medium of ∆gsh1 and ∆gsh2 mutants of P. guilliermondii decreased the riboflavin levels to that of the parental strain [32]. Hence, it is now clear that lower intracellular GSH increased riboflavin, and increased GSH brought down the requirement for excess riboflavin, indicating their reciprocal involvement. In the future, riboflavin metabolism could be blocked to find out if GSH production could be increased. GSH production is very important since it is used as a pharmaceutical compound and can be used as food additives and in the cosmetic industries. S. cerevisiae and Candida utilis are currently used to produce glutathione on an industrial scale [35].

GSH could protect cell lysis and aging, by scavenging ROS, which are most important signals in autolysis and apoptosis [36]. Exogenous addition of GSH (1–50 μM) in the medium of Picea abies could scavenge constitutive H2O2 [37], and GSH, GSSG, and γ-glutamylcysteine (γGC) at a concentration of 1 mM promoted the growth of a γ-glutamylcysteine synthetase (GCS) mutant of Candida albicans [38]. Exogenous GSH in the present study was protective by increasing the biomass and decreasing secretion of riboflavin (Table 4).

In the present study, levels of 1 mM were toxic (Table 4). A similar toxicity was observed in S. cerevisiae above 3–4 mM and in P. guilliermondii above 0.5 mM [34, 32]. The reason for GSH toxicity is due to nonspecific glutathionylation of important cellular proteins and subsequent inhibition of protein enzymatic functions [34].

Influence of H2O2 and Menadione on the Stress Metabolism at the Transcriptional Level

De novo synthesis of the stress protection enzymes, SOD and CAT, in response to paraquat and H2O2 was observed in 12 different filamentous fungal species [28]. In the present study, SOD1 and CTT1 mRNA levels were increased (Fig. 4, Table 5), indicating de novo synthesis in response to stressors as in other filamentous fungi. When the GSH levels are lowered by stressors, higher GSH1 (encoding γ-glutamyl cysteinyl synthase, the rate-limiting enzyme) level was anticipated and was observed in the present study (Fig. 4, Table 5). GSH1 levels were found to be induced in menadione-treated S. cerevisiae [22] similar to the present study. However, the present study is the first of its kind to show the increase of the mRNA levels of SOD1, CTT1, and GSH1 in A. gossypii under normal and stressed conditions. Since the involvement of these genes is now proven, they could be manipulated in the future to increase the biologically important products, riboflavin or GSH.

A direct influence of stressors on the RIB3 mRNA levels was demonstrated for the first time in the present study (Fig. 4, Table 5). A related study showed an increase in the RIB3 mRNA and an induction of the promoter activity in the production phase of riboflavin in A. gossypii [21]. An increase in RIB3 promoter activity was observed in response to nutrient limitation and growth stress in A. gossypii [2]. RIB4 expression was found to be decreased in an Agyap1 mutant, indicating the involvement of the stress transcription factor on riboflavin production in A. gossypii [5].

Yap1 is a stress transcription factor involved in the expression of the genes involved in oxidative stress defense, including GSH1 in S. cerevisiae [22, 23]. Yap1 of A. gossypii differs from S. cerevisiae Yap1 in lacking the cysteine residues essential for its functions and induction [5]. In the present study, RIB3 and GSH1 were upregulated with a simultaneous induction in YAP1 levels (Fig. 4, Table 5). The observation is very interesting and is a new avenue for exploring the molecular link on the regulation of GSH and riboflavin metabolism. Further studies can be carried out to find the possible involvement of Yap1 in the riboflavin and GSH metabolism, which could in turn be exploited for the industrial overproduction of these useful compounds. MSN2 did not show much difference in their mRNA levels in the present study, which suggest a less involvement of the general stress response pathway on supplementation of H2O2 and menadione (Fig. 4, Table 5).

The Possible Involvement of STRE/YRE in Stress and Riboflavin Metabolism

The effects of the stress transcription factors are mediated by the response elements to mediate increased expression of the stress genes [23]. The YRE and STRE are response elements on DNA to which the stress transcription factors bind. Yap1 binds to the YRE (5′-TTA(C/G)TAA-3′) in response to oxidative stress, more specifically to peroxide stress, and induces transcription of oxidative stress genes like SOD1 and GSH1 [22].

The STRE is a cis-regulatory element present on the promoters of genes involved in general stress response [23]. Msn2p and Msn4p bind specifically to STRE (5′-AGGGG-3′, 5′-CCCCT-3′) and are required for the activation of STRE- mediated transcription by different types of stress [23].

  1. (a)

    Presence of Response Elements in Stress Genes of A. gossypii

    The presence of YRE in both SOD1 and GSH1 (Table 6) indicates that the expression of these genes may be mediated by YAP1 and may be a part of the Yap regulon. The increased enzyme activity of superoxide dismutase (Table 2) and the increased mRNA levels of SOD1 and GSH1 (Fig. 4, Table 5) support the above predictions. SOD1 and GSH1 host STRE (Table 6), indicating that these genes could also be regulated by other transcription factors apart from Yap1.

    Surprisingly, CTT1 did not host either STRE/YRE (Table 6) though expected. A. gossypii does not have the cytoplasmic catalases present in S. cerevisiae, but is known to host the peroxisomal catalases [5]. The CTT1 could have hosted other elements like HSE which could have caused its increase through some other pathways in response to peroxide stress.

    GPX7 (encoding glutathione peroxidase), GLR1 (encoding glutathione reductase), and GST1 (encoding glutathione S transferase) do not host YRE, but they have STRE (Table 6). It appears that they may be regulated by the general stress response or by other transcription factors like Skn7 and Atf1.

  2. (b)

    Presence of Response Elements in RIB genes

    The presence of STRE and YRE on RIB1 (Table 6) encoding the first enzyme in the riboflavin biosynthesis indicates that riboflavin biosynthesis could be influenced by both oxidative stress and general stress. The RIB2 and RIB4 also hosted STRE (Table 6), indicating the possibility of regulation by general stress. The predictions are supported by the finding that nutrient limitation increased riboflavin production in A. gossypii [21].

    RIB3 and RIB5 had YRE-like sequences (Table 6). A similar YRE-like sequence was present in the promoter of GSH1 in S. cerevisiae [31], indicating that the RIB3 genes were similar to the GSH1 genes and could share similar regulatory mechanisms. A related study showed the presence of YRE upstream of the RIB genes in A. gossypii [5]. In contrast to the previous report, the present study shows the presence of both YRE and STRE on the RIB genes, suggesting that riboflavin production may be regulated by both oxidative stress and other general stress mechanisms, which is not yet explored at the molecular level in A. gossypii.

Conclusions

The study has identified the presence of a potential oxidative stress defense against H2O2 and menadione in A. gossypii. The stressors increased riboflavin production and secretion, supporting the hypothesis that riboflavin synthesis could be a stress defense mechanism. The lipid peroxidation and altered membrane functions due to oxidative stress increased riboflavin secretion; hence, oxidative stress could be a strategy for product recovery. This study is also the first report on the GSH metabolism in A. gossypii and its role on oxidative stress defense in A. gossypii. The study found a close relationship between GSH and riboflavin metabolism at the physiological, molecular, and regulatory levels. Physiologically, decreased GSH increased riboflavin production. The mRNA levels of GSH1 and RIB3 were increased in the presence of stressors, and both RIB3 and GSH1 hosted similar YRE, suggesting that they share common regulatory mechanisms at the molecular level. Thus, in the future, the GSH metabolism could be manipulated at the physiological and genetic levels to improve industrial riboflavin production. The study is also a prelude to a new avenue, wherein riboflavin metabolism could be manipulated, in the future, to improve the GSH production.

The study has for the first time shown the effect of stressors at the molecular level in the presence of oxidants in wild-type A. gossypii. The Yap regulon has not yet been identified in A. gossypii, and the study could be a prelude in identifying these regulons in this organism. Further studies are warranted to confirm the pathways involved in the regulation of oxidative stress protection and riboflavinogenesis.