Changing oxidoreduction potential to improve water-soluble yellow pigment production with Monascus ruber CGMCC 10910

Background Monascus pigments are widely used in the food and pharmaceutical industries due to their safety to human health. Our previous study found that glucose concentration induced extracellular oxidoreduction potential (ORP) changes could influence extracellular water-soluble yellow pigment production by Monascus ruber CGMCC 10910 in submerged fermentation. In this study, H2O2 and dithiothreitol (DTT) were used to change the oxidoreduction potential for investigating the effects of oxidative or reductive substances on Monascus yellow pigment production by Monascus ruber CGMCC 10910. Results The extracellular ORP could be controlled by H2O2 and DTT. Both cell growth and extracellular water-soluble yellow pigment production were enhanced under H2O2-induced oxidative (HIO) conditions and were inhibited under dithiothreitol-induced reductive conditions. By optimizing the amount of H2O2 added and the timing of the addition, the yield of extracellular water-soluble yellow pigments significantly increased and reached a maximum of 209 AU, when 10 mM H2O2 was added on the 3rd day of fermentation with M. ruber CGMCC 10910. Under HIO conditions, the ratio of NADH/NAD+ was much lower than that in the control group, and the expression levels of relative pigment biosynthesis genes were up-regulated; moreover, the activity of glucose-6-phosphate dehydrogenase (G6PDH) was increased while 6-phosphofructokinase (PFK) activity was inhibited. Conclusions Oxidative conditions induced by H2O2 increased water-soluble yellow pigment accumulation via up-regulation of the expression levels of relative genes and by increasing the precursors of pigment biosynthesis through redirection of metabolic flux. In contrast, reductive conditions induced by dithiothreitol inhibited yellow pigment accumulation. This experiment provides a potential strategy for improving the production of Monascus yellow pigments. Electronic supplementary material The online version of this article (10.1186/s12934-017-0828-0) contains supplementary material, which is available to authorized users.


Background
Monascus pigments are mixed pigments with three main colours: yellow, orange and red [1]. Recently, Monascus yellow pigments have been widely researched due to their anti-tumour [2,3], anti-inflammation [4], anti-obesity [5], anti-diabetic, and anti-oxidative stress capabilities [6], and their ability to improving memory and learning ability [7], counteract metabolic syndromes [8], and treat liver fibrosis [9]. A large body of evidence indicates that Monascus yellow pigments are beneficial to human health in appropriate doses and can potentially be used as nutraceutical or therapeutic agents.
In submerged fermentation, Monascus yellow pigments generally consist of a large amount of intracellular alcohol-soluble pigments and a small amount of extracellular water-soluble pigments [10]. Water-soluble yellow pigments have been a popular area of research due to their high stability and convenient utilization [11].
Water-soluble yellow pigments can be synthesized via chemical modification of intracellular alcohol-soluble pigments [12], but this strategy is a potential risk in food applications. In recent decades, considerable effort has been focused on enhancing water-soluble yellow pigment production, such as strain screening [13,14], strain mutation [15], gene modification [16], and medium optimization [17,18]. Chen has shown that both extracellular and intracellular yellow pigments are the main pigments produced during long periods of high cell density culturing by Monascus anka [19]. In addition, our previous studies found that extracellular water-soluble yellow pigments can be produced by M. ruber CGMCC 10910 using high glucose concentrations with low oxidoreduction potential (ORP) [20,21]. However, the pigment yield and glucose utilization was still low.
Oxidoreduction potential was identified as a control parameter for fermentation processes [22][23][24]. Intracellular ORP is primarily determined by the ratio of NADH/ NAD+ [25]. High NADH/NAD+ ratios can regulate intracellular metabolites to generate unusable byproducts, such as alcohol and lactate [26]. However, intracellular ORP can be influenced by changing extracellular ORP [24], while extracellular ORP can be changed by adding oxidative or reductive substances, such as H 2 O 2 , dithiothreitol (DTT), and potassium ferricyanide [24,27]. It has been reported that spinosad and pseudoaglycone yields increased 3.11-fold in Saccharopolyspora spinosa fermentation under HIO conditions, and high citric acid productivity could be achieved in Aspergillus niger cultures with defined ORP profiles [22,28]. Therefore, extracellular ORP provides an alternative parameter for the optimization of metabolic production. Monascus pigment biosynthesis occurs via a polyketide pathway and requires many primary metabolites, such as acetyl-CoA, malonyl-CoA, and NADPH [1,31,32]. The pentose phosphate pathway (PPP), a branch pathway of glycometabolism, was shown to be the primary source of NADPH for polyketide and lipid biosynthesis [29]. Recently, the biosynthetic gene cluster of Monascus pigments in the M. purpureus and M. ruber genome and the functions of some critical genes involved in the pigment biosynthetic pathway were reported [30,31]. Both the targeted inactivation of MpPKS5 (the homolog of MpigA in M. ruber) in M. purpureus and the targeted-deletion of MpigA in M. ruber resulted in the abolishment of pigment production, confirming that polyketide synthase is involved in pigment biosynthesis [31,32]. The genes MpFasA2 (the homolog of MpigJ in M. ruber) and MpfasB2 (the homolog of MpigK in M. ruber) respectively encode fatty acid synthase α-subunit and fatty acid synthase β-subunit and supply the medium-chain (C8 and C10) fatty acyl moieties for Monascus pigments biosynthetic [31][32][33][34].  [30,31,[34][35][36]. Those reports provide a frame of reference from which to investigate external factors on pigment production at the molecular level.
In this study, H 2 O 2 was added to the fermentation medium of M. ruber CGMCC 10910. Its concentration and addition time were also optimized to improve extracellular water-soluble yellow pigment production. In addition, the effect of the oxidative environment induced by H 2 O 2 on the extracellular and intracellular ORP, the expression levels of pigment biosynthesis genes, and the activities of key enzymes (PFK and G6PDH) in the glycolysis and pentose phosphate pathway (PPP) were also investigated.

Microorganism
Monascus ruber CGMCC 10910, which was deposited at the China General Microbiological Culture Collection Center (CGMCC), was used in this study, and it was cultivated on a potato dextrose agar (PDA) medium at 30 °C for 7 days and then stored at 4 °C.

Addition method of H 2 O 2 and dithiothreitol (DTT)
H 2 O 2 and DTT were filter-sterilized before being added to the medium. One, 5, 10, 15 or 20 mM H 2 O 2 was added to the medium on day 2 of fermentation to study the effect of H 2 O 2 concentration on cell growth and yellow pigment production. Ten millimolar H 2 O 2 was added to the medium on day 0, 1, 3, and 5 of fermentation to study the effect of H 2 O 2 addition time on cell growth and yellow pigment production. Three gram per liter DTT was added to the medium on day 3 to study the extracellular reducing conditions on cell growth and yellow pigment production.

Determination of glucose concentration, extracellular oxidoreduction potential (ORP), and DCW
After being cultured, the fermentation broth was vacuum-filtered through a 0.8 mm mixed cellulose esters membrane. The filtrate (extracellular broth) was appropriately diluted to determine the residual glucose concentration using the standard 3,5-dinitrosalicylic acid (DNS) method. Extracellular ORP was detected via an oxidation-reduction electrode (Leici, Shanghai) [20]. After filtration, the mycelia were washed 3 times with distilled water and dried at 60 °C in an oven for 12 h, until a constant mycelia weight was reached to determinate dry cell weight (DCW) by gravity.

Pigment analysis using UV-visible spectrophotometer
The filtrate (extracellular broth) was appropriately diluted to determine the extracellular water-soluble yellow pigment concentration. The absorbance spectrum of the water-soluble yellow pigments was recorded by a UVvisible spectrophotometer (Unico, USA) from 300 to 550 nm at 1 nm intervals, and the absorbance units (AU) at the peak wavelength of 350 nm was multiplied by the dilution ratio, which was used as an index for the watersoluble yellow pigment concentration [20].
The intracellular pigment concentration was determined according to the following the procedure: the harvested and washed mycelia were resuspended in 25 mL acidic aqueous ethanol (70% v/v, pH = 2 with hydrochloric acid) and incubated for 1 h. The suspension was filtered with filter paper, and the filtrate (intracellular extract) was appropriately diluted to determine intracellular pigment concentration. The absorbance spectrum of the intracellular pigments was recorded by a UV-visible spectrophotometer (Unico, USA) from 300 to 550 nm at 1 nm intervals, and the AU at the peak wavelength of 410 nm were multiplied by the dilution ratio, which was used as an index for the intracellular yellow pigment concentration.

Analyses of pigment compositions by HPLC
Analyses of pigment compositions were performed using an Alliance e2695 HPLC system (Waters, Milford, CT, USA) equipped with a 2998 Photodiode Array (PDA) detector (Waters, Milford, CT, USA) and a Zorbax Eclipse Plus C18 column (250 × 4.6 mm, 5 μm, Agilent, Palo Alto, CA, USA). The temperature of the column oven was set at 30 °C. A mixture of H 3 PO 4 solution (pH 2.5, phase A) and acetonitrile (phase B) were used as the mobile phase using the following gradient program: 0 min, 80% A, 20% B; 25 min, 20% A, 80% B; 35 min, 20% A, 80% B; 36 min, 80% A, 20% B; 41 min, 80% A, 20% B. The PDA was set at 200-600 nm, and the flow rate of the mobile phase was 0.8 mL/min [21].

Determination of the pigment composition using LC-MS
LC-MS (liquid chromatography-mass spectrometry) consisted of a HP1100 HPLC system (Agilent, Palo Alto, CA, USA) and a microTOF-QII mass spectrometer (Bruker, Rheinstetten, Germany). The C18 column and chromatographic conditions were the same as mentioned above, except for mobile phase A (water, 0.1% formic acid).

NADH and NAD+ determination
The intracellular NADH and NAD+ concentrations were determined using procedures described in previous studies [25,37]. Mycelia were rapidly ground into a powder using liquid nitrogen; 0.1 g of mycelia powder was quickly transferred into a sample tube containing 1 mL of acidic extracting solution (0.2 M HCl for NAD+) or alkaline extracting solution (0.4 M KOH for NADH). Next, the samples were heated at 50 °C for 10 min and then centrifuged at 12,000g, 4 °C for 10 min. The supernatant transformed into a tube and was neutralized using alkaline extracting solution or acidic extracting solution. After neutralization, the samples were centrifuged at 12,000g, 4 °C for 10 min. The supernatant was collected for NAD+ and NADH determination using the Coenzyme I NAD(H) content test kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) according to the manufacturer's instructions. The kit is based on an enzymatic cycling assay method.

Determination of PFK and G6PDH enzyme activities
Mycelia were rapidly ground into a powder using liquid nitrogen; 0.1 g of mycelia powder was quickly transferred into a sample tube containing 1 mL of crude enzyme extracting solution (25 mM Tris-HCl (pH = 7.6), 10 mM MgCl 2 , 20 mM NH 4 Cl, 0.5 mM DTT) and then fully shocked [38,39]. The protein concentration was quantified using the standard Bradford method. Activity of 6-phosphofructokinase (PFK) was determined using a PFK test kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) according to the manufacturer's instructions. Activity of glucose-6-phosphate dehydrogenase (G6PDH) was analysed as reported with minor modifications [39]. A total of 3 mL of the reaction mixture containing 600 µL of 0.2 M Tris-HCl (pH = 8.0), 150 µL of 5 mM glucose-6-phosphate, 50 µL of 5 mM NADP+, 2 µL of β-mercaptoethanol, and 20 µL of crude enzyme solution was incubated at 30 °C for 20 min. The G6PDH activity of each sample was then determined spectrophotometrically by measuring the formation of NADPH at 340 nm.

Gene expression analysis
The relative expression levels of key genes involved in pigment biosynthesis were analysed using real-time quantitative PCR as described previously [21]. Mycelia were collected for total RNA extraction using the Plant RNA Extraction Kit (TaKaRa MiniBEST). First, cDNA was synthesized using the PrimeScript ™ RT reagent Kit with gDNA Eraser (TaKaRa) according to the supplier's protocol. Primers for the amplification of MpFasA2, MpFasB2, MpPKS5, mppR1, mppA, mppB, mppC, mppD, mppE, mppR2 (GenBank Accession No. KC148521) and the actin gene (GenBank Accession No. AJ417880) are listed in Additional file 1: Table S1. The actin gene was used as a reference gene. Gene expression was monitored by RT-qPCR using SYBR Premix Ex TaqII (TaKaRa). RT-qPCR was performed using a LightCycler 96 (Roche, USA) with the following cycling program: pre-incubation at 95 °C for 30 s, followed by a two-step amplification (40 cycles of denaturation at 95 °C for 5 s, and annealing at 60 °C for 30 s) and dissociation curve analyses (at 95 °C for 10 s, annealing at 65 °C for 60 s, then collection of dissociation curves from 65 to 95 °C, with a final incubation at 97 °C for 1 s).

Statistical analysis
Each experiment was repeated in triplicate, at minimum. Numerical data are presented as the mean ± SD. The differences among the various treatments were analysed using one-way ANOVA. All statistical analyses were performed by using SPSS 22.0 software, and p < 0.05 and p < 0.01 were considered significant and highly significant, respectively.

Water-soluble yellow pigment production from different H 2 O 2 concentrations and adding times
Monascus ruber CGMCC 10910 has the potential to produce extracellular water-soluble yellow pigments with a maximum absorbance wavelength of 350 nm [20], and the extracellular water-soluble yellow pigments could reach a maximum yield of 147 AU 350 under high glucose stress fermentation [21]. Further analysis was shown that the extracellular water-soluble yellow pigments mainly contained four kinds of yellow pigments (Additional file 2: Figure S1). The cell growth and pigment production achieved by adding various amounts of H 2 O 2 to the medium on the 2nd day of M. ruber CGMCC 10910 fermentation under high glucose stress are shown in Fig. 1. The results showed that lower amounts of H 2 O 2 could promote cell growth while higher amounts of H 2 O 2 had an inhibitory effect on cell growth (Fig. 1a). Extracellular and intracellular pigments were enhanced under certain doses of H 2 O 2 but were inhibited by high concentrations (Fig. 1a-c). The optimal amount of H 2 O 2 was found to be 10 mM, which had a 19% increase in dry cell weight (DCW), and a 35 and 26% increase in extracellular and intracellular yellow pigment yield, respectively. Thus, 10 mM was chosen as the suitable concentration for the addition time experiments.
The optimized concentration, 10 mM H 2 O 2 , was added to the fermentation medium at various growth time points, including at day 0 (beginning of fermentation), the 1st day (early exponential phase), the 3rd day (middle exponential phase), and the 5th day (late exponential phase) during pigment fermentation. Cell growth and pigment production, including intracellular and extracellular yellow pigments, were inhibited when H 2 O 2 was added at the beginning of fermentation (Fig. 2), and these parameters were marginally increased when H 2 O 2 was added at the late exponential phase. However, when H 2 O 2 was added at the middle exponential phase, cell growth and pigment production were much improved. The optimal H 2 O 2 addition time was on the 3rd day (middle exponential phase), in which extracellular watersoluble yellow pigment yield reached approximately 209 AU 350 , 42% higher than that of the control (Fig. 2a, b), and intracellular yellow pigments also increased by 35% and reached a maximum of approximately 236 AU 410 (Fig. 2c). The UV-visible spectra and mass spectra of intracellular yellow pigments were compared with literature data (Additional file 3: Figure S2) [40,41], which were identified as the two well-known pigments, monascin and ankaflavin. Monascin was the mainly intracellular yellow pigment based on the HPLC profile (Additional file 4: Figure S3), and the maximum yield of monascin under HIO conditions was 467.75 μg/mL.

Water-soluble yellow pigment production under HIO conditions and DTT-induced reductive conditions
H 2 O 2 acting as an electron acceptor could cause an extracellular oxidative condition (higher extracellular ORP), while dithiothreitol (DTT) acting as a reducing agent could reduce the extracellular ORP [24]. 10 mM of H 2 O 2 was added to the medium on the 3rd day to create an extracellular oxidative environment, and 3 g/L DTT was added on the 3rd day to create an extracellular reducing environment. Extracellular ORP increased immediately after H 2 O 2 addition and reached its highest level on day 4, then decreased to a low level, similar to the control, in the later phase of the fermentation (Fig. 3a). A low extracellular ORP was also promptly obtained when DTT was added to the medium. The cell growth rate increased under an oxidative environment, which resulted in a 19% increase in dry cell weight (DCW), while the DCW under a reducing environment increased only slightly by less than 5% (Fig. 3b). The glucose consumption rate increased immediately upon addition of H 2 O 2 , and the total glucose consumption was higher than that of the control. However, the glucose consumption rate under reducing conditions increased only on day 6 and resulted in the lowest residual glucose concentration (Fig. 3b).
The extracellular water-soluble yellow pigment reaching maximum productivity on day 10 under oxidative conditions and reached 209 AU 350 , which was 42% higher than that of the control. However, the maximum productivity of the control was reached on day 8. In contrast, the yield of extracellular water-soluble yellow pigments under reducing conditions was significantly decreased (Fig. 3c). Under oxidative conditions, the yield of extracellular water-soluble yellow pigments based on glucose consumption corresponded to approximately 1.45 b Spectra of extracellular pigments. c Spectra of intracellular pigments AU/g, which was 37.6% higher than that of the control. The generation rate of intracellular yellow pigments was increased under oxidative conditions (Fig. 3d). The yield of intracellular yellow pigments under oxidative conditions based on glucose consumption corresponded to approximately 1.54 AU/g, which was 26.4% higher than that of the control. In contrast, the yield of intracellular yellow pigments under reducing conditions decreased in the later stages of fermentation and resulted in a lower overall yield (Fig. 3d). The yields of extracellular and intracellular yellow pigments per unit DCW under oxidative conditions were higher than that of the control, while those under reducing conditions were lower than the control (Table 1). Which demonstrated that the oxidative conditions improved yellow pigments productivity, while reducing condition played an opposite role. Therefore, extracellular oxidative conditions were better for yellow pigment production.

Intracellular NADH/NAD+ levels under HIO conditions
As shown in Fig. 4, the ratios of NADH/NAD+ under HIO and control conditions were increased and reached 1.3 on the 3rd day of fermentation. From the 3rd day to the 6th day, a large amount of energy was needed to transport intracellular nutrients to support the fast growth of mycelia and to biosynthesize metabolic intermediates. Consequently, the formation rate of NADH was less than its consumption rate during the growth stage, which resulted in a sharp decrease of the NADH/ NAD+ ratio during this stage. The decrease rate of the NADH/NAD+ ratio under HIO conditions was higher than that of the control and resulted in a lower NADH/ NAD+ ratio in the later exponential phase and stationary phase of the fermentation process. These results indicate that the intracellular oxidoreduction status in M. ruber CGMCC 10910 was significantly influenced by H 2 O 2 addition.

Expression of pigment biosynthesis genes under HIO conditions
The pigment biosynthetic gene cluster has been identified in Monascus spp. [31]. The expression of relative pigment biosynthetic genes under HIO conditions was analysed (Fig. 5). The MpPKS5 gene encodes a pigment polyketide synthase, which is responsible for pigment biosynthesis. MpfasA2 and MpfasB2 encode a fungal  fatty acid synthase, which produces medium-chain (C8 and C10) fatty acyl moieties, and then a trichothecene 3-O-acetyltransferase (AT), encoded by the mppB gene, transfers the medium-chain (C8 and C10) fatty acyl groups onto the polyketide chromophore to complete pigment biosynthesis. mppA, mppC, and mppE encode oxidoreductases, of which a reductive enzyme encoded by mppE controls the biosynthesis of yellow pigments. The gene mppD encodes an amine oxidase/esterase. The genes mppR1 and mppR2 are regulatory genes for pigment biosynthesis [31,32]. The expression levels of the genes MpFasA2, MpFasB2, MpPKS5, mppA, mppB, mppD, mppE, and mppR1 were significantly up-regulated (p < 0.01 or p < 0.05) under HIO conditions (Fig. 5). This corroborates findings regarding the intracellular and extracellular pigment production under oxidative conditions (Figs. 1, 2), indicating that H 2 O 2 stimulates pigment production via upregulation of transcript levels of these genes. However, the genes mppC and mppR2 were significantly down-regulated (p < 0.01 or p < 0.05) and were negatively correlated with pigment production. It has been reported that mppR2 is a negative regulatory factor [42].

Variation of PFK and G6PDH activities under HIO conditions
To understand the physiological consequences of M. ruber CGMCC 10910 caused by HIO conditions, the activities of key redox-dependent enzymes (PFK and G6PDH) in glycolysis and the pentose phosphate pathway (PPP) were analysed. As shown in Fig. 6a, PFK activity was inhibited between the 4th day and the 6th day under HIO conditions, which indicated that the glycolysis pathway was weakened when extracellular oxidative conditions were induced by H 2 O 2 . At the same time, G6PDH activity was enhanced between the 4th day and the 8th day under HIO conditions, which also indicated that the PPP was enhanced under HIO conditions (Fig. 6b). With the metabolic activities weakening, the PFK activity under HIO conditions became equal to that of the control on the 8th day, while G6PDH activity on the 8th day was also higher than that of the control and achieved levels similar to those of the control on the 10th day.
The responses of PFK and G6PDH activities to HIO conditions in this study suggest that there was a metabolic flux redirection from glycolysis to the PPP in M. ruber CGMCC 10910 fermentation, similar to that reported by other studies [28,38,39]. The higher G6PDH activity under HIO conditions suggested a higher PPP metabolic reaction rate, thus providing the pigment biosynthesis pathway with ample precursors.

Discussion
Monascus pigments are mixtures with multiple components [1,10,43]. It is difficult to determine the concentration of Monascus pigments with standard HPLC methods. Thus, the concentration of Monascus yellow pigments in this study was represented by the absorbance at their characteristic wavelengths (350 and 410 nm) [20,21]. In this study, H 2 O 2 was added at different fermentation time points to study the effects of H 2 O 2 addition time on cell growth and yellow pigment production in M. ruber CGMCC 10910. Cell growth and yellow pigment production were both inhibited when H 2 O 2 was added at the beginning of fermentation (Fig. 2), indicating the toxic effect of H 2 O 2 on the conidia germination of M. ruber [44]. Cell growth and metabolic activities were slower at the late exponential phase; thus, there was not a significant effect on cell growth and yellow pigment production when H 2 O 2 was added at this phase. However, extracellular water-soluble yellow pigment production was enhanced by more than 42%, and the DCW increased when H 2 O 2 was added at the middle exponential phase (Fig. 2), a phase in which cell growth and metabolic activity were robust. The optimal amount of H 2 O 2 for yellow pigment production was 10 mM; lower amounts of H 2 O 2 were not sufficient to enhance yellow pigment production, while higher amounts of H 2 O 2 had an inhibitory effect on cell growth and pigment production (Fig. 1). This kind of dose-dependent induction was also observed in other fermentation systems such as Taxus chinensis and Streptomyces hygroscopicus 5008 [39,45]. Because the addition of H 2 O 2 can improve the yield of yellow pigments, additional investigations using the M. ruber CGMCC10910 fermentation system were performed.
H 2 O 2 is an electron acceptor, which can induce an extracellular oxidative environment (higher extracellular ORP) when added to the fermentation medium   [24]. Meanwhile, DTT acts as a reducing agent and can decrease extracellular ORP, thereby creating reducing conditions in the extracellular environment [46]. In this study, H 2 O 2 and DTT were used to modify the extracellular ORP to study the effect of extracellular ORP changes on the water-soluble yellow pigment production in M. ruber CGMCC 10910. The results showed that the yield of yellow pigments (both extracellular and intracellular) and cell growth under oxidative conditions were significantly increased (Fig. 3b-d), and it has been reported that Monascus pigment production was coupled with cell growth [47]; thus, the HIO condition increased yellow pigment production primarily by enhancing cell growth. In contrast, the reducing environment inhibited the biosynthesis of yellow pigments (Fig. 3), and the yield of intracellular yellow pigments under reducing conditions was decreased in the later phase of fermentation, which resulted in a lower yield. Based on the molecular and mppR2 under HIO conditions. Error bars represent the standard deviation (n = 3). **p < 0.01, *p < 0.05 structure, it is possible that yellow pigments were formed by hydrogenation of orange pigments [43]. Under reducing conditions, the decrease of yellow pigments in the later phase of fermentation could be explained by their transformation into orange pigments. However, the glucose consumption under reducing conditions was enhanced (Fig. 3b), which indicated that there was less flux through the biosynthetic pathway of yellow pigments, and that useless byproducts such as alcohol and lactate were generated [26]. These results indicated that extracellular ORP could influence metabolic flux, which is consistent with reported studies [24,28,48].
The intracellular ORP, generally represented by the ratio of NADH/NAD+, has been reported to be influenced by varying the extracellular ORP [24]. In this study, the ratio of NADH/NAD+ in M. ruber CGMCC 10910 was significantly influenced by H 2 O 2 addition (Fig. 4). During the fermentation process, H 2 O 2 either accepted electrons from NADH directly, or was degraded to H 2 O and O 2 . As a result, a portion of NADH was oxidized to NAD+ by H 2 O 2 and it achieved a redox balance so that many metabolites, such as NADH dehydrogenases and inefficient terminal oxidases (cytochrome bd), were not needed to balance NADH/NAD+ metabolism under HIO conditions [49]. It was speculated that H 2 O 2 regulating NADH/NAD+ ratios maybe one of the reasons why DCW and yellow pigment yield were increased under HIO conditions (Fig. 3). During the submerged fermentation of Monascus purpureus, high stirring speed damages the mycelium and results in a lower pigments yield [50]. And the mycelium morphology of Monascus anka plays an important role in polyketides production [51]. Our study found that electron acceptors can be provided by H 2 O 2 and without increasing stirring speed, which would damage the mycelium morphology of M. ruber CGMCC 10910. Moreover, it was reported that yellow pigments are the reduction product of orange pigments [43], in which two of the C-C double bonds in the conjugated chain of two orange pigments are reduced to C-C single bonds and reducing power (NADH or NADPH) is required in this reduction reaction [12]. Thus, the ratio of NADH/NAD+ under oxidative conditions was lower than that of the control (Fig. 4).
The biosynthesis of Monascus pigment requires many primary metabolites, such as acetyl-CoA, malonyl-CoA, NADH, and NADPH [52]. NADPH acted as reducing power used for polyketide and lipid biosynthesis [31], which was mainly produced by PPP, which is a branch pathway of glycometabolism [29]. The activities of key enzymes involved in the glycolysis pathway and PPP were analysed in this study. The activity of PFK under HIO conditions was lower than that of the control group (Fig. 6a), which indicates that PFK was allosterically inhibited by higher metabolite concentrations in the glycolysis pathway under HIO conditions [53]. G6PDH activity under oxidative conditions was higher than that of the control group (Fig. 6b), which indicated that the metabolites involved in PPP under oxidative condition were significantly up-regulated. These results showed that M. ruber CGMCC 10910 redirects its metabolic flux from the glycolysis pathway to the PPP, and more energy (NADPH) could be produced for polyketide and lipid biosynthesis [28,35,36], which is consistent with reported studies that show that metabolic flux was altered from the glycolysis pathway to the PPP, and an abundance of precursors were available for validamycin A biosynthesis in Streptomyces hygroscopicus 5008 fermentation in response to various oxidant treatments [38,39]. Moreover, analysis of the relative expression levels of the pigment biosynthetic genes showed that the genes MpFasA2, MpFasB2, MpPKS5, mppA, mppB, mppD, mppE, and mppR1 were significantly up-regulated under HIO conditions (p < 0.01 or p < 0.05) (Fig. 5). Thus, HIO conditions increase yellow pigment accumulation via up-regulation of the transcript levels of relative genes and improving the precursor concentrations for pigment biosynthesis.

Conclusions
Extracellular oxidative conditions caused by H 2 O 2 addition at suitable fermentation time points can improve the yield of extracellular water-soluble yellow pigments in