The Effect of Blue Light on the Production of Citrinin in Monascus purpureus M9 by Regulating the mraox Gene through lncRNA AOANCR

Blue light, as an important environmental factor, can regulate the production of various secondary metabolites of Monascus purpureus M9, including mycotoxin-citrinin, pigments, and monacolin K. The analysis of citrinin in Monascus M9 exposed to blue light for 0 min./d, 15 min./d, and 60 min./d showed that 15 min./d of blue light illumination could significantly increase citrinin production, while 60 min./d of blue light illumination decreased citrinin production. Analysis of long non-coding RNA (LncRNA) was performed on the transcripts of Monascus M9 under three culture conditions, and this analysis identified an lncRNA named AOANCR that can negatively regulate the mraox gene. Fermentation studies suggested that alternate respiratory pathways could be among the pathways that are involved in the regulation of the synthesis of citrinin by environmental factors. Aminophylline and citric acid were added to the culture medium to simulate the process of generating cyclic adenosine monophosphate (cAMP) in cells under illumination conditions. The results of the fermentation showed that aminophylline and citric acid could increase the expression of the mraox gene, decrease the expression of lncRNA AOANCR, and reduce the yield of citrinin. This result also indicates a reverse regulation relationship between lncRNA AOANCR and the mraox gene. A blue light signal might regulate the mraox gene at least partially through lncRNA AOANCR, thereby regulating citrinin production. Citrinin has severe nephrotoxicity in mammals, and it is important to control the residual amout of citrinin in red yeast products during fermentation. LncRNA AOANCR and mraox can potentially be used as new targets for the control of citrinin production.


Introduction
Monascus is a traditional Chinese medicinal and food homologous fungus. Its secondary metabolites mainly include pigments, monacolin K, and citrinin. Monacolin K is the main lipid-lowering functional substance of Monascus and it has a competitive inhibitory effect on HMG-CoA reductase and inhibits cholesterol synthesis [1]. Monascus pigments are a mixture of compounds with a that the yield of citrinin increased, which indicated that the alternative respiratory pathway also regulates the production of citrinin. Based on the results of this study, we speculate that blue light regulates mraox to affect the production of citrinin by regulating lncRNA AOANCR.

Effect of Blue Light on Citrinin Production in M9
A culture medium inoculated with Monascus M9 seed solution was placed under blue light with a light intensity of 100 lux for 0, 15, and 60 min. on one day, with six solutions that were placed in parallel in each group. After eight days of culturing, the yield of citrinin and biomass (dry weight) in fermentation products were determined, as shown in Figure 1.
Toxins 2019, 11, x FOR PEER REVIEW 3 of 16 found that the yield of citrinin increased, which indicated that the alternative respiratory pathway also regulates the production of citrinin. Based on the results of this study, we speculate that blue light regulates mraox to affect the production of citrinin by regulating lncRNA AOANCR.

Effect of Blue Light on Citrinin Production in M9
A culture medium inoculated with Monascus M9 seed solution was placed under blue light with a light intensity of 100 lux for 0, 15, and 60 min. on one day, with six solutions that were placed in parallel in each group. After eight days of culturing, the yield of citrinin and biomass (dry weight) in fermentation products were determined, as shown in Figure 1. The effects of blue light on citrinin production and biomass. Blue light was generated by a blue-LED box which provided stable and uniform pure blue light (460 nm). The light intensity was 100 lux. The samples were treated with blue light once a day, for eight days of culturing. Citrinin was detected by high performance liquid chromatography (HPLC), and the biomass is expressed as the dry weight of mycelium per unit fermentation volume. The data represent the mean ± SD of six replicates. *** p < 0.001, * p < 0.05 compared with biomass and citrinin production for 0 min./d blue light illumination. T-test was performed at p < 0.05.
The biomass of Monascus M9 increased from 27.3 g/l in the dark to 29.5 g/l in blue light (15 min/d), however the difference was not significant. The biomass of Monascus M9 significantly decreased from 27.3 g/l in the dark to 23.27 g/l in blue light (60 min/d). This might be due to the fact that, although blue light is harmful to the growth of Monascus M9, the short illumination time may stimulate the compensatory growth of Monascus.
The citrinin yield of Monascus M9 cultured in blue light (15 min/d; 363 μg/g) was significantly higher than that of Monascus M9 cultured in dark conditions (288 μg/g). The yield of citrinin decreased (272 μg/g) after 60 min of blue light irradiation per day, however the difference was not significant. Short-term blue light irradiation can stimulate Monascus to produce more citrinin, however long-time blue light irradiation reduces citrinin production. That is, the effect of blue light on citrinin production does not increase with the extension of illumination time, which indicates that blue light cannot directly regulate the production of citrinin. It may be that the blue light signal regulates citrinin production through other regulatory modes and that the light exerts a protective mechanism on Monascus. Figure 1. The effects of blue light on citrinin production and biomass. Blue light was generated by a blue-LED box which provided stable and uniform pure blue light (460 nm). The light intensity was 100 lux. The samples were treated with blue light once a day, for eight days of culturing. Citrinin was detected by high performance liquid chromatography (HPLC), and the biomass is expressed as the dry weight of mycelium per unit fermentation volume. The data represent the mean ± SD of six replicates. *** p < 0.001 compared with biomass and citrinin production for 0 min./d blue light illumination. T-test was performed at p < 0.05.

Analysis of lncRNA in the Transcriptome of Monascus M9
The biomass of Monascus M9 increased from 27.3 g/L in the dark to 29.5 g/L in blue light (15 min/d), however the difference was not significant. The biomass of Monascus M9 significantly decreased from 27.3 g/L in the dark to 23.27 g/L in blue light (60 min/d). This might be due to the fact that, although blue light is harmful to the growth of Monascus M9, the short illumination time may stimulate the compensatory growth of Monascus.
The citrinin yield of Monascus M9 cultured in blue light (15 min/d; 363 µg/g) was significantly higher than that of Monascus M9 cultured in dark conditions (288 µg/g). The yield of citrinin decreased (272 µg/g) after 60 min of blue light irradiation per day, however the difference was not significant. Short-term blue light irradiation can stimulate Monascus to produce more citrinin, however long-time blue light irradiation reduces citrinin production. That is, the effect of blue light on citrinin production does not increase with the extension of illumination time, which indicates that blue light cannot directly regulate the production of citrinin. It may be that the blue light signal regulates citrinin production through other regulatory modes and that the light exerts a protective mechanism on Monascus.

Analysis of lncRNA in the Transcriptome of Monascus M9
After bioinformatics analysis, 1455 nucleotide sequences that conformed to the biological characteristics of lncRNA were screened from the transcriptome of monascus M9. Table 1 shows the results. Some lncRNAs are involved in the basic processes of gene regulation, including the modification and structure of chromatin and direct transcriptional regulation. Gene regulation might occur in the form of cis or trans regulation. Cis regulation usually refers to the mode of action of DNA sequences from the same chromosome in the direct regulation of the expression of other nearby genes. In this study, we used the predicted genome annotation information of lncRNA and the reference genome information of the species to identify possible cis-acting target genes of lncRNA. LncRNA, which is normally transcribed in the same direction as the target gene in the promoter region, usually promotes the expression of the target gene, while reverse transcription inhibits the expression of the target gene. In this work, we analyzed the correlation between each lncRNA and the expression of mRNA, and we selected the corresponding relationship between lncRNA and mRNA with a correlation greater than or equal to 0.7 and P value less than or equal to 0.05 was selected for further study.
According to the criterion of difference significance (i.e., more than twice the difference in expression of lncRNA, and FDR ≤ 0.05), we screened and counted the up-and down-regulation of a significant difference in expression of lncRNA under blue light illumination. Figure 2 shows the results. characteristics of lncRNA were screened from the transcriptome of monascus M9. Table 1 shows the results. Some lncRNAs are involved in the basic processes of gene regulation, including the modification and structure of chromatin and direct transcriptional regulation. Gene regulation might occur in the form of cis or trans regulation. Cis regulation usually refers to the mode of action of DNA sequences from the same chromosome in the direct regulation of the expression of other nearby genes. In this study, we used the predicted genome annotation information of lncRNA and the reference genome information of the species to identify possible cis-acting target genes of lncRNA. LncRNA, which is normally transcribed in the same direction as the target gene in the promoter region, usually promotes the expression of the target gene, while reverse transcription inhibits the expression of the target gene. In this work, we analyzed the correlation between each lncRNA and the expression of mRNA, and we selected the corresponding relationship between lncRNA and mRNA with a correlation greater than or equal to 0.7 and P value less than or equal to 0.05 was selected for further study.
According to the criterion of difference significance (i.e., more than twice the difference in expression of lncRNA, and FDR ≤0.05), we screened and counted the up-and down-regulation of a significant difference in expression of lncRNA under blue light illumination. Figure 2 shows the results.   Additionally, hierarchical clustering was used to analyze differentially expressed lncRNAs. Figure 3 shows the results. Figure 2. Numbers of lncRNAs with significant changes in expression under different blue light illumination conditions after eight days of culturing. The Monascus M9 culture was divided into three groups: dark culture conditions (D), blue light illumination for 15 min (B15), and blue light illumination for 60 min (B60) in one day; groups D, B15 and B60 had two parallels. Blue light was generated by a blue-LED box, which provided stable and uniform pure blue light (460 nm). The light intensity was 100 lux.
Additionally, hierarchical clustering was used to analyze differentially expressed lncRNAs. Figure 3 shows the results.  All of the differentially expressed lncRNAs and mRNAs were analyzed, and a target mRNA (g4769.t1) named mraox was identified, which corresponded to the negative regulation of an lncRNA (TCONS_00015111), named AOANCR. LncRNA AOANCR is 398 bp in length, is located in the mraox promoter region, has a nucleotide similarity of 79% in the range of 389 bases to 894 bases of mraox, and it has reverse transcription with mraox. Based on the characteristics and sequence similarity of lncRNA and mraox, we speculate that lncRNA AOANCR is a reverse-regulated lncRNA of mraox.
After blast analysis, mraox and Monascus ruber M7 alternative oxidase gene (GenBank: FJ640864.1) reached 100% similarity, and mraox was therefore presumed to be an alternate oxidase-encoding gene. Studies have shown that the gene encoding alternative oxidase plays an important role in the signal transduction of environmental factors, such as pH and osmotic pressure. It is speculated that this gene might also be involved in light signal transduction and citrinin regulation.

Expression of lncRNA AOANCR and Mraox Detection in Different Blue Light Conditions
In order to prove the expression relationship between lncRNA AOANCR and mraox, their expression was determined by qPCR. The expression of mraox and lncRNA AOANCR was detected in samples illuminated with blue light for 5 min., 15 min., 30 min., and 60 min. on the fourth day of cultivation. The expression of mraox and lncRNA AOANCR in a dark culture condition was taken as the reference value (value 1). The results are shown in Figure 4.
low expression of transcript. Color variation from green to red indicates incrementally higher amounts of lncRNA expression. B15: illuminated with blue light for 15 min per day; B60: illuminated with blue light for 60 min per day; D: cultured in dark conditions. Two parallel samples were performed for each group. The blue light was generated by a blue-LED box, which provided stable and uniform pure blue light (460 nm). The light intensity was 100 lux.
All of the differentially expressed lncRNAs and mRNAs were analyzed, and a target mRNA (g4769.t1) named mraox was identified, which corresponded to the negative regulation of an lncRNA (TCONS_00015111), named AOANCR. LncRNA AOANCR is 398 bp in length, is located in the mraox promoter region, has a nucleotide similarity of 79% in the range of 389 bases to 894 bases of mraox, and it has reverse transcription with mraox. Based on the characteristics and sequence similarity of lncRNA and mraox, we speculate that lncRNA AOANCR is a reverse-regulated lncRNA of mraox.
After blast analysis, mraox and Monascus ruber M7 alternative oxidase gene (GenBank: FJ640864.1) reached 100% similarity, and mraox was therefore presumed to be an alternate oxidase-encoding gene. Studies have shown that the gene encoding alternative oxidase plays an important role in the signal transduction of environmental factors, such as pH and osmotic pressure. It is speculated that this gene might also be involved in light signal transduction and citrinin regulation.

Expression of lncRNA AOANCR and Mraox Detection in Different Blue Light Conditions
In order to prove the expression relationship between lncRNA AOANCR and mraox, their expression was determined by qPCR. The expression of mraox and lncRNA AOANCR was detected in samples illuminated with blue light for 5 min., 15 min., 30 min., and 60 min. on the fourth day of cultivation. The expression of mraox and lncRNA AOANCR in a dark culture condition was taken as the reference value (value 1). The results are shown in Figure 4. Transcription levels were normalized to that of the actin gene. In order to standardize the results, we took the gene expression levels (mraox and lncRNA AOANCR) on the fourth day under dark conditions as the reference value (value 1). Data are expressed as the relative mRNA level for each gene and represent the average values of three separate experiments. The blue light was generated by a blue-LED box, which provided stable and uniform pure blue light (460 nm). The light intensity was 100 lux. Error bars represent the standard deviation (n = 3). *** p < 0.001, * p < 0.05 compared with gene ratio under dark conditions. T-test for the values was performed at p < 0.05. Transcription levels were normalized to that of the actin gene. In order to standardize the results, we took the gene expression levels (mraox and lncRNA AOANCR) on the fourth day under dark conditions as the reference value (value 1). Data are expressed as the relative mRNA level for each gene and represent the average values of three separate experiments. The blue light was generated by a blue-LED box, which provided stable and uniform pure blue light (460 nm). The light intensity was 100 lux. Error bars represent the standard deviation (n = 3). *** p < 0.001, * p < 0.05 compared with gene ratio under dark conditions. T-test for the values was performed at p < 0.05. From the measured expression levels of mraox and lncRNA AOANCR, it can be seen that the trend was opposite for mraox and lncRNA AOANCR. In the case of blue light illumination for 15 min/d, the expression of lncRNA AOANCR was the highest, and the expression of mraox was the lowest. In the case of blue light illumination for 60 min/d, the expression of lncRNA AOANCR was the lowest and the expression of mraox was the highest. These opposite trends further prove that lncRNA negatively regulates mraox. There is no linear relationship between gene expression and blue light illumination time, which might be related to the feedback inhibition of the blue light signal. Gene regulation by blue light is a hierarchical network regulation system, and different illumination times trigger different levels of the gene regulation network. In the case of Neurospora crassa, light-regulated genes can be divided into early response genes and late response genes [30]. WCC directly controls early light-regulated genes [31]. Early light-induced proteins regulate subsequent light-induced gene expression, while gene expression at the second level affects the next level. With the existence of photoadaptation, the expression of genes that was initially induced by light will be inhibited [32][33][34][35][36][37][38]. The downstream genes that were activated by the original light-regulated genes were subsequently suppressed, and the process of activation and suppression of multiple genes formed a hierarchical network of light regulation. The expression of lncRNA AOANCR and mraox determined in the present study suggests that photoadaptation also exists in Monascus.

Effects of Alternating Respiratory Pathway Mediated by Alternate Oxidase on Citrinin Synthesis in Monascus M9
From the aforementioned experimental results, it was found that blue light illumination can change the expression of lncRNA AOANCR and mraox, and the two genes have a reverse regulation relationship. However, fermentation studies are required to demonstrate whether the alternative respiratory pathway that is mediated by the mraox gene can regulate the synthesis of citrinin. 2 mm of the alternative oxidase pathway inhibitor salicylhydroxamic acid (SHAM) was added to the medium to detect the yield of citrinin after eight days of fermentation in order to study whether the alternative oxidation pathway affected the citrinin biosynthesis of Monascus M9. The experimental results are shown in Figure 5.
From the measured expression levels of mraox and lncRNA AOANCR, it can be seen that the trend was opposite for mraox and lncRNA AOANCR. In the case of blue light illumination for 15 min/d, the expression of lncRNA AOANCR was the highest, and the expression of mraox was the lowest. In the case of blue light illumination for 60 min/d, the expression of lncRNA AOANCR was the lowest and the expression of mraox was the highest. These opposite trends further prove that lncRNA negatively regulates mraox.
There is no linear relationship between gene expression and blue light illumination time, which might be related to the feedback inhibition of the blue light signal. Gene regulation by blue light is a hierarchical network regulation system, and different illumination times trigger different levels of the gene regulation network. In the case of Neurospora crassa, light-regulated genes can be divided into early response genes and late response genes [30]. WCC directly controls early light-regulated genes [31]. Early light-induced proteins regulate subsequent light-induced gene expression, while gene expression at the second level affects the next level. With the existence of photoadaptation, the expression of genes that was initially induced by light will be inhibited [32][33][34][35][36][37][38]. The downstream genes that were activated by the original light-regulated genes were subsequently suppressed, and the process of activation and suppression of multiple genes formed a hierarchical network of light regulation. The expression of lncRNA AOANCR and mraox determined in the present study suggests that photoadaptation also exists in Monascus.

Effects of Alternating Respiratory Pathway Mediated by Alternate Oxidase on Citrinin Synthesis in Monascus M9
From the aforementioned experimental results, it was found that blue light illumination can change the expression of lncRNA AOANCR and mraox, and the two genes have a reverse regulation relationship. However, fermentation studies are required to demonstrate whether the alternative respiratory pathway that is mediated by the mraox gene can regulate the synthesis of citrinin. 2 mm of the alternative oxidase pathway inhibitor salicylhydroxamic acid (SHAM) was added to the medium to detect the yield of citrinin after eight days of fermentation in order to study whether the alternative oxidation pathway affected the citrinin biosynthesis of Monascus M9. The experimental results are shown in Figure 5. In the other group, 2 mm of SHAM was added to the YES medium before fermentation. Samples were taken from the same culture every 24 h from the third to the eighth day. Values are the averages of three independent experiments. Error bars represent the standard deviation (n = 3). *** p < 0.001, * p < 0.05 compared with citrinin production under normal conditions. T-test for the values was performed at p < 0.05. The results showed that the yield of citrinin produced in the normal medium (i.e., without the addition of SHAM) reached the highest value on the fourth day of culture, i.e., 290.5 µg/L. When SHAM was added to the culture, the citrinin yield also reached the highest value on the fourth day, i.e., 362.3 µg/L, a 24.7% increase as compared to the culture in the normal medium. When the alternation oxidase pathway was inhibited, the yield of citrinin was significantly increased. This result indicates that the alternative oxidase pathway can affect the production of citrinin. Blue light can regulate the expression of the mraox gene by regulating the expression of lncRNA AOANCR, which then regulates the alternate respiratory pathway, thereby affecting the production of citrinin.

Citric Acid and Aminophylline Regulate Citrinin Production by Regulating Mraox
After blue light illumination is applied to cells, cyclic adenosine monophosphate (cAMP) is formed in the cells, which in turn produces a series of downstream reactions [39]. Previous studies found that conidia fertilization of filamentous fungi cultured under dark conditions with the exogenous addition of cAMP was similar to that under blue light illumination. However, there is no similar phenomenon with the addition of exogenous cAMP if the photoreceptor gene is knocked out [40,41]. It can be seen that blue light is transmitted to the photoreceptor gene through cAMP and then regulates other physiological functions.
The exogenous addition of citric acid and aminophylline can increase the cAMP content in fungal cells [42]. Citric acid and aminophylline were added to the medium in order to verify whether the regulatory effects of lncRNA AOANCR and mraox are also related to the cAMP pathway. The gene expressions of lncRNA AOANCR and mraox, and citrinin production, were detected. The experimental results are shown in Figure 6.
The results showed that the yield of citrinin produced in the normal medium (i.e., without the addition of SHAM) reached the highest value on the fourth day of culture, i.e., 290.5 μg/l. When SHAM was added to the culture, the citrinin yield also reached the highest value on the fourth day, i.e., 362.3 μg/l, a 24.7% increase as compared to the culture in the normal medium. When the alternation oxidase pathway was inhibited, the yield of citrinin was significantly increased. This result indicates that the alternative oxidase pathway can affect the production of citrinin. Blue light can regulate the expression of the mraox gene by regulating the expression of lncRNA AOANCR, which then regulates the alternate respiratory pathway, thereby affecting the production of citrinin.

Citric Acid and Aminophylline Regulate Citrinin Production by Regulating Mraox
After blue light illumination is applied to cells, cyclic adenosine monophosphate (cAMP) is formed in the cells, which in turn produces a series of downstream reactions [39]. Previous studies found that conidia fertilization of filamentous fungi cultured under dark conditions with the exogenous addition of cAMP was similar to that under blue light illumination. However, there is no similar phenomenon with the addition of exogenous cAMP if the photoreceptor gene is knocked out [40,41]. It can be seen that blue light is transmitted to the photoreceptor gene through cAMP and then regulates other physiological functions.
The exogenous addition of citric acid and aminophylline can increase the cAMP content in fungal cells [42]. Citric acid and aminophylline were added to the medium in order to verify whether the regulatory effects of lncRNA AOANCR and mraox are also related to the cAMP pathway. The gene expressions of lncRNA AOANCR and mraox, and citrinin production, were detected. The experimental results are shown in Figure 6. In the other two groups, 1% citric acid and 10 mM aminophylline were added to the YES medium before fermentation, respectively. Samples were taken every 24 h from the third to the eighth day. Values are the averages of three independent experiments. Error bars represent the standard deviation (n = 3). *** p < 0.001, * p < 0.05 compared with citrinin production in normal conditions. Ttest for the values was performed at p < 0.05.
The results showed that citric acid and aminophylline can reduce citrinin production. In the culture in a normal medium, the maximum citrinin yield of 290.5 μg/l was produced on the fourth day. When citric acid or aminophylline were added to the culture medium, maximum citrinin yields of 200.3 and 148 μg/l, respectively, were produced on the fifth day, 31.3% and 26.1% lower than the yields under normal conditions, respectively. In the other two groups, 1% citric acid and 10 mM aminophylline were added to the YES medium before fermentation, respectively. Samples were taken every 24 h from the third to the eighth day. Values are the averages of three independent experiments. Error bars represent the standard deviation (n = 3). *** p < 0.001, * p < 0.05 compared with citrinin production in normal conditions. T-test for the values was performed at p < 0.05.
The results showed that citric acid and aminophylline can reduce citrinin production. In the culture in a normal medium, the maximum citrinin yield of 290.5 µg/L was produced on the fourth day. When citric acid or aminophylline were added to the culture medium, maximum citrinin yields of 200.3 and 148 µg/L, respectively, were produced on the fifth day, 31.3% and 26.1% lower than the yields under normal conditions, respectively.
Taking the gene expression in the normal medium as the reference value (value 1), the gene expression levels of mraox and lncRNA AOANCR were detected on the eighth day of culture. Figure 7 shows the results. Taking the gene expression in the normal medium as the reference value (value 1), the gene expression levels of mraox and lncRNA AOANCR were detected on the eighth day of culture. Figure  7 shows the results. Figure 7. Expression of mraox and lncRNA AOANCR under different culture conditions. "normal" means fermentation with normal YES medium. In the other two groups, 1% citric acid and 10 mM aminophylline were added to the YES medium before fermentation, respectively. The cultures were sampled to detect gene expression by qPCR on the eighth day. Transcriptional levels were normalized to that of the actin gene. In order to standardize the results, we took the gene expression levels of mraox and lncRNA AOANC accumulated on the eighth day of samples cultured in normal medium as the reference value (value 1). Data are expressed as the relative mRNA level for each gene and they represent the average values from three separate experiments. Error bars represent the standard deviation (n = 3). *** p < 0.001, * p < 0.05 compared with the gene ratio of samples cultured in normal medium. T-test for the values was performed at p < 0.05.
The results of the gene expression analysis showed that, after adding citric acid and aminophylline to the culture medium, the expression of mraox increased by 137% and 229%, respectively, and the expression of lncRNA AOANCR decreased by 15% and 30%, respectively. Citric acid and aminophylline both increased the expression of mraox and reduced the expression of lncRNA AOANCR.
Aminophylline and citric acid can stimulate the production of cAMP in cells. This suggests that blue light might also regulate the expression of lncRNA and mraox through the cAMP pathway. After aminophylline and citric acid were added to the culture medium, the expression of lncRNA and mraox showed opposite trends, which further confirmed the reverse regulation relationship between lncRNA AOANCR and mraox.

Discussion
As an environmental factor, blue light plays an important role in the synthesis of secondary metabolites of Monascus. Blue light signaling regulates the metabolism of Monascus in a complex regulatory relationship that involves multiple regulatory pathways [43]. Light can affect mycelial growth, promote the sporulation [21,22] of Monascus, and can also regulate the production of citrinin, monacolin K, and pigments, and the composition of pigments in gene expression level [23,24]. Di concluded that the effect of blue light on Monascus is global [28]. Blue light treatment can reduce the basal metabolic levels of Monascus M9, for example, in carbon and nitrogen metabolism, can degrade branched-chain amino acids, and can up-regulate aromatic amino acids. Blue light can also activate IP3/Ca 2+ and DAG/PKC signaling pathways, inhibit the MAPK signaling pathway, and regulate the expression of key enzyme genes in metabolic pathways by regulating global transcription factors, thereby regulating the growth and secondary metabolism of Monascus. However, there are no reports of alternating respiratory pathways regulating citrinin production. Figure 7. Expression of mraox and lncRNA AOANCR under different culture conditions. "normal" means fermentation with normal YES medium. In the other two groups, 1% citric acid and 10 mM aminophylline were added to the YES medium before fermentation, respectively. The cultures were sampled to detect gene expression by qPCR on the eighth day. Transcriptional levels were normalized to that of the actin gene. In order to standardize the results, we took the gene expression levels of mraox and lncRNA AOANC accumulated on the eighth day of samples cultured in normal medium as the reference value (value 1). Data are expressed as the relative mRNA level for each gene and they represent the average values from three separate experiments. Error bars represent the standard deviation (n = 3). *** p < 0.001, * p < 0.05 compared with the gene ratio of samples cultured in normal medium. T-test for the values was performed at p < 0.05.
The results of the gene expression analysis showed that, after adding citric acid and aminophylline to the culture medium, the expression of mraox increased by 137% and 229%, respectively, and the expression of lncRNA AOANCR decreased by 15% and 30%, respectively. Citric acid and aminophylline both increased the expression of mraox and reduced the expression of lncRNA AOANCR.
Aminophylline and citric acid can stimulate the production of cAMP in cells. This suggests that blue light might also regulate the expression of lncRNA and mraox through the cAMP pathway. After aminophylline and citric acid were added to the culture medium, the expression of lncRNA and mraox showed opposite trends, which further confirmed the reverse regulation relationship between lncRNA AOANCR and mraox.

Discussion
As an environmental factor, blue light plays an important role in the synthesis of secondary metabolites of Monascus. Blue light signaling regulates the metabolism of Monascus in a complex regulatory relationship that involves multiple regulatory pathways [43]. Light can affect mycelial growth, promote the sporulation [21,22] of Monascus, and can also regulate the production of citrinin, monacolin K, and pigments, and the composition of pigments in gene expression level [23,24]. Di concluded that the effect of blue light on Monascus is global [28]. Blue light treatment can reduce the basal metabolic levels of Monascus M9, for example, in carbon and nitrogen metabolism, can degrade branched-chain amino acids, and can up-regulate aromatic amino acids. Blue light can also activate IP3/Ca 2+ and DAG/PKC signaling pathways, inhibit the MAPK signaling pathway, and regulate the expression of key enzyme genes in metabolic pathways by regulating global transcription factors, thereby regulating the growth and secondary metabolism of Monascus. However, there are no reports of alternating respiratory pathways regulating citrinin production.
We found that the mraox gene is sensitive to blue light signals by bioinformatic analysis of transcriptome. We also found that the addition of the alternative respiratory pathway inhibitor SHAM to the medium can stimulate the production of citrinin, which indicates that the alternate respiratory pathway regulated by the mraox gene may be an important factor in regulating the production of citrinin. In fungi, alternate respiratory pathways exist in two forms [44], namely (1) alternative NADH dehydrogenase pathways, which are mediated by alternate NADH dehydrogenases, and (2) the alternate oxidase pathway, which is mediated by alternative oxidase [45]. The alternate oxidase pathway is the main pathway that links environmental factors and microbial metabolism. N stress, temperature, illumination, and high osmotic pressure [46][47][48][49] can affect the activity of alternative oxidase (AOX).
Blue light signals contribute to the production of reactive oxygen species (ROS) in organisms and increase photooxidative stress in cells [50], while intracellular ROS is a major factor in the activation of the alternate oxidase pathway [51]. On the one hand, AOX acts as an ROS scavenger, reducing ROS production [52]; on the other hand, AOX can inhibit excessive reduction of electron transport chain complexes in mitochondria, preventing excessive ROS production and activating the ROS clearance system to reduce ROS production [53,54]. Therefore, alternative oxidase is a protective mechanism that evolved to allow organisms to adapt to harsh environments. That is, organisms regulate their metabolism by alternative oxidase pathways to adapt to environmental changes.
The alternative oxidase pathway is an important pathway for Monascus to sense external environmental signals, especially environmental stress. The gene of the alternative oxidase, mraox, can enhance the resistance of Monascus to environmental stress and promote spore germination under environmental stress [30]. In this study, it was found that mraox is also one of the genes that regulate citrinin synthesis. It is generally believed that aod-2 and aod-5 are the regulatory genes of AOX [55]. In this study, we identified lncRNA AOANCR, whose base sequence is similar to the base sequence of the mraox gene and whose expression is opposite to that of the mraox gene; therefore, lncRNA AOANCR is presumed to be a negative regulator of mraox.
The regulation of lncRNA is a widely used approach in the medical field. For example, the feedback mechanism of the virus-induced lncRNA-mediated metabolic promotion of viral infection can be used as a potential target for developing broad-acting antiviral therapeutics [56]. Additionally, lncRNA is often associated with the growth and development of plants and microorganisms [57][58][59] and their adaptation to the surrounding environment [60]. Secondary metabolites are active substances secreted by fungi in harsh environments [61]. The external environment might regulate secondary metabolism through lncRNAs.
In S. cerevisiae, 11 lncRNAs are known to be related to metabolism, including galactose metabolism [62,63], phosphate metabolism [64], asparagine catabolism [65], fatty acid metabolism [66], biosynthesis of serine and glycine, etc. [67]. LncRNA plays an important role in the regulation of fungal metabolism. Unlike other modes of regulation, lncRNA-mediated regulation is more rapid and flexible, which is probably due to the fact that lncRNA can be quickly generated without the need for translation, and it rapidly degrades [68]. Therefore, the metabolism of citrinin can be regulated during fermentation while using lncRNA AOANCR as a target.

Conclusions
Citrinin is produced during fermentation, and it is therefore very important to regulate the metabolism of Monascus during fermentation to reduce the citrinin residue in red yeast products. In recent years, studies of the regulation of the metabolism of Monascus have focused on the effects of light with different intensities, different colors, and different wavelengths on sporulation and secondary metabolism [21,69,70]. Blue light can affect the expression of individual genes in the citrinin synthesis gene cluster of Monascus, however no studies have investigated the effect of blue light on lncRNA in Monascus. In this study, 1455 hypothetical lncRNAs were identified by bioinformatic analysis of the Monascus purpureus M9 transcriptome. The lncRNA AOANCR in the promoter region of the gene mraox has high homology and negative regulation. It is speculated that blue light may regulate lncRNA AOANCR through the cAMP pathway and then negatively-regulates mraox to effect citrinin production through the alternate respiratory pathway, which is a new mechanism for the regulation of citrinin synthesis. Further research is required to determine whether the metabolic regulation of environmental factors other than blue light also involves the participation of lncRNA AOANCR. In addition to light, other environmental factors, such as pH and temperature, can significantly affect the synthesis of citrinin [71,72]. LncRNA should also be considered as an important influencing factor in studies of the mechanism of the regulation of environmental factors to Monascus.

Fungal Strain and Culture Conditions
Monascus purpureus M9 was maintained on potato dextrose agar (PDA) for five days at 30 • C in the State Key Laboratory of Food Nutrition and Safety, Tianjin, China. Spores were harvested with 3 mL sterile water and then inoculated into 75 mL seed medium (30 g rice powder/L, 2.5 g KH 2 PO 4 /L, 3 g NaNO 3 /L, and 1 g MgSO 4 ·7H 2 O/L; the initial pH of the medium was adjusted to 4.5 with lactic acid) in 200 mL flasks. The cultures were incubated at 30 • C for 20 h with shaking at 180 rpm. For citrinin production and gene expression testing, 5 mL (10% v/v) of the seed fermentation broth was inoculated into 50 mL YES medium (normal medium, consisting of 40 g yeast extract/L and 160 g sucrose/L) in a 250 mL flask and incubated under static conditions at 30 • C [60,68].

Light Exposure Conditions
Light chambers were constructed and equipped with 9 W LEDs (460 nm) with a light intensity of 100 lux.

Transcriptome Data Acquisition
Transcriptome data were shared with Di [28]. The Monascus M9 culture was divided into three groups: dark culture condition, blue light culture condition for 15 min, and blue light culture condition for 60 min in one day, with two parallel groups.

Bioinformatics Analysis
LncRNA prediction: Firstly, using the latest genome annotation information that was provided by UCSC, Ensembl, and GENCODE, the transcripts overlapping with the known protein-coding, microRNA, Transfer RNA (tRNA), small nucleolar RNA (snoRNA), Ribosomal RNA (rRNA), and pseudogene annotation regions in the transcriptome were filtered. Based on the characteristics of lncRNA, transcripts containing only one exon, transcripts less than 200 bp in length, and transcripts that were supported by less than three reads, were filtered out. The HMMER-3 software [73] was used to evaluate all possible open reading frames of transcripts and filter out transcripts containing protein domains. The Coding Potential Calculator (CPC) software [74] was used to predict protein-coding potential and filter out the transcripts with protein-coding potential.
Cis-regulated lncRNA and trans-regulated lncRNA prediction: A region within 3000 bp upstream of the target gene was selected as the promoter region of the gene, and a cis-regulatory lncRNA was searched for in the promoter region of the gene. The analysis software was performed using the Bedtools intersect method. Trans-regulation is another way in which lncRNA acts on its target genes. The positional relationships between lncRNA and target genes were not considered, and base pairing directly identified lncRNA and target genes.
Blastn was used to identify the possible trans target genes for lncRNA. Gene difference analysis was carried out while using the DESeq2 software (v1.6.3) of the Bioconductor software package (v1.6.3, Bioconductor, Boston, MA, USA) [75]. The results of the DESeq detection were screened according to the difference significance criterion (i.e., differential lncRNA expression changed more than twice and FDR ≤ 0.05).

Analysis of Citrinin Production
Citrinin production was determined by high-performance liquid chromatography (HPLC) on a C18 column at 25 • C (5 µm, 250 mm × 4.6 mm) after filtration of the fermentation broth through a Toxins 2019, 11, 536 12 of 16 0.45 µm filter. The mobile phase was acetonitrile/water/methanol (7:2:1, v/v) with the pH adjusted to 2.68 with H 3 PO 4 , running at 1 mL min −1 . The element was monitored by fluorescence; the excitation and emission wavelengths were set at 331 and 500 nm, respectively. The standard Citrinin was used from Sigma.

QPCR Analysis
First-strand cDNA was synthesized while using the PrimeScript 1st Strand cDNA Synthesis Kit (TaKaRa, Japan), with the Oligo dT Primer 15. Gene expression was monitored by RT-qPCR and carried out using the SYBR Premix Ex Taq II (TaKaRa, Japan). RT-qPCR was performed while using the Stratagen Mx3000 P (Agilent, Santa Clara city, California, USA) with the following cycling program: hold at 95 • C for 30 s, followed by a three-step PCR (42 cycles of denaturation at 95 • C for 5 s, annealing at 60 • C for 30 s, and extension at 72 • C for 30 s) and dissociation curve analysis (at 95 • C for 15 s, annealing at 60 • C for 30 s, then collecting the dissociation curve from 60 • C to 95 • C, finally at 95 • C for 15 s). Table 2 shows the sequence of the primers used.

Conflicts of Interest:
The authors declare no conflict of interest.