Identification of the high-yield monacolin K strain from Monascus spp. and its submerged fermentation using different medicinal plants

Medical plants confer various benefits to human health and their bioconversion through microbial fermentation can increase efficacy, reduce toxicity, conserve resources and produce new chemical components. In this study, the cholesterol-lowering monacolin K genes and content produced by Monascus species were identified. The high-yield monacolin K strain further fermented with various medicinal plants. The antioxidant and anti-inflammatory activities, red pigment and monacolin K content, total phenolic content, and metabolites in the fermented products were analyzed. Monacolin K was detected in Monascus pilosus (BCRC 38072), and Monascus ruber (BCRC 31533, 31523, 31534, 31535, and 33323). It responded to the highly homologous mokA and mokE genes encoding polyketide synthase and dehydrogenase. The high-yield monacolin K strain, M. ruber BCRC 31535, was used for fermentation with various medicinal plants. A positive relationship between the antioxidant capacity and total phenol content of the fermented products was observed after 60 days of fermentation, and both declined after 120 days of fermentation. By contrast, red pigment and monacolin K accumulated over time during fermentation, and the highest monacolin K content was observed in the fermentation of Glycyrrhiza uralensis, as confirmed by RT-qPCR. Moreover, Monascus-fermented medicinal plants including Paeonia lactiflora, Alpinia oxyphylla, G. uralensis, and rice were not cytotoxic. Only the product of Monascus-fermented G. uralensis significantly exhibited the anti-inflammatory capacity in a dose-dependent manner in lipopolysaccharide-induced Raw264.7 cells. The metabolites of G. uralensis with and without fermentation (60 days) were compared by LC/MS. 2,3-Dihydroxybenzoic acid, 3,4-dihydroxyphenylglycol, and 3-amino-4-hydroxybenzoate were considered to enhance the antioxidant and anti-inflammatory ability. Given that highly homologous monacolin K and citrinin genes can be observed in Monascus spp., monacolin K produced by Monascus species without citrinin genes can be detected through the complementary methods of PCR and HPLC. In addition, the optimal fermentation time was important to the acquisition of antioxidants, red pigment and monacolin K. These bioactive substances were significantly affected by medicinal plants over fermentation time. Consequently, Monascus-fermented G. uralensis had a broad spectrum of biological activities.

Information, 19 typical Monascus species have been identified. Monascus purpureus, Monasucs pilosus, and Monascus ruber are frequently used in research. Many pigments, such as rubropunctatin, monascorubrin, rubropunctamine, monascorubramine, monascin, and ankaflavin, are detected in Monascus species . They are important food colorant additives. Moreover, the pharmacological efficacy of Monascus species has attracted considerable interest for the improvement of health in lipid metabolism. Monacolin K, also known as lovastatin, has a cholesterol-lowering effect and can be found in Monascus species (Yanli and Xiang 2020). In addition to lowering lipid level, monacolin K prevents many diseases, such as colon, gastric, breast, lung, and thyroid cancer; acute myeloid leukemia; Parkinson's disease; schizophrenia; depression; and type I neurofibromatosis (Chen et al. 2015;Hong et al. 2008;Lin et al. 2015;Xiong et al. 2019;Zhang et al. 2019b). Therefore, monacolin K production through the addition of linoleic acid, non-ionic surfactant, and glutamic acid has been extensively explored Yang et al. 2021;Zhang et al. 2019a). A monacolin K biosynthetic gene cluster containing nine genes (mokA-mokI) is homologous to the lovastatin gene cluster from Aspergillus terreus (Chen et al. 2008b). Monacolin K production by Monascus pilosus can be enhanced by mokH overexpression and up-regulation of monacolin K biosynthetic genes (Chen et al. 2010b).
Probiotic-fermented food offers health benefits through the bioconversion of dairy products (Lee et al. 2021). Rice or different grains are generally used as substrates for Monascus fermentation to improve secondary metabolites. Extracts from Monascus-fermented soybean have antioxidant capacities and inhibitory activities against enzymes related to skin aging (Jin and Pyo 2017). The antioxidant activities of scavenging free radicals have been observed in Monascus-fermented coix seed (Zeng et al. 2021). Fish bone as an antioxidant-active peptide source can be fermented by M. purpureus, which increases 2,2-diphenyl-1-picrylhydrazyl (DPPH) and 2, 2′-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical scavenging ability . Additionally, Monascus-submerged fermentation using agro-industrial residues, such as rice flour and molasses, is a promising method for red pigment production (Da Silva et al. 2021). Different cereal substrates through the solid-state fermentation of M. purpureus have been developed for pigment production (Srianta et al. 2016). Besides, millet is a good candidate for monacolin K production using Monascus species (Maric et al. 2019;Zhang et al. 2018a).
Medicinal plants offer various benefits to human health. Medicinal plant bioconversion through fermentation can enhance medicinal effects by increasing efficacy, reducing toxicity, conserving resources, and producing new chemical components (Li et al. 2020a). Microbes used in the fermentation of medicinal plants include Bacillus, lactic acid bacteria, yeast, and filamentous fungi. By contrast, natural fermentation without additional microbes is a common method but less effective and specific than other known methods. In this study, a high-yield monacolin K strain from 18 Monascus spp. was screened and identified. Subsequently, medicinal plants for treating diarrhea, dementia, pain, cold, inflammation, and immune disorders were used as substrates for Monascus fermentation. The fermentation products were used in analyzing antioxidant and anti-inflammatory activities, red pigment, monacolin K, and total phenolic content. Metabolites from the fermentation product with the best efficacy were further analyzed through LC/MS.

Monascus spp., medicinal plants, and growth conditions
Monascus spp. listed in Table 1 were used in this study. To detect a high-yield monacolin K strain, the medium with 7% glycerol, 3% glucose, 3% monosodium glutamate, 1.2% polypetone, 0.2% NaNO 3 , and 0.1% MgSO 4 ·7H 2 O was performed for Monascus spp. incubation. Mycelia were harvested after 14 days of cultivation at 25 ℃ for DNA manipulation. Medicinal plants including Angelica pubescens, Pogostemon cablin, Paeonia lactiflora, Alpinia oxyphylla, Melaleuca leucadendron, Lavandula angustifolia, Osmanthus fragrans, Glycyrrhiza uralensis, Phellodendron chinense, and rice as the control were utilized for the fermentation materials (Beijing Tongrentang Traditional Chinese Medicine Co., Ltd, Chengdu, China). Two grams of medicinal plants as the sole substrate ground into a powder with 50 mL reverse osmosis water were sterilized at 121 ℃ for 30 min. Spores of Monascus spp. harvested from potato dextrose agar plate were added into the medicinal plant liquids for submerged batchfermentation. Monascus spp. with medicinal plants were stood for 60 and 120 days at 25 ℃. After 60 and 120 days of submerged fermentation, the culture was centrifugated and filtered, and the suspension was freeze-dried. The freeze-dried powders were carried out for the bioactive assays.

DNA manipulation
Approximately 0.5 g Monascus mycelia were ground using a mortar and pestle by liquid nitrogen. DNA was extracted by phenol and chloroform, and precipitated by isopropanol. DNA was finally dissolved in TE buffer. PCR was implemented according to the condition by Chen et al. (2008a). The primer sets of mokA and mokE genes involved in the monacolin K biosynthesis were mokA-F, ATC ATT CTT TCC NCGC TCC A, mokA-R, CGG GCT ATT GTC GGC CAT AG; mokE-F, GTG GTG GAC TCG ACG TTG GT, and mokE-R, TTC TCG CAG TAC ACG GTC AC. PCR was performed by an ABI 2700 PCR (Applied Biosystems, Thermo Fisher Scientific, Waltham, MA) and the reaction condition was as follows, 96 ℃ for 5 min by 1 cycle, 96 ℃ for 1 min, 50 ℃ for 1 min, and 72 ℃ for 1 or 2 min by 30 cycles, and 72 ℃ for 10 min with a final extension. The PCR products were recovered from agarose for DNA sequencing.

Monacolin K detection by HPLC
The culture of Monascus spp. fermentation was filtered by a 0.2 mm filter. Monacolin K was determined by a high-performance liquid chromatography (HPLC) (Shimadzu, Kyoto, Japan) fitted with a reverse-phase C18 column (InsertSustain, 5 μm, 4.6 × 150 mm)(GL science Inc., Tokyo, Japan). The HPLC reaction was followed by 0.1% phosphorus acid in water with 35% and methanol with 65% at a flow rate of 1 mL/min. Monacolin K was scanned by a UV spectroscopy from 210 to 400 nm.

Antioxidant analysis of fermentation product
The radical scavenging assay of DPPH and ABTS was used to estimate the antioxidant capacity of the fermentation product. The fermentation product of Monascus spp., 3,4-dihydroxyphenylglycol, and 3-amino-4-hydroxybenzoate were respectively mixed with 0.2 mM DPPH solution for 30 min and the mixture was measured by an enzyme-linked immunosorbent assay (ELISA) reader (Molecular Devices, Sunnyvale, CA) at OD 517 . The ABTS radical scavenging capacity was evaluated by the T-AOC Assay Kit (Beyotime Biotechnology, Shanghai, China). The reaction was detected by the absorbance of OD 405 . The radical scavenging activity of DPPH and ABTS (%) was calculated as follows, the radical scavenging activity (%) = ([OD 517 or 405 of control-OD 517 or 405 of fermentation product]/OD 517 or 405 of control ) × 100. The sterile H 2 O was utilized as the control.

Cell viability of fermentation product
The Raw264.7 macrophages were carried out to implement the 3-(4,5-cimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. The cell in a 96-well plate was incubated at 37 °C for 24 h with 5% CO 2 incubation. The fermentation products, 3,4-dihydroxyphenylglycol, and 3-amino-4-hydroxybenzoate were respectively added into the cells while the sterile H 2 O was utilized as the control. After 24 h incubation, the supernatant medium was removed and the fresh medium containing 10 µL MTT (5 mg/mL) was added into the 96-well plate of the cells for 4 h incubation. Then, the supernatant medium was discarded, and dimethyl sulfoxide was added to the 96-well plate. The mixture was detected by the absorbance of OD 570 .

Anti-inflammatory analysis of fermentation product
The Raw264.7 macrophages were incubated at 37 °C for 24 h and 5% CO 2 incubation in a 96-well plate. The fermentation products, 3,4-dihydroxyphenylglycol, and 3-amino-4-hydroxybenzoate were respectively added into the 96-well plate of the cells for 1 h. To stimulate the inflammation response, 1 µg/mL lipopolysaccharide was added into the cells. After 24 h incubation, the nitrite of the medium was determined by the nitrite detection kit (Beyotime Biotechnology) using the Griess reagent. The mixture was detected by the absorbance of OD 540 . The NO decreasing rate was evaluated as follows, NO decreasing rate (%) = ([OD 540 of control-OD 540 of fermentation products, 3,4-dihydroxyphenylglycol, and 3-amino-4-hydroxybenzoate]/OD 540 absorbance of control) × 100. The sterile H 2 O was utilized as the control.

Red pigment analysis of fermentation product
The freeze-dried products of Monascus spp. fermentation was dissolved by 1 mL 70% ethanol and centrifugated at 15,000×g (Hettich, Mikro 220R, Germany) for 10 min. The supernatant was obtained and detected by the absorbance of OD 505 . The red pigment was calculated as follows, red pigment (U/g) = OD 505 of fermentation product × (1/gram) × (1/volume) × dilution factor.

Monacolin K biosynthetic gene expression by RT-qPCR
Approximately 0.5 g Monascus mycelia were ground and performed by TRIzol reagent (Thermo Fisher Scientific). Total RNA was extracted by chloroform and treated with RNase-free DNase. Finally, total RNA was preserved in DEPC H 2 O. HiFi-MMLV cDNA kit (Beijing ComWin Biotech Co., Ltd., Beijing, China) with oligo (dT) and random hexamers was used to carry out the first-strand cDNA. The qPCR in a final volume of 25 µL was implemented using UltraSYBR Mixture kit (Beijing ComWin Biotech) by the Roche LightCycler ® 480 System (Roche Group, Switzerland). The primer sets of monacolin K biosynthetic genes were utilized as follows, qmokA-

Analysis of total phenol and metabolites from the fermentation product
Total phenol of fermentation product was measured by the Folin-Ciocalteu method. The fermentation product was added into Folin-Ciocalteu reagent for 5 min reaction, and then 10% sodium carbonate solution was used for color development in darkness. The gallic acid of different concentrations as a positive control was utilized for the calculation of the standard curve. The mixture was centrifugated at 15,000×g for 10 min and the supernatant was measured by the absorbance of OD 760 . Thermo Vanquish system equipped with an ACQUITY UPLC ® HSS T3 (150 × 2.1 mm, 1.8 μm, Waters Corporation, Milford, MA) column was accomplished for the analysis of fermentation product. The UPLC reaction was followed by 0.1% formic acid in acetonitrile (A) and 5 mM ammonium formate in acetonitrile (B) at a flow rate of 0.25 mL/min. The gradient of solvent A/B (v/v) was set as follows, 2% A/B from 0 to 1 min; 2-50% A/B from 1 to 9 min; 50-98% A/B from 9 to 12 min; 98% A/B from 12 to 13.5 min; 98-2% A/B from 13.5 to 14 min; 2% A from 14 to 20 min for positive model (2% B from 14 to 17 min for negative model). The ESI-MS was implemented by the Q Exactive Plus mass spectrometer (Thermo Fisher Scientific) with the spray voltage of 3.5 kV and − 2.5 kV in positive and negative modes, respectively. Data-dependent acquisition MS/MS was carried out with the HCD scan and the normalized collision energy was 30 eV. Unnecessary information in MS/MS spectra was removed by dynamic exclusion. The metabolites were screened by the accurate molecular weight (molecular weight error < 30 ppm) and identified by Metlin (

Statistical analysis
Duncan's multiple range test, Pearson correlation, and repeated-measures analysis of variances (ANOVAs) were performed by IBM SPSS Statistics v20 software package (SPSS Inc. Chicago, USA) at a confidence level of 95%. The mean and standard deviation of three replicates were shown.

Identification of monacolin K biosynthetic genes and content from Monascus spp.
mokA and mokE genes encoding polyketide synthase and dehydrogenase involved in monacolin K biosynthesis from 18 Monascus spp. and one Aspergillus terreus were amplified through PCR. The expected bands of approximately 1 and 0.5 kb were detected (Fig. 1) (Fig. 2). However, rare antioxidation was detected in rice. On the other hand, the effect of these products at 0.5 mg/ mL on the antioxidant activity of ABTS free radical were not compared because all of them had ABTS scavenging rates of over 90%. Nevertheless, the ABTS scavenging capacity of the medicinal plants had a similar antioxidant profile to DPPH result at 0.0625 mg/mL. After the 60-day fermentation of M. ruber BCRC 31535, the DPPH scavenging ability of the fermentation product was elevated, except in M. leucadendron, and L. angustifolia. In addition, the order of the DPPH scavenging capacity by fold change was rice > G. uralensis > P. lactiflora, whereas that of ABTS was A. pubescens > P. lactiflora > A. oxyphylla in the 60-day fermentation. However, the ABTS and DPPH scavenging capacities significantly declined after 120 days submerged fermentation.

Anti-inflammatory analysis of M. ruber BCRC 31535-fermented product
To analyze the anti-inflammatory capacity of M. ruber BCRC 31535-fermented product, the cell viability of the fermentation product was explored (Fig. 3). Most medicinal plants without fermentation were harmless to Raw264.7 cells, and cell viability below 80% was observed only in P. chinense. However, the cell viability of the fermentation products significantly decreased, and M. ruber BCRC 31535-fermented P. lactiflora, A. oxyphylla, G. uralensis, and rice had survival rates of over 90% after 60 days of fermentation. Thus, their anti-inflammatory effects were further investigated using their fermentation products (Fig. 4). A. oxyphylla without fermentation showed the best anti-inflammatory effect. After the fermentation of M. ruber BCRC 31535, the fermented G. uralensis significantly improved anti-inflammatory capacity. The trend of anti-inflammatory effect of the fermentation products was similar to that of their antioxidant ability. After 120 days of submerged fermentation, anti-inflammation obviously decreased.

Red pigment and monacolin K analysis of M. ruber BCRC 31535-fermented product
Red pigment and monacolin K derived through polyketide synthesis can be produced by M. ruber BCRC 31,535 and these of fermentation product were further determined. The red pigment of the fermentation product increased with fermentation time, except in A. pubescens  (Fig. 5). The red pigment of the 120-day fermentation product from rice was 8.9-fold that of the 60-day fermentation product.
Monacolin K was obviously detected in the fermentation of A. pubescens, P. cablin, P. lactiflora, M. leucadendron, G. uralensis, P. chinense, and rice (Fig. 6). It was not detected during the fermentation of L. angustifolia and O. fragrans. Monacolin K content in these fermentation products increased, similar to that of the red pigment during fermentation, except in P. chinense. The content of monacolin K derived through the 60 and 120 days fermentation of G. uralensis was the highest, which was 41.9-fold and 9.1-fold that in rice, respectively. Thus, the gene expression of monacolin K containing mokA-mokI genes was further analyzed by comparing M. ruber BCRC 31535-fermented G. uralensis and rice. The results indicated that the gene expression levels of monacolin K obtained through the 60-day fermentation of G. uralensis were higher than those in rice. However, monacolin K gene expression level after 120 days G. uralensis fermentation was obviously lower than that in rice. This result was consistent with monacolin K content. The content after 120 days G. uralensis fermentation was only 1.2 times higher than that after 60 days of fermentation. Monacolin K content in rice after 120 days of fermentation was 5.7-fold that after 60 days of fermentation.

Total phenolic content and metabolites analysis of M. ruber BCRC 31535-fermented product
The antioxidant effect of the 60-day fermentation product significantly increased relative to the antioxidant effects of medicinal plants without fermentation. Accordingly, the contribution of total phenolic content to antioxidant capacity was evaluated. The result showed that the total phenolic content was elevated after M. ruber BCRC 31535 fermentation, except in P. cablin, and M. leucadendron (Fig. 7). The highest increase in total phenolic content was found in rice after fermentation, followed by that in G. uralensis. This result was consistent with the DPPH scavenging capacity.
The anti-inflammatory capacities of 3,4-dihydroxyphenylglycol and 3-amino-4-hydroxybenzoate were explored on the basis of cell viability (Fig. 9). The survival rates of Raw264.7 cells treated with 3,4-dihydroxyphenylglycol and 3-amino-4-hydroxybenzoate below 0.1 mg/mL were over 90%, and no cytotoxicity was observed. Therefore, the anti-inflammatory capacities of the two compounds were further measured. The anti-inflammatory capacity of 3,4-dihydroxyphenylglycol was observed in a dose-dependent manner. Over 56% decrease in NO was detected at 0.05 mg/mL. However, the anti-inflammatory effects of 3-amino-4-hydroxybenzoate were not significant in a dose-dependent manner.

Discussion
Monascus species offer various health benefits to human beings. However, hepato-nephrotoxic citrinin of polyketide metabolites is present in Monascus species (de Oliveira Filho et al. 2017). Hence, many studies have focused on improving the ratio of monacolin K to citrinin  through optimal cultivation or fermentation. Given that citrinin and monacolin K biosynthetic gene clusters are present in Monascus species (Chen et al. 2008b;Shimizu et al. 2005), considerable effort has been devoted to identifying these genes in different Monascus strains. According to our previous study, pksCT, ctnA, and orf3 genes, which encode citrinin polyketide synthase, a major activator, and oxygenase, were distributed in M. purpureus and Monascus kaoliang, but not in M. pilousus, M. ruber, M. barkeri, Monascus floridanus, Monascus lunisporas, and Monascus pallens (Chen et al. 2008a). Interestingly, in this study, mokA and mokE genes were detected in M. pilousus, M. ruber, and M. barkeri only, and no monacolin K content was detected in M. barkeri. Further investigation into the citrinin and monacoin K genes of Monascus has shown that the distribution of citrinin and monacolin K is not restricted in the specific Monascus species. Some Monascus species simultaneously have two gene clusters, but some have only one (Chen et al. 2008a;Li et al. 2020b;Yang et al. 2015). Information about the evolutionary process of the polyketide gene cluster in fungi is currently limited, and thus gene duplication or horizontal gene transfer is worthy of further study (Wisecaver et al. 2014  monacolin K without citrinin can be safely applied to the food industry. Raw materials, such as rice and grains, provide carbohydrates, protein, and inorganic elements for traditional fermentation. Thus, considerable attention has been devoted to the fermentation of Monascus species with rice and grains (Jin and Pyo 2017;Maric et al. 2019;Srianta et al. 2016;Zhang et al. 2018a). Given that medicinal plants themselves have certain efficacy, fermentation through the biotransformation of microorganisms can improve the original properties of the plants and result in the production of novel chemical components (Li et al. 2020a). Scant attention has been devoted to the relationship between the fermentation of medicinal plants and Monascus species. Hence, several medicinal plants and rice were used in the fermentation of M. ruber BCRC 31535 for 60 and 120 days. Most medicinal plants and their extracts have antioxidant capacities (Chen et al. 2017;Kim et al. 2017Kim et al. , 2019Soheili and Salami 2019;Surh and Yun 2012;Wu et al. 2021Wu et al. , 2022Yuan et al. 2020;Zhang et al. 2018b). In this study, after the 60-day fermentation of M. ruber BCRC 31535 with medicinal plants, the DPPH scavenging rate significantly improved, except in M. leucadendron and L. angustifolia as raw materials. M. leucadendron and L. angustifolia contributed to antifungal activity and probably affected M. ruber BCRC 31535 fermentation (Abd Rashed et al. 2021;Valdes et al. 2008). Additionally, the long-term fermentation of medicinal plants (120 days) decreased antioxidant activity in the DPPH and ABTS assays. This effect may have promoted the formation of acids from antioxidant compounds, such as phenols, through longterm fermentation (Lee et al. 2019). Phenols are responsible for antioxidant activity, and the relationship between total phenolic content and antioxidant capacity was demonstrated (Additional file 4: Table S1, Additional file 5: Table S2, Additional file 6: Table S3). According to the Pearson correlation between total phenolic content and antioxidant activity, the significant effects of total phenolic content on DPPH and ABTS scavenging abilities were observed. Thus, the optimal fermentation time was important to the acquisition of antioxidants and prevention of further degradation by microbes.
Before the anti-inflammatory capacities of fermentations product generated by lipopolysaccharide-induced Raw264.7 cells were determined, cell viability was analyzed with the MTT method. Most medicinal plants were not cytotoxic. The effects of medicinal plants and fermentation products on cell viability across temporal change was evaluated by repeated-measures ANOVAs (Additional file 7: Table S4). The result indicated that cell viability was significantly affected by medicinal plants over time. This result was contrary to our expectation that toxicity would be reduced after fermentation (Li et al. 2020a). However, a survival rate of over 90% was still observed in the fermentation of P. lactiflora, A. oxyphylla, G. uralensis, and rice, which originally had potential anti-inflammatory activities (Okonogi et al. 2018;Xin et al. 2019;Yin et al. 2018;Zhang et al. 2018b). Thus, these medicinal plants and fermentation products were further used in the analysis of anti-inflammatory capacity. Only 60-day G. uralensis fermentation revealed a significant effect on NO decreasing rate. This benefit was consistent with the fermentation of ginger and tea by M. pilosus and M. purpureus, respectively (Chen et al. 2010a;Deng et al. 2021).
Red pigment and monacolin K derived from polyketides were the major characteristics in Monascus species. The amounts of red pigment and monacolin K increased with fermentation time in most fermentation products, and their maximum amounts were obtained after the fermentation of rice and G. uralensis, respectively. The repeated-measures ANOVAs demonstrated that the red pigment and monacolin K were significantly affected by medicinal plants over fermentation time (Additional file 8: Table S5). The trend was different from the trends of antioxidant and anti-inflammatory capacities. The stable production of red pigment and monacolin K was similar to the different studies using agro-industrial residues, cereal, and millet as fermented substrates (Da Silva et al. 2021;Maric et al. 2019;Srianta et al. 2016;Zhang et al. 2018a). In addition, the transcription level of monacolin K biosynthetic genes between the fermentation of G. uralensis and rice was further verified with RT-qPCR. As expected, the expression levels of monacolin K biosynthetic genes in the 60-day G. uralensis fermentation were higher than those in rice. Furthermore, no up-regulation was observed in the 120-day G. uralensis fermentation, and rate of increase in monacolin K content was lower than that in the 120-day rice fermentation. Moreover, MokG encoding HMG-CoA reductase may confer resistance on monacolin K (Abe et al. 2002;Chen et al. 2008b); therefore, its relative gene expression was lower than the expression levels of other genes.
LC/MS was used in analyzing difference between metabolites with or without fermentation of M. ruber BCRC 31533 in G. uralensis. The relative abundance of glutamic acid in G. uralensis without fermentation was 3.1-fold higher than that in 60 days fermentation. This result suggested that the consumption of glutamic acid promoted monacolin K production (Zhang et al. 2019a). In addition, G. uralensis, commonly known as licorice, is usually used in the food industry as a sweetener and contains various bioactive constituents, such as liquiritin, isoliquiritin, liquiritigenin, and isoliquiritigenin (Kao et al. 2014). Liquiritigenin and isoliquiritigenin are the aglycone forms of liquiritin and isoliquiritin, respectively. The relative abundance rates of liquiritin and liquiritigenin were high in G. uralensis and were 632.9and 3.8-fold those in 60-day fermentation, respectively. This result suggested that the bioactive substances were metabolized and consumed by M. ruber BCRC 31,535. This result was different from that of Kim et al., who reported that liquiritigenin and isoliquiritigenin content improved after the fermentation of Monascus albidulus (Kim et al. 2020). Furthermore, 2,3-dihydroxybenzoic acid had the highest content, which increased 143.9fold after 60-day G. uralensis fermentation. According to the previous study, 2,3-dihydroxybenzoic acid displayed good antioxidant and anti-inflammatory capacities (Adjimani and Asare 2015;Alvarez Cilleros et al. 2020). Moreover, 3,4-dihydroxyphenylglycol and 3-amino-4-hydroxybenzoate increased 291.4-and 89.1-fold through the biotransformation of M. ruber BCRC 31,535, respectively. The two compounds had high antioxidant capacities according to the DPPH and ABTS assays. The NO rate of 3,4-dihydroxyphenylglycol decreased in a dosedependent manner. These results implied that 2,3-dihydroxybenzoic acid, 3,4-dihydroxyphenylglycol, and 3-amino-4-hydroxybenzoate improved antioxidative and anti-inflammatory effects in the 60-day fermentation of G. uralensis.

Conclusions
Traditionally, the effective substance, new compound, resource conservation, and toxic reduction of medicinal plants can be enhanced by means of microbial fermentation. In this study, the antioxidant capacity and total phenolic content of medicinal plants increased after 60 days of fermentation by M. ruber BCRC31535. However, cytotoxicity was also elevated but not influenced in P. lactiflora, A. oxyphylla, G. uralensis, and rice. The positive correlation between total phenolic content and DPPH and ABTS scavenging activities was verified by repeated-measures ANOVAs. Long-term fermentation (120 days) decreased antioxidant efficacy and cytotoxicity. Red pigment and monacolin K derived from M. ruber BCRC31535 biosynthesis accumulated over time during fermentation. Among these medicinal plants, M. ruber BCRC31535-fermented G. uralensis had a broad spectrum of efficacy with antioxidative and anti-inflammatory capacities, and monacolin K contributed by the increase in 2,3-dihydroxybenzoic acid, 3,4-dihydroxyphenylglycol and 3-amino-4-hydroxybenzoate levels and glutamic acid consumption rates. To the best of our knowledge, this study is the first to report the antioxidative capacities of 3-amino-4-hydroxybenzoate for DPPH and ATBS scavenging activities.