Inactivation of mrpigH Gene in Monascus ruber M7 Results in Increased Monascus Pigments and Decreased Citrinin with mrpyrG Selection Marker

Monascus pigments (MPs) have been used as food colorants for several centuries in Asian countries and are currently used around the world via Asian catering. The MPs biosynthetic pathway has been well-illustrated; however, the functions of a few genes including mrpigH in the MPs gene cluster of M. ruber M7 are still unclear. In the current study, mrpigH was disrupted in Δmrlig4ΔmrpyrG, a highly efficient gene modification system, using mrpyrG as a selection marker, and ΔmrpigHΔmrlig4ΔmrpyrG::mrpyrG and ΔmrpigHΔmrlig4ΔmrpyrG have been obtained. Subsequently, their morphologies, biomasses, MPs and citrinin (CIT) production were analyzed, respectively. These results have revealed that the deletion of mrpigH has significant effects on the morphology and growth of M. ruber M7. Moreover, compared with M. ruber M7, the yields of MPs and CIT were drastically increased and decreased in mrpigH mutants, respectively.


Introduction
Monascus species are famous medicinal and edible filamentous fungi used in traditional fermentation in Asian countries, such as China, Japan, and the Korean Peninsula for nearly 2000 years [1,2]. At present, their fermented products, such as Hongqu, also called red fermented rice, red yeast rice and red mold rice, are widely used as food additives and nutraceutical supplements worldwide owing to their production of abundant beneficial secondary metabolites (SMs), such as Monascus pigments (MPs), monacolin K (MK) and γ-amino butyric acid (GABA) [1,3]. However, citrinin (CIT), a nephrotoxic mycotoxin produced by some strains of Monascus spp., restricts the application of Monascus fermented products [4].
M. ruber M7, which can produce MPs and CIT, without MK, was subjected to wholegenome sequencing analysis [5]. And the functions of most genes in the MPs gene cluster of M. ruber M7 have been investigated by gene manipulation [6]. However, there are a few genes in the MPs gene cluster of M. ruber M7, such as mrpigH and mrpigI, which have not been investigated [6,7]. In 2017, Balakrishnan et al. predicted that the mppE in M. purpureus KACC (highly homologous to mrpigH in M. ruber M7) encoded a reductase which can decrease orange pigments (OPs) and red pigments (RPs) in the biosynthesis of MPs [8]. In 2019, Chen et al. also guessed that MrPigH might contribute to reducing the carbon double bond of the precursor compounds to the typical yellow pigments (YPs) monascin and ankaflavin [7].
In this study, we firstly cloned mrpigH from M. ruber M7. Subsequently, mrpigH was disrupted in the highly efficient system ∆mrlig4∆mrpyrG [9] using the mrpyrG selection marker, and two mrpigH deletion strains ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG and ∆mrpigH∆mrlig4∆mrpyrG have been constructed. Finally, the morphologies, biomasses, MPs and CIT production of these mrpigH mutants were assessed. The results revealed that the deletion of mrpigH led to a dramatic reduction in biomass accumulation. Crucially, the inactivation of MrPigH resulted in an increase of MPs and a decrease of CIT.

Construction of Deletion Cassettes and Plasmids
The genomic DNA of M. ruber M7 used for PCR was isolated as described previously [12]. The mutant strains ΔmrpigHΔmrlig4ΔmrpyrG::mrpyrG and ΔmrpigHΔmrlig4ΔmrpyrG were constructed using site-directed homologous recombination. The mrpigH gene markerless deletion cassette (5′UTR-mrpyrG-5′-1UTR-3′UTR) was constructed by seamless cloning, and shown schematically in Figure 1a. The relative primer pairs were shown in Table 2.  The 5 and 3 flanking regions (993 bp and 775 bp, respectively) and the 5 -1 flanking region (531 bp) of mrpigH were amplified with the primers pigHpyrG 5F-pigHpyrG 5R, pigHpyrG 3F-pigHpyrG 3R and pigHpyrG 5F-1-pigHpyrG 5R-1, respectively. The 1.28-kb mrpyrG marker cassette was amplified from M. ruber M7 genomic DNA with the primer pair pigHpyrG pyrGeF-pigHpyrG pyrGeR. Then the four amplicons (5 and 3 regions, 5 -1 regions and mrpyrG expression fragment) were mixed at a 1:1:1:1 molar ratio and cloned into vector pBLUE-T using the seamless cloning and assembly kit (Beijing Zoman Biotechnology). Subsequently, both the cloned DNA fragment and the pCAMBIA3300 plasmid were digested with HindIII and XbaI, and ligated by T4 DNA ligase to generate plasmid pCPGPIGH for the mrpigH knock-out harbouring mrpyrG selection marker.

Deletion of mrpigH in ∆mrlig4∆mrpyrG Strain
The plasmid pCPGPIGH was transformed into Agrobacterium tumefaciens EHA105 using a freeze-thaw method [13]. ∆mrlig4∆mrpyrG, a markerless disruption strain, was used as a host strain to delete mrpigH with the mrpyrG recyclable marker [9]. The A. tumefaciens clones containing pCPGPIGH were incubated for transformation with ∆mrlig4∆mrpyrG to generate a mrpigH gene deletion mutant (∆mrpigH∆mrlig4∆mrpyrG::mrpyrG) by minimal medium without uridine/uracil. The conidia of ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG were collected and spread onto PDA with 0.75 mg/mL 5-FOA and 10 mM uridine. After incubated at 28 • C for 6 days, the surviving colonies were transferred to a new PDA under the same conditions for 4 days. The final surviving colonies were selected and verified by PCR. Selected transformants were designated as ∆mrpigH∆mrlig4∆mrpyrG.

MPs and CIT Analyses
M. ruber M7 can produce MPs and CIT, but no MK. Previous researches have shown that MPs mainly accumulate in the mycelia, while CIT exists in the media [14]. Therefore, the intracellular MPs and extracellular CIT were detected. 1 mL spores suspension (10 5 cfu/mL) of each strain were inoculated on PDA plate coated with cellophane membranes and incubated at 28 • C for 11 days. 20 mg freeze-dried mycelia or media powder were suspended in 1 mL 80% (v/v) methanol solution, and subjected to 30 min ultrasonication treatment (KQ-250B, Kunshan, China). Then, the extraction solutions were separated by centrifugation at 10,000× g for 15 min and filtered with a 0.22 µm filter membrane for further analysis.
The pigments groups concentration was measured using a UV−vis UV-1700 spectrophotometer (Shimadzu, Tokyo, Japan) at 380, 470 and 520 nm which are the maximal absorption of yellow, orange, and red pigments, respectively. The results were expressed as optical density (OD) units per gram of dried mycelia multiplied by a dilution factor [15].

Detection of the Relative Gene Expression Level in MPs and CIT Gene Clusters by RT-qPCR
To analyze the influence of mrpigH deletion on gene expression in MPs and CIT gene cluster, ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG and the wild-type strain (M. ruber M7) were selected for quantitative real-time PCR (RT-qPCR) detection. The ∆mrpigH∆mrlig4∆mrpyrG was lacked of mrpyrG and had to supply uridine, which might have had an effect on the yields of MPs and CIT.
One milliliter freshly harvested spores (10 5 cfu/mL) of each strain were inoculated on PDA plate and incubated at 28 • C and samples were taken every other day from the third day to the ninth day. RT-qPCR was performed according to the method described by Liu et al. [13]. Beta-actin was used as a reference gene. The primers used in these analyses were listed in Table S1.

Sequence Analysis of mrpigH in M. ruber M7
A 1.24-kb fragment containing the putative mrpigH homolog was successfully amplified from the genomic DNA of M.ruber M7. Sequence prediction of mrpigH by Softberry's FGENESH program has revealed that the putative mrpigH gene consists of a 1110 bp open reading frame (ORF) which consists of one exon and encodes 369-amino acids. A database search with Pfam 33.1 program has shown that MrPigH pertains to the alcohol dehydrogenase GroES-like domain. Besides, a database searched with NCBI-BLAST has been demonstrated that the deduced 369-amino acid sequences encoded by mrpigH share 65.31% similarity with the enoyl reductase (GenBank: PCH03974.1), 57.84% similarity with oxidoreductase of Glonium stellatum (GenBank: OCL03635.1), and 56.64% similarity with dehydrogenase of Hyphodiscus hymeniophilus (GenBank: KAG0646231.1). The specific function of mrpigH is still unclear.

Morphologies and Biomasses of mrpigH Mutants and M. ruber M7
To investigate whether the mrpigH was markerlessly deleted by mrpyrG in ∆mrlig4∆mrpyrG, M. ruber M7 and its mrpigH markerless mutants were cultivated on PDA supplemented the appropriate additive (10 mM uridine for the uridine/uracil auxotrophy).The results (Figure 2a) revealed that ∆mrpigH∆mrlig4∆mrpyrG showed no growth on PDA, but it was able to grow on PDA with 0.75mg/mL 5-FOA and 10 mM uridine, while the growths of M. ruber M7 and ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG were inhibited by the addition of 0.75 mg/mL 5-FOA to PDA, but could grow on PDA. Those results indicated that the mrpigH markerless mutants were successfully constructed.  To test the influence of deleting mrpigH on developmental processes, M. ruber M7 and ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG were cultivated on different media (PDA, CYA, MA and G25N) to observe their colonial and microscopic characteristics. The results showed that the colonial morphologies of mrpigH mutants (Figure 2b) were obviously different from those of M. ruber M7 on different culture plates, especially on PDA plate, the ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG showed slower growth rate and darker color. However, the microscopic morphologies, including conidia and cleistothecia, of ∆mrpigH∆mrlig4 ∆mrpyrG::mrpyrG was not dramatically different from those of M. ruber M7 on different culture plates (Figure 2c). Moreover, the biomasses of two mrpigH mutants apparently decreased compared with that of M7 on PDA in 5-11 d (Figure 2d).

Analysis of MPs and CIT Production
In order to evaluate the effect of detecting mrpigH on MPs and CITs during fermentation, the samples cultured for 3, 5, 7, 9 and 11 days on PDA were obtained and each test was repeated independently three times. OD values representing yellow, orange and red pigment production were determined using a spectrophotometer at 380 nm, 470 nm and 520 nm, respectively. As shown in Figure 3a-c, from the seventh day to the 11th day, M. ruber M7 produced much fewer MPs (including YPs, OPs and RPs) than 2 mrpigH mutants (∆mrpigH∆mrlig4∆mrpyrG::mrpyrG and ∆mrpigH∆mrlig4∆mrpyrG), whereas the ability of producing MPs among 2 mrpigH mutants showed no obvious difference. After 11 days of cultivation, the YPs, OPs and RPs production in ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG and ∆mrpigH∆mrlig4∆mrpyrG were 2.04-2.22, 8.76-10.06, 4.29-4.69 times those of M. ruber M7, respectively.
As to CIT, the UPLC has been performed to detect the production during fermentation. As shown in Figure 3d, CIT produced by M. ruber M7 and all mrpigH mutants showed an obvious difference. At the end of the 11 days of fermentation, CIT production in ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG and ∆mrpigH∆mrlig4∆mrpyrG decreased dramatically, and was two to three orders of magnitude less than that of M.ruber M7. Among 2 mrpigH mutants, CIT production in ∆mrpigH∆mrlig4∆mrpyrG was higher than that of ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG. The possible cause was that the ∆mrpigH∆mrlig4∆mrpyrG lacked mrpyrG and had to supply uridine, which might have an effect on the yields of CIT. To test the influence of deleting mrpigH on developmental processes, M. ruber M7 and ΔmrpigHΔmrlig4ΔmrpyrG::mrpyrG were cultivated on different media (PDA, CYA, MA and G25N) to observe their colonial and microscopic characteristics. The results showed that the colonial morphologies of mrpigH mutants (Figure 2b) were obviously different from those of M. ruber M7 on different culture plates, especially on PDA plate, the ΔmrpigHΔmrlig4ΔmrpyrG::mrpyrG showed slower growth rate and darker color. However, the microscopic morphologies, including conidia and cleistothecia, of ΔmrpigHΔmrlig4ΔmrpyrG::mrpyrG was not dramatically different from those of M. ruber M7 on different culture plates (Figure 2c). Moreover, the biomasses of two mrpigH mutants apparently decreased compared with that of M7 on PDA in 5-11 d (Figure 2d).

Analysis of MPs and CIT Production
In order to evaluate the effect of detecting mrpigH on MPs and CITs during fermentation, the samples cultured for 3, 5, 7, 9 and 11 days on PDA were obtained and each test was repeated independently three times. OD values representing yellow, orange and red pigment production were determined using a spectrophotometer at 380 nm, 470 nm and 520 nm, respectively. As shown in Figure 3a-c, from the seventh day to the 11th day, M. ruber M7 produced much fewer MPs (including YPs, OPs and RPs) than 2 mrpigH mutants (ΔmrpigHΔmrlig4ΔmrpyrG::mrpyrG and ΔmrpigHΔmrlig4ΔmrpyrG), whereas the ability of producing MPs among 2 mrpigH mutants showed no obvious difference. After 11 days of cultivation, the YPs, OPs and RPs production in ΔmrpigHΔmrlig4ΔmrpyrG::mrpyrG and As to CIT, the UPLC has been performed to detect the production during fermentation. As shown in Figure 3d, CIT produced by M. ruber M7 and all mrpigH mutants showed an obvious difference. At the end of the 11 days of fermentation, CIT production in ΔmrpigHΔmrlig4ΔmrpyrG::mrpyrG and ΔmrpigHΔmrlig4ΔmrpyrG decreased dramatically, and was two to three orders of magnitude less than that of M.ruber M7. Among 2 mrpigH mutants, CIT production in ΔmrpigHΔmrlig4ΔmrpyrG was higher than that of ΔmrpigHΔmrlig4ΔmrpyrG::mrpyrG.
The possible cause was that the ΔmrpigHΔmrlig4ΔmrpyrG lacked mrpyrG and had to supply uridine, which might have an effect on the yields of CIT.

The Genes' Expression in MPs and CIT Gene Clusters from mrpigH Mutants
The gene expression in MPs and CIT gene clusters in ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG and M. ruber M7, were analyzed by RT-qPCR. As shown in Figure 4, the relative expression levels of mrpigA, mrpigB, mrpigC, mrpigD, mrpigE, mrpigF, mrpigG, mrpigJ, mrpigK, mrpigM, mrpigN, mrpigO and mrpigP in ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG were obviously higher than that of M. ruber M7 on the fifth to ninth days. Therefore, the deletion of mrpigH increased the majority of MPs gene expression level, which might correspond with the enhanced MPs production.

Discussion
Monascus spp. have been widely used in food fermentation for nearly 20 centuries in East and Southeast Asian countries due to their ability of producing vivid MPs [11,19], which are a complex mixture of secondary metabolites (SMs) with a tricyclic azaphilone scaffold, produced via the polyketide pathway by a few filamentous fungi such as Monascus spp. and Penicillium spp. [3,20,21]. MPs are biosynthesized by their gene cluster, which contains 16 genes (mrpigA-mrpigP) in M.ruber M7 [6,7].
In the current study, in order to explore the function of mrpigH, we used the ∆mrlig4 ∆mrpyrG as the starting strain to generate two kinds of mrpigH disruptants, namely ∆mrpigH∆mrlig4∆mrpyrG::mrpyrG and ∆mrpigH∆mrlig4∆mrpyrG. The colonial and microbiological phenotypes and biomasses of the mrpigH mutants showed an obviously difference from those of M. ruber M7 (Figure 2b, d). In particular, the biomasses of mrpigH mutants accumulated more slowly than that of M ruber M7. Our laboratory previously constructed MPs gene knockout strains with the resistance selection markers (hph/neo) in M7, and found that the morphologies and biomasses of mrpigC, mrpigE, mrpigF, mrpigM and mrpigO knockouts were comparable to those of M. ruber M7, whereas the growth rates of mrpigA, mrpigJ, and mrpigK knockouts were increased, and the deletion of mrpigN resulted in a reduction in biomass accumulation compared with M. ruber M7 [22].
Moreover, we have discovered that the deletion of mrpigH in M. ruber M7 can enhance YPs, OPs and RPs (Figure 3a-c), which is mostly consistent with the results obtained in M. purpureus [8]. We also found that deletion of mrpigH dramatically decreased the CIT production (Figure 3d), which is a promising scheme for control of CIT. CIT is a kind of mycotoxin produced via the polyketide pathway by filamentous fungi, mainly by Monascus spp. and Penicillium spp. [23][24][25]. To this end, we investigated the changes in the expression levels of MPs and CIT genes in their gene clusters of mrpigH mutants, and found that the relative gene expression levels almost corresponded with the increase of MPs and decrease of CIT (Figures 4 and 5).
The deep study about the reason for increasing the pigment content while reducing the CIT production by knocking out mrpigH need to be investigated further. In early research, both MPs and CIT were considered to be derived from polyketide pathways [26]. Lately, two hypotheses about the biosynthetic pathways of MPs and CIT were put forward. One of them is based on metabolic pathways; the MPs and CIT shared a common pathway to a certain branch [26]. The other is based on the analysis of the whole genome sequence; their biosynthetic gene clusters had been found separately, thereby, they belonged to two different pathways [6,18].
Previous results suggested that the yields of CIT and MPs in Monascus spp. could affect each other. For example, Xie et al. (2013) and Liu et al. (2014) separately identified that the overexpression of mrpigB and mrpigE in MPs gene cluster of M. ruber M7 resulted in a reduction of CIT production [13,27]. Meanwhile, Liang et al. (2017) obtained a mutant of a putative glyoxalase (orf6) in CIT gene cluster of M. purpureus, and found that the deletion of orf6 could improve the MPs and CIT yields at the same time [28]. A recent study on the biosynthesis of MPs found that acetyl-CoA and malonyl-CoA are catalyzed by a sequence of enzymes to produce MPs precursors. Then the pathway bifurcates into two branches. First, the MPs precursors were reduced by a reductase (MrPigH and/or GME3457, which is encoded outside of the MPs gene cluster in M. ruber M7), yielding the typical YPs monascin and ankaflavin. The other branch of the pathway produces the typical OPs rubropunctatin and monascorubrin by a FAD-dependent oxidoreductase (MrPigF in M. ruber M7). Furthermore, rubropunctatin and monascorubrin were converted into RPs rubropunctamine and monascorubramine through an amination reaction [6,7]. Due to the deletion of mrpigH, the relative expression of mrpigF dramatically increased (Figure 4), which was beneficial for the OPs and RPs production, and the increase in relative yields of OPs and RPs was greater than that of YPs. Recently, Li et al. (2020) reported that MPs biosynthetic gene cluster was a composite supercluster, and the naphthoquinone (monasone) gene cluster was embedded in the MPs gene cluster, and speculated that MrPigH was essential in the biosynthesis of naphthoquinone, but it was a supplemental enzyme in the biosynthesis of MPs [29]. As a result, the mrpigH knockout's blocking of the naphthoquinone biosynthesis pathway might result in naphthoquinone reduction, and more substrates and intermediates were utilized to synthesize MPs, including YPs. In terms of CIT production, He Yi et al. [18] revealed a minimal set of conserved genes involved in CIT biosynthesis, which included nine genes (mrl7, mrl6, mrl5, mrl4, mrl3, mrl2, mrl1, mrpks and mrr1). When mrpigH was deleted, the majority of the genes in the cluster (mrl6, mrl5, mrl4, mrl2, mrl1, mrpks and mrr1) were obviously down-regulated ( Figure 5), which could explain why CIT production has decreased in this study.
In conclusion, the mrpigH gene pertains to the MPs biosynthetic gene cluster of M. ruber M7 and plays a remarkable role in the biosynthesis of MPs and CIT. The disruption of mrpigH had very little effect on the microscopic morphologies, while the mrpigH mutants showed slower biomass accumulation and darker color on PDA. Compared with M. ruber M7, the YPs, OPs and RPs production in the mrpigH mutants (∆mrpigH∆mrlig4∆mrpyrG:: mrpyrG and ∆mrpigH∆mrlig4∆mrpyrG) increased dramatically. However, the CIT production of the mrpigH mutants decreased drastically. This work will make some contribution to the regulation of MPs and CIT production in M. ruber M7.
Author Contributions: L.L. has designed and carried out the present research work, conducted experiments, analyzed the data, and written the present manuscript. N.X. performed the secondary metabolites analysis and phenotypic characterization. F.C. provided place in the laboratory, gave access to the lab facilities for experimentation, and funds for the present work. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement:
The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.