Introduction

The genus Monascus, which reproduces both sexually and asexually, belongs to the class Ascomycetes and the family Monascaceae. It has been used from ancient times in Asian countries as a natural food colorant and flavoring agent or as a medical agent. Monascus can produce various secondary metabolites such as monacolin K (Heber et al. 1999), γ-aminobutyric acid (GABA) (Kono and Himeno 1999), pigments and citrinin (Blanc et al. 1995). It also produces industrially and pharmaceutically useful substances.

As Monascus spp. are widely used in industrially fermentation and food application, genetically modified microorganisms (GMMs) used must be food-grade in accordance with a general acceptability. Only DNA from the same genus and possibly small stretches of synthetic DNA can be used in food-grade GMMs.

To further optimize Monascus for industrial exploitation, genetic modification approaches have been used. But the genetic manipulation of Monascus has many potential applications in food safety and in the development of improved food products. Many drug resistance markers, such as hygromycin B (Campoy et al. 2003) and aureobasidin A (Shimizu et al. 2006), have been successfully used in the transformation of Monascus. However, for safety considerations, antibiotic resistance markers should be avoided for food applications. The safe use of genetically modified Monascus requires the development of food-grade transformation systems. Until recently, there is little information available in literature about the food-grade selection marker used for Monascus transformation.

Food-grade selection markers are classified into two categories, complementary markers and dominant markers. As complementary markers, pyrG gene have been successfully applied to many fungi, since the uridine auxotrophic mutants for these markers can be isolated by direct selection using the pyrimidine analog 5-fluoroorotic acid (5-FOA).

We show here that pyrG can efficiently serve in Monascus as a food-grade selection marker. We verify this application by constructing a plasmid in Monascus carrying pyrG and GFP, and we successfully transform the plasmid into Monasucus uridine auxotrophic strain by using pyrG as a selection marker.

Materials and methods

Strains and growth media

Monascus aurantiacus AS3.4384 was grown on extract/sucrose medium (MES) medium (6 ºBé wort) or Czapek Dox (CD) medium (as g/l 3 NaNO3, 2 KCl, 1 KH2PO4, 0.5 MgSO4·7H2O, and 0.02 FeSO4·7H2O, 20 glucose, and, if necessary, 1 5-FOA, 20 uridine were added).

Plasmid construction

A SalI–HindI fragment (5267 bp) containing a fusion between the GFP expression cassette and pUC18 was isolated from pSGF957 (Kim et al. 2003a, b), resulting in fragment SH-GFP. Ma-pyrG containing its own promoter and terminator was amplified by primer S-pyrG and H-pyrG (S-pyrG:5′-ACGCGTCGACCAGATATGTTCAACGATCAGGAT-3′ and H-pyrG: 5′-CCCAAGCTTCGAGAGGACTATTCCGAGGGTG-3′), resulting in fragment SH-pyrG. This fragment was digested with SalI and HindI, then inserted into SH-GFP. The resulting plasmid pGFP-pyrG contains GFP as a model gene and Ma-pyrG as a selection marker.

Isolation of genomic DNA

DNA was isolated from M. aurantiacus as described in Jae-Hyuk Yu et al. (2004) in a Mini-BeadBeater-8 (BioSpec).

Amplification of Ma-pyrG by CODEHOP PCR

The CODEHOP PCR primers were designed according to the protocol reported by Rose et al. (2003). Ten fungal pyrG proteins were selected from GenBank (Table 1) and entered into the BLOCKMAKER program (http://blocks.fhcrc.org/blocks/blockmkr/make_blocks.html), which can generate blocks of similar amino acids. The output from this program, which consisted of aligned blocks of the most highly conserved pyrG regions, was next copied and pasted into the CODEHOP algorithm. The CODEHOP program was then run to identify candidate degenerate primers. Two primers were selected from the CODEHOP results, OMP-F: (5′-GCACAACTTCCTGATCTTCgargaymgnaa-3′) and OMP-R: (5′-GAGGCGGGGGTCTGGtaytgytgncc-3′), and used to amplify a fragment of the putative M. aurantiacus pyrG gene.

Table 1 Accession numbers of fungal pyrG homologs
Full size table

50 ng Monascus genomic DNA was used as template DNA for the PCR. Cycling conditions began with 94°C for 5 min; then 35 cycles at 94°C denaturing for 30 s, 60°C annealing for 60 s, and 72°C extension for 60 s, and then a final step of 72°C for 10 min. Taq (TaKaRa) was used as the polymerase and PCRs were performed in a Gene Amp PCR system 2400 thermocycler (Perkin Elmer). The PCR product was TA cloned and sequenced. The sequence was then used in a BLAST search (http://www.ncbi.nlm.nih.gov/BLAST/) and found to display strong identity to other fungal OMP decarboxylases.

PCR-based screening strategy

We performed a PCR-based screening system for the M. aurantiacus genomic fosmid library according to a previous reported method (Kim et al. 2003). In brief, 384 fosmid clones from each filter are pooled and the DNA is purified for the PCR template as a unit. When a filter with positive signal is found, then each constituent 24 columns and 16 rows pool is analyzed individually by the same PCR assay. The location of the positive clone can be identified efficiently. Primers were designed from the obtained Ma-pyrG sequence fragment, OMP-S1: (5′-ACACAGTCCAGAAACAGTACCA-3′) and OMP-S2: (5′-CAGCTTATCACCCTTTGACGAG-3′), and used to screen the Monascus fosmid library. One of the positive fosmid clones, Q12F9, was subcloned for sequencing.

Sequence analysis

Sequence assembly was performed using the program DNASTAR, and the nucleotide sequence was analyzed using the open reading frame (ORF) Finder tool (http://www.ncbi.nlm.nih.gov/gorf/gorf.html). DNA and protein sequence alignments were carried out using the blastn and blastp programs, respectively (http://www.ncbi.nlm.nih.gov/BLAST/). Multiple alignments of the protein sequences were performed using the ClustalW program (http://www.ebi.ac.uk/clustalW/).

DNA sequence accession number

The nucleotide sequence of Ma-pyrG has been submitted to GenBank under the Accession No. GU723506.

Selection and confirmation of the uridine auxotrophic strain

The spores of M. aurantiacus were collected from the MES medium plate. UV mutagenesis were performed according to standard methods (Gruber et al. 1990). Colonies that were resistant to 5-FOA were also tested for stable uridine auxotrophy by serially replica plating them onto CD agar with and without uridine. One positive colony was obtained and named UM28. The pyrG mutation of UM28 was further confirmed by amplifying and then sequencing Ma-pyrG.

Complementation of M. aurantiacus pyrG mutant with Ma-pyrG

Complementation of M. aurantiacus pyrG mutant was performed by transforming pGFP-pyrG into UM28. Plasmids were introduced into the strain UM28 using the protoplast-PEG method (Fu et al. 2007). The pUC18 empty vector was used as a negative control and the Aspergillus oryzae plasmid pAO-pyrG was used as a positive control. The A.oryzae pyrG gene has previously been shown to be functional (Jacobs et al. 1989).

Fluorescence microscopy

Conidia of pSGF957 transformants were cultured on 50 ml CD medium at 30°C for 36–40 h with shaking. The young hyphae were harvested by centrifugation at 6,000g for 5 min, and washed with and resuspended in 500 μl sterile water. Finally, the hyphae were observed using a fluorescence microscope.

Results

Isolation of the Ma-pyrG gene

Amplification of M. aurantiacus genomic DNA with the primers designed with the CODEHOP program yielded a 430 bp PCR product similar in size to other fungus. The PCR product was cloned and sequenced. A BLAST search with the translated sequences showed a highly significant degree of identity to other fungal pyrG homologues.

In order to recover the complete Ma-pyrG, a PCR-based screening strategy was used to screen the Monascus fosmid library. Specific primers were designed from the putative Ma-pyrG sequence fragment. Six pyrG-positive clones were obtained from the fosmid library by checking the PCR products. The selected clone Q12F9 was digested with EcoRI and BamHI and fragments of 1–5 kb were subcloned for sequencing. The entire sequence of the putative Ma-pyrG gene and that of the deduced amino acids of the putative ORF are shown in Supplementary Fig. 1. There are high similarities with the OMP decarboxylases from Aspergillus flavus NRRL3357 (81 % match), Neosartorya fischeri NRRL181 (79 %), Aspergillus fumigatus Af293 (78 %), Penicillium chrysogenum (73 %) and Aspergillus oryzae RIB40 (72 %). These high similarities strongly suggest that the cloned gene encodes M. aurantiacus AS3.4384 OMP decarboxylase.

Fig. 1
figure 1

Fluorescent microscopy of transformants with GFP. The transformants with pGFP-pyrG plasmid containing the GFP gene were observed under a microscope. The photographs are of DIC of UM28 (A-1) and transformant (B-1) and those of GFP of UM28 (A-2) and transformant (B-2). Scale bars = 20 μm

Full size image

Selection and confirmation of a uridine auxotrophic strain

To create Ma-pyrG deficient strains, M. aurantiacus AS3.4384 was mutagenized by UV radiation. Before the mutagenesis, the susceptibility of M. aurantiacus AS3.4384 spores to UV radiation and 5-FOA was estimated. One pyrG mutant was obtained from 108 spores on 5-FOA selective media. This strain was a uridine auxotroph with a frequency of reverse mutation of less than 10−8, it was designated UM28.

For further confirmation of the mutation, the pyrG mutation of UM28 was amplified and then Ma-pyrG was sequenced. Sequence analysis of the Ma-pyrG gene from the uridine auxotrophic strain revealed a single nucleotide mutation of C220CT → TCT caused substitution of Pro to Ser at position 63, with inactivation of OMPdecase function. We concluded that UM28 was a Ma-pyrG deficient mutant, and this auxotroph could thus be used in the transformation system with a pyrG selective marker.

Functional complementation of M. aurantiacus pyrG mutation

To determine whether the Ma-pyrG gene encodes a functional protein, we performed a complementation experiment using UM28 as a host. The plasmid pGFP-pyrG, containing the Ma-pyrG gene and a GFP gene derived from vector pSGF957, was used for transformation. It transformed UM28 to uridine prototrophy, this result showed that the Ma-pyrG gene encodes a functional OMPdecase. At the same time, expression of the transformed gene was evaluated using the GFP gene. Hyphae of the recipient strain UM28 appeared red or light green. On the other hand, hyphae of GFP-harboring transformants showed bright green signals (Fig. 1), suggesting that the transformants not only had the GFP gene but also expressed it.

Discussion and conclusions

A new food-grade selection marker is described for transformation system of M. aurantiacus: the Monascus pyrG gene was isolated by PCR using consensus-degenerated primer design with CODEHOP strategy and its complete sequence was obtained by a PCR-based strategy for screening the Monascus fosmid library. A pyrG mutant was isolated by UV mutagenesis and identified by the sequencing and plasmid complement results. The Monascus pyrG gene can be used as a safe selection marker in molecular breeding.

Although antibiotic resistance has advantages over auxotrophic complementation, as it can avoid constructing an auxotrophic mutant as the host and cloning a complementing gene, it should be avoided for food applications. The production of Monascus-fermented products could be improved by genetic engineering to reduce the content of citrinin as well as to increase the production of monacolin K and GABA, and the pyrG selective marker will be a safer choice. Furthermore, a pyrG marker recycling method was reported (Enfert 1996; Maruyama and Kitamoto 2008), it enables infinite cycles of gene manipulation, and it may be useful for future functional genomic studies.