Aspergillus oryzae-based cell factory for direct kojic acid production from cellulose

Background Kojic acid (5-Hydroxy-2-(hydroxymethyl)-4-pyrone) is one of the major secondary metabolites in Aspergillus oryzae. It is widely used in food, pharmaceuticals, and cosmetics. The production cost, however, is too high for its use in many applications. Thus, an efficient and cost-effective kojic acid production process would be valuable. However, little is known about the complete set of genes for kojic acid production. Currently, kojic acid is produced from glucose. The efficient production of kojic acid using cellulose as an inexpensive substrate would help establish cost-effective kojic acid production. Results A kojic acid transcription factor gene over-expressing the A. oryzae strain was constructed. Three genes related to kojic acid production in this strain were transcribed in higher amounts than those found in the wild-type strain. This strain produced 26.4 g/L kojic acid from 80 g/L glucose. Furthermore, this strain was transformed with plasmid harboring 3 cellulase genes. The resultant A. oryzae strain successfully produced 0.18 g/L of kojic acid in 6 days of fermentation from the phosphoric acid swollen cellulose. Conclusions Kojic acid was produced directly from cellulose material using genetically engineered A. oryzae. Because A. oryzae has efficient protein secretion ability and secondary metabolite productivity, an A. oryzae-based cell factory could be a platform for the production of various kinds of bio-based chemicals.


Background
Kojic acid (5-Hydroxy-2-(hydroxymethyl)-4-pyrone) is widely used in the food industry, pharmaceutical industry, and in cosmetics [1,2]. Furthermore, kojic acid is also used as a building block for biodegradable plastics [3,4]. Plastics have phenolic hydroxyl groups that are derived from kojic acid. Thus, it has some unique properties. Futamura et al. [5] reported that the production cost of kojic acid is about $10/kg. It must be decreased to less than $2/kg before it could be used in many applications. Thus, an efficient and cost-effective kojic acid production process would be desirable.
The utilization of biomass, particularly cellulosic materials, is desirable because it is abundant, inexpensive, renewable, and has favorable environmental properties.
However, an efficient and cost-effective method for the degradation of cellulosic materials into glucose would be required before it could be used for the production of bio-based chemicals.
Cellulose has a structure that is very rigid with a high degree of crystallinity. Thus, the degradation of cellulose requires large amounts of various types of cellulase. The degradation of cellulose requires a synergistic reaction of at least three types of cellulase: endoglucanase (EG), cellobiohydrolase (CBH), and β-glucosidase (BGL). Although many studies have reduced the cost of cellulosic material degradation using recombinant bacteria, fungi, and yeast [6], cellulase degradation efficiency has not been sufficiently improved. The filamentous fungus Trichoderma reesei degrades cellulose effectively and is known to produce more than several dozen g/L of cellulase in the medium [7], therefore, it is known for its high cellulose degradation ability. Aspergillus oryzae is known as a high protein-secreting microorganism and its genetic recombination procedures are well developed. There are many reports concerning the production of a large amount of protein by A. oryzae [8]. Thus, A. oryzae shows promise as a host strain that could produce a large amount of cellulase and bio-based chemicals from cellulose. There have been reports concerning a single cellulase gene expression by A. oryzae, but no studies have focused on the cultivation of the 3 types of cellulase expressing A. oryzae on cellulose as a sole carbon source.
Kojic acid is one of the major secondary metabolites in A. oryzae. In the 100 years since it was discovered, there has been no elucidation of the entire pathway of kojic acid biosynthesis in A. oryzae. Recently, some genes regarding the production of kojic acid in A. oryzae were reported [9]. According to the report, 14 genes from AO090113000132 to AO090113000145 including a transcription factor gene (kojR; AO090113000137), an enzyme gene (kojA, AO090113000136), and a transporter gene (kojT AO090113000138) form a cluster. These genes are engaged in kojic acid production. The kojR gene encodes a fungal-specific Zn(II) 2 Cy 6 transcription factor that is located between kojA (upstream 743 bp) and kojT (downstream 383 bp). The kojR regulates the transcription of kojA and kojT. A strain with a disrupted transcription factor gene could not produce kojic acid at all [10]. Thus, kojR seems to be a key factor in kojic acid production.
The goal of the present study was to construct an A. oryzae-based cell factory for direct kojic acid production from cellulose. First, A. oryzae over-expressing the transcription factor gene kojR was constructed, and kojic acid production from glucose and starch by the resultant strain was carried out. Then, A. oryzae co-expressing the transcription factor and 3-types of cellulase genes was constructed. Finally, direct kojic acid production from cellulose by the resultant strain was carried out. Table 1 summarizes the genetic properties of all strains used. The A. oryzae strain RIB40 [11] was used as the wild-type strain. The strain NSPlD1 [12] derived from A. oryzae RIB40 was used as a transformation host strain. Potato dextrose agar (PDA; Nissui Pharmaceutical, Tokyo, Japan) media and GPY medium (3% glucose, 0.2% KCl, 0.1% KH 2 PO 4 , 0.05% MgSO 4 · 7H 2 O, 1% peptone, and 0.5% yeast extract, pH 6.0) were used for the cultivation of A. oryzae. Czapek-Dox (CD) medium (2% glucose, 0.3% NaNO 3 (CD-NO 3 ), 0.2% KCl, 0.1% KH 2 PO 4 , 0.05% MgSO 4 · 7H 2 O, and 0.8 M NaCl, 1.5% agar, pH 6.0) with a required supplement (0.0015% methionine, 20 mM uridine, 0.2% uracil) was used for the niaD-based plasmid transformation. CD medium was used for sC-based plasmid transformation.

Construction of plasmids
The primers used in the present study are summarized in Table 2. The PCR amplification of DNA fragments was performed using KOD plus neo DNA polymerase (Toyobo, Osaka, Japan).
The transcription factor gene (kojR, AO090113000137) expression plasmid was constructed as follows. The kojR was amplified by PCR using primers ascI-kojR-F and notI-FLAG-kojR-R from A. oryzae genome DNA as a template. The amplified fragment was inserted into the AscI/NotI sites of the digested pISI, which contained the sodM promoter and glaB terminator from A. oryzae [17]. The resultant plasmid was named pISI-kojR.
The CBHI gene from Trichoderma sp. expressing plasmid pIS1-svaA-CBHI was constructed as follows. The DNA fragment encoding the Trichoderma sp. CBHI gene (GenBank accession no. X69976) was amplified by PCR using primers CBHI-F and CBHI-R from a synthesized gene (GENEWIZ, South Plainfield, NJ, USA). The fragment was inserted into the AscI/NotI site of the plasmid pISI using an In-Fusion HD Cloning Kit (Takara Bio Inc., Shiga, Japan). The resultant plasmid was named pISI-CBHI. Then, the CBHI gene was amplified by PCR using primers psvaA-ascI-CBHI-F and tsvaA-notI-CBHI-R with pISI-CBHI as a template. The svaA promoter [17] and svaA terminator were amplified by PCR using primers pIS1-psvaA-F and CBHI-ascI-psvaA-R, and CBHI-notI-tsvaA-F and pIS1-tsvaA-R, respectively, with A. oryzae genome DNA as a template. These PCR-amplified DNA fragments were simultaneously inserted into the XbaI and PstI sites of pIS1 using an In-Fusion PCR Cloning kit. The resultant plasmid was named pISI-svaA-CBHI.
An EGI gene from the T. reesei-expressing plasmid pIS1-hlyA-EGI was constructed. The EGI-expressing plasmid was constructed as follows. The DNA fragment encoding the T. reesei EGI gene was amplified by PCR using the primers EGI-F and EGI-R with T. reesei cDNA as a template. The fragment was inserted into the AscI/ NotI site of the plasmid pISI using an In-Fusion HD Cloning Kit. The resultant plasmid was named pISI-EG. The T. reesei EGI gene was then amplified by PCR using the primers phlyA-ascI-EGI-F and thlyA-notI-EGI-R with pISI-EG as a template. The hlyA promoter [18] and hlyA terminator were amplified by PCR using primers pIS1-phlyA-F and EGI-ascI-phlyA-R, and EGI-notI-thlyA-F and pIS1-thlyA-R, respectively, using the A. oryzae genome DNA as a template. These PCR-amplified DNA fragments were simultaneously inserted into the XbaI and PstI sites of pIS1 using an In-Fusion PCR Cloning kit. The resultant plasmid was named pISI-hlyA-EGI.
The BGL1 gene from A. aculeatus expressing plasmid pIS1-BGL1 was constructed as follows. The DNA fragment encoding the secretion signal from A. oryzae cutinase, the 28 amino acids from the N-terminal region of Rhizopus oryzae lipase, and the A. aculeatus BGL gene were amplified by PCR using the primers sCutL-F and sCutL-R, N28-F and N28-R, and BGL-F and BGL-R, respectively, from the genome DNA of A. oryzae, pISI-GFP [14], and pδU-PGAGBGL [15], respectively. These fragments were simultaneously inserted into the AscI/NotI site of the plasmid pISI using an In-Fusion HD Cloning Kit. The resultant plasmid was named pISI-BGL.

Transformation of A. oryzae
The transformation of A. oryzae was carried out according to the previously described method with minor modifications [19].

RNA extraction and reverse transcription
The RNA was isolated from fungus bodies cultivated in GY medium for 6 days at 30°C using NucleoSpin RNA II (Takara Bio Inc.) according to the manufacturer's protocol. The cDNA synthesis was carried out using a ReverTra Ace qPCR RT Kit (Toyobo) according to the manufacturer's protocol.

Real-time PCR analysis
The transcription level of Kojic-acid-producing genes (kojR, kojA, kojT) was quantified by real-time PCR. Quantitative real-time PCR was performed using an Mx3005P Real-Time QPCR System (Agilent Technologies, Santa Clara, CA, USA) with a Thunderbird SYBR qPCR Mix (Toyobo). The normalized transcription level was calculated using the standard curve method with β-tubulin as the house-keeping gene [20]. All primers used for realtime PCR analysis are summarized in Table 2.

Enzyme assay
A wild-type strain and 3-types of cellulase gene-expressing strains were cultivated in GY medium for 6 days at 30°C, and the culture supernatant was used for the determination of PASC degradation activity. PASC degradation activity was measured in 25 mM sodium acetate buffer (pH 5.0) at 30°C with 5 g/L of PASC. After hydrolysis, the supernatant was separated by centrifugation for 5 min at 10,000 × g and 4°C, and the produced glucose concentration was measured. One unit of PASC degradation activity was defined as the amount of enzyme producing 1 μmol/min glucose at 30°C, pH 5.0.

Analytical methods
The kojic acid concentration was determined using a colorimetric method described previously [21]. The total sugar concentration was determined by a colorimetric phenol-sulfuric acid method described previously [22]. The glucose concentration was determined using a Wako Glucose CII-Test kit (Wako Pure Chemical Industries, Osaka, Japan).

Results
Kojic acid production from glucose using the engineered A. oryzae To confirm the effect of the over-expression of the transcription factor for kojic acid production, the kojic acid productivity from the transcription factor over-expressing strain NSPlD1/pIS1-kojR from glucose was evaluated. As shown in Figure 1 the wild-type strain produced 16.4 g/L of kojic acid after 14 days of fermentation. By contrast, the transcription factor over-expressing strain NSPlD1/ pIS1-kojR produced 26.4 g/L of kojic acid after 14 days of fermentation and it was 1.6-fold higher than that of the wild-type strain.

Quantification of the transcription level of kojic-acid-producing genes by real-time PCR
To confirm the effect of the over-expression of transcription factor at the level of the kojic-acid-producing genes, the transcription level of the transcription factor, the kojic-acid-producing enzyme, and the transporter gene were evaluated by real-time PCR. As shown in Figure 2, the transcription factor of the over-expressing strain NSPlD1/ pIS1-kojR showed a 1.8, 1.8, and 5.4-fold increases in the transcription level of transcription factor, enzyme, and transporter gene, respectively, compared with the wildtype strain.

Kojic acid production from soluble starch by recombinant A. oryzae
Since A. oryzae can assimilate starch as a sole carbon source intrinsically, direct kojic acid production from starch using the transcription factor over-expressing strain NSPlD1/pIS1-kojR was carried out. As shown in Figure 3, the wild-type strain produced 6.2 g/L of kojic acid after 18 days of fermentation. In contrast, the transcription factor over-expressing strain produced 19.4 g/L of kojic acid after 18 days of fermentation, i.e. 3.1-fold gain.

Evaluation of cellulase activity
To confirm the cellulase expression of engineered A. oryzae, the PASC degradation activity of 3-types of cellulase were evaluated. As shown in Figure 4, the 3-types of cellulase gene-expressing strains NSPlD1/pIS1-kojR/ pIS1sC-CBHI-EGI-BGLI showed 29.9 U/L after 6 days of   cultivation. In contrast, the wild-type strain showed significantly lower PASCase activity (1.4 U/L).

Direct kojic acid production from PASC
To confirm direct kojic acid productivity from cellulose, kojic acid production from PASC was carried out. As shown in Figure 5, there was 0.18 g/L of kojic acid produced after 6 days of fermentation when using a transcription factor gene and the 3-types of cellulase genes co-expressing strain NSPlD1/pIS1-kojR/pIS1sC-CBHI-EGI-BGLI. In contrast, the wild-type strain did not produce a detectable amount of kojic acid.

Discussion
In the present study, a transcription factor gene overexpressing A. oryzae NSPlD1/pIS1-kojR was developed by gene recombination strategy. Furthermore, the transcription factor gene and 3-types of cellulase genes co-expressing A. oryzae NSPlD1/pIS1-kojR/pIS1sC-CBHI-EGI-BGLI was constructed and the strain successfully produced kojic acid directly from PASC. The transcription factor gene kojR, which is responsible for kojic acid production, was over-expressed in this study, and as a result, kojic acid productivity was improved. As shown in Figure 2, the transcription levels of both the kojic acid-producing enzyme kojA and the exporting transporter kojT were enhanced. This was because the kojic acid productivity was improved. Terabayashi et al. [9] reported that the kojR, kojA, and kojT genes would encode the transcription factor, the kojic acid-producing enzyme, and the transporter, respectively. In the present study, the transcription level of the enzyme and the transporter were enhanced by an over-expression of the transcription factor, which correlates well with that report. Although there are reports of improved kojic acid productivity through fermentation engineering strategies [1,5,23], to our knowledge, this study is the first to report an improvement in kojic acid productivity via genetic engineering. Kojic acid productivity could be improved even more by combining fermentation engineering and genetic engineering in future work.
Direct kojic acid production from cellulose was successfully demonstrated by A. oryzae expressing 3-types of cellulase genes. However, the yield of kojic acid remained low. This was due to the low levels of cellulase production. In previous reports, some microorganisms such as Saccharomyces cerevisiae, Kluyveromyces marxianus, Bacillus subtilis, and E. coli were used for bio-based chemical production from cellulose by conferring cellulolytic activity [6]. Although most reports did not indicate the amount of cellulase production, Zhang et al. [24] reported that about 14 mg/L of endoglucanase was produced by recombinant Bacillus subtilis. In any case, because T. reesei produces several dozen g/L of cellulase [7], the amount of cellulase expression from cellulolytic activity conferred microorganisms has not yet reached the level of T. reesei. Although A. oryzae has an extremely high protein secreting ability, cellulase expression of T. reesei is still higher than that of A. oryzae. Because the maximum protein production is approximately 20 g/L in Aspergillus organisms [25], the cellulase productivity could be improved with genetic engineering.
Three types of cellulase genes and specific transcription factors were successfully over-expressed simultaneously in recombinant A. oryzae. To this point, only a few reports have focused on the co-expression of more than 3-kinds of genes in A. oryzae. This is because the selection marker for gene recombination is limited in A. oryzae compared with other microorganisms such as yeast S. cerevisiae [26]. In the present study, a 3-cassette plasmid with different promoters was constructed and the simultaneous overexpression of 4 genes was successful. Thus, the simultaneous multi gene expression technique, which is an important technique for the construction of a cell factory in A. oryzae, has been well established. By using this technique, an A. oryzae-based cell factory could be a promising strategy for producing various bio-based chemicals.

Conclusions
The productivity of kojic acid from A. oryzae was made more efficient by the over-expression of specific transcription factors. To accomplish this, recombinant A. oryzae expressing 3 types of cellulase genes was constructed. Direct kojic acid production from PASC by genetically engineered A. oryzae was successfully demonstrated. Figure 5 Time course of kojic acid production from PASC. Open symbols, wild-type strain; closed symbols, transcription factor and 3-types of cellulase gene over-expressing strains NSPlD1/pIS1-kojR/ pIS1sC-CBHI-EGI-BGLI. The data represent the averages from three independent experiments (error bars represent SE).