2015 Volume 21 Issue 5 Pages 665-670
Two rhamnogalacturonan acetylesterase genes, designated as Asrgae1 and Asrgae2, were isolated from a shoyu koji mold, Aspergillus sojae KBN1340, and characterized. Asrgae1 comprised 793 bp, encoding 248 amino acids with a predicted molecular mass of 26,092 Da, interrupted by a single putative intron of 46 bp in length. The mature AsRgae1 of 232 amino acids had a calculated molecular mass of 24,415 Da. The coding region of Asrgae2 was determined to be 777 bp in length with no introns. The predicted protein of 258 amino acids had an estimated molecular mass of 26,902 Da. The mature AsRgae2 of 238 amino acids had a calculated molecular mass of 25,015 Da. Utilizing the A. oryzae taaG2 gene promoter and the A. oryzae taaG3 gene terminator, AsRgae1 and AsRgae2 were successfully expressed in A. oryzae and secreted into the culture medium. AsRgae1 and AsRgae2 had a molecular mass of 28.0 kDa and 33.0 kDa, respectively.
Pectin is one of the most complex polysaccharides, with a backbone composed of homogalacturonan “smooth” regions and highly rhamnified “hairy” regions of rhamnogalacturonan (Willats et al., 2001). Rhamnogalacturonans consist of repeating -(1,2)-L-Rha-(1,4)-D-GalUA disaccharide units, with many rhamnose residues substituted by neutral oligosaccharides such as arabinans, galactans and arabinogalactans. Complete degradation of rhamnogalacturonan requires the cooperation of rhamnogalacturonan hydrolase (EC. 3. 2. 1. -) and rhamnogalacturonan lyase (EC. 4. 2. 2. -) along with several accessory enzymes. These include rhamnogalacturonan acetylesterase (EC. 3. 1. 1. -), which removes acetyl esters from the rhamnogalacturonan chain. Several genes encoding for rhamnogalacturonan acetylesterase have been isolated and characterized from filamentous fungi such as Aspergillus aculeatus (Kauppinen et al., 1995) and Aspergillus niger (de Vries et al., 2000). It has also been reported that the expression of the rhamnogalacturonan acetylesterase gene in A. aculeatus is induced by the presence of soybean meal and repressed by glucose (Kauppinen et al., 1995). However, there are no reports of corresponding genes from shoyu koji molds.
The filamentous fungus Aspergillus sojae, known as the shoyu koji mold, is used for the production of shoyu or soy sauce, a traditional Japanese fermented food. A. sojae secretes various kinds of carbohydrases and proteases, which are essential for the efficient maceration and hydrolysis of soybean and wheat. Among these enzymes, pectinolytic enzymes play a key role in the complete degradation of the plant cell wall, since the depolymerization of pectin exposes other cell wall components to degradation and causes further cell wall breakdown. These enzymes are therefore considered to be vital to the soy sauce filtration process (Kikuchi, 1977). Elucidation of the pectin-degrading system of A. sojae is important to improve the filtration efficiency of soy sauce mash and decrease the amount of pressed cake. In our previous studies, the AspecA gene encoding a polygalacturonase and the AsrglA gene encoding a rhamnogalacturonan lyase were isolated from an industrial shoyu koji mold strain, A. sojae KBN1340, and their gene products were characterized (Yoshino-Yasuda et al., 2011, Yoshino-Yasuda et al., 2012).
Here, we describe the cloning and sequencing of two rhamnogalacturonan acetylesterase genes, Asrgae1 and Asrgae2, from A. sojae KBN1340. In addition, Asrgae1 and Asrgae2 were overexpressed in Aspergillus oryzae under the control of the A. oryzae taaG2 gene promoter and the A. oryzae taaG3 gene terminator (Tsukagoshi et al., 1989).
Fungal strains, culture media and transformation Industrial shoyu koji molds, A. sojae KBN1340 and A. oryzae KBN616, obtained from Bio'c (Toyohashi, Aichi, Japan) were used for DNA isolation. The pyrG, alp double deleted strain, A. oryzae P3 (Yoshino-Yasuda et al., 2012), derived from an industrial shoyu koji mold, A. oryzae KBN616, was used for transformation. SP medium containing 3% soluble starch, 1% polypeptone, 0.5% KH2PO4, 0.5% KCl, 0.1% NaNO3, and 0.05% MgSO4·7H2O was used for the production of AsRgae1 and AsRgae2 by A. oryzae transformants. Protoplast transformation of A. oryzae was carried out according to the method described previously (Kitamoto et al., 1995).
Amplification of Asrgae1 and Asrgae2 gene fragments by PCR Genomic DNA fragments encoding portions of the Asrgae1 and Asrgae2 genes were amplified by PCR using the genomic DNA of A. sojae KBN1340 as a template. Two oligonucleotide primers were synthesized based on two amino acid sequences highly conserved among A. aculeatus Rha1 (Kauppinen et al., 1995), Aspergillus niger RgaeA (de Vries et al., 2000) and Neurospora crassa rhamnogalacturonan acetylesterase (Galagan et al., 2003). The sense primer srh1 (5′-TTYGGYCAYAAYGAYGGYGG-3′) was homologous to the sense strand for amino acid residues 88 to 94 of A. aculeatus Rha1. The antisense primer srh2 (5′-GGRTARCTRGTRTGRGTRTG-3′) was complementary to the sense strand for amino acid residues 207 to 213 of A. aculeatus Rha1. The amplified 377-bp fragment designated as RGAE1-1 and 383-bp fragment designated as RGAE2-1 were cloned in pUC119 and sequenced. Sequencing was performed using a model 4000LS DNA sequencer (LI-COR, Lincoln, NE, USA).
For amplification of the 5′- and 3′- regions of RGAE1-1 and RGAE2-1, cassette-ligation-mediated PCR (CLM-PCR) was performed using an LA PCR in vitro Cloning kit (Takara Bio, Otsu, Shiga, Japan) as described for those of the A. sojae rhamnogalacturonan lyase A (Yoshino-Yasuda et al., 2012). Four specific primers: srh3 (5′-CAGACATTGTTCGGAGTCTGGCTGG-3′), srh4 (5′-GTAAGCTCCGTGGTCCACATAGTCC-3′), srh5 (5′-AAGGTCCTCATCTCCAGCCAGACTC-3′) and srh6 (5′-CCAGCCAGACTCCGAACA ATGTCTG-3′) were synthesized on the basis of RGAE1-1. The 5′-region was amplified from PstI cassette-ligated genomic DNA with two sets of primers (C1/srh4 in the primary amplification and C2/srh3 in the second round of PCR). The 3′-region was also amplified from EcoRI cassette-ligated genomic DNA with two sets of primers (C1/srh5 in the primary amplification and C2/srh6 in the second round of PCR) by a similar procedure. The amplified fragments, an 895-bp fragment for the 5′-region and a 686-bp fragment for the 3′-region, were cloned and sequenced.
Four specific primers: srh7 (5′-CATCTGAGATGCACGTCTCAGTACC-3′), srh8 (5′-CATGAGCTTGGCTGCTTGAATGACG-3′), srh9 (5′-TTCTACGTCATTCAAGCAGCCAAGC-3′) and srh10 (5′-AGCGACGATGTCACCTTTGTCGACC-3′) were also synthesized based on RGAE2-1. The 5′-region was amplified from XbaI cassette-ligated genomic DNA with two sets of primers (C1/ srh8 in the primary amplification and C2/srh7 in the second round of PCR). The 3′-region was also amplified from Sau3AI cassette-ligated genomic DNA with two sets of primers (C1/srh9 in the primary amplification and C2/srh10 in the second round of PCR) by a similar procedure. The amplified fragments, a 1,759-bp fragment for the 5′-region and a 301-bp fragment for the 3′-region, were also cloned and sequenced.
Construction of expression vectors for the Asrgae1 and Asrgae2 genes Two expression vectors, pTARA100 for the Asrgae1 gene and pTARA200 for the Asrgae2 gene, under the control of the A. oryzae taaG2 gene promoter and A. oryzae taaG3 gene terminator were constructed as follows. The 418-bp fragment of the A. oryzae taaG3 gene terminator was amplified by PCR using the chromosomal DNA of A. oryzae KBN616 as a template, and a pair of primers, taaG3T3 (5′-CCGTCGACTGAAGGGTGGAGAGTATATGATG-3′) corresponding to positions +2,041 to +2,061 and taaG3T4 (5′-AGCTCGAGCTATCTGGGCATTAGTAAGTG -3′) corresponding to positions +2,459 to +2,437, referring to the translation start site of the A. oryzae taaG3 gene as +1. The SalI and XhoI digest of the PCR-amplified DNA fragment was inserted in the SalI site on pYRMTA100 (Yoshino-Yasuda et al., 2012), resulting in the plasmid pYRMTA200. The 803-bp fragment of the coding region for the Asrgae1 gene was amplified by PCR using the chromosomal DNA of A. sojae KBN1340 as a template and a pair of primers, rgae1N2 (5′-ACATGCATTCCGTCCTTCTCCCCCTGTCCCTT-3′) corresponding to positions +7 to +30 and rgae1C (5′-ATCTCGAGCTACAGACAGTCACCCGGGAAATC-3′) corresponding to positions +793 to +771, referring to the translation start site of the Asrgae1 gene as +1. The EcoT22I- and XhoI-digested PCR-amplified DNA fragment was inserted into the EcoT22I -SalI site on pYRMTA200, generating the plasmid pTARA100.
The 784-bp fragment of the coding region for the Asrgae2 gene was also amplified by PCR using the chromosomal DNA of A. sojae KBN1340 as a template and a pair of primers, rgae2N2 (5′-TCATGCATTTCACACCATCTTCGGCGGCCTTG-3′) corresponding to positions +7 to +30 and rgae2C (5′-TACTCGAGCTACTCGTAAACAATGGTCACAGG-3′) corresponding to positions +778 to +754, referring to the translation start site of the Asrgae2 gene as +1. The EcoT22I- and XhoI-digested PCR-amplified DNA fragment was inserted into the EcoT22I -SalI site on pYRMTA200, generating the plasmid pTARA200. The resultant plasmids, pTARA100 and pTARA200, were introduced to A. oryzae KBN616-P3.
Enzyme assay Rhamnogalacturonan acetylesterase activity was determined using the method of Kauppinen et al. (Kauppinen et al., 1995) with modifications. The enzyme activity was assayed spectrophotometrically by measuring the release of p-nitrophenol from p-nitrophenyl-acetate (PNP-acetate) at 40°C for 15 min at 405 nm. The reaction mixture contained 950 µL of 1 mM PNP-acetate in 0.1 M potassium phosphate buffer (pH6.0) and 50 µL of enzyme solution. One arbitrary enzyme unit was defined as an increase of 1.0 in absorbance per min per ml of the enzyme solution.
Computational sequence analysis A homology search was performed using a BLAST search of DOGAN, the Database of the Genomes Analyzed at NITE (http://www.bio.nite.go.jp/dogan/top). The search for signal peptides was performed using SignalP 4.0 (http://www.cbs.dtu.dk/services/SignalP/). The search for glycosylation sites was performed using two online glycosylation site prediction servers, NetOGlyc 4.0 (http://www.cbs.dtu.dk/services/NetOGlyc/) for prediction of O-linked glycosylation sites and NetNGlyc 1.0 (http://www.cbs.dtu.dk/services/NetNGlyc/) for prediction of N-linked glycosylation sites.
Structural features of the Asrgae1 and Asrgae2 genes Regions of the Asrgae1 and Asrgae2 genes were amplified by PCR with genomic DNA as the template and a pair of degenerated primers corresponding to the amino acid sequences in regions highly conserved among rhamnogalacturonan acetylesterases from A. aculeatus, A. niger and N. crassa. The amplified 377-bp fragment designated as RGAE1-1 and 383-bp fragment designated as RGAE2-1 were sequenced and found to contain an open reading frame encoding 125 and 127 amino acids highly homologous to the sequences of rhamnogalacturonan acetylesterases from A. aculeatus and A. niger, respectively.
To amplify the 5′- and 3′- regions of RGAE1-1, CLM-PCR was performed with PstI or EcoRI cassette-ligated genomic DNA as the template, as described in the Materials and Methods. An 895-bp fragment for the 5′-region designated as RGAE1-2 and a 686-bp fragment for the 3′-region designated as RGAE1-3 were amplified. Based on the sequences of RGAE1-1, RGAE1-2 and RGAE1-3, the entire nucleotide sequence of the Asrgae1 gene was determined (Fig. 1). The coding region of the Asrgae1 gene was 793 bp and was interrupted by a single putative intron (46 bp in length) at the same position as the A. niger rgaseA gene. The coding sequence encoded a polypeptide of 248 amino acid residues with a molecular mass of 26,092 Da. Analysis using the SignalP 4.0 program predicted that the first 16 N-terminal amino acid of AsRgae1 function as a signal sequence. The molecular mass of the mature AsRgae1 was calculated to be 24,415 Da.
Nucleotide and deduced amino acid sequences of the Asrgae1 gene.
Numbers on the right refer to the nucleotide sequence (negative numbers refer to nucleotides upstream of the Asrgae1 ATG translation start codon) and amino acid sequence. Intron sequences are denoted in lower-case letters. The TATA box and a polyadenylation signal-like sequence are double underlined, and CreA consensus binding sites are single underlined. An asterisk (*) marks the translation stop codon. The predicted signal sequence cleavage site is indicated by a triangle. Potential O-glycosylation sites are boxed. The Asrgae1 DNA sequence is available in the DDBJ/EMBL/GenBank databases under the accession number LC015524.
For amplification of the 5′- and 3′- regions of RGAE2-1, CLM-PCR was also performed with XbaI or Sau3AI cassette-ligated genomic DNA as the template using a similar procedure as described for Asrgae1. A 1,759-bp fragment for the 5′-region designated as RGAE2-2 and a 301-bp fragment for the 3′-region designated as RGAE2-3 were amplified. Based on the sequences of RGAE2-1, RGAE2-2 and RGAE2-3, the entire nucleotide sequence of the Asrgae2 gene was determined (Fig. 2). The coding region of the Asrgae2 gene is 777 bp in length with no introns. The Asrgae2 gene encoded a polypeptide of 258 amino acid residues with a molecular mass of 26,902 Da. The signal sequence cleavage site was predicted at Ala-20. The molecular mass of the mature AsRgae2 was calculated to be 25,015 Da. One potential N-glycosylation site was found at Asn-240.
Nucleotide and deduced amino acid sequences of the Asrgae2 gene.
Numbers on the right refer to the nucleotide sequence (negative numbers refer to nucleotides upstream of the Asrgae2 ATG translation start codon) and amino acid sequence. Intron sequences are denoted in lower-case letters. The TATA box and CCAAT sequence are double underlined, and CreA consensus binding sites are single underlined. An asterisk (*) marks the translation stop codon. The predicted signal sequence cleavage site is indicated by a triangle. A potential N-glycosylation site is circled and potential O-glycosylation sites are boxed. The Asrgae2 DNA sequence is available in the DDBJ/EMBL/GenBank databases under the accession number LC015525.
The 5′- and 3′-noncoding regions of the Asrgae1 and Asrgae2 genes were searched for various consensus sequences. A potential TATA box at -97 (TATAAA) and two CreA binding sites at -153 (CTGGG) and -328 (GTGGG) before the start codon of the Asrgae1 gene were found within the determined region. A putative TATA box at -151 (TATAAC), a CCAAT sequence at -482 (CCAATT), and three CreA binding sites at -393 (GCGGGG), -472 (CCCCAG), and -520 (CCCCAG) were observed in the 5’ region of the Asrgae2 gene. A polyadenylation signal-like sequence (AATAGA) was located at 167 bp downstream from the stop codon of the Asrgae1 gene.
Comparison of the amino acid sequences of AsRgae1 and AsRgae2 with those of fungal rhamnogalacturonases The deduced amino acid sequences of AsRgae1 and AsRgae2 were compared with those of several rhamnogalacturonan acetylesterases of fungal origin. The degree of amino acid identity ranged from 50.0 to 70.0 % for the majority of the genes analyzed, with some notable exceptions (Table 1). AsRgae1 and AsRgae2 showed the highest identity (98.0 % and 98.8%) to rhamnogalacturonan acetylesterase from A. oryzae (AO 090701000556 and AO090003001268) (Machida et al., 2005), respectively, although the sequence identity between AsRgae1 and AsRgae2 was 55.7%.
| Fungal rhamnogalacturonan acetylester ase | AsRgae1 | AsRgae2 |
|---|---|---|
| A. sojae AsRgae1 | 100 | 55.7 |
| A. sojae AsRgae2 | 55.7 | 100 |
| A. aculeatus Rha1 | 69.2 | 52.9 |
| A. niger RgaeA | 62.9 | 47.0 |
| A. oryzae AO090701000556 | 98.0 | 56.1 |
| A. oryzae AO090003001268 | 56.0 | 98.8 |
| N. crassa rhamnogalacturonan acetylesterase | 55.7 | 59.4 |
Based on the results of X-ray crystallography combined with the sequence alignment of A. aculeatus Rha1, the catalytic triad and the active site have been revealed from the three-dimensional structure (Mølgaard et al., 2000); Ser-26, Asp-209 and His-212 are involved in the catalytic triad, and Ser-26, Gly-59, Asn-91 and His-212 are involved in the active site. The corresponding residues in AsRgae1 and AsRgae2 are also likely involved in the catalytic triad and the active site, i.e., Ser-25, Asp-207 and His-210 for the catalytic triad, and Ser-25, Gly-57, Asn-89 and His-210 for the active site in AsRgae1, and Ser-29, Asp-215 and His-218 for the catalytic triad, and Ser-29, Gly-63, Asn-95 and His-218 for the active site in AsRgae2.
Heterologous expression of the Asrgae1 and Asrgae2 genes in A. oryzae In order to obtain high-level production of AsRgae1 and AsRgae2 in A. oryzae, the Asrgae1 and Asrgae2 genes were efficiently expressed under the control of the A. oryzae taaG2 gene promoter and the A. oryzae taaG3 gene terminator. Two expression vectors, pTARA100 and pTARA200, were constructed as described in the Materials and Methods, and introduced into A. oryzae KBN616-P3. Thirty transformants were randomly selected and cultivated in SP medium for 4 days to screen for high-AsRgae1 or AsRgae2-producing transformants. After cultivation, all of the transformants exhibited detectable rhamnogalacturonan acetylesterase activity, although the control strain carrying the pyrG gene exhibited no rhamnogalacturonan acetylesterase activity. A high-AsRgae1-producing transformant, A. oryzae SR1-21, produced 1.91 U/mL after 4 days cultivation, whereas a high-AsRgae2-producing transformant, A. oryzae SR2-11, produced 0.55 U/mL.
After electrophoresis of the culture supernatant on a sodium dodecyl sulfate/polyacrylamide gel, rAsRgae1 with a molecular mass of 28.0 kDa and rAsRgae2 with a molecular mass of 33.0 kDa were detected as prominent bands in the culture supernatant of A. oryzae SR1-21 and A. oryzae SR2-11, respectively, when stained with Coomassie Brilliant Blue (Fig. 3, lanes 3 and 4), although Taka-amylase A (TAA) with a molecular mass of 45.0 kDa was detected as a prominent band in the culture supernatant of an A. oryzae transformant carrying pYRMTA200 (Fig. 3, lane 2). Since the calculated molecular masses were smaller than those of rAsRgae1 and rAsRgae2 by approximately 4.0 kDa and 8.0 kDa, respectively, we assumed that the differences between the theoretical and apparent molecular mass were due to protein glycosylation. This idea is supported by the presence of five putative O-linked glycosylation sites (Thr-101, Thr-106, Thr-110, Thr-113 and Thr-114) in AsRgae1 and one putative N-linked glycosylation site (Asn-240) and three O-linked glycosylation sites (Ser-108, Thr-117 and Ser-120) in AsRgae2, as predicted by NetNGlyc 1.0 and NetOGlyc 4.0, respectively.
SDS-polyacrylamide gel electrophoresis of AsRgae1 and AsRgae2 secreted by A. oryzae transformants.
The A. oryzae transformants SR1-21 and SR2-11 were cultivated as described in the Materials and Methods. The culture filtrates (9 µL) were subjected to SDS-PAGE on a 12.5% gel and stained with Coomassie Brilliant Blue. Lane 1, molecular-mass markers [rabbit muscle phosphorylase b (97.4 kDa), bovine serum albumin (66.3 kDa), rabbit muscle aldolase (42.4 kDa), bovine carbonic anhydrase (30.0 kDa), and soybean trypsin inhibitor (20.1 kDa)]; lane 2, the culture filtrate of an A. oryzae transformant carrying pYRMTA200; lane 3, the culture filtrate of an A. oryzae transformant SR1-21; lane 4, the culture filtrate of an A. oryzae transformant SR2-11.
We are now evaluating the effects of AsRgae1 and AsRgae2 on the degradation of soybean pectin.
Acknowledgments We thank T. Tsuzuki for her technical assistance. This work was partially supported by a study on biorecycling of wastes from the agriculture, forestry and fisheries sector from the Ministry of Agriculture, Forestry and Fisheries.
