Isolation of Four Pepsin-Like Protease Genes from Aspergillus Niger and Analysis of the Eﬀect of Disruptions on Heterologous Laccase Expression

Available online at www.sciencedirect.com

Fungal Genetics and Biology 45 (2008) 17–27 www.elsevier.com/locate/yfgbi

Isolation of four pepsin-like protease genes from Aspergillus niger and analysis of the eﬀect of disruptions on heterologous laccase expression

Yongchao Wang a, Wei Xue a, Andrew H. Sims b, Chuntian Zhao a, Aoquan Wang a,*, Guomin Tang a, Junchuan Qin c, Huaming Wang b

a Key Laboratory of Systematic Mycology and Lichenology, Institute of Microbiology, Chinese Academy of Sciences, 100101 Beijing, PR China b Genencor, A Danisco Division, Palo Alto, CA 94304, USA c School of Life Science, Nanjing University, 210093 Nanjing, PR China

Received 10 February 2007; accepted 24 September 2007 Available online 29 September 2007

Abstract

Four new aspartic protease genes pepAa, pepAb, pepAc and pepAd from Aspergillus niger were identified using a comparative genomic approach. All four gene products have highly conserved attributes that are characteristic of aspartic proteases; however, each one has novel sequence features. The PEPAa protease appears to represent an ortholog of a pepsin-type aspartic protease previously identified from Talaromyces emersonii and Scleotinia sclerotiorum. The PEPAb protease appears to be an ortholog of an aspartic protease previously identified from BcAP1 of Botryotinia fuckeliana. The PEPAc protease also appears to be an ortholog of BcAP5 from B. fuckeliana. These four genes appear to be conserved in many species of filamentous fungi, all except PEPAb contain a predicted signal peptide. Tran- scriptome analysis revealed that transcripts of the pepAa gene of Aspergillus nidulans were significantly up-regulated due to recombinant chymosin secretion, suggesting that silencing these genes may lead to improved yields of secreted proteins. To establish the effects of reduced protease activity on the stabilities of secreted proteins, three of the four genes were individually disrupted by double crossover, although we were unable to disrupt the pepAc gene. The secretion level of heterologous laccase in the pepAa, pepAb and pepAd disruption mutants were increased by about 21%, 42% and 30%, respectively. And their total glucogenic enzymes secretion were also increased by about 18.7%, 37.0% and 5.20%, respectively. Ó 2007 Elsevier Inc. All rights reserved.

Keywords: Aspergillus niger; Pepsin-like protease; Disruption; Heterologous laccase expression

1. Introduction tein production upon deletion of the protease genes (Archer et al., 1992; van den Hombergh et al., 1997b). Filamentous fungi represent excellent hosts for secre- Four extracellular proteases with acid pH optima (an tion of homologous and heterologous proteins (Conesa aspartic protease, PEPA, a glutamic protease, PEPB and et al., 2001). However, proteolytic degradation is a major two serine carboxypeptidases, PEPF and PEPG) have pre- problem during protein production even in a well-estab- viously been characterized (Berka et al., 1990; Inoue lished organism such as Aspergillus niger (Archer et al., et al., 1991; Krishnan and Vijayalakshimi, 1985; Mattern 1992). Proteolytic degradation aﬀects mainly heterologous et al., 1992; van den Hombergh et al., 1994). A ﬁfth pro- proteins evidenced by improvement of heterologous protease gene, encoding an extracellular subtilisin-type serine protease, PEPD, has been cloned based on conserved * Corresponding author. Fax: +86 010 64807505. amino acid sequences within subtilisins (Jarai et al., E-mail address: [email protected] (A. Wang). 1994a). Furthermore, three proteases that are homologous

1087-1845/$ - see front matter Ó 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.fgb.2007.09.012 18 Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27 to yeast vacuolar proteases have been cloned from A. 2. Materials and methods niger; these are PEPE, a pepsin-type aspartyl endoprotease that is the homologue of the vacuolar pep4 gene 2.1. Strains and growth conditions product in yeast (Jarai et al., 1994b), PEPC, a subtilisin- type serine endoprotease (Frederick et al., 1993) and A. niger GICC2773 (Valkonen et al., 2003) was used for CPY, a serine carboxypeptidase (Yaver et al., 1995). Fol- the disruption experiments of the putative aspartic protease lowing the whole-genome sequencing of A. niger (Pel genes. This strain contains the disruption mutant of the et al., 2007), approximately 200 predicted proteases have pepA gene and integrated heterologous gene (lcc1 of been identified, with similar numbers also seen following theTrametes versicolor laccase gene) expressed as fusion whole-genome sequencing of Aspergillus nidulans (Gala- to the glucoamylase. Escherichia coli DH5a served as host gan et al., 2005), Aspergillus oryzae (Machida et al., for routine DNA manipulations (Sambrook et al., 1989). 2005) and Aspergillus fumigatus (Nierman et al., 2005). Aspergillus niger GICC2773 strains were maintained on Pepsin-like enzymes such as aspartic proteases are solid sporulation medium, GMP (2% glucose, 2% maltose members of the A1 family of peptidases (Rawlings extract, 0.1% peptone, 1% agar) supplemented with 0.3% et al., 2004). This family comprises proteins with a casein when necessary. Disruption mutants were grown three-dimensional structure close to that of pepsin. Gen- on GMP supplemented with 200 lg hygromycin B per erally, these enzymes form two domains with different ml. Seed cultures of the A. niger strains in 30 ml of S3Y2 amino acid sequences, but basically similar folds. The medium (3% soluble starch, 2% yeast extracts, 0.5% catalytic site of the pepsin-like enzymes is formed at KH2PO4, 0.5% corn powder) were incubated at 30 °C with the junction of the two domains. It contains two aspartic 200 rpm agitation in a rotary G10 incubator (New Bruns- acid residues, Asp32 and Asp215 (human pepsin number- wick Scientific, NJ) for 14 h. For laccase production study, ing), one in each domain (Blundell et al., 1998). In accor- A. niger strains were grown in 30 ml of modified Promosoy dance with the accepted mechanism of the pepsin-like medium (Ward et al., 2004), where the promosoy was enzyme function (James, 2004), the Asp215 has to be substituted with tryptic soy broth (Difco, Detroit, MI). charged, whereas Asp32 has to be protonated. A remark- able property of the catalytic center is its adaptation for 2.2. Protease identification and sequence analysis the action in a wide range of pH, from 4.0 to 7.0. Aspar- tic proteinases have been identified from Botrystis cinerea The predicted pepsin-like proteases were identified as (ten Have et al., 2004) and A. oryzae (Machida et al., described previously (Sims et al., 2004a,b). Briefly, files 2005). containing all the predicted open reading frames (ORFs) Berka et al. showed that deletion of the pepA gene from species of filamentous fungi with recently completed increased bovine prochymosin production by more than whole-genome sequences (Table 2) were downloaded 66% (Berka et al., 1990). Van den Hombergh et al. dis- from their respective databases and filtered for sequences rupted three different acid-protease genes in A. niger, of 100 or more amino acids beginning with a methionine pepA, pepB and pepE, to determine the contribution of start codon. These were then searched for putative aspar- each protease to the overall protease spectrum using tic protease domains (PF00026) using the protein families bovine serum albumin as a broad substrate. They con- and HMM database (Pfam) release 11.0 (Bateman et al., cluded that the PEPA and PEPB proteases appeared to 2004). The results were compared to known protease constitute 84% and 6% of the extracellular acidic proteo- families in the peptidase database, MEROPS release 6.4 lytic activity, respectively, whereas the PEPE protease is (Rawlings et al., 2004). The coding sequences for pre- responsible for 68% of the intracellular acidic protease dicted proteases were manually compared using the Clu- activity (van den Hombergh et al., 1997a). Moralejo stalW program (Thompson et al., 1994) with default et al. also showed that a defect in the pepA gene reduced settings. Additional sequences of genes from other spe- degradation of overexpressed thaumatin in A. niger cies that have previously been cloned and characterized (Moralejo et al., 2000). were added to demonstrate that the ORFs represent Here, we report the isolation and characterization of putative orthologs of known genes. The Treeview pro- four ‘new’ genes (pepAa, pepAb, pepAc and pepAd) gram was used to draw and visualize hierarchical trees encoding pepsin-like aspartic proteases from A. niger based upon amino acid sequence alignments (Page, and demonstrate that these genes are conserved within 1996). The SignalP (Nielsen et al., 1997) and ProP the Aspergillus genus. The proteins possess different char- (Duckert et al., 2004) programs were used to identify acteristics and may have distinct localizations and func- prepro signal sequences and propeptide cleavage sites, tions. Transcription of the pepAa gene during respectively. The big-PI (http://mendel.imp.univie.ac.at/ recombinant protein production was observed with our sat/gpi/fungi_server.html) fungal predictor was also used A. nidulans cDNA microarray (Sims et al., 2005). There- to identify likely GPI modification sites (Eisenhaber fore, we examined the effect of disrupting these proteases et al., 2004). The disulfide bonds were predicted using on the secretion of the homologous and the heterologous DIpro 2.0 (http://contact.ics.uci.edu/bridge.html)(Cheng proteins in A. niger. et al., 2006). Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27 19

2.3. Microarray analysis sequence of the pepAx genes of A. niger. At the 50 end of each primer, a restriction enzyme recognition site of HpaI Transcriptome analysis was performed as described pre- was introduced. The PCR was carried out with Pfu DNA viously (Sims et al., 2005, 2004b,c). Briefly, a pyrG- single- polymerase and consisted of one cycle at 94 °C for 4 min copy recombinant chymosin-producing A. nidulans strain and 30 cycles at 94 °C for 1 min, 55 °C for 1 min, 72 °C (Cullen et al., 1987) was compared to its parental strain for 2.5 min, and then a final extension at 72 °C for transformed with an empty vector. Shake flask cultures 10 min. Following the PCR, amplified DNA was cloned were grown on 500 ml of SCM media (Ward et al., 1990) into a pBluescript KS(+) vector to yield the plasmid in 2 l flasks. BLASTx was used to identify ESTs (expressed pBS-TMx. A 1.5 kb BamHI/SalI fragment from pMW1, sequence tags) representing predicted cDNA (ORF) containing the PpcbC-hph expression cassette, was filled sequences from the A. nidulans genome sequence in with Klenow fragment and then inserted into the (www.broad.mit.edu). Differential gene expression was cal- intended disruption site of pBS-TMx to construct the final culated after normalisation in MaxDView (www.bio- disruption plasmid pAxS-T. inf.man.ac.uk/microarray/) and significance values determined as described previously (Sims et al., 2004c). 2.5. Transformation of A. niger

2.4. Plasmid construction Protoplasts of A. niger strains were prepared with refer- ence to Campbell et al. (1989) and transformed by the poly- For construction of gene disruption plasmid, the four ethylene glycol method (Yelton et al., 1984). For the pepAx genes (x represents a, b, c, or d) were ampliﬁed from disruption of the pepAx genes in A. niger (strain genomic DNA of A. niger using primers PAx1 and PAx2 GICC2773), the plasmid pAxS-T was digested with HpaI. listed in Table 1, which were designed according to the The resulting DNA disruption fragment and the 3.0 kb

Table 1 List of primers used in the disruption experimentsa 50 Primer 30 Primer 0 0 0 0 PAa1 5 -GCACTTCTTTCCCCTTTTTGTTTAC-3 PAa2 5 -AGGTTAACTTGAATTGTAGATACAGCCAC-3 0 0 0 0 PAb1 5 -ACGTTAACCATATCACAGCTATATCCCC-3 PAb2 5 -ACGTTAACGCCAGGTCCTCCTTCTGC-3 0 0 0 0 PAc1 5 -TGGTTAACGAGGGATTGCTCTATTG-3 PAc2 5 -TGGTTAACTGTGCTATGCTATTGGTG-3 0 0 0 0 PAd1 5 -TGGTTAACTCGTAAGTAGGTAGGCTG-3 PAd2 5 -ATGTTAACCCGAGGTGCTGCTTG-3 0 0 Phph 5 -GAGGGCAAAGGAATAGAGTAGATG-3 0 0 PoutAa n.d. PoutAa 5 -TCATGGATTAGGGTTAGAAAGAGTG-3 0 0 PoutAb 5 -GGAGAGATAGGACGTAAACTTCATG-3 PoutAb n.d. 0 0 PoutAc n.d. PoutAc 5 -TCATGTGAATATGTGCACGAAGGTG-3 0 0 PoutAd n.d. PoutAd 5 -AGAGCAGAGAAGAAATACTGAGGAG-3 0 0 0 0 PAa3 5 -TCCTCCAGTCCCTCATTGTTGCC-3 PAa4 5 -AACTCGTACTCGCCCACTCCGTC-3 PAb3 Same as PAb1 PAb4 Same as PAb2 0 0 PAc3 Same as PAc1 PAc4 5 -TCTGCTCGTCGGTGGTTGTG-3 PAd3 Same as PAd1 PAd4 Same as PAd2 a The introduced restriction sites of HpaI are indicated in bold.

Table 2 Predicted aspartic proteases in the species of filamentous ascomycetes Species No. predicted Sequenced by Website ORFs Aspergillus fumigatus 7 Consortium (Nierman et al., 2005) http://www.cadre.man.ac.uk/ 8 NRRL 3—Integrated genomics/genencor http://www.jgi.doe.gov/ Aspergillus niger 8 ATCC 1015—Joint Genome Institute http://www.jgi.doe.gov/ 8 CBS 513.88—DSM (Pel et al., 2007) http://www.jgi.doe.gov/ Aspegillus oryzae 12 National Institute of Technology and Evaluation (Machida http://www.bio.nite.go.jp/ngac/e/rib40- et al., 2005) e.html Aspergillus nidulans 9 Broad Institute (Galagan et al., 2005) http://www.broad.mit.edu/ Magnaporthe grisea 16 Broad Institute http://www.broad.mit.edu/ Neurospora crassa 16 Broad Institute http://www.broad.mit.edu/ Fusarium 15 Broad Institute http://www.broad.mit.edu/ graninearum Trichoderma reesei 16 Joint Genome Institute http://www.jgi.doe.gov/ Open-reading frames were found to contain the Pfam domain PF00026, suggesting that they are members of the peptidase A1 family (pepsin family). 20 Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27 vector fragment were used to transform the A. niger strain 0.1 M HOAc buffer at pH 4.6) as substrate and incubated GICC2773. Hygromycin B resistant transformants were at 37 °C for 30 min. The reaction was stopped by boiling. selected in GMP with 0.8 M sucrose as osmotic stabilizer The glucose liberated was quantified by the glucose oxi- supplemented with 200 lg/ml hygromycin B. dase/peroxidase method using a glucose assay kit (Zhongsheng Biotechnology and Science Inc. product). 2.6. DNA isolation The optical density at 525 nm was measured and the amount of glucose produced was calculated. The expres- Genomic DNA of the A. niger strains was obtained by a sion level of total glucogenic enzymes was indicated in modification of the benzyl chloride extraction method (Zhu the units of total enzymes activity secreted per gram dry et al., 1993) as follows: the fungi were grown in S3Y2 medium for 16 h at 30 °C and harvested by filtration on Nytal 30 filters. To each sample, 500 ll extraction buffer (100 mM Tris–HCl, pH 9.0, 40 mM EDTA), 100 ll 10% SDS and 300 ll benzyl chloride were added. The tube was vortexed and incubated in 50 °C for 1 h with shaking or repeatedly vortexing at 10 min interval to keep the two phases thor- oughly mixed. Then 300 ll 3 M NaOAc, pH 5.0, was added to the tube and kept on ice for 15 min. After centrifugation at 12,000 rpm, 4 °C for 10 min, the supernatant was collected, and DNA was precipitated with isopropanol.

2.7. PCR screening of transformants for homologous recombination

The organization of the targeted gene and its neighbour- ing regions were examined by PCR for the disruption of pepAx genes using oligonucleotide primer pairs as follows: (i) primers annealing to the hph gene and the sequence outside homologous fragment: Phph and PoutAx (see Table 1); (ii) primers annealed outside the disrupted site of pepAx genes: PAx3 and PAx4 (see Table 1). PCR consisted of one cycle at 94 °C for 4 min and 30 cycles at 94 °C for 1 min, 55 °C for 1 min, 72 °C for 2.5 min, and then a ﬁnal extension at 72 °C for 10 min.

2.8. Laccase expression assay

Aspergillus niger strains GICC2773, DpepAa, DpepAb and DpepAd were grown in modified Promosoy medium at 28 °C and 200 rpm for 144 h. Culture filtrate and mycelium were collected by centrifugation. For laccase assay a 30 ll amount of appropriately diluted sample solution was mixed with 2.9 ml 0.1 M HOAc buffer (pH 4.6) and 70 ll 20 mM ABTS substrate and incubated at 37 °C for 30 min. The optical density at 420 nm was measured. One unit of laccase activity is defined as the amount of enzyme oxidizing 1 lmol of ABTS per min. For determination of mycelium dry weight the collected mycelium was washed, squeezed, blotted dry and dried at 80 °C for 3 h before weighing. The expression level of laccase was indicated in Fig. 1. Hierarchical clustering of sequence alignments of the ORFs the laccase activity (international units) produced per gram predicted to represent genes of the A1 aspartic protease family (pepsin dry weight of mycelium ( IU/g ). family) in the Aspergillus genus. Clustering performed using the ClustalW (Thompson et al., 1994) program with default settings. The tree was drawn 2.9. Assay of total glucogenic enzymes activity with Treeview program (Page, 1996) and is rooted with human pepsin A sequence. Previously isolated fungal aspartic proteases in the MEROPS database (Rawlings et al., 2004) are in bold. GenBank Accession numbers A60ll amount of culture filtrate collected above was are presented in brackets, Aspergillus hypothetical proteins are defined by mixed with 140 llH2O and 2.8 ml 2% soluble starch (in locus ID accession numbers, for details of genome sequences (see Table 2). Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27 21 weight of mycelium (U/g). One unit of glucogenic enzyme ome sequencing (Sims et al., 2004a), a number of putative activity was defined as the amount of enzymes that liber- novel pepsin-like aspartic proteases were identified. Within ated 1 lmol glucose per min from soluble starch. the filamentous ascomycetes, between 7 and 16 open-reading frames for each species were predicted to represent 3. Results aspartic proteases based upon their conserved sequence to the Pfam domain PF00026 (Bateman et al., 2004) for 3.1. Identification and characterization of four pepsin-like members of the peptidase A1 (pepsin) family of proteases proteases (Table 2). The Eurotiales class (including three species of Aspergillus) contained approximately half as many pre- In a survey of all the predicted proteases encoded by 22 dicted aspartic proteases as the Pezizales and Sordariales species of fungi that have recently undergone whole-gen- classes (including Magnaporthe grisea, Neurospora crassa,

Fig. 2. Alignment of the predicted amino acid sequences of the four A. niger aspartic proteases (PEPAa–PEPAd) with that of aspergillopepsin A (PEPA) using ClustalW (Thompson et al., 1994). Signal peptides for all sequences (except PEPAb) are boldfaced. The possible KexB cleavage sites found in pepAa (RR), pepAc (KR) and pepAd (KR) were marked with an arrow. Hallmark motifs in aspartic proteases are indicated by black shading blocks. The active site motifs were shown in italics. A serine-rich region at the C-terminus was boxed. The predicted GPI-modiﬁcation site was double underlined. Locations of the cysteine residues for the predicted disulﬁde bonds are shown with grey shading blocks. The insertion sites for PEPAa, PEPAb and PEPAd were shown in ‘‘.’’. 22 Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27

aspartyl protease 4 from Coccidioides posadasii (ABA54909, unpublished). The pepAa orthologs have an extra A. oryzae one, as previously demonstrated for the pepA and pepE ‘clusters’ (Machida et al., 2005). The pepAx genes encode proteins of 394, 426, 453 and 480 amino acids, respectively, which are in agreement with other aspartic proteinases of fungal origin. PEPAa, PEPAc and PEPAd appear to contain a signal peptide (boldfaced in Fig. 2). Possible KexB cleavage sites (Arginine and Lysine) were detected in PEPAa, PEPAc and PEPAd (see Fig. 2). Alignment of the predicted amino acid sequences of the four A. niger proteins (PEPAx) with the previously isolated aspartic protease, aspergillopepsin A (PEPA) dem- onstrates that functional regions of these proteins are highly conserved (Fig. 2), such as the active site motif Asp101-Thr102-Gly103 of aspergillopepsin A number- Fig. 3. The A. nidulans putative ortholog (AN2517.1) of the A. niger pepAa pepsin-like protease is significantly differentially expressed during ing in Fig. 2 (James, 2004). Asp residue (Asp156 in asperg- increasing heterologous chymosin secretion. Expression in chymosin- illopepsin A) was not conserved since the PEPAc (see producing strain is relative to the parental strain. Error bars are equal to Fig. 2) has a Glu residue at this position as seen in other one standard deviation of the mean (three repeats) and the values above aspartic protease (Capasso et al., 1998). PEPAd has a long and below the dotted lines were previously shown to represent significant C-terminal extension consisting of about 70 residues. This changes in gene expression (Sims et al., 2005). extension contains lengthy stretches of hydrophilic, pre- dominantly serine residues, terminating with roughly 20 Fusarium graninearum and Trichoderma reesei). Eight pre- hydrophobic residues, which may facilitate extracellular dicted ORFs from the NRRL3 strain of A. niger sequenced attachment. An algorithm for identifying fungal glycosyl- by Genencor/Integrated were found to contain the Pfam phosphatidylinositol (GPI) modification motifs (Eisenhab- domain PF00026 for members of the peptidase A1 (pep- er et al., 2004) predicted that GPI modification in PEPAd sin). All eight ORFs with 100% amino acid sequence occurs at Gly456 with high probability scores of S >15 matches were found in the ‘best protein models’ download (see Fig. 2). A disulfide bond predictor (DIpro 2.0) (Cheng of the whole-genome sequencing of the ATCC 1015 strain et al., 2006) suggested that PEPAa had one disulfide bond of A. niger by the Joint Genome Institute. Eight predicted between residues 382–389, in the same region as one pre- genes were also identified to contain this Pfam domain in dicted for PEPA. In addition to conserved predicted disul- the whole-genome sequencing of the CBS 513.88 strain of fide bonds in this region (between residues 365–402 and A. niger by DSM (Pel et al., 2007). Four of these genes 339–370, respectively), PEPAc and PEPAd were also pre- (referred to from here on as pepAa, pepAb, pepAc and dicted to have an N-terminal disulfide bond (between resi- pepAd) encode pepsin-like proteases. Putative orthologs dues 130–144 and 137–142, respectively), however, PEPAb for all four genes were also identified in A. nidulans, A. ory- was predicted to have none (see Fig. 2). zae and A. fumigatus, suggesting that these genes are well conserved within the Aspergillus genus (Fig. 1). However, 3.2. Microarray results the sequence of A. nidulans AN6686.1 appears to be dis- tinctly different from the putative A. niger and A. fumigatus One of the unknown transcripts found to be significantly pepAb orthologs. It may be due to incorrect gene designa- up-regulated during recombinant chymosin secretion (25– tion which has been a problem with the A. nidulans whole- 50 h) in A. nidulans shake flask cultures (Sims et al., genome sequence (Sims et al., 2005). Comparison of the 2005) was putatively identified by sequence similarity to predicted aspartic protease sequences listed in the peptidase represent a pepsin-like peptidase (Fig. 3). BLASTx was database (MEROPS) using ClustalW (Thompson et al., used to establish that this transcript represents the hypo- 1994) demonstrated that the predicted pepAx sequences thetical ORF AN2157.1, which appears to encode the A. were likely to be orthologs of previously isolated genes nidulans ortholog of the PEPAa protease (Fig. 1). from different species of filamentous fungi. The first, denoted PEPAa protease seems to represent an ortholog 3.3. Disruption of pepAa, pepAb, pepAc and pepAd genes of the pepsin-type aspartic protease previously identified from Talaromyces emersonii (AF439995, unpublished) We disrupted pepAa, pepAb and pepAd genes by homol- and Scleotinia sclerotiorum (Poussereau et al., 2001). Fur- ogous recombination. Fig. 4a shows disruption strategies ther conserved sequence, PEPAb and PEPAc appear to for construction of the disruption plasmids and for identi- be orthologs of aspartic proteases (BcAP1 and BcAP5) pre- fication of disruption strains. In all three cases, the corre- viously identified from Botryotinia fuckeliana (ten Have sponding genes were amplified by PCR (data not shown) et al., 2004). The closest known ortholog to PEPAd is and the hygromycin resistant gene was inserted into the Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27 23 gene at a restriction site so the hygromycin resistant gene respectively (Section 2). Screening of fungal transformants 0 was flanked by homologous AB fragment at the 5 end by PCR using the primer pair of Phph and PoutAx (Table 1) and homologous BC fragment at the 30 end (Fig. 4a). identified transformant #23 as a disruption strain for the Aspergillus. niger GICC2773 protoplasts were transformed pepAa gene (Fig. 4b), transformant #17 as a disruption with HpaI fragment of pAaS-T, pAbS-T and pAdS-T, strain for the pepAb gene (Fig. 4c) and transformant #29

Fig. 4. (a) Disruption strategy for the pepAx (pepAa, pepAb, pepAc and pepAd) gene of A. niger. The expected pattern of PCR products is indicated for the parental strain (fragment I) and for the disruption mutants (fragment II) using the indicated primer pairs of PAx3 and PAx4. No PCR product is expected for the parental strain. Fragment III is expected for the disruption mutants using the indicated primer pairs of Phph and PoutAb. Fragment IV is expected for the disruption mutants using the indicated primer pairs of Phph and PoutAa,PoutAc and PoutAd. (b) PCR analysis and identiﬁcation of the pepAa disruption strain. From left to right, lane 1 is the DNA molecule weight marker, lanes 2 and 3 used template DNA from the parental strain (GICC2773), lanes 4 and 5 used template from a transformant with ectopic integration (AN#6), and lanes 6 and 7 are the disruption mutant (A. NDpepAa #23). The

PCR primers used for lanes 2, 4 and 6 are PAa3 and PAa4. The PCR primers used for lanes 3, 5 and 7 are Phph and PoutAa. (c) PCR analysis and identiﬁcation of the pepAb disruption strain. From left to right, lane 1 is the DNA molecule weight marker, lanes 2 and 3 used template DNA from the parental strain (GICC2773), lanes 4 and 5 used template from a transformant with ectopic integration (AN#1), and lanes 6 and 7 are the disruption mutant

(A.NDpepAb#17). The PCR primers used for lanes 2, 4 and 6 are PAb3 and PAb4. The PCR primers used for lanes 3, 5 and 7 are Phph and PoutAb. (d) PCR analysis and identification of the pepAd disruption strain. From left to right, lane 1 is the DNA molecule weight marker, lanes 2 and 3 used template DNA from the parental strain (GICC2773), lanes 4 and 5 used template from a transformant with ectopic integration (AN#8), and lanes 6 and 7 are the disruption mutant (A.NDpepAa#29). The PCR primers used for lanes 2, 4 and 6 are PAd3 and PAd4. The PCR primers used for lanes 3, 5 and 7 are Phph and PoutAd. 24 Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27 as a disruption strain for the pepAd gene (Fig. 4d). The spe- within the Aspergillus genus. Both the sequence align- cific bands (1.14 kb PCR product from the pepAa disrup- ment/clustering program, ClustalW and Pfam/MEROPS tion strain, 1.19 kb PCR product from the pepAb classification system are dependent upon the level of disruption strain and 1.08 kb PCR product from the pepAd sequence similarity. Therefore, the clusters that are gener- disruption strain) resulted from homologous recombina- ated correspond to discrete members of the same MER- tion were detected only in the disruption strains, not in OPS and Pfam families. At the most conserved level, the wild type strain (GICC2773) or the epitopic integration within individual protein families, clustering by sequence strain (AN#6 of Fig. 4b, AN#1 of Fig. 4c and AN#8 of similarity can identify orthologs and paralogs (Sims Fig. 4d). The specific PCR fragments were also sequenced et al., 2004a). Whole-genome sequencing of closely related to confirm their identity (data not shown). In addition, species has resulted in the identification of predicted prote- using primer pair of PAx3 and PAx4 (Table 1), the amplified ases that represent orthologs of previously isolated genes PCR fragments showed differences of 1.5 kb when compar- from different species. Using this approach, it is also possi- ing the disruption strains to the wild type strain or the epi- ble to identify orthologs which have not been isolated pre- topic integration strain. In the pepAa disruption strain a viously and may also lead to the re-classification of some 2.32 kb PCR fragment was detected compared to a proteases. For example, A. niger PEPA is denoted 0.82 kb fragment in the wild type strain (Fig. 4b). Similarly, A01.016 in MEROPS peptidase database (Rawlings et al., the pepAb disruption strain had a 3.67 kb rather than a 2004) and yet it is sandwiched between A01.026 peptidase 2.17 kb PCR fragment (Fig. 4c) and the pepAd disruption F from A. oryzae and A. fumigatus (see Fig. 1). strain had a 3.66 kb rather than the 2.16 kb PCR fragment The Aspergillus PEPAd sequences appear to have a con- in wild type strain (Fig. 4d). In an attempt to disrupt the served C-terminal modification that may facilitate their pepAc gene in A. niger GICC2773 strain, we generated association with the cell membrane by means of GPI 286 transformants in four experiments and screened all anchor (Hamada et al., 1998). The Saccharomyces cerevisi- transformants by PCR with primers Phph and PoutAc. How- ae aspartic proteases yapsin 1, 2 and 3 have all been shown ever, no transformant could be identified containing the to be GPI-anchored proteins (Ash et al., 1995; Cawley pepAc disruption pattern of a 1520 bp specific band result- et al., 1995) and B. cinerea BcAP3 and 4 contains similar ing from homologous recombination. domain (ten Have et al., 2004). Moreover, PEPAd contains serine-rich stretches just upstream of the GPI modification 3.4. Effect of gene disruptions on laccase production site (Fig. 2). These stretches might be targets for O-glyco- sylation, which may facilitate adherence to the extracellular We tested laccase expression of all disruption strains by glucan matrix of A. niger. An aspartic protease that is growing them in shake flasks for 144 h in modified Promo- attached to the cell membrane by a GPI anchor, or embed- soy medium and then assaying the laccase activity in the ded in the hyphal matrix, might support various functions supernatants. We observed a 42% average increase of the such as the maturation of other fungal hydrolytic enzymes, laccase production in the strain A.NDpepAb#17 with the proteolysis of host cell wall protein in the vicinity of the respect to the parental strain GICC2773 in two parallel hyphal tip. experiments. Analysis of secretion level of the laccase also Perhaps the most distinctive features of these four revealed that a 21% average increase of the laccase expres- A. niger sequences are those of PEPAb. While it does contain sion in the strain A.NDpepAa#23 and a 30% average all the residues necessary to function as an active proteo- increase in the strain A.NDpepAd#29 (Table 3). lytic enzyme, the sequence of PEPAb is highly atypical; it is not predicted to contain any disulfide bonds. In particu- 3.5. Effect of gene disruptions on production of native protein lar, the C-terminal disulfide bond conserved in virtually all other eukaryotic aspartic protease sequences is absent, the The secretion of total glucogenic enzymes in culture fil- precursors are commonly synthesized on membrane-bound trate was analyzed in the three disruption mutants. The ribosomes and the nascent polypeptide chains are directed strain A.NDpepAa#23 and A.NDpepAb#17 showed signifi- into the endoplasmic reticulum where the disulfide bonds cant increase in the expression level of their glucogenic are introduced by post-translational modification to enzymes by approximately 19% and 37%, respectively. enhance stability. Predicted orthologs of PEPAb from a However, there was a only slight increase (5.2%) in the number of filamentous fungal species were also predicted strain containing the disruption of the pepAd gene (Table 3). to lack disulfide bonds and a conventional signal peptide. These enzymes have sequences with very acidic isoelectric 4. Discussion points; the predicted pI value for the putative N. crassa ortholog of PEPAb is almost as acidic as that of pig pepsin Four pepsin-like protease genes were identified and which operates in a very acidic environment (pH < 2) in the characterized from A. niger, the products of which have stomach. Thus, whatever the mechanism of trafficking of characteristics that are novel among aspartic proteases these proteins, the cellular compartment in which they from filamentous ascomycetes. Sequence analysis demon- finally reside can be predicted to be very acidic, otherwise strated that these genes appear to be highly conserved each polypeptide would unfold and denature. Although a Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27 25 signal peptide sequence could not be identified for PEPAb, The effect of silencing pepAa, pepAb and pepAd expres- the absence of a ‘typical’ signal peptide sequence in a sion on homologous and heterologous protein secretion secreted protein is not unprecedented. The superoxide dis- in A. niger was analyzed. The laccase activities were mutase BcSOD1 in B. cinerea also lacks a signal peptide increased by 21%, 42% and 30%, respectively, in the pepAa, sequence but was unequivocally demonstrated to be a pepAb and pepAd gene disruption strains. The effect of the secreted protein (Rolke et al., 2004). pepAb gene disruption, despite the lack of a signal peptide Real-time quantitative RT-PCR was performed to and disulfide bonds, suggesting the PEPAb is likely to be establish whether the pepAx genes are stimulated by heter- another important protease leading to low production of ologous protein production. All four aspartic protease heterologous proteins, though the substrate specificity of genes were transcribed in the laccase expressing strain different proteases should be considered. The increased (Fig. 5), although there was no significant difference in production of laccase by disruption of the pepAd gene, is the level of gene expression compared with parental strain relatively small compared with that obtained from pepAb (data not shown), suggesting that transcription of the pro- removal, which could support the hypothesis deduced from teases does not directly respond to the expression of heter- the sequence characterization that PEPAd is likely to be a ologous laccase. Perhaps these proteins are constitutively GPI-anchored protease. The proteolytic effect of GPI- expressed under given certain culture conditions or only anchored or membrane-embedded protease may be limited, induced by certain heterologous proteins. The expression and may only degrade heterologous proteins in the vicinity of a transcript representing the A. nidulans ortholog of pep- of the cell wall as compared to extracellular proteases. Aa was observed to up-regulated in a chymosin-producing The effect of disruptions of these three proteases on strain compared to its parent (Fig. 3). homologous proteins production was also studied. In comparison with the parental strain, there was no significant change in glucogenic enzymes production in the pepAd disruption mutant which could represent a protective mechanism to prevent native proteins from degradation by nascent proteases (Calmels et al., 1991). However, in the pepsins and the majority of other members of the family show specificity for the cleavage of bonds in peptides of at least six residues with hydrophobic amino acids in both the Pl and Pl0 positions. So the proteolytic activity of the same protease to different heterologous proteins will vary and the relative degradation ability compared with other aspartic proteases needs to be examined with a number of different substrates due to the substrate specificities for enzymes. To disrupt the pepAc gene in A. niger, a total of 286 transformants were generated in four experiments. All transformants were screened by PCR, but none was identified to contain the pepAc gene knockout. It is not known why no disruptant was obtained, However, a number of pepsinogen genes are reported to be located in the centro- Fig. 5. Detection of pepAx transcripts by RT-PCR. 215, 100, 236 and meric region of human chromosome 11 (Taggart et al., 176 bp amplified products were detected for pepAa, pepAb, pepAc and 1985), where homologous recombination in such area in pepAd genes, respectively. RNA was extracted from GICC2773 strain. A. niger is so infrequent.

Table 3 Extracellular laccase and glucogenic enzyme activity secreted by protease-deﬁcient and wild type strains Strain Mycelium dry Laccase Glucogenic enzymes a,c weight (g/L) a,b,c a,b,c Activity Activity relative to Activity relative to Activity Activity relative to Activity relative to (IU/L) mycelium dry weight parental strain (%) (U/L) mycelium dry parental strain (%) (IU/g) weight (U/g) DpepAa#23 8.48 ± 0.36 479 ± 20.7 56.5 ± 1.22 121.0 7.96 ± 0.20 954.0 ± 70.2 118.7 DpepAb#17 8.37 ± 0.40 555 ± 31.0 66.3 ± 3.01 142.0 8.07 ± 0.28 119.2 ± 77.6 137.0 DpepAd#29 8.12 ± 0.30 491 ± 17.8 60.7 ± 1.76 130.0 6.98 ± 0.23 842.0 ± 66.9 105.2 GICC2773 10.22 ± 0.78 476 ± 10.7 46.7 ± 2.72 100 7.74 ± 0.13 800.6 ± 15.2 100 a Data are averages ± standard deviation of eight determinations (four replicates for two independent experiments). b The supernatant was used as crude enzyme for the enzyme activity assay. c The cultivation was carried out for 144 h. For all cultures the ﬁnal pH was approximately 4.6. 26 Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27

Residual proteolytic activity (other than that accredited tion of a novel aspartic proteinase from Antarctic fish. Biochim. to identified proteases) remains a big challenge for the pro- Biophys. Acta 1387, 457–461. duction of heterologous protein such as laccase (Archer Cawley, N.X., Wong, M., Pu, L.P., Tam, W., Loh, Y.P., 1995. Secretion of yeast aspartic protease 3 is regulated by its carboxy-terminal tail: et al., 1992; Conesa et al., 2001). The roles of these prote- characterization of secreted YAP3p. Biochemistry 34, 7430–7437. ases and their contribution to laccase degradation are Cheng, J., Saigo, H., Baldi, P., 2006. Large-scale prediction of disulphide unclear at this time. Some of them might have intracellular bridges using kernel methods, two-dimensional recursive neural proteolytic activities required for adequate protein process- networks, and weighted graph matching. Proteins 62 (3), 617–629. ing and thus might be indispensable. On the other hand, we Conesa, A., Punt, P.J., van Luijk, M., van den Hondel, C.A., 2001. The secretion pathway in filamentous fungi: a biotechnological view. do not know the substrate specificity for each of these pro- Fungal Genet. Biol. 33, 155–171. teases, especially for the laccase molecule. Knowing the Cullen, D., Gray, G.L., Wilson, L.J., Hayenga, K.J., Lamsa, M.H., Rey, specificities of the proteases will undoubtedly broaden M.W., Norton, S., Berka, R.A., 1987. Controlled expression and our methods to improve strains for heterologous protein secretion of bovine chymosin in Aspergillus nidulans. Biotechnology 5, production by selectively disrupting one or more genes. 369–376. Duckert, P., Brunak, S., Blom, N., 2004. Prediction of proprotein In summary, removal of proteolytic degradation due to convertase cleavage sites. Protein Eng. Des. Sel. 17, 107–112. the presence of PEPAx is a useful method for increasing Eisenhaber, B., Schneider, G., Wildpaner, M., Eisenhaber, F., 2004. A laccase production in A. niger. This strategy, in association sensitive predictor for potential GPI lipid modification sites in fungal with the removal of aspergillopepsin A (van den Hombergh protein sequences and its application to genome-wide studies for et al., 1997a) and other aspartic proteases by combining Aspergillus nidulans, Candida albicans, Neurospora crassa, Saccharo- myces cerevisiae, and Schizosaccharomyces pombe. J. Mol. Biol. 337, multiple gene disruptions, by classical mutagenesis and 243–253. by physiological optimization of culture medium, repre- Frederick, G.D., Rombouts, P., Buxton, F.P., 1993. Cloning and sents a good method to result in large increases in heterol- characterisation of pepC, a gene encoding a serine protease from ogous protein production in A. niger. In future, it would be Aspergillus niger. Gene 125, 57–64. interesting to look at the effects of disrupting multiple pep- Galagan, J.E., Calvo, S.E., Cuomo, C., Ma, L.J., Wortman, J.R., Batzoglou, S., Lee, S.I., Basturkmen, M., Spevak, C.C., Clutterbuck, Ax genes or generating strains that silence all proteases that J., Kapitonov, V., Jurka, J., Scazzocchio, C., Farman, M., Butler, J., are known to affect yields of recombinant proteins. Purcell, S., Harris, S., Braus, G.H., Draht, O., Busch, S., D’Enfert, C., Bouchier, C., Goldman, G.H., Bell-Pedersen, D., Griffiths-Jones, S., Acknowledgments Doonan, J.H., Yu, J., Vienken, K., Pain, A., Freitag, M., Selker, E.U., Archer, D.B., Penalva, M.A., Oakley, B.R., Momany, M., Tanaka, T., Kumagai, T., Asai, K., Machida, M., Nierman, W.C., Denning, D.W., This work was supported by grants from Genencor, a Caddick, M., Hynes, M., Paoletti, M., Fischer, R., Miller, B., Dyer, P., Danisco Division, Palo Alto, California, USA. Authors Sachs, M.S., Osmani, S.A., Birren, B.W., 2005. Sequencing of would like to thank Michael Ward for his critical reading. Aspergillus nidulans and comparative analysis with A. fumigatus and A. oryzae. Nature 438 (7071), 1105–1115. Hamada, K., Fukuchi, S., Arisawa, M., Baba, M., Kitada, K., 1998. References Screening for glycosylphosphatidylinositol (GPI)-dependent cell wall proteins in Saccharomyces cerevisiae. Mol. Gen. Genet. 258, 53–59. Archer, D.B., MacKenzie, D.A., Jeenes, D.J., Roberts, I.N., 1992. Inoue, H., Kimura, T., Makabe, O., Takahashi, K., 1991. The gene and Proteolytic degradation of heterologous proteins expressed in Asper- deduced protein sequences of the zymogen of Aspergillus niger acid gillus niger. Biotechnol. Lett. 14, 357–362. proteinase A. J. Biol. Chem. 266, 19484–19489. Ash, J., Dominguez, M., Bergeron, J.J., Thomas, D.Y., Bourbonnais, Y., James, M.N.G., 2004. Catalytic pathway of aspartic peptidases. In: 1995. The yeast proprotein convertase encoded by yap3 is a Barrett, A.J., Rawlings, N.D., Woessner, J.F. (Eds.), Handbook of glycophosphatidylinositol-anchored protein that localizes to the Proteolytic Enzymes. Elsevier, London, pp. 12–19. plasma membrane. J. Biol. Chem. 270, 20847–20854. Jarai, G., Kirchherr, D., Buxton, F.P., 1994a. Cloning and characteriza- Bateman, A., Coin, L., Durbin, R., Finn, R.D., Hollich, V., Griffiths- tion of the pepD gene of Aspergillus niger which codes for a subtilisin- Jones, S., Khanna, A., Marshall, M., Moxon, S., Sonnhammer, E.L., like protease. Gene 139, 51–57. Studholme, D.J., Yeats, C., Eddy, S.R., 2004. The Pfam protein Jarai, G., Van Den Hombergh, J.P.T.W., Buxton, F.P., 1994b. Cloning and families database. Nucleic Acids Res. 32, D138–D141. characterization of the pepE gene of Aspergillus niger encoding a new Berka, R.M., Ward, M., Wilson, L.J., Hayenga, K.J., Kodama, K.H., aspartic protease and regulation of pepE and pepC. Gene 145, 171–178. Carlomagno, L.P., Thompson, S.A., 1990. Molecular cloning and Krishnan, S., Vijayalakshimi, M.A., 1985. Purification of an acid protease deletion of the gene encoding aspergillopepsin A from A. awamori. and a serine carboxypeptidase from Aspergillus niger using metal- Gene 86, 153–162. chelate affinity chromatography. J. Chromatogr. 329, 165–170. Blundell, T.L., Guruprassas, K., Albert, A., Williams, M., Sibanda, B.L., Machida, M., Asai, K., Sano, M., Tanaka, T., Kumagai, T., Terai, G., Dhanarai, V., 1998. The aspartic proteinases. an historical overview. Kusumoto, K., Arima, T., Akita, O., Kashiwagi, Y., Abe, K., Gomi, K., Adv. Exp. Med. Biol. 436, 1–13. Horiuchi, H., Kitamoto, K., Kobayashi,T.,Takeuchi,M.,Denning,D.W., Calmels, T.P.G., Martin, F., Durand, H., Tiraby, G., 1991. Proteolytic Galagan, J.E., Nierman, W.C., Yu, J., Archer, D.B., Bennett, J.W., events in the processing of secreted proteins in fungi. J. Biotechnol. 17, Bhatnagar, D., Cleveland, T.E., Fedorova, N.D., Gotoh, O., Horikawa, 51–66. H., Hosoyama, A., Ichinomiya, M., Igarashi, R., Iwashita, K., Juvvadi, Campbell, E.I., Unkles, S.E., Macro, J.A., van den Hondel, C., Contreras, P.R., Kato, M., Kato, Y., Kin, T., Kokubun, A., Maeda, H., Maeyama, N., R., Kinghorn, J.R., 1989. Improved transformation efficiency of Maruyama, J., Nagasaki, H., Nakajima, T., Oda, K., Okada, K., Paulsen, Aspergillus niger using the homologous niaD gene for nitrate reductase. I., Sakamoto, K., Sawano, T., Takahashi, M., Takase, K., Terabayashi, Y., Curr. Genet. 16, 53–56. Wortman, J.R., Yamada, O., Yamagata, Y., Anazawa, H., Hata, Y., Capasso, C., Riggo, M., Scudiero, R., Carginale, V., di Prisco, G., Kay, J., Koide,Y.,Komori,T.,Koyama,Y.,Minetoki, T., Suharnan, S., Tanaka, Kille, P., Parisi, E., 1998. Molecular cloning and sequence determina- A., Isono, K., Kuhara, S., Ogasawara, N., Kikuchi, H., 2005. Genome Y. Wang et al. / Fungal Genetics and Biology 45 (2008) 17–27 27

sequencing and analysis of Aspergillus oryzae. Nature 438 (7071), French bean, whereas a glucose oxidase (BcGOD1) is dispensable. 1157–1161. Mol. Plant Pathol. 5, 17–28. Mattern, I.E., van Noort, J.M., van den, B.P., Archer, D.B., Roberts, Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A I.N., van den Hondel, C.A., 1992. Isolation and characterization of Laboratory Manual. Cold Spring Harbor Laboratory Press, New mutants of Aspergillus niger deficient in extracellular proteases. Mol. York. Gen. Genet. 234, 332–336. Sims, A.H., Dunn-Coleman, N.S., Robson, G.D., Oliver, S.G., 2004a. Moralejo, F.J., Cardoza, R.E., Gutierrez, S., Sisniega, H., Faus, I., Glutamic protease distribution is limited to filamentous fungi. FEMS Martıń, J.F., 2000. Overexpression and lack of degradation of Microbiol. Lett. 239, 95–101. thaumatin in an aspergillopepsin A-defective mutant of Aspergillus Sims, A.H., Gent, M.E., Lanthaler, K., Dunn-Coleman, N.S., Oliver, awamori containing an insertion in the pepA gene. Appl. Microbiol. S.G., Robson, G.D., 2005. Transcriptome analysis of recombinant Biotechnol. 54, 772–777. protein secretion by Aspergillus nidulans and the unfolded protein Nielsen, H., Engelbrecht, J., Brunak, S., Heijne, G., 1997. Identification of response in vivo. Appl. Environ. Microbiol. 71, 2737–2747. prokaryotic and eukaryotic signal peptides and prediction of their Sims, A.H., Gent, M.E., Robson, G.D., Dunn-Coleman, N.S., Oliver, cleavage sites. Protein Eng. 10, 1–6. S.G., 2004b. Combining transcriptome data with genomic and cDNA Nierman, W.C., Pain, A., Anderson, M.J., Wortman, J.R., Kim, H.S., sequence alignments to make confident functional assignments for Arroyo, J., Berriman, M., Abe, K., Archer, D.B., Bermejo, C., Aspergillus nidulans genes. Mycol. Res. 108, 853–857. Bennett, J., Bowyer, P., Chen, D., Collins, M., Coulsen, R., Davies, R., Sims, A.H., Robson, G.D., Hoyle, D.C., Oliver, S.G., Turner, G., Prade, Dyer, P.S., Farman, M., Fedorova, N., Fedorova, N., Feldblyum, R.A., Russell, H.H., Dunn-Coleman, N.S., Gent, M.E., 2004c. Use of T.V., Fischer, R., Fosker, N., Fraser, A., Garcia, J.L., Garcia, M.J., expressed sequence tag analysis and cDNA microarrays of the Goble, A., Goldman, G.H., Gomi, K., Griffith-Jones, S., Gwilliam, R., filamentous fungus Aspergillus nidulans. Fungal Genet. Biol. 41, 199– Haas, B., Haas, H., Harris, D., Horiuchi, H., Huang, J., Humphray, 212. S., Jimenez, J., Keller, N., Khouri, H., Kitamoto, K., Kobayashi, T., Taggart, R.T., Mohandast, T.K., Showst, T.B., Bell, G.I., 1985. Variable Konzack, S., Kulkarni, R., Kumagai, T., Lafon, A., Latge, J.P., Li, numbers of pepsinogen genes are located in the centromeric region of W., Lord, A., Lu, C., Majoros, W.H., May, G.S., Miller, B.L., human chromosome 11 and determine the high-frequency electropho- Mohamoud, Y., Molina, M., Monod, M., Mouyna, I., Mulligan, S., retic polymorphism. Proc. Natl. Acad. Sci. USA 82, 6240–6244. Murphy, L., O’Neil, S., Paulsen, I., Penalva, M.A., Pertea, M., Price, ten Have, A., Dekkers, E., Kay, J., Phylip, L.H., van Kan, J.A., 2004. An C., Pritchard, B.L., Quail, M.A., Rabbinowitsch, E., Rawlins, N., aspartic proteinase gene family in the filamentous fungus Botrytis Rajandream, M.A., Reichard, U., Renauld, H., Robson, G.D., cinerea contains members with novel features. Microbiology 150, Rodriguez de Cordoba, S., Rodriguez-Pena, J.M., Ronning, C.M., 2475–2489. Rutter, S., Salzberg, S.L., Sanchez, M., Sanchez-Ferrero, J.C., Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTALW: Saunders, D., Seeger, K., Squares, R., Squares, S., Takeuchi, M., improving the sensitivity of progressive multiple sequence alignment Tekaia, F., Turner, G., Vazquez de Aldana, C.R., Weidman, J., White, through sequence weighting, position-specific gap penalties and weight O., Woodward, J., Yu, J.H., Fraser, C., Galagan, J.E., Asai, K., matrix choice. Nucleic Acids Res. 11, 4673–4680. Machida, M., Hall, N., Barrell, B., Denning, D.W., 2005. Genomic Valkonen, M., Ward, M., Wang, H., Penttila, M., Saloheimo, M., 2003. sequence of the pathogenic and allergenic filamentous fungus Asper- Improvement of foreign-protein production in Aspergillus niger var. gillus fumigatus. Nature 438 (7071), 1151–1156. awamori by constitutive induction of the unfolded-protein response. Page, R.D.M., 1996. TREEVIEW: an application to display phylogenetic Appl. Environ. Microbiol. 69 (12), 6979–6986. trees on personal computers. Comput. Appl. Biosci. 12, 357–358. van den Hombergh, J.P.T.W., Jarai, G., Buxton, F.P., Visser, J., 1994. Pel, H.J., de Winde, J.H., Archer, D.B., Dyer, P.S., Hofmann, G., Schaap, Cloning, characterization and expression of pepF, a gene encoding a P.J., Turner, G., de Vries, R.P., Albang, R., Albermann, K., Andersen, serine carboxypeptidase from Aspergillus niger. Gene 151, 73–79. M.R., Bendtsen, J.D., Benen, J.A., van den Berg, M., Breestraat, S., van den Hombergh, J.P.T.W., Sollewijn Gelpke, M.D., van de Vonder- Caddick, M.X., Contrera, R., Cornell, M., Coutinho, P.M., Danchin, voort, P.J.I., Buxton, F.P., Visser, J., 1997a. Disruption of three acid E.G., Debets, A.J., Dekker, P., van Dijck, P.W., van Dijk, A., proteases in Aspergillus niger—effects on protease spectrum, intracel- Dijkhuizen, L., Driessen, A.J., d’Enfert, C., Geysens, S., Goosen, C., lular proteolysis, and degradation of target proteins. Eur. J. Biochem. Groot, G.S., de Groot, P.W., Guillemette, T., Henrissat, B., Herweijer, 247, 605–613. M., van den Hombergh, J.P., van den Hondel, C.A., van der Heijden, van den Hombergh, J.P.T.W., van de Vondervoort, P.J.I., Fraissinet- R.T., van der Kaaij, R.M., Klis, F.M., Kools, H.J., Kubicek, C.P., van Tachet, L., Visser, J., 1997b. Aspergillus as a host for heterologous Kuyk, P.A., Lauber, J., Lu, X., van der Maarel, M.J., Meulenberg, R., protein production: the problem of proteases. Trends Biotechnol. 15, Menke, H., Mortimer, M.A., Nielsen, J., Oliver, S.G., Olsthoorn, M., 256–263. Pal, K., van Peij, N.N., Ram, A.F., Rinas, U., Roubos, J.A., Sagt, Ward, M., Wilson, L.J., Kodama, K.H., Rey, M.H., Berka, R.M., 1990. C.M., Schmoll, M., Sun, J., Ussery, D., Varga, J., Vervecken, W., van Improved production of chymosin in Aspergillus by expression as a de Vondervoort, P.J., Wedler, H., Wosten, H.A., Zeng, A.P., van glucoamylase-chymosin fusion. Biotechnology (NY) 8, 435–440. Ooyen, A.J., Visser, J., Stam, H., 2007. Genome sequencing and Ward, M., Lin, C., Victoria, D., Fox, B., Fox, J., Wong, D., Meerman, H., analysis of the versatile cell factory Aspergillus niger CBS 513.88. Nat. Pucci, J., Heng, M., Tsurushita, N., Gieswein, C., Park, M., Wang, H., Biotechnol. 25 (2), 221–231. 2004. Characterization of humanized antibodies secreted by Aspergil- Poussereau, N., Gente, S., Rascle, C., Bilon-Grand, G., Fevre, M., 2001. lus niger. Appl. Environ. Microbiol. 70, 2567–2576. aspS encoding an unusual aspartyl protease from Sclerotinia sclero- Yaver, D.S., Thompson, S.A., Golightly, E.J., Nelson, B., 1995. tiorum is expressed during phytopathogenesis. FEMS Microbiol. Lett. Proceeding of the 18th Fungal Genetics Conference. Pacific Grove, 194, 27–32. CA, p. 62. Rawlings, N.D., Tolle, D.P., Barrett, A.J., 2004. MEROPS: the peptidase Yelton, M.M., Hamer, J.E., Timberlake, W.E., 1984. Transformation of database. Nucleic Acids Res. 32, 160–164. Aspergillus nidulans by using a trpC plasmid. Proc. Natl. Acad. Sci. Rolke, Y., Liu, S., Quidde, T., Williamson, B., Schouten, A., Weltring, USA 81, 1470–1474. K.-M., Siewers, V., Tudzynski, B., Tudzynski, P., 2004. Functional Zhu, H., Qu, F., Zhu, L.H., 1993. Isolation of genomic DNAs from

analysis of H2O2-generating systems in Botrytis cinerea: the major plants, fungi and bacteria using benzyl chloride. Nucleic Acids Res. 21, Cu-Zn-superoxide dismutase (BcSOD1) contributes to virulence on 5279–5280.