WO1997046689A1 - Fungal metallo protease genes - Google Patents

Abstract

The present invention discloses protease deficient filamentous fungus mutants with reduced extracellular acid protease activity, which in addition also contains a site selected disruption of DNA that results in the filamentous fungus having reduced metallo protease activity. Preferably, the filamentous fungus is of the genus Aspergillus. The invention further discloses DNA sequences encoding fungal metallo proteases. Said DNA sequences are used for disruption of metallo protease genes in filamentous fungi as well as for recombinant production of fungal metallo proteases.

Description

FUNGAL METALLO PROTEASE GENES

Field of the invention

The present invention relates to the field of molecular biology. In particular the invention relates metallo proteases obtainable from filamentous fungi and the genes encoding said proteases.

Background of the invention

Aspergillus species, and in particular Aspergillus niger are used for the industrial production of enzymes used in the food processing industry. A. niger has advantages as a host for the production of recombinant proteins because of its large capacity for secretion of proteins, and because systems are available for its molecular genetic manipulation. However, the presence of proteases in the culture fluid, periplasmic space, endoplasmic reticulum and Golgi apparatus has proven deleterious to the expression of homologous and heterologous proteins in A. niger. In fact Aspergilli are used commercially to produce proteases. A number of extracellular proteases from Aspergilli have been described in the literature. The gene pepA encoding aspergillopepsin A from Aspergillus awamori has been cloned. The pepA gene product accounts for a major part of the secreted acid proteases of A. niger and strains in which the pepA gene has been deleted have allowed increased expression of heterologous proteins in A. niger var. awamori (EP-A-0 429 490) . Other protease genes have also been recently cloned from Aspergilli and these include an alkaline aspartic protease of A. oryzae, an alkaline aspartic protease of A. fumigatus, a non- pepsin type acid protease from A. niger var. macrosporus, (pepB) an aspartic protease of A. niger (pepE), a metallo protease called neutral protease II from A. oryzae, the metallo protease from A. fumigatus and A. flavus, two serine proteases from A. niger (pepC, pepD) and two serine carboxypeptidases from A. niger (pepF and cpy) . Also an alkaline protease (prtA) from A. nidulans has been cloned.

Isolated and mutated protease genes of A. niger can be used for gene disruption experiments, i.e. the preparation of mutant strains in which the corresponding natural gene is destroyed. For example, the pepA gene from Aspergillus awamori has been destroyed by gene disruption in order to prepare aspergillopepsin A deficient strains. Similarly the pepB and the pepE gene have been destroyed by gene disruption and a recombinant strain carrying all three disruptions has also been made. However, as mentioned above Aspergilli produce a large number of different proteases and, thus, there is a continuing need for Aspergillus strains deficient in other proteases for the industrial production of proteins. For this purpose there is also a need for other protease genes which can be used for the preparation of protease deficient strains by in vitro mutagenesis, e.g. gene disruption. Moreover, there is also a need for recombinant protease proteins which can be industrially applied for protein processing .

Another major constituent of the secreted protease activities in A. niger is a metallo protease. Metallo proteases have been cloned in some of the fungi. Neutral protease II encoded by a metallo protease has been cloned from A. oryzae and from A. fumigatus and A. flavus whereas a neutral protease I encoding gene was cloned from the A. fumigatus. It is now found that some A. niger protease deficient mutants (prt mutants) isolated by mutagenesis and strains in which one or all of the acidic protease genes have been disrupted, have still metallo protease activity which surprisingly is even increased as compared to the parental wild type A. niger strain, apparently in an attempt to compensate for the loss of the acidic protease activity . This is for example the case in the prtF and prtD mutants of A. niger.

Aspergillus nidulans is also a suitable host for the expression of heterologous proteins. Isolated and mutated protease genes of A nidulans can also be used for gene disruption experiments. The number of A nidulans protease genes cloned is quite limited and therefore there is a need to establish the protease spectrum, to clone the protease genes and to destroy the corresponding natural genes.

Brief description of the Figures

Figure 1 a: shows the partial nucleotide sequences of A niger pepH, and the deduced partial amino acid sequence of PEPH.

Figure 1 b: shows a partial alignment of the amino acid sequences of PEPH and

MEP.

Figure 2a: shows the nucleotide sequences of A. nidulans pepl and pepJ and the deduced amino acid sequences of PEPI and PEPJ .

Figure 2b: shows the homologies between the amino acid sequences of the A. nidulans metallo proteases and several other metallo proteases.

Figure 3a and 3b: present a summary of data which indicate that the A. nidulans

PEPI and PEPJ belong to a novel class of proteases in filamentous fungi.

Description of the invention

The present invention provides protease deficient mutants of filamentous fungi which contain a mutation which is a site selected disruption of DNA that results in the filamentous fungus having a reduced metallo protease activity. Said filamentous fungi are useful for the production of proteins, both heterologous and homologous, because reduced protease activity will minimize both the chance that, and rate at which, such proteins will be degraded during production.

The filamentous fungi of the invention are preferably fungi that can be used for the industrial production of proteins. Such fungi are non-pathogenic, produce virtually no mycotoxins and have often aquired GRAS status in conjunction with enzymes produced by the fungus. Examples of such fungi belong to the genera of Trichoderma, Penicillium, Fusarium, Mucor, Rhizopus, and most preferably Aspergillus. Within the genus Aspergillus, the preferred fungi belong to the Aspergillus niger group, the Aspergillus nidulans group and the Aspergillus flavus group, whereby these "groups " are defined as in Raper and Fennell ( 1 965, In: The Genus Aspergillus, The Williams & Wilkins Company, Baltimore) , thereby comprising all species and variants included in a particular group by these authors. Hence, the terms A. niger, A. nidulans and A. oryzae will herein refer to the entire Aspergillus niger group, the Aspergillus nidulans group and the Aspergillus flavus group, respectively.

In a further embodiment, the filamentous fungi according to the present invention also have a reduced extracellular acid protease activity. A mutant filamentous fungus with a reduced extracellular acid protease activity is herein understood to comprise any fungal mutant of which the activity of one or more of the extracellular acid proteases, which are not metalloproteases, is reduced as compared to the activity in the wild type filamentous fungus. An acid protease is herein understood to mean a protease of which the pH optimum is more acidic than pH 7.0. Extracellular acid protease activity levels of mutant and wild type fungi can be determined by the methods described in van den Hombergh et al. ( 1 995, Curr. Genet. 28: 299-308) .

In a preferred embodiment, the reduced extracellular protease activity is the result of reduced activity of one or more gene products encoded by the A.niger pep A, pepB, pepC, pepD, pepE, pepF, cpy genes or the A. nidulans prtA gene. The skilled person will understand that in fungi other than A.niger or A. nidulans, the corresponding homologues of these genes (genetic variants) are intended. Preferably the reduction in activity of these gene products is achieved by site selected disruption of the corresponding genes.

Alternatively, the reduced extracellular acid protease activity is the result of one or more of the prt mutations as defined by van den Hombergh et al. supra. Preferably the reduced extracellular acid protease activity is the result of a prtD or prtF mutation.

Surprisingly we have found that some of the fungal mutants with reduced extracellular acid protease activity display increased levels of metallo protease activity, possibly in an attempt to compensate for the loss of the acid protease activity. As a consequence, proteins to be produced by these mutant fungal hosts can still be subject to proteolytic degradation by these metallo proteases.

In a preferred embodiment, the mutant filamentous fungi of the invention have reduced activity levels of both extracellular acid proteases and metallo proteases.

A fungal metallo protease is herein understood as (i) being derived from a filamentous fungus, (ii) exhibiting protease activity due to a metal ion, usually Zn^{2 +}, which is usually evidenced by the fact that the protease is sensitive to chelating agents such as EDTA, EGTA or phenanthroline, (iii) having an amino acid sequence which shares sufficient positional identity with amino acid sequences of known metallo proteases for being grouped into one of the metallo protease families.

In order to be able to reduce the metallo protease levels in filamentous fungi we have cloned metallo protease genes from A.niger (pepH) and A. nidulans (pepI and pepJ) using heterologous hybridization as described in Examples 1 , 2, and 3. Further characterization and sequence analysis of these cloned metallo protease genes is presented in Examples 4 and 5. The results of the sequences analysis are presented in the Sequence Listing: SEQ ID NO: 1 , partial A.niger pepH gene; SEQ ID NO:2, partial A.niger pepH cDNA; . SEQ ID NO:3, partial A.niger PEPH amino acid sequence; SEQ ID NO:4, A. nidulans pepI gene; SEQ ID NO:5, A. nidulans pepl cDNA; . SEQ ID NO:6, A. nidulans PEPI amino acid sequence; SEQ ID NO: 7, A. nidulans pepJ gene; SEQ ID NO:8, A. nidulans pepJ cDNA; SEQ ID NO:9, A. nidulans PEPJ amino acid sequence.

The present disclosure of the metallo protease genes in A.niger and A. nidulans provides an incentive for the identification of additional metallo protease genes in these and other fungi which are not known at present. The preferred candidates in this respect are the industrially important fungi which have been mentioned above.

The novel metallo protease sequences of the present invention can be used in conjunction with the already available metallo protease sequences to more accurately define the conserved regions in the metallo protease amino acid sequences and type of substitutions occurring therein. This will facilitate the design of improved degenerate oligonucleotides which will increase the chance of obtaining new metallo protease genes in PCRs or hybridization experiments.

Even though the preferred method for cloning new metallo protease genes is the method of the present invention, i.e. the use of degenerate oligonucleotides in a PCR on genomic DNA (or cDNA) , other methods can also be used for the cloning of new metallo protease genes . Such methods may include heterologous hybridization, hybridization with (degenerate) oligonucleotides, (heterologous) complementation of metallo protease-negative mutants, or even screening of expression-libraries with suitable antibodies.

The invention thus provides isolated DNA sequences encoding proteins having metallo protease activity, which are obtainable from a filamentous fungus. The preferred DNA sequences are the DNA sequence of the pepH gene of Aspergillus niger as partially shown in SEQ ID NO: 1 and 2, the DNA sequence of the pepl gene of Aspergillus nidulans as shown in SEQ ID NO:4 and 5, the DNA sequence of the pepJ genes of Aspergillus nidulans as shown in SEQ ID NO:7 and 8, as well as genetic variants of these DNA sequences. In a further embodiment, the invention also provides DNA sequences encoding proteins having metallo protease activity which are obtainable from filamentous fungi and which are capable of hybridizing to one or more of the pepH, pepl and pepJ sequences.

The metallo protease DNA sequences of the invention are used to prepare mutant filamentous fungi with disrupted metallo protease genes resulting in reduced metallo protease levels. For this purpose DNA constructs containing a metallo protease DNA sequence which comprises a disruption that results in the DNA sequence being incapable of encoding an active metallo protease are provided. A number of methods are available to the skilled person for disruption of DNA sequences encoding an active protein. Such disruptions may include deletions, insertions, substitutions, reversions and truncations of the active coding sequences in order to functionally inactivate the metallo protease coding sequence. Preferably the disruption is nonrevertable, e.g . by deletion of (most of) the coding sequence.

The DNA constructs for disruption of the metallo protease genes will further comprise a selectable marker gene for selection of transformation. A variety of such genes are available to the skilled person, e.g . the fungal amdS, argB, trpC, niaD and pyrG genes as well as several antibiotic resistance genes.

Upon transformation the DNA constructs for disruption of the metallo protease genes will insert at the genomic metallo protease locus through homologous recombination. The various elements in the disruption construct are arranged to optimize for such events as one-step or two-step gene replacement. Examples 9 and 1 0 describes the construction and use of disruption constructs for the pepH, pepl and pepJ metallo protease genes.

The invention thus provides a process for the preparation of a filamentous fungus which has a reduced metallo protease activity. The process comprises the steps of a) transforming filamentous fungus with a DNA construct for disruption of a metallo protease gene, and b) selecting a transformed filamentous fungus with reduced metallo protease activity.

As a product from this process the invention thus provides filamentous fungi wherein the reduced metallo protease activity is the result of a site selected disruption of a DNA sequence encoding a metallo protease. The DNA sequence encoding the metallo proteases preferably are the DNA sequences of the pepH, pepl, and pepJ genes as disclosed in the Sequence Listings or are genetic variants thereof. In a further embodiment the DNA sequences to be disrupted are capable of hybridizing to one of the pepH, pepl, and pepJ genes or their genetic variant.

In a preferred embodiment the disruption constructs are constructed such that the selection marker gene can be removed from the fungal genome after that the disruption of the metallo protease gene has taken place. Methods and examples of this approach are disclosed in EP-A-0 635 574. The advantage of this approach is that multiple (metallo)protease genes can be disrupted in successive transformation rounds in one strain using a single selection marker gene.

In the context of the present invention the term "genetic variants", both when applied to metallo protease genes as well as genes coding for extracellular acid proteases, is understood to comprise hybrid DNA sequences containing a protease-encoding sequence coupled to regulatory regions, such as promoter, secretion and terminator signals, originating from homologous or heterologous organisms. Genetic variants also include DNA sequences encoding mutant protease proteins and degenerate DNA sequences wherein the protease activity of the enzyme is retained. The present invention also includes DNA sequences which are capable of hybridizing to the protease-encoding DNA sequences and genetic variants thereof, as described above, but which may differ in codon sequence due to the degeneracy of the genetic code or cross-species variation.

The term "gene" is used herein to indicate any coding DNA sequence, irrespective of whether the sequence is a genomic sequence (possibly containing introns), a cDNA sequence or a synthetic (isocoding) sequence.

In a further embodiment the present invention provides DNA constructs wherein the DNA sequence encoding the metallo protease is operable linked to regulatory regions suitable for expression of the DNA sequence in a suitable host. The suitable hosts are preferably microorganisms, i.e. bacteria such as E.coli or Bacillus species, yeasts such as Saccaromyces cerevisiae and Kluyvermyces lactis, or filamentous fungi of which are preferred fungi from the genera Trichoderma, Penicillium, Fusarium, Mucor, Rhizopus, and most preferred are Aspergillus species.

A wide variety of regulatory regions suitable for regulating the expression of the metallo protease genes in the fore-mentioned hosts are available to the skilled person. Of these regulatory regions the promoters are most important. Examples of strong constitutive and/or inducible promoters which are preferred for use in fungal expression hosts are the ATP-synthetase, subunit 9 (oliC), pyruvate kinase (pki), triose phosphate isomerase (tpi) , alcohol dehydrogenase (alcA), α-amylase (amy), amyloglucosidase (AG) , acetamidase (amdS) and glyceraldehyde-3-phosphate dehydrogenase (gpd) promoters. Examples of strong yeast promoters are the alcohol dehydrogenase, lactase, 3- phosphoglycerate kinase and triosephosphate isomerase promoters. Examples of strong bacterial promoters are the α-amylase and Spo2 promoters as well as promoters from extracellular protease genes.

The invention therefore discloses host organisms capable of heterologous or overexpression of the filamentous fungal metallo protease genes of the invention, as well as a process for the preparation these metallo proteases. This process comprises the steps of a) culturing the host organisms capable of heterologous or overexpression of the filamentous fungal metallo protease genes under conditions conducive to the expression of the metallo protease gene, and b) recovering the metallo protease.

Several uses for the metallo proteases produced according to the invention can be envisaged, e.g. the obtained metallo proteases can be used to asses in vitro whether or not a certain protein of interest, to be produced in a fungal host is susceptible to the protease in question . This allows to determine which metallo protease genes need to inactivated in the fungal host.

Alternatively, metallo proteases may find application in many industrial processes where proteases are applied . The advantage of the metallo protease being that the protease activity can in principle be switched on or off by the addition of the metal ion or a chelating agent respectively.

Included within the meaning of the term "fungal metallo protease" as used in the present invention are also fragments of such enzymes which retain metallo protease activity, however, the full length enzymes are preferred embodiments. It is understood that also fusion proteins containing a fungal metallo protease of the invention attached to additional amino acids, peptides, or proteins are part of the present invention.

The following examples are provide so as to give those of skill in the art a complete disclosure and description of how to make and use the invention and are not intended to limit the scope of what the inventors regard as their invention.

Examples

1 . Preparation of metallo protease probes.

1 .1 Preparation of an A. oryzae probe.

Two specific DNA primers (CGGGATCCCGCGTGAGTGAACTCGTG and CGGAATTCCGGTCACCGACTGCAAGG, containing a BamHI and an EcoRI restriction site, respectively) derived from the sequence for the NPII encoding gene (Tatsumi et al., 1 991 ) are used in PCR reactions on A. oryzae ATCC 20386 genomic DNA (isolated as described by de Graaff et al., 1 988) . The PCR conditions used are; an initial melting step, (94°C, 3 min), followed by 30 cycles of 1 min annealing (68°C), 1 min extension (72 °C) and 1 min melting (94°C) . After a final extension at 72 °C for 5 min, the PCR mixtures are analyzed on 1 .0% (w/v) agarose gels in 1 xTAE. The resulting 0.41 kb PCR fragment is cut out from the agarose gel and the DNA is obtained using the GeneClean kit (Westburg) as described in the instruction manual. The 0.41 kb PCR fragment is digested with restriction enzymes EcoRI and BamHI in a reaction mixture composed of the following solutions; ( 1 0 μl (approx. 2 μg) DNA solution; 5 μl of the appropriate 10 x React, buffer (Life Technologies); 1 0 U of both restriction enzymes (Life Technologies) and sterile water to give a final volume of 50 μl . After digestion the restriction enzymes were removed via fenol, fenol-chloroform and chloroform extraction and subsequent ethanol precipitation of the DNA. 50 ng of the restriction digested PCR fragment and 50 ng of BamHI-EcoRI digested pUC1 9 are mixed with 4 μl of 5 x ligation buffer (Life Technologies; composition: 500 mM Tris-HCl, pH 7.6, 1 00 mM MgCI₂, 1 0 mM ATP, 1 0 mM dithiothreitol, 25 % (w/v) PEG-6000) and 1 μl ( 1 .2 U) T₄ DNA ligase (Life technologies) . After overnight incubation at 1 4 °C the mixture was diluted five times with sterile water. 1 0 μl of the diluted mixture was used to transform E. coli DH5α competent cells, as described by Sambrook et al. (1989) . From the resulting (white) colonies five were grown overnight on liquid LB medium (LB medium per 1 000 ml: 1 0 g trypticase peptone (BBL) , 5 g yeast extract (BBL) , 10 g NaCl, 0.5 mM Tris-HCl pH 7.5) containing 100 μg/ml ampicillin. From the cultures plasmid DNA was isolated by the alkaline lysis method as described by Maniatis et al. ( 1 982), which was used in restriction analyses to select a clone harbouring the desired plasmid plM664. The 0.41 kb A. oryzae PCR fragment in the resulting plasmid, plM664, was sequenced to prove cloning part of the A. oryzae NPII encoding gene. 50 ng of the PCR fragment is labelled with ³²P-dATP via random priming as described by Sambrook et al. ( 1 989) and used immediately for either Southern or plaque lift probings.

1 .2 Preparation of an A. fugimatus probe.

Plasmid pMTL21 -H5 contains a 5.0 kb Hindlll kb fragment of A. fumigatus DNA, that encodes the A. fumigatus MEP gene (described by Jaton- Ogay et al., 1 994) . The part of the coding region that encodes the mature MEP protease can be excised by EcoRI. The plasmid pMTL21 -H5 is therefore digested with EcoRI and the fragments are separated on a 0.7% (w/v) agarose gel in 1 x TAE. The 1 .3 kb EcoRI fragment is cut out and the DNA is obtained using the GeneClean kit (Westburg) as described in the instruction manual. 50 ng of this fragment is labelled with ³²P-dATP via random priming as described by Sambrook et al. ( 1 989) and used immediately for either Southern or plaque lift probings.

2. Screening genomic libraries.

2.1 Southern analysis of A. niger and A. nidulans genomic DNA

Total genomic DNA is isolated from A. niger and A. nidulans wild type strains (N400 and WG096, respectively) is restriction digested with several restriction enzymes, fragments are separated on 0.7% (w/v) agarose gels in 1 xTAE and subsequently blotted to nylon membranes. In a series of Southern hybridization experiments (Sambrook et al., 1989 ) both the A. fumigatus and the A. oryzae metallo probes are used to screen for the presence of related genes in A. niger and A. nidulans. In A. niger it is observed that at 60° C after washing down to 2 x SSC, 0.5% (w/v) SDS the A. fumigatus metallo protease probe revealed clear specific hybridizing fragments in all digests tested . Under identical conditions no hybridizing signals were detected using the A. oryzae metallo protease probe. Hybridization of A. nidulans genomic DNA with the A. oryzae probe (at 56°C, 4 x SSC, 0.5% (w/v) SDS) however resulted in clear hybridizing signals, indicative for the existence of NPII encoding gene(s) in A. nidulans. Probing A. niger genomic DNA under these conditions (again) resulted in no hybridizing signals.

After isolation and characterization of the A. niger pepH gene and the A. nidulans pepI and pepJ genes, the Southern analysis are repeated under stringent washing conditions (final wash 0.2 x SSC, 0.1 % (w/v) SDS at 68° C) to establish copy numbers the individual protease genes. These analyses indicate that all three protease genes are present in single copy in the genome.

2.2 Screening of an A. niger N400 genomic library.

For screening of the A.niger genomic library, constructed as described by Harmsen et al., 1 990, for the pepH gene 5 x 1 0³ pfu per plate were plated in NZCYM top-agarose containing 0.7% agarose on five 85-mm-dιameter NZCYM ( 1 .5 % agar) plates as described (Sambrook et al., 1989 ) using E. coli LE392 as plating bacteria. After overnight incubation of the plates at 37 °C two replicas of each plate were made on HybondN ⁺ filters (Amersham) as described in Maniatis et al. ( 1982 ) . After wetting the filters in 5 x SSC the filters were washed for 30 mm. at room temperature in 5 x SSC. Hybridisation using a ³²P- labelled 1 .3 kb EcoRI fragment form A. fumigatus MEP metallo protease gene, prepared as described by Sambrook et al. ( 1989 ) , was done according the following procedure; prehybridisation in 6 x SSC (20 x SSC per 1 00ml: 1 75.3 g NaCl, 107.1 g sodium citrate.5.5 H₂O, pH 7.0), 0.5% (w/v) SDS, 5 x (Denhardt's solution ( 100 x Denhardt's solution per 500 ml: 10 g Ficoll-400, 10 g polyvinylpyrrolidone, 1 0 g Bovine Serum Albumin (Boehringer, fraction V) and 1 00 μg/ml denatured heering sperm DNA at 60°C for 3-5 hrs and hybridisation in an identical buffer which contained the denatured radiolabelled probe at 60°C for 1 5- 1 8 hrs, followed by three washes in 2 x SSC, 0.5% (w/v) SDS at 60°C. The membrane was covered with Saran wrap and autoradiographed overnight at -70° C using Konica X-ray films and Kodak X-omatic cassettes with regular intensifying screens.

This screening resulted in about 20 positive phages, of which ten were purified . Each positive phage was picked from the plate using a Pasteur pipette and the phages were eluted from the agar plug in 0.5 ml of SM buffer containing 10 μl chloroform, as described in Maniatis et al . ( 1982 ) . The phages obtained were purified by repeating the procedure described above using filter replicas from plates containing 50-100 plaques of the isolated phages.

2.3 Screening of an A. nidulans genomic library.

For screening of the A. nidulans genomic library, constructed as described by Orr and Timberlak ( 1 982), the library was plated as described in 2.2. The replicas prepared from this library (Sambrook et al., 1 989) were prehybridised in hybridization buffer (see 2.2) at 56 °C for 3-5 h. Hybridisation was performed in an identical buffer which contained additionally the denatured ³²P radiolabelled 0.41 kb A. oryzae derived metallo protease PCR fragment. After 1 5- 1 8 h hybridisation, the replicas were washed three times in 4 x SSC, 0.5 % (w/v) SDS at 56°C. The replicas were covered were saran wrap and autoradiographed at - 70°C using Konica X-ray films and X-omatic cassettes with regular intensifying screens. This screening resulted in at least 35 positive phages, of which ten were purified as described in 2.2. 3. Characterization of the lambda clones.

3.1 Characterization of A. niger lambda clones.

After purification, the phages were propagated by plating 5 x 1 0³ phages on NZCYM medium. After overnight incubation at 37 °C confluent plates were obtained, from which the phages were eluted by adding 5 ml SM buffer and storing the plate for 2 h at 4°C with intermittent shaking . After collection of the supernatant using a pipette, the bacteria were removed from the solution by centrifugation at 4,000 x g for 1 0 min. at 4° C. To the supernatant 0.3% (w/v) chloroform was added and the number of pfu is determined. These phage stocks contain approximately 1 0⁹ pfu/ml.

DNA of 1 0 selected phages (lambda 1 - 10) , isolated as described in Sambrook et al. ( 1 989) was analyzed by Southern analysis. The DNA was digested for 4 h. at 37 °C in a reaction mixture composed of the following solutions; 5 μl (approx. 1 μg) DNA solution; 2 μl of the appropriate 1 0 x React, buffer (Life Technologies); 10 U Restriction enzyme (Life Technologies) and sterile water to give a final volume of 20 μl. The samples were incubated for 1 0 min. at 65 °C and rapidly cooled on ice, before adding loading buffer and then loaded on a 0.6 % (w/v) agarose gel in TAE buffer. The DNA fragments were separated by electrophoresis at 25 V for 15 -18 h (at RT) .

After electrophoresis the DNA was transferred and denatured by alkaline blotting (VacuGene XL, Pharmacia LKB) to nylon membrane (Hybond N, Amers- ham) as described in the VacuGene XL instruction manual (pp . 25-26) and subsequently prehybridised and hybridised as described in 2.2. The hybridisation pattern was obtained by exposure of Kodak XAR-5 X-ray film for 1 8 h. at -70 °C using an intensifying screen. In all these clones fragments originating from the same genomic region were found.

3.2 Characterization of A. nidulans lambda clones.

Pure phage stocks were generated for the 10 A. nidulans purified metallo phages, similar to the procedure described in 3.1 . After isolation of phage DNA, this DNA was digested with several restriction enzymes and the resulting fragments were separated by electrophoresis in 1 x TAE agarose gels. The resulting separated DNA fragments were electroblotted to nylon membranes (see 3.1 ) and subsequently prehybridised and hybridised as described for the replicas in 2.1 .

In all clones hybridizing fragments are identified and as these fragments can be divided into two groups this Southern analysis indicated that two classes of phages, originating from two independent genomic loci, were isolated . Class I of the A. nidulans metallo phages was identified by the presence of a 2.4 kb hybridizing PstI fragment whereas class II metallo phages were identified by the presence of a 0.7 kb EcoRI fragment.

4. Cloning of pepH into a plasmid, its sequencing and its characterization.

4.1 Construction of plM676.

Based on the results obtained during the characterization of the lambda phages a 10 kb EcorW was selected for subcloning . 25 ng pUC 1 8 FcoRI digested fragment was mixed with 100 ng 1 0 kb EcoRI fragment and 4 μl of 5 x ligation buffer (composition: 500 mM Tris-HCl, pH 7.6; 1 00 mM MgCI₂; 1 0 mM ATP; 1 0 mM dithiothreitol; 25 % (w/v) PEG-6000) and 1 μl ( 1 .2 U/ml) T₄ DNA ligase (Life Technologies) was added to this mixture in a final volume of 20 μl. After incubation for 1 6 h at 14°C the mixture was diluted to 1 00 μl with sterile water. 1 0 μl of the diluted mixture was used to transform E. coli DH5α competent cells, prepared as described by Sambrook et al. ( 1 989) . 1 0 of the resulting (white) colonies were grown overnight in LB medium (LB medium per 1 000 ml: 1 0 g trypticase peptone (BBL) , 5 g yeast extract (BBL), 1 0 g NaCl, 0.5 mM Tris-HCl pH 7.5) containing 100 pg/ml ampicillin. From the cultures plasmid DNA was isolated by the alkaline lysis method as described by Maniatis et al. ( 1 982) , which was used in restriction analysis to select a clone harbouring the desired plasmid plM676. Plasmid DNA was isolated on a large scale from 200 ml cultures E. coli DH5σ containing plM676 grown in LB medium containing 1 00 μg/ml ampicillin (Sambrook et al., 1 989) . The plasmid was purified using the Nucleobond AX ion exchange silica purification system as detailed in the Nucleobond AX instruction manual (p p. 8) . The yield was approximately 200 μg .

4.2 Nucleotide sequence of pepH.

The sequence of the structural part of the A. niger pepH gene, was determined by subcloning fragments from plM676 in pUC 1 8/1 9, in combination with the use of specific oligonucleotides as primers in the sequencing reactions.

For nucleotide analysis restriction fragments were isolated and cloned in pUC 1 8/1 9 DNA vectors, digested with the appropriate restriction enzymes. The nucleotide sequences were determined by the dideoxynucleotide chain- termination procedure (Sanger et al., 1 977) using the Pharmacia T7 DNA polymerase sequencing kit. Computer analysis was done using the PC/GENE program (Intelligenetics) . The sequence determined is given SEQ ID NO: 1 . The sequence of the pepH gene is not yet complete because at the 3'-end of the coding region several ambiguities were found due to compressions. We are still in the process of solving these ambiguities.

The deduced PEPH protein sequence shows high homologies with the A. fumigatus MEP metallo protease (see appendix I and III). Detailed homology comparisons with members of the individual metallo protease classes demonstrated only limited homology. However, both the A. niger and the A. fumigatus metallo proteases contain a region which to some extent resembles the third ligand region in thermolysin type metallo proteases.

5. Cloning of pepl and pepJ into plasmids, their sequencing and characterization.

5.1 Construction of plM667, plM668 and plM669.

To verify whether indeed two metallo protease genes were isolated from A. nidulans, from lambda clone 1 a 2.4 kb hybridising Pstl fragment and from lambda clone 5 a 0.7 kb hybridising fcoRI fragment are initially subcloned following a similar procedure as described for the cloning of the A. niger pepH gene in 4.1 . This results in plasmids plM665 and plM666, respectively. Both the inserts in plM665 and plM666 are sequenced and the deduced amino acid sequences are both found to have homologies with the A oryzae NPII protease. After confirmation that both A. nidulans phage classes contain metallo protease sequences, larger fragments are subcloned to complete the sequencing of both the genes. From phage class I two EcoRI fragments ( 1 .4 kb and 2.0 kb) and from phage class II a 6.5 kb Hindlll-Xhol fragment are subcloned in pUC 1 9 resulting in plasmids plM667, plM668 and plM669, respectively.

5.2 Nucleotide sequence of pepl.

For nucleotide analysis specific restriction fragments are isolated and cloned in pUC 1 8/1 9 DNA vectors, digested with the appropriate restriction enzymes. The nucleotide sequences are determined by the dideoxynucleotide chain-termination procedure (Sanger et al . , 1 977) using the Pharmacia T7 DNA polymerase sequencing kit. Computer analysis is done using the PC/GENE program (Intelligenetics) . The sequence determined is given in SEQ ID NO:2.

5.3 Nucleotide sequence of pepJ.

For nucleotide analysis restriction fragments are isolated and cloned in pUC1 8/1 9 DNA vectors, digested with the appropriate restriction enzymes. The nucleotide sequences are determined by the dideoxynucleotide chain-termination procedure (Sanger et al. , 1 977) using the Pharmacia T7 DNA polymerase sequencing kit. Computer analysis is done using the PC/GENE program (Intelligenetics) . The sequence determined is given SEQ ID NO:3.

5.4 A. nidulans metallo proteases, a novel class of enzymes

Both the A. nidulans metallo proteases contain a region HEXXH in which two histidine residues, in analogy with other metallo proteases are involved in coordinating the Zn^{2 +} cation in the active center, and a glutamate involved in the bond-breaking process (Jongeneel et al., 1 989) . All metallo proteases characterized sofar have been assigned to several independent classes based on the homologies in the proximity of the metal ion ligands (Jiang and Bond, 1 992) . The first two ligands, which in all metallo proteases studied sofar are two histidine residues, are present in either a HEXXH or a HXXEH motif (Appendix V-A) . In the HEXXH-containing metallo proteases, the thermolysin-type metallo proteases, the third metal ion ligand is a glutamic acid residue which is located in a GAXNEAFSD sequence whereas the fourth ligand has been shown to be a water molecule (Kester and Matthews, 1 977) . For the other metallo proteases containing a HEXXH motif, the third ligand is also a histidine residue and the fourth ligand is a tyrosine residue. The amino acid present after the third metal ion ligand and the four amino acids present immediately before the fourth ligand did allow further differentiation into four classes, Astacin, Serratia, Matrixin and Snake Venom, respectively (Jiang and Bond, 1 992) . The metallo proteases from A. nidulans and A. oryzae do not belong to any of the classes mentioned. Clearly, these metallo proteases do not have a histidine as a third ligand as, apart from the two histidine residues in the HEFTHA sequence, no additional histidine residues are present in the mature proteases . Also no GAXNEAFSD sequence, which could indicate a thermolysin-type metallo protease, was found . The A.niger and A. fumigatus metallo protease encoding genes (pepH and MEP, respectively) has also been cloned and sequenced and the deduced amino acid sequence contains in addition to the HEYTH zinc ligands a region resembling the third ligand region in thermolysins, ESGGMGEGWSD (see appendix III) . This type of metallo protease has twice the size of the A. nidulans metallo proteases, shows no homology to the A. nidulans PEPI and PEPJ metallo proteases and is probably a thermolysin-type metallo protease (see also 4.2) . Based on these observations it appears that the metallo proteases from filamentous fungi can be divided into two separate classes; the thermolysin like and the acid metallo proteases. The third and fourth metal ion ligands are unknown for the acid metallo proteases but (compared to other metallo protease classes) at approximately the same C-terminal distance of the HEFTH box a conserved sequence SYALY is found, which resembles partially the region preceeding the fourth ligand in the Astacin, Serratia, Matrixin and Snake venom classes (Appendix III) .

6. Overexpression of pepH into A. niger.

6.1 Multiple copies.

A. niger NW21 9 (cspA1 , leuA 1 , pyrA6, nicA 1 ) is transformed with 1 μg pGW635 plus 10 μg plM676 to yield uridine prototrophs. Colonies are purified and DNA prepared as described above. Southern blots probed with the internal Hindlll fragment from pepH show that some transformants have a single copy of plM676 integrated into their genome whereas others have up to and above 1 0 extra copies in their genome. These strains produce correspondingly more proteolytic activity and are mitotically stable.

6.2 Gene fusions.

Expression cassettes based upon the A. niger pkiA promoter and terminator are prepared for the A. niger pepH gene as described in Bartling et al. ( 1 996) . The correct structure of the expression cassette cloned in pUC 1 8 (and resulting in plasmid pPKIPEPH) is confirmed by restriction digestion and sequencing . pPKIPEPH contains a fragment inserted into pUC 1 8, which contains an expression cassette consisting of the pyruvate kinase promoter of A. niger fused to the ATG start codon of the pepH gene of A. niger, which is terminated by the pyruvate kinase terminator. pPKIPEPH is used with pGW635 (containing the A. niger pyrA gene) to cotransform A. niger NW21 9 to uridine prototrophy.

The presence of the pki-pepH fusion is confirmed by purifying DNA from individual purified transformants and using it for Southern analysis using probes from the pki and the pepH genes. Strains with one or more copies of this gene fusion integrated into their genome are shown to produce more proteolytic activity when the cells are grown rapidly on glucose as C source. This approach allows expression of the pepH protease gene under conditions (high glucose concentrations) which repress the expression of other extracellular protease encoding genes.

7. Overexpression of pepl and pepJ in A. nidulans.

7.1 Multiple copies.

For construction of multiple copy tranformants for the A. nidulans pepJ and pepl genes, the plasmids plM669 and plM667-8 are used respectively. Plasmid plM667-8 is generated by combining the 1 .4 kb and 2.0 kb EcoRI fragments into a 3.4 kb fragment which contains the promoter, coding region and terminator sequences of the pepl gene. Both plasmids are individually co- transformed with plasmid pGW635 to uridine auxotrophic A. nidulans WG096 strain. Selected and purified transformants are analyzed as described for the pepH multicopy transformants (6.2) . Finally, single and multicopy transformants are selected to produce the A. nidulans PEPI and PEPJ proteins as described for the A. niger PEPH protein.

7.2 Gene fusions.

Similar to the procedures described in 6.2 gene fusions for the A. nidulans pepI and pepJ genes can be constructed. Upon transformation of the uridine auxotrophic strain WG096, uridine prototrophic transformants are isolated and purified. Via Southern analysis the copy number of the expression cassettes in the isolated transformants can be determined . Selected transformants can be grown on high glucose containing media to produce the A. nidulans metallo proteases. Similar to the production of PEPH in A.niger the growth conditions used repress the expression of other homologous extracellular proteases in A. nidulans.

8. Expression of pepH, pepl and pepJ in other organisms.

8.1 .1 . pepH in A. nidulans

(see explanation in 7)

8.1 .2 pepl pepJ in A. niger

The expression cassettes constructed for the A. nidulans pepl and pepJ genes which are used to generate multicopy transformants both for the non- fused situation (see 7.1 ) and the gene-fusions (see 7.2) are used to generate multi copy transformants in A. niger NW21 9 (as described in 6.1 ) . Selected transformants are analyzed via Southern analysis (see 6.1 ) to select single and multicopy transformants which are subsequently used to generate the PEPI and PEPJ proteins.

9. Residual metallo protease activity in A. niger prt mutants.

Van den Hombergh et al. ( 1 995) described the isolation and genetical characterization of a set of protease deficient (prt) mutants, comprising at least seven complementation groups. All prt complementation groups result in reduced extracellular proteolytic activities and have been tested in vitro for reduced proteolytic degradation of proteins which are very susceptible to proteolytic degradation. The residual proteolytic spectrum in these prt mutants has been characterized using specific protease enzyme assays in cultures which are induced for protease expression using wheat bran . Enzyme assays in combination with specific inhibitors for metallo proteases (EDTA and 1 , 1 0 phenanthroline) demonstrate increased EDTA and 1 , 1 0 phenanthroline inhibitable activities in the cleared culture supernatants of protease mutants NW228 (bioA 1 , prtF28) , NW229 (bioA 1 , prtF29) and NW232 (bioA1 , prtD32) compared to wild type N400.

In vitro degradation experiments as described by van den Hombergh et al . ( 1 995) are repeated with culture supernatants of the prtF28 protease (strain NW228) mutant and its parental wild type strain (N 573; bioA 1 ) , both in the presence and in the absence of EDTA and 1 , 1 0 phenanthroline to inhibit metallo protease activities . Both in N573 and NW228 a reduction of the degradation of the purified PELB protein, which due to its susceptibility to proteolytic degradation is used as tester protein, is observed in the presence of these inhibitors.

9.1 Construction of a pepH disruption construct.

The HEYTHG encoding sequence in A. niger pepH contains a sequence GAATaC which resembles an EcoRI restriction site (GAATTC) . Using the ' altered sites II in vitro mutagenesis system' (Promega) the GAATaC sequence is changed into GAATTC thus generating an EcoRI site within the PEPH active site region which is involved in binding a Zn^{2 +} ion. After generating this internal EcoRI site the 1 0 kb EcoRI fragment is digested with EcoRI to obtain two (2.5 and 7.5 kb) EcoRI fragments. The 7.5 kb EcoRI fragment is cloned into pUC 1 8 (linearized with EcoRI) and a plasmid which contains the EcoRI insert in the correct orientation is selected and designated plM676-3' . Plasmid plM676-3' is digested with Xbal and the 6.3 kb XbaI fragment (containing the left flanking region of pepH) is isolated .

Plasmid plJ 1 6, containing the A. nidulans argB gene (Johnstone et al . , 1 985) , is digested with BamHl and using a specific primer this BamHl site is converted into a Notl site, generating plasmid plJ 1 6-Not. Plasmid plJ 1 6-Not is digested with Pstl and using a specific primer the Psfl site is converted into a Xbal site, resulting in plasmid plJ 1 6-NotXba . Plasmid plJ 1 6-NotXba is linearized with EcoRI and the 2.5 kb EcoRI fragment obtained from the initial 1 0 kb pepH containing EcoRI fragment is cloned into it. The resulting plasmids are analyzed for the orientation of the EcoRI insert and the correct oriented plasmid is designated plM676-5'-dιs. plM676-5'-dιs is linearized with XbaI and the 6.3 kb fragment from plasmid plM676-3' is cloned into it, resulting in plM676-dis. plM676-dιs contains 8.8 kb from the original 1 0 kb EcoRI pepH containing fragment in which at an introduced EcoRI site in the HEYTHG encoding sequence the A. nidulans argB gene is inserted. Digestion of plM676-dis with Notl and Xbal generates the 1 1 .5 kb linear disruption fragment. This disruption fragment is designed to function as a one-step gene-disruption construct. The linear fragment will upon transformation to A. niger and after a double crossover event at the homologous pepH locus result in a pepH disruption strain .

9.2 Transformation A. niger.

The disruption construct for the A. niger pepH gene is transformed in A. niger strain NW205 (cspA 1 , argB1 3, pyrA6, nicA 1 ) and strain NW1 55 (cspA 1 , argB 1 3, pyrA6, prtF28, nicA 1 ) . Arginine prototrophic transformants are selected and purified .

9.3 Identification gene disruptions.

DNA is isolated from purified transformants and analyzed in a Southern analysis using specific pepH and A. nidulans argB probes to identify double cross-over integration of the disruption construct at the homologous (pepH) locus, thus disrupting the pepH gene.

Selected disruptants are grown on media optimized for protease expression (as described by van den Hombergh et al. , 1 995) and analyzed via specific enzyme assays and in vitro degradation assays to demonstrate reduction of extracellular EDTA- and 1 , 1 0 phenanthroline inhibitable activities and to demonstrate a reduction of in vitro PELB degradation compared to strains in which the pepH gene is not disrupted.

10. Metallo protease activity in A. nidulans.

A. nidulans wild type strain WG096 is grown on liquid medium containing 0.5 g MgSO₄, 1 .5 g KH₂PO₄, 0.5 g KCI, trace elements (Vishniac and Santer, 1 957), 1 % (w/v) BSA (Boehringer, fraction V) and 1 % (w/v) elastm (Fluka) for 72 h at 37 °C. Proteolytic activities in the cleared culture supernatants were determined as described by van den Hombergh et al ( 1 995) using BSA (fraction V) as a substrate and using 1 , 1 0 phenanthroline as specific inhibitor of metallo protease activities. Comparison of non-inhibited and inhibited protease activities in the A nidulans wild type culture supernatants indicate the presence of metallo protease activities.

1 1 . Disruption of A. nidulans metallo protease genes.

1 1 .1 Construction of an A. nidulans pepl disruption construct

Plasmid plJ1 6, containing the A. nidulans argB gene (Johnstone et al., 1 985), is linearized with PstI. A primer containing the unique restriction sites Xhol and Pstl is ligated into the linearized plJ 1 6, to generate plasmid plJ 1 6-Xho. Plasmid plM668 is digested with EcoRI and the 2.0 kb EcoRI insert is isolated . Plasmid plJ 1 6-Xhol is linearized with EcoRI and the 2.0 kb EcoRI fragment, isolated from plasmid plM668, is ligated into this plasmid, resulting in plasmid plM668-5'dis. Upon analysis of the orientation of the 2.0 kb insert in plM668-5'dis a plM668- 5' plasmid in which the 2.0 kb fragment is inserted in the correct orientation is selected and subsequently digested with Pstl . Plasmid plM667 is digested with Pstl and the kb Pstl fragment is isolated . This Pstl fragment is ligated into plasmid plM668-5'dis to generate plM668-dis. In plasmid plM668-dis the internal EcoRI-Pstl fragment (encoding part of the PEPI protease) is replaced with the A. nidulans argB gene, thus generating an inactive A. nidulans pepl gene. Restriction enzyme digestion with Xhol and BamH l allows removal of the 2.7 kb vector sequence and results in the linear disruption fragment. This disruption fragment which is designed to function as a one-step gene-disruption construct will upon transformation to A. nidulans and after a double cross-over event at the homologous pepl locus result in a pepl disruption strain.

1 1 .2 Construction of an A. nidulans pepJ disruption construct

Plasmid plM669 is digested with EcoRV and Smal (to remove the polylinker Pstl site) and the 9.5 kb fragment is isolated . After backligation of this fragment plasmid plM669-Pst is generated. Plasmid plM669-dPst is restriction digested with Psfl and BamHI to remove the complete pepl coding region. From plasmid plJ 1 6 (Johnstone et al., 1 985) the 2.7 kb BamHl-Pstl fragment containing the A. nidulans argB gene is isolated . After ligation of this argB containing fragment into the Pstl- BamHl digested plM669-dPst, plasmid plM669-dis is obtained . Restriction enzyme digestion with Xhol and EcoRI allows removal of the 3.0 kb pBluescript vector sequence and results in the linear disruption fragment. This disruption fragment which is designed to function as a one-step gene-disruption construct contains the A. nidulans argB gene inserted between the 5' and 3' flanking regions of the A. nidulans pepJ gene. This linear fragment will upon transformation to A. nidulans and after a double cross-over event at the homologous pepJ locus result in a pepJ disruption strain. Microorganism deposits

E.coli strains containing plasmids pIM 676, pIM 667, pIM 668 and pIM 669 have been deposited 5 June 1996 at the Centraal Bureau voor Schimmelcultures, Baarn, The Netherlands, under the Budapest Treaty with accession numbers CBS 619.96, CBS 620,96, CBS 621.96 and 622.96, respectively.

References

Bartling, S., van den Hombergh, J.P.T.W., Olsen, 0., von Wetttein, D. and Visser, J. (1996) Expression of an Erwinia pectate lyase in three species of Aspergillus. Curr. Genet.29:474-481

de Graaff, L. van den Broeck, H. and Visser, J. (1988) Isolation and characterization of the Aspergillus niger pyruvate kinase gene. Curr. Genet. 133:315-321

Harmsen, J.A.M., Kusters-van Someren, M.A. and Visser, J. (1990) Cloning and expression of a second Aspergilus niger pectin lyase gene (pe/A) - indications of a pectin lyase gene family in Aspergillus niger. Curr. Genet. 18:161-166

van den Hombergh, J.P.T.W., van de Vondervoort, P.J.I., van der Heijden, N.C.B.A. and Visser, J. (1995) New protease mutants in Aspergillus niger result in strongly reduced in vitro degradation of target proteins; genetic and biochemical charaterization of seven complementation groups. Curr. Genet. 28:299-308

Jaton-Ogay, K., Paris, S., Huerre, M., Quadroni, M., Falchetto, R., Togni, G., Latge, J.P. and Monod, M. (1994) Cloning and disruption of the gene encoding an extracelllular metallo protease of Aspergillus fumigatus. Mol. Microbiol. 14:917-928

Jiang, W. and Bond, J.S. (1992) Families of metalloendopeptidases and their relationships. FEBS Lett.312:110-114

Johnstone, I.L., Hughes, J. and Clutterbuck, A.J. (1985) Cloning an Aspergillus nidulans developmental gene by transformation. EMBO J. 4:1307- 1311

Jongeneel, C.V., Bouvier, J. and Bairoch, A. (1989) A unique signature identifies a family of zinc-dependent metalloproteinases. FEBS Lett. 242:211- 214

Kester, W.R. and Matthews, B.W. (1977) Crystallographic study of the binding of dipeptide inhibitors in thermolysin: implications for the mechanism of catalysis. Biochemistry 16:2506-2516 Orr, W.C. and Timberlake, W.E. (1982) Clustering of spore-specific genes in Aspergillus nidulans. Proc. Natl. Acad. Sci. USA 79:5976-5980

Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular cloning. A laboratory manual, 2nd edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Sanger, F., Nicklen, S. and Coulson, A.R. (1977) DNA sequencing with chain terminators. Proc. Natl. Acad. Sci. USA 74:5463-5467

Tatsumi, H., Murakami, S., Tsuji, R.F., Ishida, Y., Murakami, K., Masaki, A., Kawabe, H., Arimura, H., Nakano, E. and Motai, H. (1991) Cloning and expression in yeast of a cDNA clone encoding Aspergillus oryzae neutral protease II, a unique metallo protease. Mol. Gen. Genet.228:97-103

Vishniac, W. and santer, M. (1957) The thiobacilli. Bacteriol. Rev. 21:195-213.

Claims

Claims 1. A protease deficient filamentous fungus which is characterized in that the filamentous fungus contains a site selected disruption of DNA that results in the filamentous fungus having reduced metallo protease activity.

A filamentous fungus according to claim 1 wherein the filamentous fungus is suitable for the industrial production of proteins.

3. A filamentous fungus according to claim 2 further characterized in that the fungus has reduced extracellular acid protease activity.

4. A filamentous fungus according to any one of claims 1 - 3 wherein the filamentous fungus belongs to the genus Aspergillus.

5. A filamentous fungus according to claim 4 wherein the filamentous fungus is selected from the group consisting of the Aspergillus niger group, the Aspergillus nidulans group and the Aspergillus flavus group .

6. A filamentous fungus according to any one of claims 3 - 5 wherein the reduced extracellular acid protease activity is the result of reduced activity of a gene product encoded by a gene selected from the group consisting of pep A, pepB, pepC, pepD, pepE, pepF, cpy, and prtA .

7. A filamentous fungus according to any one of claim 3 - 5 wherein the reduced extracellular acid protease activity is the result of a prt mutation.

8. A filamentous fungus according to any one of claims 1 -7, wherein the reduced metallo protease activity is the result of a site selected disruption of a DNA sequence encoding a metallo protease selected from the group consisting of:

a) pepH, pepl, and pepJ; and

b) genetic variants of the DNA sequences of part a); and

c) DNA sequences capable of hybridizing to one of the sequences of parts a) and b), above.

9. An isolated DNA sequence encoding a protein having metallo protease activity, which is obtainable from a filamentous fungus and which is selected from the group consisting of:

a) the DNA sequence of the pepH gene of Aspergillus niger as partially shown in SEQ ID NO:X; and

b) genetic variants of the DNA sequence of part a); and

c) DNA sequences capable of hybridizing to one of the sequences of parts a) and b), above.

1 0. An isolated DNA sequence encoding a protein having metallo protease activity, which is obtainable from a filamentous fungus and which is selected from the group consisting of:

a) the DNA sequences of the pepl and pepJ genes of Aspergillus nidulans as shown in SEQ ID NO:X and SEQ ID NO:X, respectively; and

b) genetic variants of the DNA sequences of part a); and

c) DNA sequences capable of hybridizing to one of the sequences of parts a) and b), above.

1 1 . A DNA construct comprising a DNA sequence according to claims 7 or 8.

1 2. A DNA construct according to claim 1 1 wherein the DNA sequence comprises a disruption that results in the DNA sequence being incapable of encoding an active metallo protease.

1 3. A process for the preparation of a metallo protease deficient filamentous fungus comprising the steps of:

a) transforming a filamentous fungus mutant with a DNA construct according to claim 1 2; and

b) selecting a transformed filamentous fungus with reduced metallo protease activity.

14. A DNA construct according to claim 1 1 wherein the DNA sequence is operable linked to regulatory regions suitable for expression of the DNA sequence in a filamentous fungal host.

1 5. A filamentous fungus comprising a DNA construct according to claim 14.

16. A process for the preparation of a filamentous fungal metallo protease, wherein the process comprises the steps of:

a) culturing the filamentous fungus of claim 1 5 under conditions conducive to the expression of the DNA sequence, and b) recovering the metallo protease.