CA2378016A1 - Use of aspergillus sojae as host for recombinant protein production - Google Patents

Abstract

A recombinant Aspergillus sojae comprising an introduced acetamidase S (amdS ) gene as selectable marker is disclosed. An Aspergillus sojae exhibiting grow th with medium comprising uracil and fluoro-orotic acid, said Aspergillus sojae further not exhibiting growth on medium comprising uridine and fluoro-orotic acid i.e. said Aspergillus sojae exhibiting uracil auxotrophy, said Aspergillus sojae being unable to utilize uridine, said Aspergillus sojae being pyrG negative, said Aspergillus sojae exhibiting resistance to fluoro- orotic acid, said uracil auxotrophy and said fluoro-orotic acid resistance being relievable upon complementation with an active introduced pyrG gene, i s described. The Aspergillus sojae further comprises a nucleic acid sequence encoding a phytase or a protein having phytase activity or any other heterologous protein or polypeptide and can be used for the biotechnological production of said phytase or said other heterologous proteins or polypeptides. Additional mutants exhibiting amended morphology are also disclosed. Methods of producing such expression hosts are described.

Description

Novel means of transformation of fungi' and their use for heterologous protein production.
SUMMARY OF THE INVENTION
The invention relates to novel means of transformation of fungi and to their use for production of heterologous proteins. The means involve genetically engineered fungi belonging to the taxonomic group Aspergillus sojae. Suggestions have been provided in the past to use Aspergillus sojae as a host strain for transformation. However to date no data are provided on successful transformation and/or expression of heterologous proteins.
In addition it has been found, that so far certain proteins, such as phytase which were difficult to express in large amounts, due to several reasons including proteolytic degradation in expression hosts other than Aspergillus sojae, can surprisingly be expressed in Aspergillus sojae. Production levels for heterologous proteins in Aspergillus sojae have been found to exceed those levels achieved for the same proteins in Aspergillus niger and Aspergillus awamori. In addition to the above, the subject of the invention further covers a process for obtaining improved Aspergillus sojae strains for expression purposes, characterized by, on the one hand, a decreased proteolytic activity, and, on the other hand improved fermentation characteristics related to the morphology of the fungi.
BACKGROUND OF THE INVENTION
Suggestions have been provided in the past to use Aspergillus sojae as a host for transformation. However, to date no data are provided on successful transformation and/or production of heterologous proteins and, more specifically, nothing is revealed concerning expression of phytase. Previously, expression levels were too low in expression hosts other than Aspergillus sojae, mainly due to proteolytic degradation. We have now found expression levels for the protein in Aspergillus sojae that exceed those levels achieved for the same protein in other strains, e.g. Aspergillus niger, Aspergillus awamori and Trichoderma. It is surprising to find such an improvement in closely related strains. Thus, prior art disclosures concerning phytase production exhibit shortcomings.
Prior art disclosures on the use of Aspergillus sojae for expressing heterologous proteins or polypeptides were inadequate.
The fact that until now hardly any successful attempts for A. sojae transformation have been reported is remarkable in view of the fact that numerous successful transformations of closely related strains of the taxonomic group Aspergillus oryzae have been described in the past. On the basis of this close relationship the skilled person would anticipate (and in fact did anticipate) that analogous methods to those used for Aspergillus oryzae are also applicable for Aspergillus sojae strains.
W097/04108 for example describes the isolation of a protease encoding nucleic S acid sequence, specifically a leucine aminopeptidase encoding sequence, and the transformation of a variety of host organisms, i.a. Aspergillus sojae, with a leucine aminopeptidase encoding sequence. However no illustration of this particular transformation actually having been carried out is provided. It is merely suggested as one possibility among many other strains such as Trichoderma reesei, Aspergillus niger, Aspergillus awamori, Aspergillus foetidus, Aspergillus japonicus, Aspergillus phoenicis, and Aspergillus oryzae as a potential host strain to be used for transformation. In the cited document, 3 transformation protocols readily used in the art are suggested for the strains.
Specifically any of the selection markers acetamidase S (=amdS) (e.g. as maintained on vector p3SR2), argB or hygromycin B (e.g. using the vector pAN7-1) are suggested as being suitable markers to be used according to the transformation protocols described therein.
The use of vector p3 SR2 with the amdS marker has frequently been described in the literature as being useful for transforming various strains, for example Aspergillus oryzae (in EP 0.238.023), Trichoderma reesei (in EP 0.244.234) and Aspergillus niger (EMBO Journal 4, pages 475-479). Consequently, the analogous use for transforming Aspergillus in general is put forward in W097/04108 on the basis of these previous publications.
Quite specifically on page 17 of W097/04108 it is described that Aspergilli and Trichoderma that prior to transformation grew slowly on minimal medium comprising solely the substrate acetamide as source of nitrogen could be selected after transformation with the vector p3SR2 due to a clear growth advantage. Subsequently, the thus obtained transformants would need to be further subjected to selection for leucine aminopeptidase (=LAP) productivity in order to fmd a desired transformant. As stated above this is merely put forward as speculative means of transformation applicable over the two aforementioned genera in toto based on a few successful transformations of strains other than Aspergillus sojae.
The suggested transformation protocol is, however, unsuccessful with Aspergillus sojae. The selection criteria described in the prior art are insufficient to ensure practical selection of desirable transformants when using the vector p3SR2. We have conducted the . .
experiments and found the described method inoperable due to excessive background growth eliminating practical selectability.
Another routinely used selection method for fungal transformants is that of transformation of orotidine-5-monophosphate decarboxylase (=Pyre) mutants.
Mattern et al. in Mol. Gen. Genet. 210, pages 460-461 disclose transformation of Aspergillus oryzae using the Aspergillus niger pyre gene. Standard practice is to isolate pyre mutants based on direct resistance to fluoro-orotic acid as a positive selection marker.
This has resulted in isolation of numerous pyre mutants for a variety of fungi to date.
From experience with a number of different filamentous fungi, the auxotrophic pyre-based system has many favourable characteristics. Experiments were carried out to obtain A. sojae pyre mutant strains, using a standard procedure based on direct selection for resistance to fluoro-orotic acid (FOA) on plates containing uridine to support growth of the mutant strain (Van Hartingsveldt et al. in Mol. Gen. Genet. (1987) 206, pages 71-75).
However, use of the analogous method on Aspergillus sojae strains did not lead to pyre mutants. The usual method did lead to fluoro-orotic acid resistant strains but all the strains were able to grow without uridine. Thus, none of these strains were pyre mutants.
Normally, the isolation of the pyre mutants can be done directly from the fluoro-orotic acid resistant strains on a uridine selection medium. For Aspergillus sojae this method however turned out to be inoperable.
Clearly, Aspergillus sojae exhibits different traits than the closely related Aspergillus oryzae when it comes down to transformation. The standard protocols using amdS
or pyre as selectable markers do not suffice. Unfortunately, the method of argB as selectable marker is not an attractive option either, since this requires isolation of a corresponding argB mutant for every host strain one wishes to use. This is an arduous task based on trial and error. The required argB mutant can be obtained through random mutagenesis followed by screening of tens of thousands of colonies. The situation for pyre is better in that the mutant is itself selectable. In the case of amdS no mutant is required as the presence of amdS works as dominant selectable marker.
Additional problems dissuading the skilled person from use of Aspergillus sojae as expression host for recombinant proteins or polypeptides exist. In JP-A-02-234666 for example an ArgB based selection of Aspergillus sojae is described using an analogous protocol to that described for other fungi. Such a process has been described for Aspergillus oryzae in Biotechnology (1988) 6, pages 1419-1422. The cited article also refers to successful analogous transformation of Aspergillus nidulans and Aspergillus niger. However, when the Aspergillus sojae strain ATCC42251 disclosed in the Japanese patent application was analysed, an undesirable protease profile was found.
The protease profile of this strain is incompatible for application as a production host.
So even though a transformation protocol has been suggested in the prior art for this particular Aspergillus sojae strain it could not possibly lead to a high level of expression of heterologous protein even if the protocol for transformation was successful.
It is in fact due precisely to the explicit characteristics of Aspergillus sojae strains to produce excessive amounts of alkaline proteases and amylases that they currently find application in practice. They are used specifically in processes requiring degradation of complex polymeric substrates. It was thus at best to be expected that any transformants of Aspergillus sojae that are finally successful will not lead to good expression levels unless the product is an Aspergillus sojae protein that is impervious to its own proteases.
In summary the problems facing the skilled person in fording a means to use Aspergillus sojae strains for expressing heterologous recombinant proteins on an industrial scale are manifold. Firstly, a number of processes for introducing the desired nucleic acid material to be expressed are not applicable in the manner used for other fungi. This includes pyre- and amdS-based processes that are useful for the closely related Aspergillus oryzae. Secondly, it remained to be seen whether high level production of heterologous proteins would be feasible despite the known excessive proteolytic activity of the host strain Aspergillus sojae.
Unexpectedly, it has been found that the problems addressed above can be solved, thus resulting in novel expression hosts for producing proteins and novel methods of production of heterologous proteins. We describe transformation of A. sojae strains with the amdS and pyre selection markers. In addition efficient gene expression is described, including expression of a phytase gene.
DESCRIPTION OF THE INVENTION
As stated the subject invention is directed at Aspergillus sojae strains and the application thereof for production of recombinant proteins and polypeptides.
Firstly, a description of Aspergillus sojae strains is provided.
Aspergillus sojae determination.
The fungal taxonomy is a complex issue. The Aspergillus genus comprises Aspergillus sojae in the Flavi/Tamarii section (see Table 1). A. sojae is clearly shown to be distinct from A. oryzae which is located in the same section (see Table 2).
Currently, strains belonging to Aspergillus sojae can be distinguished from taxonomically closely related Aspergillus oryzae and also closely related Aspergillus parasiticus strains in a number of manners recognised in the art. Reference is made to the random PCR
fragments, 5 ver-1, aflR and rDNA sequences as described, respectively, in Ushijama et al. (1981), Chang et al. (1995), Yuan et al. (1995), Kusomoto et al. (1998) and Watson et al. (1999).
In addition it has been found that Aspergillus oryzae further differs from Aspergillus sojae upon comparison of the alpA sequence of these strains. Inter alia (there are other sequence differences between A. oryzae and A. sojae alpA which could be used as a determination tool), it has been found that Aspergillus sojae comprises an XmnI restriction site at a specific location in the alpA gene. The corresponding location in the alpA
gene of several Aspergillus oryzae strains does not possess such a restriction site. Thus, this provides an additional discrimination point between the two types of fungal strains.
Consequently, numerous methods are available to the skilled person to assess whether a strain is an Aspergillus sojae. Currently more than 10 strains are deposited with the ATCC
that are defined as Aspergillus sojae. The 10 oldest deposits have been analysed. Two out of 10 did not pass the lastly mentioned determination test. One of them is the ATCC20235 which according to Ushijama et al. (1981) also did not fulfil the requirements for classification as an Aspergillus sojae on the basis of morphological parameters. The other is ATCC46250.
The definition of Aspergillus sojae as used throughout the patent application is meant to imply a strain that preferably fulfils all the requirements described in the cited references in combination with the presence of the XmnI restriction site in the alpA gene.
Specific homologous primers for both the Aspergillus oryzae and Aspergillus sojae sequences are also provided. They can be used to test for the presence of the XmnI
restriction site by way of example of a screening test useful for distinguishing Aspergillus oryzae from Aspergillus sojae (Primer sequences are SEQ ID No.l MBL1784: 5'-CGGAATTCGAGCGCAACTACAAGATCAA-3' and SEQ ID No.2 MBL1785: 5'-CGGAATTCAGCCCAGTTGAAGCCGTC-3'). They are derived from the coding region of the alpA gene. It will be obvious to the skilled person on the basis of the known sequence data that alternative probes or primers are conceivable. PCR
amplification using these primers on Aspergillus DNA, followed by restriction enzyme digestion of the resulting DNA fragments with XmnI provides a way to discriminate A. sojae strains from A. oryzae strains. Having established the definition of Aspergillus sojae strains we can proceed further with the detailed description of the invention.

The invention in one aspect covers a recombinant Aspergillus sojae comprising an introduced acetamidase S (amdS) gene as a selectable marker. Such an A. sojae is selectable on a medium comprising a substrate for the introduced amdS protein as sole source of nitrogen, said medium further comprising a carbon substrate and said medium being free of endogenous amdS inducing substrate. A suitable medium comprises acrylamide as substrate for the introduced amdS as sole source of nitrogen. A
suitable medium at least further comprises minimum substrates required for growth of Aspergillus sojae. A suitable category of A. sojae according to the invention is formed by A. sojae that are not selectable on acetamide comprising medium. An A. sojae according to the invention is suitably an A. sojae selectable on a medium free of glucose, i.e.
a medium wherein the carbon source is not glucose. Such a medium can be a medium having sorbitol as carbon source. Best results in the case of sorbitol are achieved when sorbitol is the sole carbon source.
An Aspergillus sojae according to the invention may comprise a further introduced nucleic acid sequence, said further introduced sequence preferably encoding a protein or polypeptide. The further introduced sequence may be adapted for optimised codon usage to the host strain codon usage or may have the original codons from the host from which it has been derived. The introduced sequence is in principle any sequence the skilled person wishes to express. The introduced sequence can suitably be heterologous, i.e.
foreign to the Aspergillus sojae into which it is introduced. It can also be native but introduced in the form of one or more additional copies.
One of the subjects of the invention is aimed at expressing phytase or proteins having phytase activity. Numerous sequences are known to the skilled person concerning sequence data of phytases. We refer to and incorporate by reference the contents of EP
684.313, EP 897.010, WO 99/49022, EP 911.416 and EP 897.985. These documents describe various natural and modified phytase sequences. They also describe a consensus sequence. A suitable embodiment is formed by phytase sequences from Peniophora being either the natural sequences or modified versions thereof. The new system is more flexible than prior systems and thus heterologous sequences, including heterologous sequences encoding phytase or proteins having phytase activity that were difficult to express in the prior art fungal systems can be expressed in the novel system according to the invention.
An Aspergillus sojae according to the invention as defined in any of the embodiments defined above comprising an introduced amdS gene as selectable marker may suitably have no active endogenous amdS gene. The Aspergillus sojae according to such an embodiment may by way of example have an endogenous amdS gene comprising an endogenous amdS
inactivating mutation. Any type of inactivating mutation known or conceivable to the skilled person may have occurred. A suitable example of such inactivating mutation may be a deletion or disruption. The mutation may inactivate the gene or the gene product. The skilled person will realise that numerous options are available to achieve this and that they can readily be achieved.
In an alternative embodiment the invention is also directed at a recombinant Aspergillus sojae free of an active endogenous amdS gene and further comprising an introduced amdS gene as selectable marker. The recombinant Aspergillus sojae according to the invention is selectable on a medium comprising a substrate for the amdS
as sole source of nitrogen, said medium further comprising a carbon substrate. A
suitable medium at least further comprises minimum substrates required for growth of A. sojae.
In a suitable embodiment the endogenous amdS gene can for example have been inactivated.
This inactivation can be any type of inactivation known or conceivable to a person skilled in the art that still leaves the A. sojae viable. By way of example the endogenous amdS gene can comprise an inactivating mutation in the form of a substitution, deletion or insertion of the gene or part thereof, or by virtue of a mutation affecting expression of the gene such as to render it inactive. The complete endogenous amdS gene can also be absent.
An Aspergillus sojae in any of the described embodiments according to the invention may be an A. sojae into an amdS gene has been introduced. This can be achieved e.g. by transformation or transfection. The resulting Aspergillus sojae according to the invention must then subsequently have been separated from non transformed or transfected A. sojae.
Any of the embodiments described above as such or in combination are covered by the invention.
The invention not only covers Aspergillus sojae as such, but also covers a method of introducing a nucleic acid sequence into A. sojae. The method comprises subjecting Aspergillus sojae to introduction of a nucleic acid sequence in a manner known per se for introduction of a nucleic acid sequence into a fungus. Such a manner can e.g.
be trans-formation or transfection of the A. sojae. The method comprises the introduction of the amdS gene as the nucleic acid sequence followed by selection of the resulting transformed or transfected A. sojae on a medium free of endogenous amdS inducing substrate, said medium further comprising a substrate for the introduced amdS as sole source of nitrogen and said medium further comprising a carbon substrate, said medium enabling the desired A. sojae comprising introduced amdS gene to grow whilst eliminating growth of A. sojae devoid of a functional amdS gene. A suitable embodiment of such a method involves applying a medium comprising a substrate for amdS other than acetamide.
Suitably, such a medium comprises acrylamide as substrate for the introduced amdS as sole source of nitrogen. Suitably, a medium for the method according to the invention comprises a carbon source other than glucose. Suitably, a medium for use in a method according to the invention comprises sorbitol as carbon source, preferably as sole carbon source. A suitable medium at least further comprises minimum substrates required for growth of A.
sojae.
A method according to the invention as defined above in any of the embodiments comprises introduction of an additional nucleic acid sequence besides the amdS
gene. The additional nucleic acid sequence for example encodes a protein or polypeptide, such as a phytase or proteins having phytase activity. The sequence does not necessarily have to be a non Aspergillus sojae sequence, but can also include A. sojae derived sequences It is however intended to indicate that the sequence that is introduced is absent in the non transformed strain or else is present in a lower copy number than in the A.
sojae according to the invention.
Naturally, the subject invention also covers any Aspergillus sojae obtained by the method described above. Basically, the method is directed at introducing a sequence capable of realising the presence of sufficient active amdS to function as selectable marker as opposed to the A. sojae into which the sequence is introduced which cannot for some reason or another produce sufficient active amdS to enable growth on a substrate for amdS
as sole source of nitrogen.
A method of selecting transformed or transfected A. sojae also falls within the scope of the invention. The method comprises subjecting A. sojae (with no active endogenous amdS
gene as defined according to any of the embodiments described) to a method of transformation or transfection of the A. sojae in a manner known per se for transformation or transfection of fungi with a nucleic acid sequence. The method comprises the introduction of an amdS gene as the nucleic acid sequence, followed by selection of the resulting transformed or transfected A. sojae, said selection occurring on a medium comprising a substrate for the introduced amdS as sole source of nitrogen, said medium further comprising a carbon substrate, said medium enabling the desired A.
sojae to grow whilst eliminating growth of non transformed or transfected A. sojae due to inability of such to grow without the introduced amdS gene on the selection medium. A
suitable medium at least further comprises minimum substrates required for growth of A.
sojae.
The invention is also directed at a method for producing recombinant Aspergillus sojae. This method comprises introducing a desired nucleic acid sequence e.g.
by transformation or transfection in a manner known per se into an A. sojae, said desired nucleic acid sequence being flanked by sections of the endogenous amdS gene of a length and homology sufficient to ensure recombination. The introduction is followed by selection of the recombinant A. sojae having the desired nucleic acid sequence. The selection occurs for a selectable marker comprised in or transformed in cotransformation with the desired nucleic acid sequence, said selectable marker being absent in the A. sojae prior to introduction of the desired nucleic acid sequence. The flanking sequences may also be sequences corresponding to the endogenous amdS gene sufficient to ensure recombination. The skilled person can readily assess which sequences will suffice on the basis of hybridisation knowledge and the sequence data of the endogenous amdS
gene. The recombination event eliminates the endogenous amdS activity in both cases. The selectable marker can quite suitably be pyre, with, however, uracil instead of uridin in the selection medium.
1 S A further embodiment of the invention comprises Aspergillus sojae exhibiting growth on medium comprising uracil and fluoro-orotic acid, said A. sojae further not exhibiting growth on medium comprising uridine and fluoro-orotic acid. This means that the A. sojae exhibits uracil auxotrophy, is unable to utilize uridine, is pyre negative and exhibits resistance to fluoro-orotic acid. The uracil auxotrophy and the fluoro-orotic acid resistance are relievable upon complementation with an active introduced pyre gene. Such an A.
sojae according to the invention can be free of active endogenous pyre genes.
The pyre negative A. sojae according to the invention may comprise an endogenous pyre gene with a mutation inactivating it. The mutation can be any mutation known or conceivable to a person skilled in the art, said mutation inactivating a pyre gene or the expression product thereof. Such a mutation can by way of example be in the form of an insert of a nucleic acid sequence in the gene, a substitution of a part of the encoding sequence of the gene, a deletion of a part of the encoding sequence of the gene or a deletion of the whole encoding sequence of the gene. The mutation can also occur in the regulating part of the gene. In the case of Aspergillus sojae according to the invention with a mutated pyre gene, said Aspergillus sojae can have a nucleic acid sequence for the mutated pyre gene different to that of the wild type A. sojae pyre gene. A further embodiment comprises pyre negative A. sojae according to the invention as described in any of the above embodiments which further comprise any of the characteristics described for any of the amdS
variant A. sojae according to the invention as such or in combination.

A method of selecting transformed or transfected Aspergillus sojae also falls within the scope of the invention. The method comprises subjecting A. sojae of the pyre negative type according to any of the embodiments of the invention as described above to a method of transformation or transfection with a nucleic acid sequence, said method comprising 5 introducing an active pyre gene into the pyre negative A. sojae in a manner known per se for transformation or transfection. The introduction step is then followed by selection of the resulting transformed or transfected A. sojae on a medium free of uracil and fluoro-orotic acid, said medium at least further comprising minimum substrates required for growth of A. sojae, said medium enabling the desired A. sojae to grow whilst eliminating 10 growth of non-transformed or -transfected A. sojae due to inability of such to grow without uracil due to the inactivated pyre gene. In a suitable embodiment of such a method the active pyre gene that is introduced is flanked by identical nucleic acid sequence fragments, and the pyre positive A. sojae resulting from the introduction of the pyre gene and the flanking sequences is selected on a medium free of uracil and fluoro-orotic acid.
Subsequently the pyre positive A. sojae is cultivated on medium comprising uracil and fluoro-orotic acid, thereby eliminating the pyre gene that had been introduced and thus resulting in a pyre negative A. sojae that is selectable by growth on uracil comprising medium and fluoro-orotic acid resistance. In a suitable embodiment of the aforementioned method the flanking sequences and the pyre gene are further flanked by sequences that direct integration of the pyre gene and the flanking sequences into a specific location, due to the fact that the integration directing sequences are homologous to a specific sequence of the A. sojae to be transformed. This enables knock out (if desired) of the gene associated with the specific sequence. The process of knock-out mutant creation as such is well known to the person skilled in the art. Any of the embodiments of the selection method just described may further comprise the step wherein the Aspergillus sojae is transformed or transfected with a further heterologous nucleic acid sequence. The further heterologous nucleic acid sequence preferably encodes a protein or polypeptide and the same remarks are valid here as made elsewhere in this description for the nature of such further nucleic acid sequences for the other embodiments of Aspergillus sojae and fungi in general according to the invention. The further sequence can be introduced with the active pyre gene either on the same vector or by cotransformation with the active pyre gene that is introduced. The method of selecting transformed or transfected A. sojae as described may also be carried out in combination with the method for introducing a nucleic acid comprising introduction of a heterologous amdS gene in any of the embodiments according to the invention disclosed therefore above. Naturally, the invention covers any recombinant A. sojae obtained by the method of selecting transformed or transfected A.
sojae according to the invention.
The invention is also directed at a method for producing recombinant Aspergillus sojae, said method comprising transformation or transfection in a manner known per se of a pyre positive Aspergillus sojae with a nucleic acid sequence comprising the sequence to be introduced flanked by sections of the pyre gene or corresponding sequences of a length and homology sufficient to ensure recombination eliminating the pyre gene and introducing the desired sequence, followed by selection of the recombinant Aspergillus sojae with the desired sequence by selecting for the A sojae with a pyre negative phenotype. Determination of the corresponding sequences lies within the reach of the skilled person by virtue of their knowledge of hybridisation processes with nucleic acid sequences and their knowledge of required sequence data of the pyre genes.
In particular the invention also covers such Aspergillus sojae exhibiting the characteristics of the amdS variant A sojae according to the invention as defined above.
Thus any Aspergillus sojae strain obtained by either the amdS and/or pyre introduction method according to the invention is a novel strain falling within the scope of the invention as is any subsequent use of such a novel strain. Such a novel strain can comprise nucleic acid sequences that do not occur in the original corresponding Aspergillus sojae strain or even do not occur in Aspergillus sojae, Aspergilli or fungi. The sequences can be of mammalian origin or derived from any animal, plant or microbe. Nucleic acid sequences can also be expressed that are naturally present in the Aspergillus sojae strain but that are present in a lower copy number in the corresponding non-transformed A sojae.
Thus the production of homologous proteins is also covered by the invention when pyre and/or amdS Aspergillus sojae strains according to the invention are involved. A
preferred embodiment is that wherein the particular protein or polypeptide to be produced is absent in the corresponding non-treated A. sojae and/or is present in a lower copy number in the corresponding non-treated A. sojae, i.e. the A. sojae prior to introduction of the nucleic acid sequence. Expression of heterologous proteins by any of the novel strains of Aspergillus sojae in a manner known per se for producing protein or polypeptide in a fungus thus covers both sequence native to the strain and foreign to the strain. Basically, only the native non-transformed or -transfected A sojae is excluded from protection. A
process of production comprises cultivating the fungus under suitable conditions for expression of the desired sequence to occur. The process of production optionally includes the step of isolation of the resulting polypeptide or protein in a manner known per se for protein or polypeptide production by fungi. Preferably the protein or polypeptide will be secreted into the culture medium.
A preferred protein or polypeptide is a protein or polypeptide susceptible to S degradation upon expression by Aspergillus niger or Aspergillus awamori. A
number of such proteins and polypeptides have already been disclosed in the prior art and a large number remain yet to be determined. Such determination is however a matter of routine for the skilled person. Another preferred embodiment of the protein or polypeptide to be expressed is one whereby the protein or polypeptide differs from an Aspergillus sojae protease and amylase. A preferred embodiment involves a non Aspergillus sojae protein or polypeptide.
A particularly interesting embodiment comprises a combination of the two processes for introducing nucleic acid sequences according to the invention as described above. The advantage thereof lies in the fact that the frequency of transformation obtained with the pyre marker is clearly much higher than that of the amdS marker.
However, secondary screening of the pyrG+ strains for the best growth on acrylamide selective plates allows the identification of those recombinant Aspergillus sojae showing the highest copy number and thus most likely the highest level of gene expression.
As indicated in the examples homologous and heterologous expression regulating sequences can be used by Aspergillus sojae i.e. natively occurring sequences of the strain itself or sequences foreign to the strain can be used. Thus the transformants according to the invention can comprise any such regulatory sequences. The selection of the suitable regulatory region is a matter of choice that lies well within the range of the standard capabilities of the skilled person and will depend on the particular application. The regulating sequences can be constitutive or inducible. The regulating sequences can be fungal or non-fungal. A broad range are exemplified in the examples. A large number of expression regulating sequences are regularly used in the art for other systems, in particular fungal systems such as Aspergilli, and can routinely be applied without undue burden in the Aspergilli according to the invention.
For introducing the desired nucleic acid sequences into Aspergillus sojae any vector may be used that is suitable for introducing nucleic acid sequences into fungal host cells. Numerous examples are available in the art. In particular vectors that have been found suitable for transformation, transfection or expression in Aspergilli such as Aspergillus niger, Aspergillus awamori and Aspergillus oryzae can suitably be applied.

In addition to the above the subject invention describes efficient protein production for recombinant Aspergillus sojae. Such efficient production is disclosed in those strains having a protease profile superior to ATCC42251 or at least as good as any of ATCC9362, ATCC11906 and ATCC20387. The subject description thus reveals that some known strains of A. sojae are well suited already as such for production of proteins, polypeptides and metabolites. These Aspergillus sojae strains exhibit a lower proteolytic activity than the reference strain A. sojae ATCC42251. In particular the two known strains and ATCC20387 are well suited. So preferred A. sojae strains for production of proteins, polypeptides and metabolites will be those expressing equal to or less proteolytic activity than the two preferred strains. Strain ATCC 11906 is the best embodiment of the deposited ATCC A. sojae strains according to the prior art. Suitable proteins or polypeptides will be produced. Now that the subject invention has enabled introduction of nucleic acid sequences, such can serve to provide any protein or polypeptide of choice using an A. sojae as expression host.
The subject invention offers an improvement over existing expression systems.
A
number of existing protein production systems exhibit expression problems due to proteolysis. In particular the new system is better than the currently frequently applied expression systems Aspergillus niger and Aspergillus awamori. The subject invention now renders it possible to provide a recombinant Aspergillus sojae comprising a introduced nucleic acid sequence encoding a protein or a polypeptide for expression, said protein or polypeptide being susceptible to degradation upon expression by A. niger or A.
awamori.
The invention also provides a recombinant A. sojae comprising an introduced nucleic acid sequence encoding a protein or polypeptide for expression, said protein or polypeptide being other than A. sojae protease and amylase. A preferred embodiment is that wherein the introduced nucleic acid sequence encodes a non-A. sojae protein or polypeptide. Such recombinant A. sojae strains also fall within the scope of the invention.
In addition, illustration of Aspergillus sojae strains that have been modified in order to enhance their suitability as expression hosts is currently provided. These modifications can be reduced proteolytic activity as induced by any means. Specifically, the use of UV
random mutagenesis is illustrated. Also specific mutation of one or more protease genes is illustrated. The means by which mutations can be introduced are common knowledge to the skilled person, and numerous alternative embodiments are thus readily available to arrive at the desired mutants. A suitable embodiment is formed by mutants in which alkaline proteolytic activity has been reduced. In particular elimination of activity of specifically the major 35 kDa alkaline protease is illustrated as ensuring increased expression of proteins and polypeptides. Specifically the invention thus also covers novel strains exhibiting reduced proteolytic activity, specifically reduced alkaline proteolytic activity. Such strains are obtainable using any specific mutation route known or conceivable to the skilled person. A preferred embodiment of such expression hosts exhibiting reduced proteolytic activity as described above further comprises a selectable marker. Quite suitably the selectable marker will be amdS, pyre or a combination thereof.
The invention in particular covers a method of producing protease deficient mutants of A. sojae by knocking out the 35 kDa alkaline protease gene. There are numerous ways in which this can potentially be achieved on the basis of the sequence data provided for this gene. In particular a method using recombination with a pyre selection marker linked to two flanking regions eliciting cross over of the 35 kDa alkaline protease gene, whereby the resulting strain has the pyre selection marker and misses the 35 kDa alkaline protease gene is an elegant one. Subsequently the pyre selection marker can be eliminated, thus providing a 35 kDa alkaline protease negative Aspergillus sojae mutant that can be used for expression purposes of any desired sequence to be introduced therein.
Naturally, the sequence to be introduced can have been incorporated in the previous steps already either on the same vector as the pyre marker or in a cotransformation event. Also the method can be carried out analogously where a different protease gene than the 35 kDa alkaline protease gene is to be knocked out. The analogous measures to be taken are obvious to the skilled person on the basis of the illustration provided herein in combination with knowledge of other protease sequences. Also analogously the amdS
selectable marker can be used in accordance with the invention as described elsewhere in this description.
Mutant fungi exhibiting improved fermentation characteristics are also provided as an additional aspect of the subject invention. Specifically, the invention is directed at a fungus comprising a mutation inhibiting the activity of proprotein convertase or an equivalent protein. Numerous proprotein convertases are known in the art. In particular we refer to Figure 1 providing sequence data of a number of such proteins. A
fungus according to the invention is suitably selected from Agaricus, Aspergillus, Trichoderma, Rhizopus, Mucor, Phanerochaete, Trametes, Penicillium, Cephalosporium, Neurospora, Tolypocladium and Thielavia. Particularly suitable fungi are Aspergillus niger, Aspergillus foetidus, Aspergillus sojae, Aspergillus awamori, Aspergillus oryzae, Trichoderma reesei, Penicillium chrysosporum, Cephalosporium acremonium, Neurospora crassa, Tolypocladium geodes and Thielavia terrestris. A preferred embodiment covers the mutant WO 01/09352 PCT/N_ L00/00544 when it is an Aspergillus sojae, most particular preference is extended to Aspergillus sojae as defined above according to the invention, i.e. comprising heterologous nucleic acid sequences, e.g. in combination with the selectable markers amdS and/or pyre.
A suitable equivalent of a proprotein convertase is a protein or polypeptide 5 exhibiting an amino acid sequence with more than 40%, preferably more than 45%
similarity or identity with the inferred amino acid sequences of the DNA
sequences given in SEQ ID No. 3 (= gene fragment encoding A. niger proprotein convertase amino acid sequence), SEQ ID No. 4 (= partial gene fragment encoding Aspergillus sojae proprotein convertase amino acid sequence) or with any of the sequences given in Seq ID
Nos. 5 to 9.
10 The functionally equivalent protein may suitably have a nucleic acid sequence capable of hybridising under stringent conditions to a nucleic acid sequence according to SEQ ID
Nos. 3 to 9. Stringent hybridisation conditions can readily be determined by the skilled person. A suitable example of stringent hybridisation conditions are hybridisation at 50°C
and preferably at 56°C and final washes at 3xSCC. PE4, PCLI and PCL2 are specifically 15 mentioned as examples of suitable oligonucleotide mixtures corresponding to the coding strand (i.e. SEQ ID Nos. 10, 11 and 12). For the noncoding strand PE6, PCL2-rev, PCL3 and PCL4 are mentioned (i.e. SEQ ID Nos. 13, 14, 1 S and 16, respectively).
Use of these primers in amplification procedures common in the art will provide equivalent sequences and such use and the resulting newly found sequences and application thereof in the manner analogous to that described in the subject description fall within the scope of the invention. The sequences for which the oligonucleotides were made are well conserved as could be determined from comparison of the various amino acid sequences for the proteins provided (see Figure 1). Any other nucleic acid sequences exhibiting the same or higher degree of identity, similarity or homology with the sequences provided in the subject patent application for the proteins or relevant active parts thereof are covered by the invention as is the use thereof as primers or probes to find other proprotein convertase or equivalent protein encoding sequences and/or for subsequently introducing mutations in such protein encoding sequences. By way of example Maniatis et al. (1982) Molecular Cloning, A Laboratory manual, Cold Spring Harbor Laboratory, New York or any other handbook on cloning and/or screening nucleic acid sequences has been referred to. The equivalent protein or polypeptide will exhibit the activity of a proprotein convertase as the one having an amino acid sequence according to SEQ ID Nos. 3 to 9. The mutant fungus can comprise a substitution, insertion or deletion in the encoding sequence of the proprotein convertase or equivalent protein. The mutant fungus can suitably comprise a mutation in the regulation of the expression of the gene encoding proprotein convertase or equivalent protein. A mutant fungus according to the invention in a suitable embodiment exhibits reduced viscosity vis a vis the corresponding non mutated fungus under equivalent cultivation conditions. A mutant fungus according to any of the above embodiments exhibiting increased expression of a desired introduced nucleic acid sequence encoding a protein or polypeptide is included within the scope of the invention, said fungus exhibiting increased production of a protein or polypeptide under equivalent conditions vis a vis the corresponding wild type fungus. The activity site for the A. sojae proprotein convertase has been ascertained to be comprised within the amino acid sequences inferred by SEQ ID
Nos. 3 and 4.
A process for producing a phytase or protein having phytase activity or any other protein or polypeptide, preferably a recombinant phytase or any other heterologous protein or any other polypeptide, said process comprising cultivating a mutant fungus according to any of the embodiments described above falls within the scope of the invention. A process for obtaining the resulting protein or polypeptide either from the cell as such or after secretion thereof from the cell is also included.
The use of any of the described novel strains for transformation of any nucleic acid sequence encoding a phytase or protein having phytase activity or any other protein or polypeptide thereto and any subsequent expression of any nucleic acid sequence introduced therein and also optionally any following processing and/or secretion and/or isolation is covered by the invention.
Any phytase or phytase-like or any other heterologous protein or polypeptide encoding sequence can suitably be used. This can be of fungal or non-fungal origin. A
preferred embodiment is formed by acid labile protein or polypeptide encoding sequences.
Suitably the protein encoding sequence encodes non protease-like proteins. The examples show a phytase sequence and a number of heterologous sequences suitable for use in transformation and also for expression in Aspergillus sojae hosts. Further examples of suitable proteins to be expressed are obvious to a person skilled in the art.
The invention is further illustrated by the examples below. The examples are not to be considered restrictive to the interpretation of the scope of the invention.
Alternative embodiments are readily envisageable to the skilled person on the basis of the description and knowledge of the relevant field of technology. The content of references cited in the description are incorporated by reference. The claims serve to illustrate the intended scope of the invention.

EXPERIMENTAL DETAILS CONCERNING THE INVENTION
CONSTRUCTION OF AN Aspergillus sojae GENE LIBRARY.
Genomic DNA of A. sojae was isolated from protoplasts obtained from ATCC 11906 using a previously described protocol (Punt, van den Hondel, 1992).
After isolation, DNA was extracted from the protoplasts using the protocol described by Kolar et al., 1988. Subsequently the DNA was partially digested with MboI to result in DNA
fragments of an average size of 30-50 kb.
Vector pAOpyrGcosarpl, which was used for the construction of the gene library, was constructed by ligation of a 3 kb BamHI-HindIII fragment from pANsCosl (Osiewacs, 1994) and a 3.2 kb Acc65I-HindIII fragment from pA04.2 (De Ruiter-Jacobs, 1989) in Acc65I-BamHI digested pHELPI (Gems et al., 1991). This cosmid vector carnes the A.
oryzae pyre selection marker and is self replicating in filamentous fungi.
MboI digested genomic DNA was ligated to BamHI-digested pAOpyrGcosarpl, and the ligation mixture was packaged into phage particles using the Stratagene Supercosl vector kit. In total 30,000 individual clones were obtained representing an approximate 30-fold representation of the A. sojae genome. Stocks (in 15% glycerol) of pools of the resulting clones were stored at -80°C for later use.
AMDS TRANSFORMATION METHOD AND TRANSFORMANTS.
Two currently used protoplasting protocols and transformation protocols [the modified OM-method (Melton et al., P.N.A.S. 81 (1984) 1470-1474) and the NaCI-method (Punt and Van den Hondel, Meth. Enzym. 216 (1993) 447-457)] were tested on the Aspergillus sojae strain ATCC9362. Both methods resulted in protoplasts, but the yield of viable protoplasts with the OM-method was clearly better. The overall yields were lower than normally obtained for A. niger. A pilot protoplasting/transformation experiment was carried out with all A. sojae strains using the OM method.
For transformation, vector p3SR2 (carrying the amdS marker) was used in combination with pAOpyrGcosARPl. This latter vector is a derivative of the autonomously replicating Aspergillus vector Arpl, which in all Aspergillus species tested so far, resulted in highly increased numbers of (instable) transformants when used as a cotransforming vector. For nearly all strains sufficient protoplasts (about 10E6-10E7 per transformation) were obtained. Analysis of appropriate AmdS selection conditions for the various A. sojae strains revealed vigorous growth of most strains on the commonly used selective acetamide medium. Clearly, the acetamide selection conditions proposed for A.
sojae amdS transformants as reported in W097/04108, were not appropriate for the selection of A. sojae transformants. Our experiments revealed, surprisingly, that AmdS+
transformants could only be selected with acrylamide selection. Even on selective acrylamide plates, a considerable background from non-transformed protoplasts was observed. Selection of primary transformants requires around three weeks and many of the initially selected putative transformants turned out to be false positives, only showing background growth after transfer to fresh selective acrylamide plates. To optimize selection of transformants attempts had to be made to reduce this background growth.
Improved results were obtained by omitting glucose from the selective plates.
In Table 3 the composition of the improved selection medium and the usual media is given.
Figures 2a, b and c show the background growth observed for selected strains on the selection medium described in W097/04108 and the improved acrylamide selection medium described in Table 3.
Further transformation experiments with the three selected A. sojae strains revealed that protoplasting efficiencies for ATCC11906 and ATCC20387 were better using the NaCI-method. Successful protoplasting was obtained using various commercially available protoplasting enzyme preparations such as NOVOZYM, Caylase, Glucanex, etc.
Based on the NaCI transformation protocol the three selected A. sojae strains were transformed with amdS selection vector p3SR2 or derivatives thereof. Using the modified acrylamide selection plates a number of vigorously growing transformants were obtained, while no growth was observed in a control transformation without DNA. Another approach to circumvent background growth of non-transformed mycelium is the elimination of the activity of the wild type A. sojae amdS gene. This can be achieved for example by disruption of the A. sojae amdS gene. As a first step specific DNA fragments carrying ATCC11906 amdS sequences were PCR-amplified using primers derived from published A. oryzae amdS sequences (Gomi et al.; 1991, Gene 108, 91-98). Previous experiments had shown that cloning by stringent hybridisation would be unsuccessful due to a low level of sequence conservation between A. nidulans and A. sojae amdS sequences. The expected fragment of about 1.6 kb, which should carry most of the coding region of the amdS gene, was obtained. Sequence analysis from both ends of the cloned PCR fragment (Figures 3a and 3b) confirmed the cloning of a part of the A. sojae amdS gene. The stringent hybridisation occurred at 56°C with final washes at 3xSSC. The cloned sequence was very similar to the published A. oryzae amdS sequence. Several hybridising clones (7 out of 10.000) were isolated from the ATCC11906 cosmid library in pAOpyrGcosarpl using the cloned ATCC 11906 amdS fragment as a probe. After subcloning a fragment carrying the complete amdS gene, a part of the amdS gene was replaced by a re-usable pyre selection marker to generate an amdS replacement vector. Transformation of this vector to Aspergillus sojae ATCC11906PyrG resulted in pyrG+ transformants. After subsequent analysis of these transformants on acetamide and acrylamide selection plates several of these transformants showed reduced background growth. Southern analysis of a few of these strains revealed that the expected gene replacement had occurred. One of these strains was used for subsequent transformation with the A. nidulans amdS gene using acrylamide selection plates and resulted in a number of amdS+ transformants.
PYRG TRANSFORMATION METHOD AND TRANSFORMANTS.
(1) Initial experiments For A. sojae, the standard experiments used in the prior art for other fungi to generate pyre mutants as described in the introduction resulted in numerous fluo-orotic acid (FOA)-resistant strains. However, all of these strains were able to grow on medium without uridine and were therefore not considered pyre mutants. With our final aim to isolate the appropriate mutant strains a number of alternative approaches were followed.
(2) Near-homologous gene disruption Based on the expectation that the pyre genes from A. sojae and A. oryzae are very similar in sequence (which was confirmed by Southern hybridisation carried out under stringent conditions), experiments were carried out to disrupt the A. sojae pyre gene with a mutant version of the A. oryzae pyre gene using an approach previously described by Gouka et al. (1996). The stringent hybridisation occurred at 65°C with final washes at 0.3 x SSC. An A. oryzae pyre disruption vector was constructed in which an 0.5 kb CIaI
fragment carrying part of the pyre coding region was deleted (Figure 4). The XbaI pyre fragment from this new vector was used for transformation and direct selection for FOA
resistant transformants. None of the FOA resistant colonies obtained was uridine requiring.

(3) UV mutagenesis and filtration enrichment Another approach to improve the yield of specific mutant strains is the use of a filtration-enrichment step (Bos et al. 1986, Thesis, Agricultural University Wageningen).
The UV mutagenized spores are used for inoculation of a minimal medium (MM) liquid culture. From the resulting repeated overnight culture those spores unable to germinate in minimal medium (a.o. pyrG mutant spores) are separated from the grown mycelium by filtration through myracloth. The spores obtained after several enrichment steps were tested for their Pyre phenotype, by inoculating the spores on plates containing FOA.
5 Again none of the resulting FOA resistant colonies was uridine requiring.
Also none of the colonies obtained after this enrichment on MM plates containing uridine was shown to be uridine requiring.

(4) Modified selection conditions 10 Our previous attempts to isolate pyre mutants from A. sojae had failed suggesting the inability of the required pyre mutants to utilize exogenous uridine, which is used in the FOA selection medium for the analysis of uridine auxotrophy. A modified selective FOA
medium, now containing uracil next to uridine, was used in a new isolation attempt. From this approach several FOA resistant mutants were obtained which were uracil requiring.
15 Retesting of these strains showed that these were unable to grow on uridine supplemented minimal medium. Subsequent transformation experiments with some of the uracil-requiring strains showed that these mutants could indeed be complemented with a fungal pyre gene (e.g. vector pAB4.l; A. niger pyrG). The inability of pyre mutants to grow in minimal medium supplemented with only uridine was an unprecedented observation for 20 related Aspergillus species (A. nidulans, A. niger, A. oryzae) and various other fungal species.

(5) Re-usable selection marker Versatile genetic modification of A. sojae requires the possibility to modify, disrupt and express a number of different genes in a single fungal strain, which would require the availability of a (series) of different selection markers. However, the availability of a marker such as pyre, which allows selection of both the mutant (FOA selection) and the transformant (Uracil-less medium), provides the possibility of repeated use of the same marker in subsequent experiments. For this approach a pyre marker gene was designed, in which the complementing sequence was flanked by a direct repeat sequence originating from the 3' flanking end of the pyre gene. The resulting plasmid is pAB4-lrep.
The construction of this vector is detailed in Figure 5. The full sequence of the vector is given in SEQ ID No. 17. Transformation of A. sojae pyre mutants with this vector results in a similar number of PyrG+ transformants as with the vector pAB4-1. However, subsequent plating of spores of selected pAB4-1 and pAB4-rrep traristormants to r'UA
selection plates resulted in many more FOA resistant/uracil requiring colonies for the pAB4-lrep transformant. Southern analysis of these FOA resistant/uracil requiring clones showed that in most of the pAB4-lrep strains the A. niger pyre marker gene had been deleted leaving only the small 0.7 kb repeat region at the locus of integration, while in the pAB4-1 strains the A. niger gene was still present and had presumably acquired a mutation resulting in the pyre-negative phenotype.
EXPRESSION HOSTS: STRAIN SELECTION.
Protease production Very important characteristics of a fungal expression system are the level and type of fungal proteases produced under various culture conditions. Sometimes strains which can be readily transformed are not suitable as expression hosts due to production of proteases or acidification of the culture media which is detrimental to the expressed product. Analysis of the growth behaviour of the various A. sojae strains revealed that, in contrast to what was observed for A. niger, acidification of the culture medium did not occur either on agar based plates (MacConkey) or in shake flask cultures. In fact in shake flask cultures, irrespective of the three medium types analyzed (Table 4), in most cases even an alkaline pH was obtained in the cultures. Based on these results and literature data it is thus expected that primarily alkaline proteases will be present in the A. sojae culture fluid. To analyse protease activity of the culture fluids of the various strains, a milk clearing assay was performed. In addition medium samples were incubated with different proteins (e.g. bovine serum albumin (BSA)), and degradation of these proteins was followed in time in order to assess the suitability of the tested strains as expression hosts for a range of products. BSA was chosen as in our previous experiments with A.
niger.
This protein was shown to be very susceptible to proteases. A. terreus phytase was chosen as example of another proteolytically instable protein. Degradation of milk proteins as shown by the formation of a milk clearing zone at the periphery of growing colonies is a generally accepted criterion for protease activity. Detection of BSA was carried out by Coomassie staining of SDS-PAGE gels. For phytase, Western analysis using specific antibodies, was carried out. As shown in Table 4, clear differences of degradation in A.
niger culture fluid are evident when this is compared with that in A. sojae culture fluid. In A. niger culture fluid (pH 3-4) rapid degradation of BSA occurs. In A. sojae culture fluids from richer media, degradation of BSA occurs, albeit less than in A. niger culture fluid. In most A. sojae culture fluids (pH 7-8) rapid degradation of A. terreus phytase occurs, with the exception of ATCC9362, ATCC11906 and ATCC20387 culture fluids. In general, the strains with the lowest phytase degradation also show low BSA degradation under the conditions tested. In particular the two A. oryzae strains ATCC20235 and show much higher proteolytic activity than most A. sojae strains.
To exclude that differences in the pH of the culture fluid cause the observed effects, similar degradation experiments were also carried out with culture fluids of which the pH
was adjusted to pH 4.5 (50 mM Na/HAc), pH 5.8 (50 mM Na/HAc) and pH 8.3 (50 mM
Tris/HCl). Table 5 gives the degradation data obtained with these samples. As can be seen in the table A. oryzae ATCC20235, which had the highest proteolytic activity at pH 7-8 also shows high proteolysis at other pH values. Degradation of A. terreus phytase occurs primarily at pH 8. Similarly to what was found before, ATCC11906 and ATCC20387 showed low phytase degrading activity. ATCC9362 showed phytase degradation in rich media. BSA degradation by A. sojae showed no significant differences with the data presented in Table 4.
In conclusion, these protease assays resulted in the identification of three low protease A. sojae strains, namely ATCC9362, ATCC11906 and ATCC20387. Thus, A.
sojae can cleary be used as expression host for a range of proteins and provides a series of advantages over prior art transformation and expression systems.
STRAIN IMPROVEMENT
Once the potential of transformability and expression had been ascertained for Aspergillus sojae, means by which additional strains could be created with enhanced characteristics for expression were considered. Two different approaches which can be used as such or in combination were developed to provide novel improved strains for expression of proteins.
On the one hand the possibility of developing protease deficient mutants was investigated and the impact of such on levels of expression was assessed. On the other hand strains with amended morphology were developed with a view to improve fermentation characteristics. To achieve this a hitherto non-disclosed or suggested route was followed which is applicable not only to Aspergillus sojae but to Aspergilli and in fact to fungi in general.

Development of protease deficient mutants To obtain protease deficient A. sojae strains two approaches were followed. In a first approach spores from ATCC 11906 and ATCC 11906-derived strains were mutagenized with UV. In a second approach gene disruption of the major alkaline protease S was carried out.
UV mutants Freshly harvested spores from A. sojae ATCC 11906 or one of its pyre derivatives was UV-mutagenized in a Biorad UV-chamber with a dose resulting in 20-SO %
survival.
Serial dilutions were plated onto skim-milk plates (Mattern et al., 1992).
From 5000 UV
surviving strains four mutant strains with a considerably reduced milk-clearing halo were obtained AIpA gene disruption In this approach the endogenous alpA (alkaline protease) gene cloned from ATCC11906 was disrupted using a disruption vector carrying the re-usable pyre selection marker as described in this description.
An ATCC 11906 cosmid library (in a Pyre cosmid) was constructed. From 10.000 independent cosmid clones initially 4 were found to hybridize under homologous conditions with an A. sojae alpA fragment obtained by PCR with primers MBL1784 and MBL1785. Rescreening of the 4 clones revealed only strong hybridisation with one clone.
A 4 kb EcoRI and a 2.5 kb HindIII fragment from this clone, together expected to carry the complete gene, were subcloned and characterised by restriction enzyme digestion and sequence analysis. Based on these subclones a new gene-replacement vector was constructed as described in Figure 6. For transformation of an ATCC 11906pyrG
derivative the vector was digested with EcoRI, and the 8.7 kb alpA deletion fragment was used for transformation (see Figure 6). Transformation of the replacement cassette to ATCC 11906pyrG5 resulted in a number of transformants with a reduced milk-halo.
Southern analysis of these strains revealed the successful deletion of the alpA gene. To allow subsequent use of the pyre marker for transformation of one of these strains, spores from this strain were plated on FOA containing medium selective for pyre mutants. From strains with the correctly integrated disruption cassette with the re-usable pyre marker a large number of FOA resistant colonies were obtained. In contrast to the results obtained for spontaneous FOA resistant mutants of wild type strains, the FOA strains obtained from these disruption strains were virtually all uracil requiring and turned out to be Pyre negative again. Southern analysis was used to confirm the desired removal of the pyre marker gene at the alpA locus, leaving only the 700 by "footprint".
Analysis of protease activity in UV and disruption mutants To analyse the levels of protease production in the different low protease derivatives of ATCC 11906 controlled batch fermentation experiments were carried out.
From the resulting culture supernatants protease activities were determined at various pH
values. Deletion of the alpA gene resulted in a strong reduction of proteolytic activity at alkaline pH. Analysis of the protease activity in one of the UV mutants showed almost complete absence of proteolytic activity at both pH 6 and pH 8. Consequently the level of proteolysis towards secreted proteins produced in these strains was considerably less than observed for the parent strain.
Development of low viscosity mutants Initial controlled batch or fed batch fermentation trials with A. sojae resulted in considerable biomass yield, however both the culture viscosity and sporulation phenomena in the fermenter vessel represented less favourable characteristics.
Therefore attempts were made to improve these characteristics in the desired host strain. Various patent applications teach that low viscosity mutants can be isolated by various ways of screening. W096/02653 and W097/26330 describe non defined mutants exhibiting low viscosity. However here we describe a new unexpected case of a completely characterised and fully defined low viscosity mutant from A. sojae. It was found that a proprotein processing mutant from this organism had an unexpected aberrant growth phenotype (hyper-branching) while no detrimental effect on the productivity of proteins was observed. Controlled fermentation experiments with this strain revealed that increased biomass concentrations were obtained at considerably lower viscosity values.
The observed characteristics were not only present in A. sojae but other fungi as well, e.g. in A.
niger.
(1) Construction of an A. niger proprotein processing mutant To clone the proprotein convertase encoding gene from A. niger, heterologous hybridisation using specific probes from the Saccharomyces cerevisiae KEX2, Schizosaccharomyces pombe KEXl and the Xenopus laevis PC2 genes was carried out.

However, no specific hybridisation signals were obtained even at very low stringency hybridisation conditions (47°C, washes at 6xSSC), precluding the use of this approach to clone the corresponding A. niger gene.
As an alternative approach to clone the proprotein convertase encoding gene from S A. niger, PCR was used. Based on the comparison of various proprotein convertase genes from various yeast species and higher eukaryotes (Figure 1) different PCR
primers were designed (SEQ ID Nos. 10, 13 and 18-23) which are degenerated, respectively, 2048, 49152, 4, 2, 2, 512, 2048, and 4608 times. From the amplification using primers PE4 and PE6, two individual clones were obtained of which the encoded protein sequence did show 10 significant homology to the S. cerevisiae KEX2 sequence (SEQ ID No. 24).
These clones were used for further experiments.
Based on the observed homology to other proprotein convertase genes of the cloned PCR fragment, the corresponding A. niger gene was designated pclA (from proprotein convertase-like). Southern analysis of genomic digests of A. niger revealed that the pclA
15 gene was a single copy gene with no closely related genes in the A. niger genome, as even at stringent hybridisation conditions (50°C; washes at 6xSSC), no additional hybridisation signals were evident. A first screening of an EMBL3 genomic library of A.
niger N401 (van Hartingsveldt et al., 1987) did not result in any positively hybridising plaques although about 10-20 genome equivalents were screened. In a second screening a full 20 length genomic copy of the pclA gene was isolated from an A. niger N400 genomic library in EMBL4 (Goosen et al., 1987). Of the 8 hybridising plaques which were obtained after screening 5-10 genome equivalents, 6 were left after a first rescreening. All these 6 clones most likely carried a full copy of the pclA gene, as in all clones (as was observed for the genomic DNA) with the PCR fragment two hybridising EcoRV fragments of 3 and 4 kb 25 were present (note that the PCR fragment (SEQ ID No. 24) contains an EcoRV
restriction site). Based on a comparison of the size of other proprotein convertases, together these fragments will contain the complete pclA gene with 5'- and 3'-flanking sequences. The two EcoRV fragments and an overlapping 5 kb EcoRI fragment were subcloned for further characterisation. A detailed restriction map of the DNA fragment carrying the pclA gene is given in Figure 7.
Based on the restriction map given in Figure 7 the complete DNA sequence of the pclA gene was determined from the EcoRI and EcoRV subclones (SEQ ID No. 3).
Analysis of the obtained sequence revealed an open reading frame with considerable similarity to that of the S. cerevisiae KEX2 gene and other proprotein convertases. Based on further comparison two putative intron sequences (SEQ ID No. 3, from position 1838 to 1889 and from 2132 to 2181) were identified in the coding region. Subsequent PCR
analysis with primers flanking the putative introns, on a pEMBLyex based A.
niger cDNA
library revealed that only the most 5' of these two sequences represented an actual intron.
The general structure of the encoded PcIA protein was clearly similar to that of other proprotein convertases (SEQ ID No. 25 and Figure 8). The overall similarity of the PcIA
protein with the other proprotein convertases was about 50% (Figure 1 ).
To demonstrate that the cloned pclA gene is a functional gene encoding a functional protein, the construction of strains devoid of the pclA gene was attempted.
Therefore, pPCLIA, a pclA deletion vector, in which a large part of the pclA coding region was replaced with the A. oryzae pyre selection marker gene, was generated.
Subsequently, from this vector the 5 kb EcoRI insert fragment was used for transformation of various A.
niger strains. From these transformations (based on pyre selection) numerous transformants were obtained. Interestingly, a fraction of the transformants (varying from 1-50%) displayed a very distinct aberrant phenotype (Figure 9). Southern analysis of several wild-type and aberrant transformants revealed that these aberrant transformants, which displayed a severely restricted growth phenotype, had lost the pclA gene. All strains displaying wild-type growth were shown to carry a copy of the replacement fragment integrated adjacent to the wild-type pclA gene or at a non-homologous position.
Analysis of protein production in selected pclA mutant strains carrying various glucoamylase fusion genes revealed the production of unprocessed fusion protein. The production of high levels of unprocessed glucoamylase-interleukin-6 fusion protein in a pclA mutant was achieved. Protein analysis revealed that in pclA mutant strains also no fully processed endogenous glucoamylase is formed but only pro-glucoamylase.
To further improve the yields of fusion proteins controlled batch and fed-batch fermentations were also carried out. Surprisingly the fermentation characteristics of pclA
mutant strains were clearly superior to those of the parent strain resulting in a much reduced viscosity/biomass ratio, without loss of productivity.
(2) Construction of an A. sojae proprotein processing mutant To construct the corresponding mutant in A. sojae, functional complementation of the low-viscosity mutant of A. niger genomic cosmid clones from the ATCC 11906 cosmid library were isolated, which comprised the A. sojae proprotein processing protease pclA
gene. Partial sequence analysis of the isolated sequences SEQ ID No. 4 confirmed the cloning of the A. sojae pclA gene. Figure 10 shows the comparison of the DNA
sequences of part of the A. niger and A. sojae pclA genes. Based on the A. sojae sequence and a partial restriction map with the coding region of the A. sojae pclA gene, a replacement vector was generated using the EcoRV-site in the A. sojae pclA gene to clone the re-usable pyre marker as a SmaI fragment inside (Figure 11). The resulting vector was partially digested with CIaI to obtain the delta-pcl fragment of 10.5 kb (see Figure 11). This fragment was isolated to be used for transformation of A. sojae pyre strains.
The gene replacement vector was used to generate pclA mutants in ATCC 11906 and ATCC

derivatives. The resulting strains were used for the expression of homologous and heterologous proteins. Controlled fermentation experiments with some of the resulting transformants revealed improved fermentation characteristics, in particular a lower viscosity/biomass ratio of the culture.
(3) Cloning offungal genes homologous to Aspergillus pclA
Based on the comparison of amino acid sequences inferred from the A. niger and A.
sojae pclA genes with those of other proprotein processing enzymes (Figure 1 ) several oligonucleotide mixtures corresponding to the coding or non-coding strand of well conserved sequences were designed (SEQ ID Nos. 10 to 16).
These oligonucleotide mixtures were used in PCR with chromosomal DNA from Trichoderma reesei QM9414, Fusarium venenatum ATCC20334, Penicillium chrysogenum P2, Trametes versicolor, Rhizopus oryzae ATCC200076, and Agaricus bisporus HORST. Depending on the template DNA used, PCR amplifications (30 cycles; 1 min 94°C; 1 min 40°C; 2 min 68°C) with one or more of the combinations of coding and non-coding strand oligonucleotides resulted in specific PCR products. Table 6 gives the results of the various amplification reactions. Sequence analysis was carried out with a number of the obtained PCR fragments using either of the two oligonucleotide mixtures used for amplification. These analyses resulted in the identification of the various pclA
homologues from these different fungi. Figure 12 gives the inferred aminoacid sequences corresponding with the various DNA fragments (SEQ ID Nos. 5 to 9).
(4) Examples of biomass and viscosity determinations The following operating parameter data ranges have been determined for fungal fermentations using a number of different fungal strains.

Viscosi Viscosity is determined on a Haake Viscotester VT500 using sensor system MV
DIN (vessel number 7), operated at 20°C. Obtain a fresh sample of fermentation broth and place 40 ml of the broth in the measuring cell. After a small period of equilibration (4 min) at a set spindle speed the viscosity is measured. This measurement is repeated for ten different spindle speeds. Multiplication of the spindle speed with the measuring cell factor results in the shear rate. Viscosity r1 (in centipoise = cP) is plotted against shear rate y (1/s) and gives the viscosity profile of the fermentation broth.
Viscosity ranges have been determined for fermentations using the specified fungal organism using the above procedure (Table 7).
Biomass:
Biomass is determined by the following procedure:
Preweigh 5.5 cm filter paper (Whatman 54) in an aluminium weighing dish.
Filter 25.0 ml 1 S whole broth through the 5.5 cm paper on a Buchner funnel, wash the filter cake with 25.0 ml deionised water, place the washed cake and filter in a weighing pan and dry overnight at 60°C. Finish drying at 100°C for 1 hour, then place in desiccator to cool. Measure the weight of dried material. Total biomass (g/1) is equal to the difference between the initial and final weights multiplied by 40.
Protein:
Protein levels were determined using the BioRad Assay Procedure. The data presented above represent values determined 48 hours into the fermentation process until fermentation end; all values of Aspergilli and Trichoderma are for commercially relevant fungal organisms and reflect actual commercial data.
A fizngal strain such as A. sojae lfvA and A. sojae pclA has the advantage that the low viscosity permits the use of lower power input and/or shear the in the fermentation to meet oxygen demands for those cases where shear stress on the product may be detrimental to productivity due to physical damage of the product molecule. The lower biomass production at high protein production indicates a more efficient organism in the conversion of fermentation media to product. Thus A.sojae mutants provides better biomass and viscosity data whilst also delivering at least as much protein, and in fact a lot more protein than the two commercially used systems which obviously are better than for typically deposited Aspergillus or Trichoderma reesei strains in general public collections.

The high protein production with low biomass concentration produced by A.
sojae ~A would allow development of fermentation conditions with higher multiples of increase in biomass, if increasing biomass results in increased productivity, for the desired product before reaching limiting fermentation conditions. The present high levels of biomass and viscosity produced by the T. longibrachiatum and A. niger organisms restrict the increase of biomass as the present levels of biomass and viscosity are near limiting practical fermentation conditions.
EFFICIENT GENE EXPRESSION
(1) Heterologous regulatory sequences The three selected A. sojae strains were cotransformed with three GUS reporter vectors carrying different fungal expression signals (A. nidulans PgpdA;
pGUS54, A. niger PglaA; pGUS64, A. niger PbipA; pBIPGUS) and the amdS selection vector p3SR2 or derivatives thereof. Expression of the GUS gene was analysed in representative transformants (Table 8). From the results it is clear that under the conditions tested the gpdA promoter was by far the best promoter resulting in about 5000 U GUS/mg protein.
This corresponds to about 5% of the total amount of cellular protein. The bipA
promoter results in about 30% of the gpdA activity, which corresponds to expression data obtained in A. niger. Surprisingly, the glaA promoter which is very active in A. niger (at least as active as gpdA) results in less than 1 % of the gpdA activity in A. sojae.
(2) A. sojae regulatory sequences We also isolated an Aspergillus sojae homologous promoter and assessed the applicability of such in an expression system. In some instances of expression it will be preferable to use a homologous promoter rather than a heterologous promoter.
It was also interesting to assess whether the homologous promoter would be more efficient than a heterologous one.
PCR cloning of three efficiently expressed A. sojae genes, namely alpA
(alkaline protease; inducible), amyA (amylase; inducible) and gpdA (glyceraldehyde-3-phosphate dehydrogenase; constitutive) was attempted using primers based on sequences available from A. oryzae (SEQ ID Nos. 26 to 31). Figures 13 a, b and c give the sequences and the position in published A. oryzae sequences of the various PCR primers used for this approach. Genomic template DNA from A. sojae ATCC 11906 was used for PCR
amplification. Initial PCR amplifications (30 cycles; 1 min 94°C; 1 min 40°C; 2 min 68°C) resulted in a specific PCR product of the expected size (400 bp) for the gpdA.
For the other two PCR reactions no product was obtained. Therefore, for alpA PCR conditions were made less stringent (10 cycles; 1 min 94°C; 1 min 25°C; 2 min 68°C + 20 cycles; 1 min 94°C; 1 min 40°C; 2 min 68°C), which resulted in a specific alpA PCR product of about 1000 bp.
5 The complete sequence of the cloned genes was determined. As shown in Figure the A. sojae ATCC 11906 gpdA promoter region has a very high homology with other gpdA
promoter sequences and the alpA promoter was virtually identical to the A.
oryzae alpA
promoter (SEQ ID Nos. 32 and 33). Expression vectors carrying expression cassettes comprising these A. sojae promoters show significant levels of gene expression.
HETEROLOGOUS PROTEIN PRODUCTION
A number of heterologous proteins were tested which were known to be susceptible to acidic proteolysis and thus could not be expressed efficiently in other well known expression systems. Also proteins that are already efficiently expressed in alternative systems were tested in order to assess by way of comparison the levels of expression achieved with Aspergillus sojae vis a vis other known expression systems such as Aspergillus niger.
Phytase production DNA fragments carrying various fungal phytases (Wyss et al. (1999) Appl.
Environ. Microbiol. 65, 359-366) were ligated as 5' NcoI or BspHI sites introduced at the ATG codon - 3' blunt-ended fragments downstream of the A. nidulans gpdA
promoter in pAN52-lNotI. The resulting vectors were used in cotransformation experiments of A.
sojae using the amdS and or pyre selection marker. Phytase producing transformants were screened using the described phytase plate-assay.
Further improved phytase expression vectors were generated using a multicopy cosmid approach. In this approach several copies of a phytase expression cassette, recloned in a multiple cloning site vector (pMTL series, Chambers et al., (1988) Gene 68, 139-149) to allow its isolation as a EcoRI fragment. Several copies of these EcoRI
fragments were cloned into cosmid vector pAN4cos1 through packaging (Verdoes et al. (1993) Transgenic Research 2, 84-92), resulting in cosmid clones carrying a number of expression cassettes.
The resulting clones were introduced into A. sojae using the amdS selection marker.
AmdS+ clones were screened for phytase production using the phytase plate-assay.

Further phytase expression vectors were generated using the GLA fusion approach (e.g. Broekhuijsen et al. (1993) J. Biotech. 31, 135-145). To this end phytase gene fragments, encoding the mature A. fumigatus phytase protein were fused, using convenient restriction sites and fusion PCR, to the 3'-end of the glucoamylase carrier gene in vector pAN56-1 (Genbank accession number Z32700). Between the glucoamylase and phytase part of the gene-fusion a sequence encoding a proprotein processing site (Asn-Val-Ile-Ser-Lys-Arg) was introduced. The resulting vectors were used in cotransformation experiments of A. sojae using the amdS and/or pyre selection marker. Phytase producing transformants were screened using the described phytase plate-assay.
Shake flask fermentation was carried out resulting in significant levels of active phytase. Yield were significantly higher than those reported in literature for A. niger (van Hartingsveldt et al. (1993) Gene 127, 87-94; Van Gorcom et al. (1991) EP420358). On average, the levels obtained with the multicopy cosmid vectors were higher than those obtained with the single copy vectors. Phytase levels obtained with the glucoamylase-phytase fusion vectors resulted in high levels of both glucoamylase and phytase.
Controlled batch and fedbatch fermentations from a selected number of phytase producing A. sojae transformants revealed a further increased level of phytase.
Glucoamylase production An example of an efficiently produced fungal protein is provided by the expression of the A. niger glaA gene. Vector pGLA6S (Figure 15) is derived from pGLA6 (Punt et al.
(1991) J. Biotech. 17, 19-334) by introducing a 5 kb EcoRI fragment carrying the A.
nidulans amdS gene as selection marker into the unique EcoRI site of pGLA6.
Vector pGLA6S (Figure 15) carrying the amdS selection marker and the glucoamlyase gene under control of the A. nidulans gpdA promoter was introduced into A. sojae ATCC11906pyrG
using cotransformation with vector pAB4.l. Starch plate-assays demonstrated the production of increased levels of amylolytic activity in these transformants.
From the resulting transformants those showing proficient growth on acrylamide medium were analysed for glucoamylase production. On a Coomassie Briljant Blue-stained SDS
PAGE
gel from the culture supernatant of some of these transformants a single dominant protein band corresponding to glucoamylase was observed . Western analysis using a monoclonal antibody against glucoamylase (Verdoes et al. (1993) Transgenic Research 2, 84-92) was used to confirm the identity of this protein band.

Interleukin-6 production Production of interleukin-6, which is an example of a highly sensitive protein for proteolytic degradation, was shown to be virtually impossible in A. niger without the use of the gla-fusion strategy and protease deficient strains. Even with all these improvements the yields of IL-6 were only a few mg per litre culture fluid. Introduction of the IL-6 vector pAN56-4 (Broekhuijsen et al. (1993) J. Biotech. 31, 134-145) into A. sojae by cotransformation with the pyre or amdS marker resulted in transformants expressing the IL-6-fusion gene present in this vector. From the resulting transformants a few were selected for further analysis. Shake flask fermentation experiments were carried out with these transformants. SDS-PAGE and Western analysis of culture supernatants of several of these strains surprisingly showed levels of correctly processed IL-6 which were about 5-10 fold higher than the levels obtained in the best reported cases in A. niger.
The use of the various types of protease deficient and fermentation-optimized mutants from A.
sojae further increased the level of IL-6 production to be obtained from controlled fermentations (Broekhuijsen et al. (1993) J. Biotech. 31, 134-145).
Green fluorescent protein (GFP) Another type of acid labile protein we have attempted to produce in A.sojae is GFP
from the jelly fish Aequoria victoria. This protein is not only proteolytically sensitive but furthermore it loses its activity at acid pH. Vectors carrying GFP or GLA-GFP
fusion genes (driven by the A. nidulans gpdA promoter) were introduced into A. sojae by cotransformation using either the pyre or amdS selection marker. Expression resulted in brightly fluorescent A. sojae transformants for both vector types. Based on the observed fluorescence and the subsequent analysis of culture supernatants from selected, shakeflask-cultured transformants using SDS-PAGE and Western analysis it was ascertained that the yields of intact cytoplasmic GFP and secreted GLA-GFP are much higher than those obtained in A. niger protease deficient hosts (Siedenberg et al. Biotechn.
Prog. (1999) 15, 43-50; Gordon et al., Microbiology (2000) 146, 415-426). In contrast to the situation in A.
niger culture supernatants also the secreted GFP showed significant fluorescence.
DESCRIPTION OF THE FIGURES
Figure 1: This figure provides a comparison of amino acid sequences of KEX2-like processing proteases from X. laevis (XENPC2 and XNFURIN), S. cerevisiae (SCKEX2), K. lactis (KLKEX1), C. albicans (CAKEX2), S. pombe (SPKRP) and Y. lipolytica (YLKEX2). The primers, which encode for the amino acid sequences with the highest overall identity (indicated with lightblue boxes), are indicated: MBL793, MBL1208, MBL
794, MBL1158, PE6, PCL1, PCL2(rev), PE6, PCL3, MBL789, PCL4 and MBL1219.
Regions of overall identity (4 out of 7 entries) are indicated with purple boxes. Gaps are indicated with . ; no sequence data are indicated with ~ ; asteriks indicate the stop codon of the protein.
Figure 2: This consists of 2a, b and c Figure 2a provides the background growth of the A. sojae strain described in patent W097/04108 after 5 days of incubation at 33°C. The top picture reveals growth on non selection medium. The bottom left picture shows selection medium according to W097/04108 and the bottom right picture shows the results using improved medium (acrylamide) according to the invention.
1 S Figure 2b provides the background growth of the A. sojae strain ATCC 11906 after 5 days of incubation at 33°C. The top picture reveals growth on non selection medium. The bottom left picture shows selection medium according to W097/04108 and the bottom right picture shows the results using improved medium (acrylamide) according to the invention.
Figure 2c provides the background growth of the A. sojae strain ATCC20387 after 5 days of incubation at 33°C. The top picture reveals growth on non selection medium. The bottom left picture shows selection medium according to W097/04108 and the bottom right picture shows the results using improved medium (acrylamide) according to the invention.
Figure 3 (a and b): This figure provides a comparison of A. sojae ATCC11906 and A.
oryzae amdS sequences from both ends. A and B indicate the two ends. The cloned 1.6 kb A. sojae sequence was used. Underlined bold bases differ between species/strains. Intron I
sequences are indicated in small letters.
Figure 4 (a and b): This figure illustrates the construction of a pyre disruption vector via pA04-13 and pA04-l3deltaCla.
Figure S: This figure illustrates the construction of pAB4-lrep going from pAB4-1 via isolation of XhoI fragment and HindIII fragment followed by cloning into pMTL24.
Figure 6: The construction of the alpA gene replacement vector is disclosed in this figure.
A 4.4 kb EcoRI-StuI fragment from pASI-1 with the ATCC11906 genomic fragment, the 2.6 kb SmaI-NcoI fragment from pAB4-lrep and the 4.4 kb NcoI-EcoRI fragment from pASI-2A are ligated in a 3 way ligation thus providing pASI-deltaalp.
Figure 7: This figure provides the restriction map of the DNA fragment carrying the A.
niger pclA gene.
Figure 8: This figure provides the structure (functional organisation) of the A. niger pclA
encoded protein. It shows pre, pro, activity and P domains from left to right.
The light coloured triangles indicate KR sites. The dark coloured triangles indicate glycosylation sites. The vertically striped light box is an S/P/T rich region. The dark weavepatterned box at the right end is a D/E rich region.
Figure 9: This figure illustrates growth phenotype of an A. niger pclA mutant strain.
Figure 10: This figure provides a DNA sequence comparison between the A. sojae and A.
niger pclA genes. A vertical bar indicates identity; : indicates 5; ~
indicates 1. 72.139%
similarity and 72.073% identity were found.
Figure 11: The construction of the pclA gene replacement vector is disclosed in this figure.
A 7.6 kb CIaI fragment, which is a ATCC 11906 genomic fragment, was cloned into pMTL23p. In this construct the 2.6 kb SmaI fragment from pAB4-lrep was cloned into the EcoRV-site, thus providing pAS2-delta pcl.
Figure 12: This figure shows the amino acid sequence comparison of the various PcIA
homolous from S. cerevisiae (Sckex2), K. lactis (Klkexl), A. sojae (Aspcla), A. niger (A.
niger), P. chrysogenum (Penpcll), A. bisporus (Agarmb1129), T. reesei (Trichpcll), R.
oryzae (Rhizpcll), F. venenatum (Fuspcll), S. pombe (Spkrp), C. albicans (Cakex2) and Y.
lipolytica (Ylkex2). Regions of overall identity (8 out of 12 entries) are indicated with yellow boxes. Gaps are indicated with .. ; no sequence data are indicated with ~ .

Figure 13: Sequence data are provided in figure 13a for the A. oryzae alpA
promoter sequences (Q11755). The primer position for PCR cloning is indicated. In figure 13b the sequence data are provided for the A. oryzae amyA promoter sequences also including primer positions (A02532). Figure 13c provides the ATCC42149 A. oryzae derived gpdA
5 promoter sequences (EP0.436.858 al) also including primer positions.
Figure 14: This figure provides a comparison between various gpdA promoter sequences of Aspergillus: From top to bottom, A sojae ATCC 11906, A. oryzae, A. niger and A. nidulans.
Asterisks indicate the putative intron present in the 5' untranslated region of the promoters.
10 Arrowhats indicate the CT rich regions. Bold underlined letters indicate the differences between the A. oryzae and A. sojae sequences.
Figure 15: This figure shows a map of the vector pGLA6S of 12700bp.

SEQ ID No. 1 MBL 1784: 5'-CGGAATTCGAGCGCAACTACAAGATCAA-3' SEQ ID No. 2 MBL 1785: 5'-CGGAATTCAGCCCAGTTGAAGCCGTC-3' SEQ ID No. 3 The sequence of the Aspergillus niger gene encoding proprotein convertase The start codon and the stop codon are indicated with bold underlined letters The intron is indicated with underlined small letters 1801 GCTATGTTGA TTGGACCTTC ATGAGCGTTG CTCACTGgta agtaaaaact 1851 ttttctcggt tgtcggttct tctgctaata catatctagG GGCGAGTCCG

WO 01/09352 PCT/NLOO/0_0544 SEQ ID No. 4 The partial sequence in the coding region of the Aspergillus sojae gene encoding proprotein convertase SEQ ID No. 5 The partial sequence in the coding region of the Trichoderma reesei QM9414 gene encoding proprotein convertase SEQ ID No. 6 The partial sequence in the coding region of the Fusarium venenatum ATCC20334 gene encoding proprotein convertase SEQ ID No. 7 The partial sequence in the coding region of the Penicillium chrysogenum P2 gene encoding for proprotein convertase SEQ ID No. 8 The partial sequence in the coding region of the Rhizopus oryzae ATCC200076 gene encoding proprotein convertase SEQ ID No. 9 The partial sequence in the coding region of the Agaricus bisporus HORST gene encoding proprotein convertase 43' SEQ ID No. 10 coding strand BamHI-site is underlined PE4 5'- CG CGGATC CA(T/C) GGX ACX (C/A)GX TG(T/C) GCX GG -3' degenerated 2048 times SEQ ID No. 11 coding strand PCL1 5'-CA(T/C) GGX ACX (C/A)GX TG(T/C) GCX GGX GA-3' degenerated 8192 times SEQ ID No. 12 coding strand PCL2 5'-AT(C/T/A) TA(T/C) TCX TG(T/C) TCX TGG GGX CC-3' degenerated 768 times SEQ ID No. 13 non coding strand BamHI-site is underlined PE6 5'- CGC GGA TCC XCC (A/G)TT XCC X(C/G)(A/G) XGC
(G/A/C)(C/A)A XAC -3' degenerated 49152 times SEQ ID No. 14 non coding strand PCL2rev 5'-GG XCC CCA XGA (A/G)CA XGA (A/G)TA (A/T/G)AT-3' degenerated 768 times SEQ ID No. 15 non coding strand PCL3 5'-(A/G)TT XGT (A/G)TA XCC (A/G)TC (A/G)(A/T)A (A/G)TT-3' degenerated 1024 times SEQ ID No. 16 non coding strand PCL4 S'-GC XGC XGA XGT XCC XCC (A/G)TG-3' degenerated 2048 times SEQ ID No. 17 The sequence of pAB4-lrep ........59-499 bp...................................................: 0.4 kb HindIII fragment of pAB4-1 1-58 bp........500-513 bp.............2873-2930 by : polylinker sequence of pMTL24 (indicated with underlined small letters) ...................................514-2872 bp....................: 2.3 kb XhoI fragment of pAB4-1 1 ggccagtgaa ttcgagctcg gtacccgggg atcctctaga gtcgacctgc 51 aggcatgcAA GCTTGGTCAG CAGTACCAGA CGCCCGGATC GGCTATCGGC

451 GCGTAGAGAA AATGGCGACG GGTGGGCTGA TAAGGGCGGT GATAAGCTTg 501 catgcctgca gc~cCTCGAGC TAACATACAT TCCGAACCGT GCAGCCCAAG

1701 AGGACCGCAA GtTCATTGAC ATCGGCAACA CGGTCCAGAA GCAATACCAC

2851 CTAGTTTAGT TGGATGCTCG AGatctccat GGacgcgtga cgtcgactct 2901 gaggatcccc dggtaccgag ctcgaattcg SEQ ID No. 18 MBL 789 EcoRI is underlined 5'- GGAA TTC (A/G)GA ATA (T/A)GG AGG ATG TAG -3' degenerated 4 times SEQ ID No. 19 MBL 793 BamHI is underlined 5'- CGGATCCG CAG TGG CAC TTG (G/A)TC AAT CCA A -3' degenerated 2 times SEQ ID No. 20 MBL 794 EcoRI is underlined 5'- GGA ATT CTT AAA A(T/G)C CCA AGA ACC TTC A -3' degenerated 2 times SEQ ID No. 21 MBL 1158 EcoRI is underlined 5'- G GAA TTC (T/C)TC (T/G)CC (T/G)GC (A/G)CA (C/G)C(T/G) (C/G)GT (T/G)CC (A/G)TG -3' degenerated 512 times SEQ ID No. 22 MBL 1208 CIaI is underlined 5'- CGG ATC GA(T/C) GGX ACX (C/A)GX TG(T/C) GCX GG -3' degenerated 2048 times SEQ ID No. 23 MBL 1219 BamHI is underlined 5'- CGG ATC (C/T)TG XA(G/T/C) (A/G)TC XC(T/G) CCA XGT
(C/A/G)AG -3' degenerated 4608 times SEQ ID No. 24 Restriction sites are bold Primers are underlined BamHI PE4 primer GGATCCATGG CACGAGATGT GCAGGTGAAA TCGGTGCGGC GAAAGAAAAC

GGGTTGGTGT TGCGTATGAT AGTCGCATCG CTGGTATTCG GATTCTCTCC

EcoRV
ATGACACTGA TGAGGCTGCG GCTATTAACT ACGCCTATCA GGAGAACGAT

GTTCCTGGGG TCCCTATGAT GATGGCGCCA CAATGGAAGC CCCGGGCACT

GGGCCATGGT CAATGGTATC CAAAATGGTC GAGGTGGAAA AGGCTCGGTT

PE6 primer CCCCCGGAAA TGGTGGATCC

BamHI
SEQ ID No. 25 Aspergillus niger PcIA protein sequence ~9~
1 Met Arg Leu Thr Gly Gly Val Ala Ala Ala Leu Gly Leu Cys Ala 16 Ala Ala Ser Ala Ser Leu His Pro His Arg Ser Tyr Glu Thr His 31 Asp Tyr Phe Ala Leu His Leu Asp Glu Ser Thr Ser Pro Ala Asp 46 Val Ala Gln Arg Leu Gly Ala Arg His Glu Gly Pro Val Gly Glu 61 Leu Pro Ser His His Thr Phe Ser Ile Pro Arg Glu Asn Ser Asp 76 Asp Val His Ala Leu Leu Asp Gln Leu Arg Asp Arg Arg Arg Leu 91 Arg Arg Arg Ser Gly Asp Asp Ala Ala Val Leu Pro Ser Leu Val 106 Gly Arg Asp Glu Gly Leu Gly Gly Ile Leu Trp Ser Glu Lys Leu 121 Ala Pro Gln Arg Lys Leu His Lys Arg Val Pro Pro Thr Gly Tyr 136 Ala Ala Arg Ser Pro Val Asn Thr Gln Asn Asp Pro Gln Ala Leu 151 Ala Ala Gln Lys Arg Ile Ala Ser Glu Leu Gly Ile Ala Asp Pro 166 Ile Phe Gly Glu Gln Trp His Leu Tyr Asn Thr Val Gln Leu Gly 181 His Asp Leu Asn Val Thr Gly Ile Trp Leu Glu Gly Val Thr Gly 196 Gln Gly Val Thr Thr Ala Ile Val Asp Asp Gly Leu Asp Met Tyr 211 Ser Asn Asp Leu Arg Pro Asn Tyr Phe Ala Ala Gly Ser Tyr Asp 226 Tyr Asn Asp Lys Val Pro Glu Pro Arg Pro Arg Leu Ser Asp Asp 241 Arg His Gly Thr Arg Cys Ala Gly Glu Ile Gly Ala Ala Lys 10 Asn 256 Asp Val Cys Gly Val Gly Val Ala Tyr Asp Ser Arg Ile Ala Gly 15 271 Ile Arg Ile Leu Ser Ala Pro Ile Asp Asp Thr Asp Glu Ala Ala 286 Ala Ile Asn Tyr Ala Tyr Gln Glu Asn Asp Ile Tyr Ser Cys Ser 301 Trp Gly Pro Tyr Asp Asp Gly Ala Thr Met Glu Ala Pro Gly Thr 316 Leu Ile Lys Arg Ala Met Val Asn Gly Ile Gln Asn Gly Arg Gly 331 Gly Lys Gly Ser Val Phe Val Phe Ala Ala Gly Asn Gly Ala Ile 346 His Asp Asp Asn Cys Asn Phe Asp Gly Tyr Thr Asn Ser Ile Tyr 361 Ser Ile Thr Val Gly Ala Ile Asp Arg Glu Gly Asn His Pro Pro 376 Tyr Ser Glu Ser Cys Ser Ala Gln Leu Val Val Ala Tyr Ser Ser 391 Gly Ala Ser Asp Ala Ile His Thr Thr Asp Val Gly Thr Asp Lys 406 Cys Ser Thr Thr His Gly Gly Thr Ser Ala Ala Gly Pro Leu Ala 421 Ala Gly Thr Val Ala Leu Ala Leu Ser Val Arg Pro Glu Leu Thr 436 Trp Arg Asp Val Gln Tyr Leu Met Ile Glu Ala Ala Val Pro Val 451 His Glu Asp Asp Gly Ser Trp Gln Asp Thr Lys Asn Gly Lys Lys 466 Phe Ser His Asp Trp Gly Tyr Gly Lys Val Asp Thr Tyr Thr Leu 481 Val Lys Arg Ala Glu Thr Trp Asp Leu Val Lys Pro Gln Ala Trp 496 Leu His Ser Pro Trp Gln Arg Val Glu His Glu Ile Pro Gln Gly 511 Glu Gln Gly Leu Ala Ser Ser Tyr Glu Val Thr Glu Asp Met Leu 526 Lys Gly Ala Asn Leu Glu Arg Leu Glu His Val Thr Val Thr Met 541 Asn Val Asn His Thr Arg Arg Gly Asp Leu Ser Val Glu Leu Arg 556 Ser Pro Asp Gly Arg Val Ser His Leu Ser Thr Pro Arg Arg Pro 571 Asp Asn Gln Glu Val Gly Tyr Val Asp Trp Thr Phe Met Ser Val 586 Ala His Trp Gly Glu Ser Gly Ile Gly Lys Trp Thr Val Ile Val 601 Lys Asp Thr Asn Val Asn Glu His Thr Gly Gln Phe Ile Asp Trp 616 Arg Leu Asn Leu Trp Gly Glu Ala Ile Asp Gly Ala Glu Gln Pro 631 Leu His Pro Met Pro Thr Glu His Asp Asp Asp His Ser Tyr Glu 646 Glu Gly Asn Val Ala Thr Thr Ser Ile Ser Ala Val Pro Thr Lys 661 Thr Glu Leu Pro Asp Lys Pro Thr Gly Gly Val Asp Arg Pro Val 676 Asn Val Lys Pro Thr Thr Ser Ala Met Pro Thr Gly Ser Leu Thr 691 Glu Pro Ile Asp Asp Glu Glu Leu Gln Lys Thr Pro Ser Thr Glu 706 Ala Ser Ser Thr Pro Ser Pro Ser Pro Thr Thr Ala Ser Asp Ser 721 Ile Leu Pro Ser Phe Phe Pro Thr Phe Gly Ala Ser Lys Arg Thr 736 Glu Val Trp Ile Tyr Ala Ala Ile Gly Ser Ile Ile Val Phe Cys 751 Ile Gly Leu Gly Val Tyr Phe His Val Gln Arg Arg Lys Arg Ile 766 Arg Asp Asp Ser Arg Asp Asp Tyr Asp Phe Glu Met Ile Glu Asp 781 Glu Asp Glu Leu Gln Ala Met Asn Gly Arg Ser Asn Arg Ser Arg 796 Arg Arg Gly Gly Glu Leu Tyr Asn Ala Phe Ala Gly Glu Ser Asp 811 Glu Glu Pro Leu Phe Ser Asp Glu Asp Asp Glu Pro Tyr Arg Asp 826 Arg Gly Ile Ser Gly Glu Gln Glu Arg Glu Gly Ala Asp Gly Glu 841 His Ser Arg Arg SEQ ID Nos. 26 to 31 1 S PCR-primers for A.sojae promoter cloning Restriction sites are underlined Primer Sequence (5' - 3') SEQ Alp-1 GGAATTCGCGGCCGCGGTTATTCTGCGGAAGC
ID

No. G

EcolU Notl SEQ Alp-2 GGAATTCCCATGGTGAGAAGATTGTAAAG
ID

No.27 E~o~ Nco' SEQ Amy-1 GGAATTCGCGGCCGCAGATCTGCCCTTATAAA
ID

No. TCTCC

EcoRl Notl SEQ Amy-2 GGAATTCCCATGGATGCCTTCTGTGGGG
ID

No.29 E~~~ Nor SEQID AOGPDA GGAATTCGCGGCCGCCTATGAAACCGGAAAG

No.30 -1 EcoRl Notl SEQ AOGPDA GGAATTCTAGCCATGGTTTAGATGTG
ID

No.31 -2 EcoRl Ncol SEQ ID No. 32 The sequence of the Aspergillus sojae gpdA promoter region SEQ ID No. 33 The sequence of the Aspergillus sojae alpA promoter region SS

Table 1. Taxonomic scheme of the genus Aspergillus (Samson, 1992) GENUS SUBGENUS SECTION SELECTED SPECIESa "SUBSPECIESES~~b Aspergillus Circumdati Wentii A. wentii (glucosidase) Flavi/Tamarii A.oryzae (amylase, protease) A. tamarii'o"
A. sojae (fermented food, protease) A. parasiticus'o"
A. flavus'~"
Nigri A.J_~er -----------------------~A. pulverulentes (fermented food, A. phoenicis various proteins, A. awamori organic acids) A. foetidus A 1_______L__ A. usamii A. ficuum A. japonicus----------------~ A. aculeatus (endoglucanase) (glucosidase, galactanase) A. ellipticus A. tubin eg nsis -------------~ "A. niger"
Circumdati A. ochraceus'~" (xulanase) A. alliaceus'°"
Candidi A. candidus (lipase, glucosidase) Cremei A.itaconicus (organic acid) Sparsi A. sparsus Aspergillus Aspergillus AspergillusA. glaucus (fermented food) Restricti A. restrictus'"

Fumigati Fumigati A. fumi -ga~s'oX

Cervini Ornati Clavati Clavati A. ~iganteus'o"

Nidulantes Nidulantes A. nidulanstoX
Versicolores A. sydowii (lipase) Usti Terrei A. terreust°" (glucansae) Flavipedes a For the species selected for this list either the production of proteins/organic acids/fermented foods (indicated between brackets) and/or a DNA-mediated transformation procedure (indicated by underlining) is described, except for A. tamarii, A.
sparsus and A. ellipticus. Species recorded to produce toxins are indicated by ~o"
b Based on various methods the listed names may be considered synonymous to the given SPECIES name.

Table 2. The classification of the different ATCC strains Strains Morphology''Aflatoxin RAPD'~ PCRo~,R''PCRapA Classification"' Production ATCC ND no'' A. sojaeA. sojaeA. sojaeA. sojae 93625 type I

ATCC A. sojae no''' A. sojaeND A. sojaeA. sojae 119066 type I

ATCC A. oryzae no'' A. sojaeND A. A. oryzae 20235 type oryzae II

ATCC A. sojae no''' A. sojaeA. sojaeA. sojaeA. sojae 20245 type I

ATCC A. sojae no'' ND ND A. sojaeA. sojae ATCC A. sojae no'' ND ND A. sojaeA. sojae ATCC A. sojae no'' A. sojaeND A. sojaeA. sojae 42249 type II

ATCC A. sojae no'' ND ND A. sojaeA. sojae ATCC A. sojae no'' ND A. sojaeA. sojaeA. sojae ATCC ND ND ND ND A. A. oryzae 46250 oryzae IFO 4177 A. oryzae No' ND ND A. A. oryzae (CBS oryzae 205.89) Legend : ND = not determined REF: Ushijima S, Hayashi K and Murakami H (1982) The current taxonomic status ofAspergillus sojae used in Shoyu fermentation. Agric. Biol. Chem., 46:2365-2367, 1981.
~~ REF: Yuan GF, Liu CS and Chen CC (1995) Differentiation ofAspergillusparasiticus from Aspergillus sojae by Random Amplification of Polymorphic DNA. Appl. Environm.
Microbiol., 61:2384-2387.
'~ REF: Chang PK, Bhatnagar D, Cleveland TE and Bennett JW (1995) Sequence variability 1 ~ in homologs of the aflatoxin pathway gene aflR distinguishes species in Aspergillus section Flavi.
Appl. Environm. Microbiol., 61:40-43.
'~ Conclusion on classification drawn by TNO based on data presented in this table 5~ This strain was deposited at ATCC as A. oryzae, but later reclassified as A. sojae based on Yuan et al, 1995z~ and Chang et al, 1995' 15 6' This strain was deposited at ATCC as A. parasiticus, but later reclassified as A. sojae based on Ushijima et al, 1981' and Yuan et al, 19952 '~ REF: ATCC catalogue e~ REF: Liu BH, Chu FS (1998) Appl. Env. Microbiol., 64:3718-3723.

Table 3. Composition of selection media Non-selectionSelection Acrylamide Improved medium medium selection mediumacrylamide omposition (W097/041 selection medium 08) KHzP04 1.5 g/1 1.5 g/l 1.5 g/1 1.5 g/1 KCl 0.5 g/1 0.5 g/1 0.5 g/1 0.5 g/1 MgS04.7Hz0 0.5 g/1 0.5 g/1 0.5 g/1 0.5 g/1 NaN03 6 g/1 ---- ---- ----glucose 10 g/1 10 g/1 10 g/1 ----sorbitol 1.2 M ---- 1.2 M 1.2 M

saccharose ---- 1 M ---- ----mineral solution's0.1% v/v 0.1% v/v 0.1% v/v 0.1% v/v acetamide ---- 10 mM ---- ----acrylamide ---- ---- 10 mM 10 mM

CsCI ---- 15 mM 15 mM 15 mM

agar 15 g/1 15 g/1 15 g/1 15 g/1 5 ~~ mineral solution : CuSO4.SHz0 0.16 g/1 FeSOa.7Hz0 0.5 g/1 ZnSOa.7Hz0 2.2 g/1 MnCIz.4Hz0 0.5 g/1 CoClz.6Hz0 0.17 g/1 10 NazMo0a.2H20 0.15 g/1 H3BO3 1.1 g/1 EDTA 5 g/1 + + + + + + + +

+ + + + + + +

H ~
w + + . + . ~ + +

GA

00 ,~ o Two ~ ~ ~no ~ v, x ~O I~ l~ I~l~ ~ I~ C~I~ I~ M

U

U
N + + + + + + +
~

G1.

O
~ W + ~ + ~ ~ ~ ~ -~ +
,O

'b U
b x ~ N ~ ~ '~ v~v~ v~v~ O

Q. l~ 00 00 00~' I~l~ r l~ 00 it U + ~ + + + + +
f-~r .,.., a v~ o o ~n~n o 0 0 0 ~n ~n x [~ O V O ~ N M M ~ l~ 00 SOS ~ ~D l~ 00 00I~ 0000 0000 I~ M

O M ~ 00 00~ O ~ O
O_~d N M M N N N N
y ~ U N N N N ~ ~ ~ N
U a U U U V U U ~V
a a ~N V
~, ~ a a a a a a V a a ..

Legend : + (partial) degradation of proteins after 4 hours incubation, large milk clearing zone - no degradation of proteins after 4 hours incubation, small/no milk clearing zone Incubation at 30°C: 27 ~1 medium sample 2.5 ~.1 BSA (25 mg/ml) 0.5 ~1 Phytase (A.terreus, 3-4 g/1) BSA and phytase were added after the mediumsample was taken from the culture. This sample was incubated at 30°C and after certain timepoints the sample was analysed for the degradation of BSA and phytase.

Table 5. Protease activity at different pH values STRAINS Degradation of proteins in Minimal Medium +

Trusoy after incubation pH=4.5 pH=6 pH=8 BSA Phytase BSA Phytase BSA Phytase (A. terreus) (A. terreus) (A.

terreus) ATCC 9362 - - - - - +

ATCC - - - - - -ATCC + - + + + +

(= A. oryzae) ATCC - - - - - -Legend : + (partial) degradation of proteins alter 4 hours incubation - no degradation of proteins after 4 hours incubation Incubation at 30°C: 25 ~1 mediumsample SOr"Nt Na,a~ pH~.2 2 ~tl buffer (SOmM) "'~ SOmMNaAc pH=5.8 2.5 ~.tl BSA (25 mg/ml) SOmM Tris~Cl PH=8.3 0.5 ~l Phytase (A.terreus, 3-4 g/1) BSA, phytase and buffer were added after the mediumsample was taken from the culture.
This sample was incubated at 30°C and after certain timepoints the sample was analysed for the degradation of BSA and phytase.

Table 6. PCR results for cloning fungal pclA genes Expected size Primercombination PCR product 1 pcll + pcl2rev 180 by 2 pcll + pcl3 350 by 3 pcl l + pcl4 500 by 4 pcll + MBL1372 300 by pcl2 + pcl3 200 by 6 pcl2 + pcl4 350 by 7 pcl2 + MBL1372 150 by 8 MBL1298 + pcl2rev180 by 9 MBL1298 + pcl3 350 by MBL1298 + pcl4 500 by Strain Primercombination Trichoderma reesei QM9414+ -.y.- - - - - + - -=

Penicillium chrysogenum + +~ + - + - - + - -P2 ' Fusarium venenatum ATCC20334+ + + - - + - + + -:

Trametes versicolor TV - - - - - - - - - -Rhizopus oryzae ATCC200076'"~ - - - - - - + -:~~

Agaricus bisporus HORST - + - - - - - - ~+ -' Aspergillus sojae ATCC + + + - - + - + -positive control + + + + + + + + + +

Legend: + specific PCR product aspecific or no PCR product ~~'.'~ this PCR product was used for sequencing Table 7. The viscosity ranges of the various A. sojae strains Viscosity Strains (cP) Biomass (g/1) Shear rate Shear rate Shear rate 644.41/s 6.5 1/s 83.21/s A. sojae wild 2000 1505 155 8.8 and type 16.6 A. sojae pclA 2000 751 76 7.6 and 17.2 A. sojae lfvA 1565 94 18 6.9 and 18.8 10 Table 8. Promoter strength in A. sojae transformants Transformants PromoterGUS activity (U/mg) in Minimal Medium 5% 5% 5% 2% 5%

xylose glucose maltodextrinstarch trusoy ATCC 11906 wild --- 0 0 0 0 0 type ATCC11906[pGUSS gpdA 9141 6291 6667 6937 3391 4]

ATCC11906[pGUS6 glaA 33 50 25 S1 176 4]

ATCC11906[pBIPG bipA 2914 2849 1642 493 2083 US]

Claims (35)

1. A recombinant Aspergillus sojae comprising an introduced acetamidase S
(amdS) gene as selectable marker.

2. An Aspergillus sojae according to claim 1, said Aspergillus sojae being selectable on a medium comprising a substrate for the introduced amdS as sole source of nitrogen, said medium further comprising a carbon substrate and said medium being free of endogenous amdS inducing substrate.

3. An Aspergillus sojae according to claim 1 or 2, wherein the source of nitrogen is acrylamide.

4. An Aspergillus sojae according to any of the preceding claims wherein the Aspergillus sojae has no active endogenous amdS gene, for example because the endogenous amdS gene comprises an endogenous amdS inactivating mutation, e.g. a deletion or a disruption.

5. A method of introducing a nucleic acid sequence into Aspergillus sojae, said method comprising subjecting Aspergillus sojae to a method of introduction of a nucleic acid sequence e.g. transformation or transfection of the Aspergillus sojae in a manner known per se for introduction of a nucleic acid sequence into fungi, said method comprising the introduction of the amdS gene as the nucleic acid sequence (henceforth the introduced amdS gene) followed by selection of the resulting transformed or transfected Aspergillus sojae on a medium free of endogenous amdS inducing substrate, said medium further comprising a substrate for the introduced amdS
as sole source of nitrogen and said medium further comprising a carbon substrate, said medium enabling the desired Aspergillus sojae comprising the nucleic acid sequence to grow whilst eliminating growth of Aspergillus sojae free of the socalled introduced nucleic acid sequence due to inability of such Aspergillus sojae to grow without the introduced amdS gene on the selection medium, said medium suitably comprising a substrate for amdS other than acetamide, for example acrylamide as substrate for the introduced amdS as sole source of nitrogen.

6. An Aspergillus sojae obtained by the method of claim 5.

7. A method of selecting transformed or transfected Aspergillus sojae said method comprising subjecting Aspergillus sojae according to any of claims 1-4 and 6 to a method of transformation or transfection of the Aspergillus sojae in a manner known per se for transformation or transfection of fungi with a nucleic acid sequence, said method comprising the introduction of the amdS gene as the nucleic acid sequence followed by selection of the resulting transformed or transfected Aspergillus sojae on a medium comprising a substrate for the introduced amdS as sole source of nitrogen and said medium further comprising a carbon substrate, said medium enabling the desired Aspergillus sojae to grow whilst eliminating growth of non-transformed or -transfected Aspergillus sojae due to inability of such to grow without the introduced amdS gene on the selection medium.

8. A method for producing recombinant Aspergillus sojae, said method comprising introduction of a nucleic acid sequence into an Aspergillus sojae e.g. by transformation or transfection in a manner known per se according to any of claims 1-4 and 6, said nucleic acid sequence comprising a desired sequence to be introduced flanked by sections of an endogenous amdS gene or corresponding sequences of a length and homology sufficient to ensure recombination thus simultaneously eliminating the endogenous amdS gene and introducing the desired sequence, followed by selection of the recombinant Aspergillus sojae with the desired sequence by selecting for a selectable marker comprised in or transformed in cotransformation with the desired sequence, said selectable marker being absent in the Aspergillus sojae prior to introduction of the nucleic acid sequence, suitably the selectable marker being pyrG.

9. An Aspergillus sojae exhibiting growth with medium comprising uracil and fluoro-orotic acid, said Aspergillus further not exhibiting growth on medium comprising uridine and fluoro-orotic acid, i.e. said Aspergillus sojae exhibiting uracil auxotrophy, said Aspergillus sojae being unable to utilize uridine, said Aspergillus sojae being pyrG negative, said Aspergillus sojae exhibiting resistance to fluoro-orotic acid, said uracil auxotrophy and said fluoro-orotic acid resistance being relievable upon complementation with an active introduced pyrG gene, suitably said Aspergillus sojae being free of active endogenous pyrG genes; e.g. the Aspergillus sojae endogenous pyrG gene comprises a mutation in the form of an insertion, substitution or deletion in the gene or in a gene regulating sequence, e.g. a deletion of the whole coding sequence of the gene.

10. An Aspergillus sojae according to claim 9 in combination with the characteristics of an Aspergillus sojae according to any of claims 1-4 and 6.

11. A method of selecting transformed or transfected Aspergillus sojae, said method comprising subjecting Aspergillus sojae according to claim 9 or 10 to a method of transformation or transfection with a nucleic acid sequence, said method comprising introducing an active pyrG gene into the Aspergillus sojae in a manner known per se for transformation or transfection of fungi followed by selection of the resulting transformed or transfected Aspergillus sojae on a medium free of uracil and fluoro-orotic acid, said medium at least further comprising minimum substrates required for growth of Aspergillus sojae, said medium enabling the desired Aspergillus sojae to grow whilst eliminating growth of non-transformed or -transfected Aspergillus sojae due to inability of such to grow without uracil due to the inactivated pyrG
gene.

12. A method according to claim 11, wherein the active pyrG gene that is introduced is flanked by identical nucleic acid sequence fragments, and the pyrG positive Aspergillus sojae resulting from the introduction of the pyrG gene and the flanking sequences is selected on a medium free of uracil and fluoro-orotic acid and subsequently the pyrG positive Aspergillus sojae is cultivated on medium comprising uracil and fluoro-orotic acid thereby eliminating the pyrG gene that had been introduced thus resulting in a pyrG negative Aspergillus sojae that is selectable by growth on uracil comprising medium and by fluoro-orotic acid resistance, suitably the flanking sequences and the pyrG gene being further flanked by sequences that direct integration of the pyrG gene and the flanking sequences into a specific location due to the fact that the integration directing sequences are homologous to a specific sequence of the Aspergillus sojae to be transformed, thereby enabling knock out, if desired, of the gene associated with the specific sequence.

13. A method according to claim 11 or 12, wherein the Aspergillus sojae according to claim 9 or 10 has a further nucleic acid sequence introduced therein, preferably said further nucleic acid sequence encoding a protein or polypeptide, said further nucleic acid sequence being introduced with the active pyrG gene either on the same vector or by cotransformation with the active pyrG gene that is introduced.

14. A method of selecting transformed or transfected Aspergillus sojae by carrying out the method according to any of claims 11-13 in combination with the method of claim S.

15. A method for producing recombinant Aspergillus sojae, said method comprising introducing a nucleic acid sequence into a pyre positive Aspergillus sojae, e.g. by transformation or transfection in a manner known per se, said nucleic acid sequence comprising the desired sequence flanked by sections of the pyre gene or corresponding sequences of a length and homology sufficient to ensure recombination eliminating the pyre gene and introducing the desired sequence, followed by selection of the recombinant Aspergillus sojae with the desired sequence by selecting for Aspergillus sojae with a pyre negative phenotype.

16. A recombinant Aspergillus sojae obtained by a method according to any of claims 11-15, optionally further comprising the characteristics of an Aspergillus sojae according to any of claims 1-4, 6, 9 and 10.

17. A recombinant Aspergillus sojae comprising an introduced nucleic acid sequence encoding a protein or polypeptide for expression, said protein or polypeptide being susceptible to degradation upon expression by Aspergillus niger or Aspergillus awamori.

18. A recombinant Aspergillus sojae comprising an introduced nucleic acid sequence encoding a protein or polypeptide for expression, said protein or polypeptide being other than Aspergillus sojae protease and amylase, said protein or polypeptide preferably being a non-Aspergillus sojae protein or polypeptide.

19. A mutant or recombinant Aspergillus sojae comprising a mutation inactivating a protease gene, suitably an alkaline protease gene.

20. A mutant or recombinant Aspergillus sojae comprising a mutation inactivating the major protease gene, suitably a mutation inactivating the major alkaline protease gene, e.g. the gene encoding major alkaline protease gene of 35 kDa.

21. A method for producing recombinant Aspergillus sojae, said recombinant A.
sojae exhibiting reduced proteolytic activity, said method comprising introduction into an A.
sojae, e.g. by transformation or transfection in a manner known per se, of a nucleic acid sequence comprising a selectable marker encoding sequence to be introduced flanked by sections of the protease gene to be eliminated and further said flanking sequence and the selectable marker encoding sequence being comprised within sequences of a length and homology sufficient to ensure recombination at the protease gene thus simultaneously eliminating the protease gene and introducing the desired selectable marker encoding sequence, the introduction being followed by selection of the recombinant A. sojae by selecting for the selectable marker, whereby the A. sojae prior to the introduction of the nucleic acid sequence, e.g. by transformation or transfection, is free of the selectable marker to be introduced, e.g. the A.
sojae prior to the introduction of the nucleic acid sequence being mutated such that the A.
sojae cannot produce active selectable marker, suitably the selectable marker being the pyre gene, suitably the method being carried out together with the method according to any of claims 11-15.

22. A recombinant Aspergillus sojae obtained according to the method of claim 21.

23. A mutant or recombinant Aspergillus sojae according to any of claims 17-20 or 22 comprising a selectable marker, preferably amdS as defined in any of claims 1-4 or 6 and/or pyre as defined in claims 9, 10 or 16.

24. A recombinant Aspergillus sojae according to any one of claims 1-4, 6, 9, 10, 16-20, 22 and 23, comprising an introduced nucleic acid sequence encoding phytase or a protein having phytase activity.

25. A process of expression of an introduced nucleic acid sequence encoding a protein or polypeptide comprised in a recombinant or mutant Aspergillus sojae as defined in any of the claims 1-4, 6, 9, 10, 16-20, 22-24 or obtained via a method according to any of claims 5, 7, 8, 11-15 and 21, said process comprising cultivating the recombinant or mutant A. sojae, suitably the introduced nucleic acid sequence encoding a protein or polypeptide being absent in the corresponding non-transformed or wild-type A.
sojae and/or being present in a lower copy number.

26. A recombinant fungus comprising a mutation in a gene encoding a proprotein convertase or a functionally equivalent protein.

27. A fungus according to claim 26 exhibiting increased production of a protein, polypeptide or metabolite under equivalent conditions when compared to the corresponding wild-type fungus.

28. A fungus according to claims 26 or 27 said mutation being obtained by specific gene modification using transformation or transfection in a manner known per se.

29. A fungus according to claims 26-28, said proprotein convertase or functionally equivalent protein being encoded by a nucleotide sequence of which a fragment can be amplified by in vitro DNA amplification using any of two mixtures of nucleotides given in SEQ ID Nos. 10 to 16.

30. A fungus as described in claim 27, said proprotein convertase or functionally equivalent protein being encoded by a nucleotide sequence allowing functional complementation of the growth phenotype of an Aspergillus niger mutant comprising a mutation which inhibits the activity of a proprotein convertase or a functionally equivalent protein.

31. A fungus according to any of claims 26-30, said fungus being an Aspergillus sojae.

32. A fungus according to any of claims 26-30, said fungus further containing an introduced amdS gene or pyre gene.

33. A process for expressing a protein or polypeptide, preferably a recombinant protein or polypeptide, encoded by a nucleotide sequence, said process comprising cultivating a fungus according to any of the claims 26-32.

34. A process for producing a protein or polypeptide, preferably a recombinant protein or polypeptide, said process comprising a process of expression according to claim 33, optionally including processing and/or secretion and/or isolation of the expressed protein or polypeptide.

35. A process for producing a phytase or a protein having phytase activity, preferably a recombinant phytase or recombinant protein having phytase activity, said process comprising a process of expression according to claim 33, optionally including processing and/or secretion and/or isolation of the expressed phytase or protein having phytase activity.