WO1997011187A1 - Subtilisin inhibitor of streptomyces venezuelae, and use of the gene sequences for expression and/or secretion of heterologous proteins in streptomyces - Google Patents

Abstract

The present invention relates to a polynucleic acid coding for a protein of Streptomyces venezuelae with subtilisin inhibitor activity. The gene regulating sequences (promoter, secretion signal sequence etc.) can be used in a vector for the expression and secretion of heterologous proteins in Streptomyces.

Description

SUBTILISIN INHIBITOR OF STREPTOMYCES VENEZUEUE, AND USE OF THE GENE SEQUENCES FOR EXPRESSION AND/OR SECRETION OF HETEROLOGOUS PROTEINS IN STREPTOMYCES

The present invention relates to a novel subtilisin inhibitor protein of Streptomyces venezuelae and uses thereof, polynucleic acids encoding this protein, probes and primers derived from the same, the use of these polynucleic acid sequences in the expression and secretion of heterologous proteins and processes for production of heterologous proteins based thereon.

Protein protease inhibitors are widespread in nature They occur in animal and plant cells as well as in microorganisms. They can be classified into at least 10 families on the basis of their topological structure (Laskowski & Kato, 1980). Recently, it has been shown that proteinase inhibitors are also widely distributed among Streptomyces species (Taguchi et al. , 1993) and they secrete these compounds into the medium Most of these are inhibitors of serine proteases with a strong inhibitor activity towards subtilisin BPN' , but some also interact with trypsin or chymotrypsin. In addition, an inhibitor of metallo-proteases has been described in Streptomyces (Kumazaki et al., 1993)

To date, the complete amino acid sequence of several Streptomyces subtilisin inhibitors (SSI) has been determined These comprise plasminostreptin from Streptomyces antήibrinolyticus (Sugmo et al., 1978) (PSN), alkaline protease inhibitor (API-2C) from Streptomyces griseoincarnatus (Suzuki et al. , 1981 ), SSI from Streptonnces albogriseoliis S- 3253 (Obata et al., 1989), SLP1 (synonymous for STH or LEP 10) and LT1 (synonymous tor STI2) from Streptomyces lividans and Streploim'ces longispoms respectively (Strickler el ai , 1992; WO 88/01278), SIL2, SIL3 and SIL4 from Streptomyces parvuius, Streptomyces coelicolor and Streptomyces lavendulae, respectively (Taguchi et al. , 1994) According to the primary sequence similarity, these eight inhibitors can be divided into three groups (Taguchi el al. , 1994). In addition a number of other Streptomyces proteinase inhibitors were characterized, namely SIL1 from Streptomyces cacaoi (Kojima et al 1994), SIL8 from Streptomyces virginae (Terabe et al. , 1994a), SIL10, SIL 13 and SIL14 produced by Streptomyces thermotolerans, Streptomyces galbus. and Streptomyces azureus respectively (Terabe et al. , 1994b) and SIL15 from Streptomyces bikiniensis (Terabe et al , 1995) The total length of these different mature protease inhibitors from Streptomyces varies between 107 and 116 amino acids and a molecular weight of 10.9 and 1 1 .8 kDa. They form stable dimers (I₂) composed of two identical subunits and for subtilisin BPN' (E) it has been shown that they form an E₂I₂ complex (Takeuchi et al 1991 ) Structure function relationship of SSI has been investigated by crystallographic studies (Mitsui et al. , 1979 Takeuchi et al. , 1992) and by site-directed mutagenesis (Kojima et al. , 1990)

Streptomyces has already been extensively studied as a host for heterologous expression (for a review, see Brawner et al. ( 1991 ) and Anne and Van Mellaert ( 1993)) As a gram-positive bacterium, it offers the advantage of possible protein secretion directly in the culture medium Moreover, extensive fermentation experience is available with this organism for antibiotic production

However, as compared to E coli, the development of Streptomyces as a host for foreign expression is still in its very early stages Although there are many examples of heterologous expression in Streptomyces, the expression levels for all but a few proteins have been too low for most industrial applications There is still a need for nov el expression and/or secretion signals which may lead to higher expression levels and/or more efficient secretion of heterologous proteins.

Because of the efficient expression and secretion of subtilisin inhibitors, their promoters and signal peptides have been investigated for the secretion of heterologous proteins in Streptomyces including apidaecin 1b (Taguchi et al. , 1992) and soluble forms of human T-cell receptor CD4 (Fornwald et al. , 1993)

It is the aim of the present invention to provide for the polynucleic acids encoding a novel subtilisin inhibitor protein of Streptomyces venezuelae and the use of these new polynucleic acid sequences or fragments thereof for the heterologous expression of proteins in host cells, more particularly in Streptomyces

It is more specifically an aim of the present invention to provide for new polynucleic acid sequences which can be used to direct expression and/or secretion of heterologous proteins in host cells, more particularly in Streptomyces

It is more specifically an aim of the present invention to provide for new regulatory sequences which can be used to direct expression and/or secretion of heterologous proteins in host cells, more particularly in Streptomyces

It is in particular an aim of the present invention to prov ide for a new promoter sequence which can be used to express heterologous proteins in host cells more particularly in Streptomyces.

It is in particular another aim of the present invention to provide for a new ribosome binding site sequence which can be used to express heterologous proteins in host cells, more particularly in Streptomyces.

It is in particular another aim of the present invention to provide for new sequences encoding a secretion signal, capable of directing and/or optimalizing the secretion of heterologous proteins in host cells, more particularly in Streptomyces

It is more particularly an aim of the present invention to provide for specifically mutated (secretion signal encoding)-sequences, resulting in varying efficiencies of secretion of heterologous proteins in host cells, more particularly in Streptonwces . Said signal sequences may be used to optimize the secretion efficiency of heterologous proteins in Streptomyces.

It is also an aim of the present invention to provide for recombinant DNA molecules, more particularly recombinant vectors, for use in heterologous expression of proteins in host cells, more particularly in Streptomyces.

It is moreover an aim of the present invention to provide for a process for producing heterologous proteins in host cells, more particularly in Streptomyces .

It is in addition an aim of the present invention to provide for probes hybridizing specifically to the above-mentioned new polynucleic acid sequences.

It is in addition an aim of the present invention to provide for primers capable of specific amplification of the above-mentioned new polynucleic acid sequences.

It is also an aim of the present invention to provide for a novel subtilisin inhibitor protein isolated from Streptomyces venezuelae, functional analogues and fragments thereof, and the use of these proteins or protein fragments in the medical and/or non-medical field .

All of the above-mentioned aims have been achieved by the following embodiments of the invention.

As described in more detail in the Examples section, the current invention identifies a novel subtilisin inhibitor protein (VSI-protein) with a primary structure that is rather different from other known Streptomyces subtilisin inhibitors. It is secreted in high amount by Streptomyces venezuelae CBS 762.70. The isolation of the gene and its sequence analysis is disclosed. The deduced primary amino acid sequence is compared to the already known sequences of other known Streptomyces subtilisin inhibitor proteins In addition, the current invention shows that the promoter and signal sequence of the vsi-gene can be used for the highly efficient secretion of cytokines, such as mouse tumor necrosis factor (mTNF-α), human IL-2, human IL-10, and human GM-CSF by Streptomyces lividans.

The invention describes a polynucleic acid in substantially isolated form comprising a sequence of at least 10, more prefer ably at least 15, most preferably at least 20 contiguous nucleotides selected from:

(a) the polynucleic acid sequence as shown in Fig. 1 (SEQ ID NO 1 ). or

(b) the polynucleic acid sequences which are degenerate as a result of the genetic code to the polynucleic acid sequence as defined in (a) and which code for a polypeptide showing a subtilisin inhibitory activity, or

(c) the polynucleic acid sequences which hybridize to the polynucleic acid sequences as defined in (a) or (b).

Fig. 1 represents the sequence of the vsi-gene of this invention, isolated from Streptomyces venezuelae CBS 762.70, encoding a novel subtilisin inhibitor protein.

The term "polynucleic acid" corresponds to either double-stranded or single-stranded cDNA or genomic DNA or RNA, containing at least 10, 20, 30 40 or 50 contiguous nucleotides. A polynucleic acid which is smaller than 100 nucleotides in length is often also referred to as an oligonucleotide. Single stranded polynucleic acid sequences are always represented in the current invention from the 5' end to the 3 ' end

The term "hybridize to" refers to preferably stringent hybridization conditions, allowing hybridization between sequences showing at least 70 %, 80% , 90% , 95 % or more homology with each other.

The terms "homologous" and "homology" are used in the current application as synonyms for "identical" and "identity": this means that polynucleic acid (or amino acid) sequences which are said to be 80% homologous show 80% identical base pairs (or amino acids) in the same position upon alignment of the sequences

The Streptomyces polynucleic acids as disclosed abov e are preferably more than 70% , more preferably more than 80% , most preferably more than 90% homologous to the polynucleic acid sequence as shown in SEQ ID NO 1

The term "degenerate" refers to possible variations in the polynucleic acid sequence encoding the same protein, due to the possible occurrence of different codons for the same amino acid.

The expression "in substantially isolated form" refers to the tact that the polynucleic acid has been isolated from its natural environment, e.g. by purification techniques, or by recombinant DNA techniques, or by techniques of chemical synthesis of DNA It implies that said polynucleic acid is 90%, more preferably 95 % , most preferably 98% pure as measured by its weight versus the weight of possible contaminants .

The expression "subtilisin inhibitory acitivity" refers to an inhibitory action of the protein of the invention towards the activity of subtilisin. a serine protease. It is to be understood however that said inhibitory activity is not limited to subtilisin, and that it is also applicable to other serine proteases, and possibly also to trypsin and/or α-chymotrypsin.

In a more specific embodiment, the invention provides for a polynucleic acid in substantially isolated form comprising or consisting of at least part of the sequence as shown in Fig 1 (SEQ ID NO 1).

In another embodiment the invention provides for a recombinant DNA molecule comprising any of the polynucleic acid sequences as described above or at least part of them.

The term "recombinant DNA molecule" refers to any DNA molecule which has been made in vitro by DNA cloning techniques, and which may contain DNA originating from different sources ligated to each other.

Another embodiment of the invention provides for a recombinant DNA molecule comprising the gene expression unit of any of the polynucleic acid sequences as described above, or at least part of said gene expression unit.

The expression "gene expression unit" refers to those sequences involved in the expression of the gene to protein, i.e. its transcription and/or translation and/or the regulation thereof . The gene expression unit may thus comprise at least the promoter sequence, including possible promoter regulation sequences, the ribosome binding site, the sequence coding for the secretion signal, the sequence coding for the mature protein, and the termination sequence.

Another specific embodiment of the invention provides for a recombinant DNA molecule comprising the gene expression unit of the polynucleic acid sequence as represented by SEQ ID NO 1 , or part(s) of it, said gene expression unit comprising:

- the promoter sequence located in the region extending from pos (-686) to pos (-38) in Fig. 1 (SEQ ID NO 1), more specifically located in the region extending from position (-81 ) to pos (-38),

- the ribosome binding site (RBS) located in the region extending from pos (-37) to pos (-1) in Fig. 1 (SEQ ID NO 1 ), more specifically located in the region extending from pos (-15) to pos (-1) in Fig. 1 ,

- the sequence coding for the secretion signal, extending from pos ( + 1 ) to pos ( + 84) in Fig. 1 (SEQ ID NO 1),

- the sequence coding for the mature protein, extending from pos ( +85) to pos ( +439) in Fig. 1 (SEQ ID NO 1), and/or

- the termination sequence located in the region extending from pos ( +442) to pos ( + 976), more specifically located in the region extending from pos ( + 518) to pos ( +558) in Fig. 1 (SEQ ID NO 1).

It is to be understood that some elements of the gene expression unit (promoter, RBS, termination sequence) are localized in a sequence region, determined by way of sequence comparison with other gene sequences. The numbers delineating these regions therefore do not have to be taken exactly, but they can differ by one or several units.

Another specific embodiment of the invention provides for a recombinant DNA molecule as described above, comprising the regulatory sequence of SEQ ID NO 1 , or part of it, operably linked to a heterologous coding sequence.

The expression "regulatory sequence" can encompass the promoter sequence, and/or the ribosome binding site and/or the nucleotide sequence encoding the secretion signal.

Another specific embodiment provides for the use of any of the regulatory sequences of this invention in a method to direct and/or optimalize the expression and/or secretion of heterologous proteins in recombinant host cells.

The expression "operably linked to" refers to a juxtaposition where the components are configured so as to perform their usual function. Thus, regulatory sequences operably linked to a coding sequence are capable of directing the expression and eventually secretion of the coding gene.

The term "heterologous" means originating from another source organism than the host cell in which the protein is expressed. It the context of the present invention, where the host cell is preferably belonging to the genus Streptomyces. heterologous may refer to any organism other than Streptomyces, or to a Streptomyces strain which is different from the host strain which is used for expression.

Another specific embodiment of the invention provides for a recombinant DNA molecule as described above, comprising the promoter sequence of SEQ ID NO 1 . or part of it, operably linked to a heterologous coding sequence. The term "promoter sequence" refers to the nucleotide sequence in the DNA to which RNA polymerase binds to start transcription. In the context of the present invention, it may also include upstream sequences which may be involved in the regulation of said binding process.

Still another specific embodiment provides for a recombinant DNA molecule as described above, comprising the ribosome binding site of SEQ ID NO 1. or part of it, operably linked to a promoter sequence and to a second nucleotide sequence encoding a heterologous protein.

The term "ribosome binding site" refers to the sequence upstream of the initiation codon determining the site where the ribosomes bind to the mRNA to initiate protein translation. Said ribosome binding site is complementary to the 3'end of the 16S rRNA of the same organism.

Still another specific embodiment provides for a recombinant DNA molecule as described above, comprising the sequence encoding the secretion signal of SEQ ID NO 1 , or part of it, operably linked to a promoter sequence and to a second nucleotide sequence encoding a heterologous protein.

The term "secretion signal" or "signal peptide" refers to an oligopeptide, in most cases localised at the N-terminus of a polypeptide. said oligopeptide governing the secretion of said polypeptide, upon which cleavage of the secretion signal follows The resulting cleaved protein, which will in most cases be secreted, is also referred to as the "mature protein".

Still another specific embodiment provides for a recombinant DNA molecule as described above, comprising the regulatory sequence of SEQ ID NO 1 . or part of it, operably linked to the coding sequence of SEQ ID NO 1. or part of it wherein said coding sequence is fused in frame with a second heterologous coding sequence. This recombinant DNA molecule aims at the production of a fusion protein

The term "fusion protein" refers to a hybrid protein consisting of a sequence originating from the VSI-protein, or part of it, linked to the sequence of a heterologous protein.

In another embodiment, the invention provides for a recombinant DNA molecule as described above, wherein the sequence encoding the signal peptide of SEQ ID NO 1. i.e. the sequence extending from position (+ 1) to position ( + 84) in the polynucleotide sequence as represented in figure 1 , is specifically mutated resulting in one at least of the following amino acid substitutions:

R substituted by Q at amino acid position 2 of Figure 1 , and/or

R substituted by Q at amino acid position 3 of Figure 1 , and/or

T substituted by R at amino acid position 4 of Figure 1 , and/or

L substituted by R at amino acid position 5 of Figure 1. and/or

K substituted by Q at amino acid position 6 of Figure 1.

As will be exemplified further on in the Examples section, the above-specified mutations in the signal sequence lead to varying efficiencies of secretion of heterologous proteins in Streptomyces.

Another embodiment of the invention provides for a recombinant vector capable of transforming a host cell, and capable of replicating and/or directing expression in said host cell , wherein said recombinant vector comprises a recombinant DNA molecule as described above.

The term "vector" refers to any DNA molecule that replicates on its own in a host cell and can be used as a vehicle for replicating and/or directing the expression of another type of DNA (insert), and may include a plasmid, a phage, a cosmid or a virus. Preferably said vector is a plasmid.

The term "recombinant vector" refers to a vector containing DNA from different sources ligated together using DNA cloning-techniques.

The invention more particularly provides for a recombinant vector as described above wherin said host cells can be procaryotic or eucaryotic, but preferably belong to the genus Streptomyces, and more preferably to the species Streptomyces Iividans In the latter case, the vector is most preferably chosen from the pIJ-type plasmids e g pIJ702, pIJ486/7 (Hopwood et al. , 1987).

The invention also provides for a host cell transformed by a recombinant vector as described above, wherein said host cell can be procaryotic, e.g a bacterium like Eschericia coli or Streptomyces , or eucaryotic, such as plant cells, animal cells , or yeast cells.

In a specific embodiment the invention provides for a transformed host cell belonging to the genus Streptomyces, and being preferably Streptomyces Iividans .

In another embodiment the invention provides for a process for producing a heterologous protein from a host cell, comprising the steps of - growing a culture of a host cell transformed by a recombinant DNA vector as described above under conditions which enable the expression and possibly secretion of the heterologous protein,

- recovering said heterologous protein from said culture .

In a specific embodiment the invention provides for a process as described above, with said host cell belonging to the genus Streptomyces, and being most preferably Streptomyces Iividans.

In another embodiment, the invention provides for a recombinant polypeptide being the product of a process as described above.

In a subsequent embodiment the invention provides for a polypeptide in substantially pure form comprising an amino acid sequence, or part of it encoded by any of the polynucleic acid sequences as defined above.

More specifically, the invention provides for a polypeptide in substantially pure form, comprising a sequence of at least 9 contiguous amino acids selected from

(a) an amino acid sequence encoded by any of the polynucleic acid sequences as defined above, or

(b) the sequence of 146 amino acid residues as represented in Fig 1 (SEQ ID NO 2). or

(c) polypeptide sequences resulting from the modification of one or several amino acids in the polypeptide sequence as represented by SEQ ID NO 2, said modifications comprising substitution and/or deletion and/or addition of one or several amino acids, provided that said modficiation do not abolish the subtilisin inhibitory activity of the polypeptide encoded by SEQ ID NO 2.

The sequence as represented in SEQ ID NO 2 refers to a novel subtilisin inhibitor protein isolated from Streptomyces venezuelae CBS 762.70 (VSI-protein) as described in the examples section. It is characterized by an inhibitory activity towards proteases like subtilisin.

The subtilisin inhibitor activity can be determined by any method know n by the man skilled in the art, and more specifically by a method as described in the examples section, or as described by Kojima et al. (1990).

The term "polypeptide" designates a linear series of amino acids connected one to the other by peptide bonds between the alpha-amino and carboxy-groups of adjacent amino acids

Polypeptides can be in a variety of lengths, either in their natural (uncharged) forms or in forms which are salts, and either free of modifications such as glycosylation, side chain oxidation, or phosphorylation or containing these modifications It is well understood in the art that amino acid sequences contain acidic and basic groups, and that the particular ionization state exhibited by the polypeptide is dependent on the pH of the surrounding medium when the protein is in solution, or that of the medium from which it was obtained it the protein is in solid form. Also included in the definition are proteins modified by additional substituents attached to the amino acids side chains, such as glycosyl units, lipids. or inorganic ions such as phosphates, as well as modifications relating to chemical conversions of the chains, such as oxidation of sulphydryl groups Thus, "polypeptide" or its equivalent terms is intended to include the appropriate amino acid sequence referenced. sub|ect to those of the foregoing modifications which do not destroy its functionality.

The polypeptides of the invention can be prepared by classical chemical synthesis. The synthesis can be carried out in homogeneous solution or on solid phase. For instance, the synthesis technique in homogeneous solution which can be used is the one described by Houbenweyl in the book entitled "Methode der organischen chemie " (Method of organic chemistry) edited by E. Wunsh, vol. 15-1 et II. THIEME. Stuttgart 1974.

The polypeptides of the invention can also be prepared in solid phase according to the methods described by Atherton and Shepard in their book entitled "Solid phase peptide synthesis" (IRL Press, Oxford, 1989).

The polypeptides according to this invention can also be prepared by means of recombinant DNA techniques as described by Sambrook el al. , (Molecular Cloning: A

Laboratory Manual, 2nd Edition, New York, Cold Spring Harbor Laboratory Press, 1989).

The expression "in substantially pure form " signifies that the polypeptide has been purified from its natural environment, e.g. by protein purification techniques, or by production in vitro via recombinant DNA techniques, or by chemical protein synthesis techniques. It implies that said polypeptide has a degree of purity of at least 90% , preferably

95 % and more preferably of 98% of the protein expressed in its weight versus the weight ot possible contaminants, as determined by one or two dimensional PAGE (polyacrylamide gel electrophoresis).

The present invention also relates to proteins characterized by an amino acid sequence showing a homology of at least 60% , more preferably at least 70% more preferably at least

80% and most preferably at least 90% to the amino acid sequence as described in SEQ ID NO 2. Said "related" proteins may also be referred to as "muteins" of the protein shown in SEQ ID NO 2. The term "muteins" (or "functional analogues") may be defined as proteins containing substitutions and/or deletions and/or additions of one or several amino acids, provided that said muteins have retained the subtilisin inhibitor activity characteristic for the

Streptomyces protein of the invention. It should be evident that such muteins might have a slightly different molecular weight from the protein described by SEQ ID NO 2 as determined by SDS-PAGE. An overview of the amino acid substitutions which could form the basis for the above-mentioned muteins is shown in Table 1.

It is known from the prior art (e.g. Kojima el al. 1990) that specific mutations in the active site of subtilisin inhibitor proteins may cause alterations of the substrate specificity. It is therefor to be understood that the above-mentioned muteins may also show a different substrate specificity than the VSI-protein as isolated from Sirepiomyces venezuelae CBS

762.70.

In a very specific embodiment the invention provides for the novel VSI-protein of

Streptomyces venezuelae CBS 762.70, as shown in Fig. 1 (SEQ ID NO 2).

This invention thus provides for a polypeptide in substantially pure form comprising or consisting of at least part of the amino acid sequence as represented by SEQ ID NO 2.

In addition, the invention also provides fragments of said polypeptides, with said fragments still showing subtilisin inhibitory activity and/or with said fragments being able to elicit antibodies binding specifically to the above-mentioned polypeptides.

Most preferably, said fragments are located within the region of the polypeptide sequence extending from amino acid pos. 55 to amino acid pos. 1 18 as shown in Fig. 1 (SEQ

ID NO 2), and most preferably from amino acid pos. 55 to amino acid pos. 80.

The invention also provides for a fusion protein comprising a polypeptide of the invention, or a fragment thereof, fused to a heterologous protein.

Another embodiment of the invention provides for the above-mentioned polypeptides or polynucleic acids for use as a medicament, or for use as an additive in the non-medical field, for instance food/feed processing.

The polypeptides of the invention can be used in applications of different kind. In the medical and/or pharmaceutical field, the subtilisin inhibitory activ ity of the above-mentioned polypeptides may be applied to treat medical distresses e.g. in the modulation of blood coagulation (inhibition of plasminogen activator), in the dow nregulation of inflammatory reactions, in the treatment of invading tumors (inhibition of urokinase) . Moreover, in the non- medical field (e.g. food and fermentation industry) the above-mentioned polypeptides may be used to prevent proteolytic degradation (e.g. as additives)

Another embodiment of the invention provides for the use of the polypeptides or the polynucleic acids of the invention for the manufacture of a medicament for therapeutic use as specified above.

According to another embodiment, the present invention also relates to an oligonucleotide comprising at least 10 contiguous nucleotides selected from any of the polynucleic acid sequences as described above, for use as a specific hybridization probe detecting the polynucleic acids of the invention.

The term "probe" refers to single stranded sequence-specific oligonucleotides which have a sequence which is sufficiently complementary to hybridize to the target sequence to be detected. Preferably, these probes are about 5 to 50 nucleotides long more preferably from about 10 to 25 nucleotides.

According to the hybridization solution (SSC, SSPE. etc. ) , these probes should be stringently hybridized at their appropriate temperature in order to attain sufficient specificity.

However, by slightly modifying the DNA probes, either by adding or deleting one or a few nucleotides at their extremities (either 3' or 5'), or substituting some non-essential nucleotides ( i.e . nucleotides not essential to discriminate between types) by others (including modified nucleotides or inosine) these probes or variants thereof can be caused to hybridize specifically at the same hybridization conditions (i.e . the same temperature and the same hybridization solution). Also changing the amount (concentration) of probe used may be beneficial to obtain more specific hybridization results. It should be noted in this context that probes of the same length, regardless of their GC content, will hybridize specifically at approximately the same temperature in TMAC1 solutions (Jacobs et al., 1988).

The latter implies that variant probes contemplated within this aspect of the present invention can be defined as probes hybridizing with the same specificity as the probe they are derived from under different, but stringent, hybridization and wash conditions (different solutions, different concentrations of butter, different concentrations of probe, and/or different temperatures).

The oligonucleotides used as primers or probes may also contain or consist of nucleotide analoges such as phosphorothioates (Matsukura et al., 1987), alkylphosphoro- thiates (Miller et al., 1979) or peptide nucleic acids (Nielsen et al., 1991 Nielsen et al., 1993) or may contain intercalating agents (Asseline et al.. 1984).

As most other variations or modifications introduced into the original DNA sequences of the invention, these variations will necessitate adaptions with respect to the conditions under which the oligonucleotide should be used to obtain the required specificity and sensitivity. However, the eventual results of hybridization will be essentially the same as those obtained with the unmodified oligonucleotides.

The introduction of these modifications may be advantageous in order to positively influence characteristics such as hybridization kinetics, reversibility of the hybrid-formation, biological stability of the oligonucleotide molecules, etc.

In a very specific embodiment the oligonucleotide probe comprises a sequence chosen from among the following group of sequences:

SEQ ID NO 3: CGCTCTACGCCCCCTC, or

SEQ ID NO 4: CTTCATSGTGCACTGGTTG.

According to another embodiment, the present invention also relates to an oligonucleotide comprising at least 10 contiguous nucleotides selected from any of the polynucleic acid sequences as described above, for use as a primer for specific amplification of any of the polynucleic acids of the invention.

The term "primer" refers to a single stranded DNA oligonucleotide sequence capable of acting as a point of initiation for synthesis of a primer extension product which is complementary to the nucleic acid strand to be copied. The length and the sequence of the primer must be such that they allow to prime the synthesis of extension products. Preferably the primer is about 5-50 nucleotides long, more preferably from about 10 to 30 nucleotides long. Specific length and sequence of the primer will depend on the complexity of the required DNA or RNA targets, as well as on the conditions of primer use such as temperature and ionic strength.

The fact that amplification-primers do not have to match exactly with the corresponding template sequence to warrant proper amplification, is amply documented in the literature (Kwok et al. , 1990).

In a specific embodiment, the oligonucleotide primer comprises a sequence chosen from among the following group of sequences:

SEQ ID NO 7: TTCGCCAACCAGTGCACGAT or

SEQ ID NO 8: CTTGAGGGTGCGACGCATGG or SEQ ID NO 9: ACGAAAGGTCACCGGGACAT or

SEQ ID NO 10: CATTTCAGATCTGCACCC.

In a specific embodiment, the present invention also relates to the optimalisation of the secretion of heterologous proteins in host cells by modifying the amino acid sequence of said secretion signal through site-specific mutagenesis. As further illustrated in the examples section, said site-specific mutagenesis may be achieved using at least one of the following oligonucleotide primers:

SEQ ID NO 16: GCTCCCACGGCCTGCAGGGTCTGCTGCATGGTGAACTCTCC SEQ ID NO 17: GCTCCCACGGCCTGCAGGGTCTGACGCATGGTGAAC

SEQ ID NO 18: GCTCCCACGGCCTGCAGGGTGCGACGC

SEQ ID NO 19: CCACGGCCTTGAGACGTCGACGCATGGTGAAC

SEQ ID NO 20: CCCACGGCCTTGCGACGTCGACGCATGGTGAAC

resulting respectively in modification of the following amino acids in the signal sequence:

R substituted by Q at position 2 of Figure 1 ,

R substituted by Q at position 3 of Figure 1 ,

T substituted by R at position 4 of Figure 1 ,

L substituted by R at position 5 of Figure 1 ,

K substituted by Q at position 6 of Figure 1.

Table Legends

Table 1 : Amino acid substitutions which may form the basis of the muteins according to the present invention. Table 2: E. coli strains used and their genotype.

Table 3: 3a: E. coli plasmids used in the current invention.

3b: Streptomyces plasmids used in the current invention. Table 4: Primers used in the DNA sequencing reactions.

Table 5: Comparison of mTNF yield detected in supernatants of transformed S. Iividans cultures.

Table 6: Oligonucleotides used for in vitro mutagenesis.

Figure legends: Fig 1:

Nucleotide sequence of the cloned 1.6 kb fragment and the deduced amino acid sequence of the S.venezuelae subtilisin inhibitor. The amino acid residues identified by N-terminal amino acid sequence analysis are underlined. Direct and inverted repeats are indicated by specific arrows above the lines and the -35 and -10 region of the vsi putative promoter are underlined, is the transcription initiation site and "↓" is the signal peptidase cleavage site. Fig_2:

(a) SDS-PAGE of the culture filtrate of S. venezuelae CBS 762.70.

standard: low molecular weight marker (3 μg/protein: Pharmacia)

lane 1 : about 2 μg total protein

lane 2: about 10 μg total protein

(b) SDS-PAGE of fractions 47, 49, 51 , 53 and 55 following elution from the Sephadex G50 column. Fig 3:

(a) Restriction map of the 5kb PvuII fragment of S.venezuelae CBS 762.70 chromosomal DNA hybridizing with the deoxyoligonucleotides 5'-CGCTCTACGCCCCCTC-3' (SEQ ID

NO 3) and 5'-CTTCATSGTGCACTGGTTG-3' (SEQ ID NO 4) complementary to the DNA sequence of the N- and C-terminal part of VSL

(b) Sequencing strategy of the 1.6 kb subfragment of the 5.0 kb PvuII fragment containing the vsi gene.

(c) FRAME analysis (Bibb et al. , 1984) of the 1 .6 kb nucleotide sequence with potential start

( | at the bottom line) and stop ( | in the middle) codons.

Fig 4:

Comparison of the signal peptides of subtilisin inhibitors from S venezuelae (VSI), S. Iividans (SLP1), S.longisporus (STI2) and S. albogriseolus (SSI). Italic sequences represent possible pro-protein sequences which are possibly cleaved off during maturation of the protein.

Fig_5 :

Comparison of the different promoters of Slreptomyces subtilisin inhibitors of S albogriseolus ( ssiPI and ssiP2), S.Iividans 1362-K12 (stiPI) and S.venezuelae vsiPI . Short direct repeats are indicated by arrows.

Fig 6:

Alignment of amino acid sequences of the mature forms of 18 protease inhibitors of Streptomyces spp. of which the complete amino acid sequence is known β-strands and α- helices known from crystallographic studies (Takeuchi et al., 1992 ) are indicated. The proteinase inhibitors are from S.venezuelae CBS 762.70 (VSI) , S.albogriseolus (SSI), S. gr iseoincarnatus (API-2c'), S. Iividans (STI1), S.coelicolor (SIL3), S. ambofaciens (SIL-), S.parvulus (SIL2), S. antifibrinolyticus (PSN), S.galbus (SIL 13). S.longisporus (STI2), S. lavendulae (SIL4), S. thermotolerans (SIL10), S.azureus (SIL 14), S.fradiae (SIL5),

S.hygroscopicus (SIL12), S.virginae (SIL8), S.bikiniensis (SI L 15 ), and S.cacaoi (SIL 1 ) Bold characters: residues common to eight or more proteinase inhibitors among the 18 alligned; "*": perfectly conserved residues, "↓" : reactive site of inhibitors," .": well conserved residues

Fig 7:

N-terminal amino acid sequence analysis of the abundantly secreted proteins in 48 h cultures of S.lividans TK24[pCBS(2)mTNF].

Fig 8:

SDS-PAGE [12.5 %(w/v)gels] (A) and immunoblot analysis (B) of culture filtrate of S.lividans TK24 [pCBS(2)mTNF] grown for 48h in NM-medium.

Samples in panel A are from cultures" of S.lividans TK24 [pIJ486| (lane 1 ), (250μl), and S.lividans [pCBS(2)mTNF] (lane 2-4) (250μl, 125μl, 62μl, resp .) . Lane 5 is rmTNF ( 1μg) and lane 6 is Gibco-BRL low molecular weight protein standard Panel B represents 50ng rmTNF (50ng) and 10μl of S.lividans TK24[pCBS(2)mTNF]. All culture filtrates were precipitated with 20% trichloroacetic acid and dissolved in gel loading butter,

protein content of culture filtrate is 60-80μg/ml.

Fig, 9:

Phylogenetic tree of the subtilisin inhibitor proteins of the 18 Streptoimyces spp. for which the amino acid sequence of the complete mature protein was available . Abbreviations see legend to Fig. 6.

Fig , 10:

Immunoblot analysis of culture filtrate of S. Iividans TK24 [pCBS2hlL 10] grown for 30h in NM medium.

Lane 1. Gibco-BRL low molecular weight protein standard, lane 2 1 ml of S Iividans TK24 [pCBS2hIL10] culture filtrate (protein content 80 μg/ml )) precipitated with 20% trichloroacetic acid and dissolved in gel loading butter No immunoreactive bands were visible in the S. Iividans TK24[pIJ486] culture filtrate (not show n).

Figure 1 1. Schematic representation of

A the construction of the vsi-mTNF secretion cassette from the plasmid pCBS2mTNF and B. the mutations introduced in the original VSI signal peptide (vsi-P l : promoter of the vsi-gene, SSvsi: signal sequence of the vsi-gene, SD: Shine Dalgarno sequence, SPase: signal peptidase cleavage site, N: hydrophilic region of the signal peptide. H : hydrophobic region of the signal peptide, C: C-terminal region of the signal peptide).

Figure 12. A. Intracellular, extracellular and total amounts of mTNF produced by S. iividans under the influence of the different signal peptide mutants as obtained by biological activity measurements. B. Coomassie staining of secreted proteins of S. Iividans cultures, harbouring the different signal peptide mutants, after TCA precipitation and SDS-PAGE. Equivalents of 50 μl culture fluid were applied on gel. C. Graphical representation of secretion efficiencies calculated as secreted mTNF/(secreted mTNF + intracellular mTNF) × 100.

Figure 13. Immunoblot of secreted (a) and intracellular mTNF (b). For each mutant, an equivalent of 20 μL culture fluid ar.d a corresponding amount of cell lysate were used for

SDS-PAGE and subsequent blotting.-: S.lividans [plJ486]

Figure 14. Slot blot of total RNA isolated from S. lividans strains, containing the different signal peptide mutants, hybridised with A: a mTNF cDNA specific probe and B: a chromosomal sti l -probe. Next to the blots the average OD. obtained by scanning, is indicated.

Abbreviations:

MCS = multiple cloning site, Tc^R = tetracycline resistance, Ap^R = ampicillin resistance, Cm^R = chloramphenicol resistance, Neo^R = Neomycin resistance, Tsr^R = thiostrepton resistance, sBLA= synthetic Bovine alpha-LActalbumine

Table 5. Comparison of mTNF yield detected in supernatants of 48 h cultures of S.lividans with mTNF under the control of (a) S.venezuehae α-amylase (Van Mellaert et al., 1994) or ( b) vsi regulatory sequences and signal peptide.

In both cases (a) and (b) below. N-terminal sequence analysis was canied out on 2 distinct immunoreactive bands. The higher band contained the correctly cleaved fusion protein (+1) or a truncated form (+4): the lower band consisted of truncated forms of mTNF. the numbering of which starts at the first amino acid of mTNF.

EXAMPLES

Example 1: Materials and Methods

Strains and plasmids

Streptomyces venezuelae CBS 762.70 (= ATCC 14583) was used as source for the isolation of the subtilisin inhibitor gene. S.lividans TK24 (Hopwood et al., 1985) was the host strain for maintenance and isolation of the Streptomyces plasmids pIJ486 (Ward et al., 1986) and the newly constructed secretion vectors (table 3). For cloning, several Escherichia coli strains have been used (Table 2). Clones and subclones were constructed with the E.coli vectors summarized in Table 3.

Media and Buffers

S. venezuelae and S.lividans were grown at 27°C. 300rpm in NM medium, i.e. Nutrient Broth N°2 (Lab B Bury, U.K.) at pH 7.0 with 0.05M MOPS (= 3-[-N-motpholino]-propane sulfonic acid). S. lividans transformants were obtained and grown on RII -medium as described by Hopwood et al. (1985). E.coli was cultured at 37°C in Luria-Bertani (LB) medium (Miller, 1972). where necessary supplemented with ampicillin ( 100μg/ml ). chloramphenicol ( 10μg/ml). or tetracycline (15μg/ml) for selection, and containing 15g agar/L. when applicable. Purification of extracellular proteins

The proteins in the supernatant of a 48h culture of S.venezuelae. obtained by centrifugation (3000xg, 10min, 4°C), were first precipitated by the addition of solid (NH₄)₂SO₄ to 80% saturation. The precipitate was collected by centrifugation ( 12000xg. 10min. 4ºC) and subsequently dissolved in 7ml of water. Next, the solution was applied on a Sephadex superfine G50 column (2.4×280cm) and the proteins were eluted with 50mM NH₄HCO₃ in the presence of 0.2% NaN₃ at a flow rate of 20mL/h. The effluent was collected in 2ml fractions and detection of the protein content was at A₂₈₀ and A₂₃₀. Fractions containing proteins were controlled by SDS-PAGE ( 12.5%) (Laemmli. 1970) to identify the protein content of the tractions.

Endoproteinase Lys-C digest of subtilisin inhibitor was separated by HPLC on an

Aquapore C8RP column (2.1×220mm) (Applied Biosystems). equilibrated with a solution of 0. 1 % trifluoro- acetic acid in water. The column was eluted with a linear gradient from 0 to 100% with solvent (0.1% trifluoro-acetic acid and 80% acetonitrile in water) at a rate of 400μl/min and 400μl fractions were collected.

Amino acid sequence analysis

The N-terminal amino acid sequence of the relevant proteins were determined by

Edman degradation on an automatic 477A-120A protein sequencing system (Applied Biosystems). To determine internal amino acid sequences, the purified subtilisin inhibitor protein was digested in a total volume of 1ml using 1 μg of endoproteinase Lys-C (sequencing giade, Boehringer Mannheim) for 18 h at 37°C using a protocol as described by the manufacturer. After digestion the peptides were separated on HPLC as described above, and particular fractions were applied to the automatic protein sequencer

DNA manipulations

Chromosomal DNA of S.venezuelae was extracted from cells by a modified method of Womble et al. (1977). Briefly, S.venezuelae was grown in S medium supplemented with 0.5% glycine inoculated with a seed culture (0.1%). After growth (27°C. 350rpm) for 24h. 40ml cultures were collected by centrifugation (3000xg. 10min) and washed twice with buffer A (0. 15M NaCl 0.1M EDTA. 10mM HEPES. pH8.0) prior to incubation with lysozyme ( 500μ g/ml) in 5ml of buffer A. After 10min at 37ºC. the cells were lysed by the addition of 50μl of 25% (w/v) SDS. Subsequently, 0.5 ml pronase solution (20mg pronase/ml H₂O and placed at 30°C for 10 min for autodigestion) was added and the suspension was further incubated for 2h (37°C) followed by 30min at 65ºC and shearing by repeated suckening in a syringe. Chromosomal DNA was purified by EtBr-CsCl gradient centrifugation ( lg CsCl/ml) (45h. 100.000xg). Purified chromosomal DNA was digested with several restriction endonucleases and separated on a 1% (w/v) agarose gel. The DNA digest was blotted onto

Hybond-N (Amersham) using a Vacugene blotting system ( Phaimacia). Restriction endonucleases and DNA modifying enzymes were from Boehringer Mannheim. Biolabs and Gibco-BRL and they were used according to the manufacturers' protocol.

The DNA fragment containing the subtilisin inhibitor gene was isolated by hybridization experiments with synthetic deoxy-oligonucleotides as probe. These deoxy-oligonucleotides consisted of 5'-CGCTCTACGCCCCCTC-3' (SEQ ID NO 3 ) (sense strand) (T_m - 58°C) and 5'-CTTCATSGGTGCACTGGTTG-3' (SEQ ID NO 4) (antisense strand) (T_m = 56°C), corresponding to the N-terminal sequence of the N- and C-terminal part of the protein, respectively. They were labeled with γ-[³²P] ATP (3000Ci/mmol ) ( Amersham) using T4-polynucleotide kinase (Boehringer- Mannheim). Hybridization of these probes with chromosomal DNA cross-linked to Hybond-N was canned out at 48ºC overnight in the hybridization solution as described in the Amersham protocol. After hybridization, the membranes were washed in 2xSSC, 0.1% SDS twice at room temperature, and subsequently, at the T_m -value of the deoxyoligonucleotide.

DNA sequence analysis

DNA sequencing was carried out using the chain terminating method developed by

Sanger et al. (1977) either with [³⁵S]-labeled fragments or using fluoro-labeled primers, when using the automatic ABI DNA sequencing system (Applied Biosystems Inc.). Fragments were separated on a 6% polyacrylamide gel [acrylamide:bis-acrylamide ( 19. 1 )] with 7.0M urea in TBE buffer (60mM Tris, 50mM boric acid 1.5mM EDTA) at 1500V (35 W. 35mA). Prior to sequencing, several restriction fragments, obtained by digesting the original approx. 5kb Pvull fragment of S. venezuelae chromosomal DNA. containing the subtilisin inhibitor gene with several restriction enzymes, were subcloned in pBluescript KS ( + ). p ACYC184 or pUC19. All DNAs for sequencing were purified by CsCl-EtBr gradient centrifugation. Sequencing primers were the Ml 3 universal and reverse primer for subclones in pUC 19. T3 and T7 primers for those in pBluescript and primers 1565 (SEQ ID NO 1 1 ). 1566 (SEQ ID NO 12). 93F17 (SEQ

ID NO 5) and 93F18 (SEQ ID NO 6) (Table 4) for subclones in p.ACYC 184. In addition, a number of subtilisin inhibitor specific primers were used: 93G234 (SEQ ID NO 7). 93G314 (SEQ ID NO 8). 93G31 (SEQ ID NO 9). 93K145 (SEQ ID NO 10) (Table 4).

Computer aided analysis

The sequencing data were processed using the Gel Assembler system of PC Gene (Genofit. IntelliGenetics. Inc.) and the protein coding sequences were identified by virtue of the similarity of their codon usage to a codon frequency table by FRAME analysis (Bibb et ai. 1984). A phylogenetic tree was constructed using the CLUSTAL program of PC Gene. Primer extension studies

Total RNA was extracted as described by Hopwood et al. ( 1 985 ) from 24h cultures of S. venezuelae CBS 762.70 grown in NM medium. Seventy μg of RNA in a total volume of 30μl was hybridized for 2h at 60°C with a [³²P]-y-ATP labeled (6.6× 10⁵cpm) antisense oligonucleotide 93G314 5'-CTTGAGGGGTGCGACGCATGG-3' ( SEQ I D NO 8) (T_m = 66°C) complementary to the sequence of the 5' end of the subtilisin inhibitor gene. .After hybridization, the solution was immediately cooled to 0°C. Next, primer extension was canied out by the addition of 30 u AMV-reverse transcriptase (AMY-RT) ( Boehringer Mannheim) in 70μl AMV-RT reaction buffer (Vögtli and Hütter. 1987 ) and incubation at 42°C for 1h. The primer extension product was purified by consecutive treayment with phenol: chloroform

( 1 : 1 ) and precipitation with ethanol. Further purification was by elution from a NAP- 10 column (Pharmacia). By comparison of the labeled primer extension product with a sequence ladder on a 6% polyacrylamide gel containing 7.0M urea in TBE-buffer. the nucleotide corresponding to the mRNA start could be determined.

Protease inhibitor test

Inhibitory activity against subtilisin BPN (7.81U/mg) ( Sigma. Protease Type XXV II), trypsin (10.600U/mg) (Sigma) and α-chymotrypsin (44U/mg) ( Sigma ) were tested using Azocoll (Sigma) as a substrate for protease activity. Therefore. 150μl supernatant of S. venezuelae cultures grown for 33h in NM-medium were incubated with 0.04 to 10 units of protease in 10mM sodium maleate buffer (pH 6.2 ) in a total volume of 500μl. After 10min incubation at room temperature, 5ml of buffer solution containing Azocoll ( 5mg/ml ) was added and slowly shaken at 27°C. At regular time intervals the suspension was centrifugated and OD₅₂₃ of the supernatant measured. Inhibitory acitivity against subtilisin BNP' was also tested using N-succinyl-L-Ala-L-Ala-L-Pro-L-Phe-p-nifroanilide as described by Kojima et al. ( 1990).

Construction of the pCBS2mTNF and pCBSmTNF secretion vectors

For the construction of the pCBS(2)mTNF secretion vector, a 574 bp (nucl. pos. 319- 893) Smal -fragment (see Fig. 3a) containing part of the subtilisin inhibitor gene including the promoter. RBS. signal sequence and the coding region of the N-terminal end of the mature protein was cloned in the Smal site of pBSDK. resulting in pBSDK-CBSS The 5^,-piOtruding ends of the DraII digested (unique site) pBSDK-CBSS were filled in using a Klenow polymerase fill-in reaction (by which manipulation a blunt end site was fonned 2 ttiplets downstream the vsi signal sequence). Subsequently. pBSDK-CBSS was digested with EcoRI. A 0.7kb SnaBI/EcoRI fragment of pIG2mTNF-m7 containing the coding region of recombinant mTNF (in pIG2mTNF the first amino acid of mTNF was changed from L to Y) could thus be fused in frame in such a manner that the second amino acid of the subtilisin inhibitor gene was followed by the first amino acid (Y) of mTNF. A direct fusion between the vsi signal sequence and the mTNF cDNA was also achieved by introducing a unique NsiI site at the 3' end of the vsi signal sequence using PCR mutagenesis. The resulting pBS-CBSΝ vector was digested with NsiI. Subsequently, the 3'-protruding ends were removed in a T4

DΝA polymerase reaction. After EcoRI digestion of the vector, the above-mentioned 0.7 kb SnaBI/EcoRI fragment of pIG2mTΝF-m7 was cloned in the vector The correctness of both constructs, i.e. pCBS2mTNFEc and pCBSmTNFEc. was verified by DNA sequence analysis. From both plasmids the module containing the subtilisin inhibitor transcription and translation start signals and the mTNF gene fused, either directly or indirectly, in flame with the subtilisin inhibitor signal sequence was ligated in pIJ486 after proper digestion with BamHI and EcoRI.

Construction of the pCBS2hIL10 secretion vector

For the construction of pCBS2hIL10, the pBSDK-CBSS v ector was digested with DraII. The

5'-recessed ends were subsequently filled in using a Klenow polymerase fill-in reaction. In this vector the 530 bp NsiI (blunted in a T4 DNA poKmerase reaction ) HindIII fragment of pIGFH20hIL 10 containing the WL10 cDNA could be cloned

In pCBS2hIL10Ec plasmid the second codon of mature vsi was followed by the first hIL10 cDNA codon. The conectness of the fusion was checked by restriction analysis of the plasmid since an in-frame fusion resulted in a unique StuI site. The module from pCBS2hIL10. containing the vsi regulatory sequences and the hIL 10 gene fused in frame with the vsi signal sequence, was cloned in a BamHI digested pI J486 vector . Construction of the pCBS2hIL2 and pCBShIL2 secretion vectors

In order to obtain the pCBS2hIL2 secretion vector. pBSDK-CBSS was digested with DroII. After filling in the DraII 5'-protruding ends using a Klenow polymerase fill-in reaction, the vector was digested with EcoRI. Subsequently, the 412bp BspMI (with filled in 5'-protruding ends)/EcoRI fragment from pUC 18hIL2 containing the hIL2 cDNA was ligated to the vector resulting in pCBS2hIL2 Ec in which the hIL2 cDNA was fused downsfream the first 2 codons of mature vsi.

For the construction of the pCBShIL2 secretion vector, the mentioned BspMI/ EcoRI fragment was ligated to pBS-CBSN digested with N siI (3 -protruding ends blunted in a T4 DNA polymerase reaction) and EcoRI. These manipulations resulted in the plasmid pCBShIL2Ec in which the hIL2 cDNA was directly fused to the vsi signal sequence.

The fusions in pCBS2hIL2Ec and pCBShIL2Ec were controlled by DNA sequence analysis. From each plasmid the module containing the vsi regulatoiy sequences and the hIL2 cDNA fused to the vsi coding sequence was cloned in the BamHI site of pl.1486.

Construction of the pCBS2hGM-CSF secretion vector

For the construction of pCBS2hGM-CSF. the 525bp hspl Hindlll tiagment of pIGH2hGM-CSF-m5 containing the hGM-CSF cDNA was ligated to pBSDK-CBSS. Therefore. this vector was digested with DraII and the 5 '-protruding ends blunted using a Klenow polymerase fill-in reaction. Subsequently, the vector was digested with Hindlll. The ligation reaction resulted in a fusion of the hGM-CSF cDN A 2 triplets down-stream the vsi signal sequence. The conectness of the fusion was v erified by DN A sequence analysis. The module of the plasmid designated pCBS2hGM-CSF Ec. containing the vsi regulatory sequences and the hGM-CSF cDNA fused in-frame downstream the first 2 codons of the mature vsi sequence was cloned in a SacI/ HindIII digested pl.1486 v ector.

Construction of the pCBS2sBLA and pCBSsBLA secretion vectoi

As in the other examples, a direct and indirect fusion with the vsi signal sequence was established. The sbla gene located on a 399bp Mnll/EcoRI fragment of pEPsBLA was. therefore, ligated either to pBSDK-CBSS digested with DraII ( blunted using a tfll-in Klenow polymerase reaction and EcoRI or to pBS-CBSN digested with N si I ( treated with T4 DNA polymerase) and EcoRI. In this way, the plasmids pCBS2sBL.AEc and pCBSsBLAEc were constructed with the sb/a gene fused 2 triplets downstream the vsi signal sequence or directly to the vsi sequence, respectively. The correctness of these fusions was verified by DNA sequence analysis.

From both plasmids the module containing the vsi regulatory sequences and the sbla gene fused directly or indirectly in-frame with the vsi signal sequence was inserted in pl.1486 digested with BamHI and EcoRI.

Construction of the pCBS2hIL10Ec secretion vector

For the construction of pCBS2hIL10. the pBSDK-CBSS vector was digested with DraII. The 5'-recessed ends were subsequently filled in using a Klenow poKmerase fill-in reaction. In this vector the 530 bp NsiI (blunted in a T4 DNA polymerase reaction)/HindIII fragment of pIGFH20hIL10 containing the hIL10 cDNA could be cloned.

In pCBS2hIL10Ec plasmid the second codon of mature vsi was followed by the fiist hIL10 cDNA codon. The correctoess of the fusion was checked by restriction analysis of the plasmid since an in-frame fusion resulted in a unique StuI site. The module from pCBS2hIL10Ec, containing the vsi regulatory sequences and the hIL 10 gene fused in frame with the vs i signal sequence, was cloned in a BamHI digested pI J486 vector.

Biological activity determination, immunoblot assay and N-teiminal amino acid acid sequence analysis of mTNF

Methods for biological activity determination and immunoblot analysis have been previously described (Van Mellaeit et al., 1994). N-tenninal amino acid sequence analysis has been carried out on proteins in culture filtrates separated on 12.5.% SDS-PAGE gels stained with Coomassie blue and blotted on PVDF membranes ( BioRad)

Biological activity determination and immunoblot assay of hIL 10

Biological activity determination of hIL10 is carried out using a I L- 10 sensitive mouse cell line MC9. which proliferates in the presence of IL- 10. as described by Thompson-Snipes et al., ( 1991). ³H-Thymidine is added after 24 h of culture and counting of cell proliferation is measured in a γ-counter after another 18 h of incubation. One unit (U) of biological activity is defined as the amount of hIL10 required to mediate half maximal proliferation of MC9 cells. The immunoblot analysis was carried out in the same

way as described for mTNF using rabbit hIL10 antibodies (Innogenetics) as first antibody

Immunoblot assay of hIL2 and hGM-CSF

Immunoblot analysis of hIL2 and hGM-CSF was canied out as described for mTNF using rabbit antibodies (Innogenetics) against hIL2 and hGM-CSF. respectiv ely.

Detection of sBLA

The detection of sBLA occurred by a sandwich immunoradiometric assay (iRMA. Viaene et al., 1991). Therefore, the wells of a microtiter plate were coated with polyclonal anti-BLA antibodies. After contact with the samples. ¹²⁵I-labeled anti-BLA antibodies were added. Binding of those antibodies - and thus the presence of sBLA - was detected by measurement of the radioactivity.

Example 2: Cloning and sequence analysis of the vsi-gene of Streptomyces venezuelae CBS 762.70

I N-terminal determination of an abundantly secreted 13kPa protein

During a search for Streptomyces spp. secreting high lev els of a protein, we have identified following SDS-PAGE and Coomassie staining, S.venezuelae CBS 762.70 as a strain producing high amounts of a protein appealing as a double band of approximately 13kDa in size. Besides this 13kDa protein, only a few other proteins, larger in size and at much lower quantities were observed (Fig. 2a). Quantitative estimations indicated that this 13kDa protein was secreted using shake flask cultures in amounts of up to 300mg'L. B

y separation on a Sephadex G50 column, the proteins in the double band eluted in separate fractions (tract. 47 and 55) (Fig. 2b). N-terminal amino acid sequence analysis of the proteins in fraction 47 and 55 indicated that the proteins in both fractions were identical at their N-terminal end except that in fraction 47 the protein was truncated (+7 to + 10). with the +9 form being predominant, comprising 45 pet. of the total amount. More extensiv e determination of the 13kDa protein was in a first instance by N-terminal sequencing of sev eral fragments obtained following endoproteinase Lys-C digestion and separation ov er HPLC ( see in Fig. 1 ). A comparison in the SWISS-PROT protein database of the amino acid sequences obtained with those of other proteins revealed that the 13kDa protein belongs to the family of subtilisin inhibitors. The identification of the 13kDa protein secreted by S. venezuelae as a subtilisin inhibitor (VSI) was also proven by protease inhibitor tests, which showed that YSI had inhibitory activity towards subtilisin BPN and trypsin, but not for α-chymotrypsin (data not shown). Furthennore, with respect to the heterogeneous N-terminal end of VSI. similai heterogeneous N-tennini were observed for STI2. when produced in S.lividans (Strickler et al. , 1992). They were supposed to be the result of a protease activity. It could howev er, also be the result of an incorrect processing of the signal peptidase, since pre-SSI of S.alhogriseolus S-3253 was shown to have different signal peptidase processing sites when cloned in S.lividans 66 (Obata et ai. 1989) or E.coli (Taguchi et al. , 1990). On the other hand it might be that the S. venezuelae subtilisin inhibitor is secreted as a pro-protein. 2. Isolation and sequence analysis of the vsi-gene

We used the information of the N-teiminal amino acid sequence analysis of the abundantly secreted 13kDa protein and of one of its internal fragments to develop 2 oligonucleotides. in order to isolate the vsi-gene from chromosomal DNA of S.venezuelae. These oligodeoxvnucleotides were designed taking into account the codon bias of Sireptomyces (Wright & Bibb, 1992). They were radioactively labeled and probed to the S. venezuelae chromosomal DNA digested with Apal. BssHII. EspI. KpnI. NaeI. NarI, NcoI. NolI, NruI. PstI, PvuII, SalI, SmaI or StuI, and separated on an agarose gel. A PvuII restriction fragment of 5kb hybridized with both probes and was. therefore, supposed to contain the complete vsi gene. This fragment was subcloned into the EcoRV site of pACYC184 and multiplied in Ecoli XLl-Blue MRF. In case the high copy number pBluescriptII KS (+) was used to clone the 5kb PvuII fragment and with Ecoli HB101 as host, no transformants could however, be detected. A restriction map of the PvuII insert was constructed for Apal, NcoI. NotI, NruI. PvuI and SmaI ( Fig. 3a) and fragments were subcloned in pBluescript II KS (+), pUC19 or pACYC 184 for sequencing. Fig. 1 shows the nucleotide sequence of the 1.662bp fragment, obtained following sequence analysis on both strands using a sequencing strategy outlined in Fig. 3b. Potential protein-coding sequences were predicted by the computer-aided "FRAME" analysis dev eloped by Bibb et al ( 1984). based on the extremely biased codon usage for a G or C in the third and most degenerate position of the codon (Wright & Bibb, 1992). The analysis rev ealed 1 complete ORF (from 687 to 1 124) and 2 incomplete ORFs (from nt 1 to 324 and from 1282 to 1662) (Fig. 3c). Translation of the nucleotide sequence in ORF2 revealed a complete subtilisin inhibitor gene. It is assumed that the ATG at nt687-689. which was situated 8nt downstream a consensus

RBS-sequence ('5-AAGGAG-). is the translational start codon. The most probable signal sequence is formed from 28 amino acids containing 3 positiv e charges (Arg at position +2 and +3 and Lys at +6) and with Ala-Gb-Ala at the end of the signal sequence (Fig. 4). This is in agreement with the general role that signal peptidases recognize a (-3. - 1 ) pattern with small uncharged residues, usually alanine. in position -3 and - 1. relative to the signal peptidase cleavage site (von Heijne & Abrahamsen. 1989) A comparison of the VSI signal peptide with those of the other Streptomyces protease inhibitors shows that the former is quite different from the subtilisin inhibitor SSI of S.albogriseolus and the trypsin inhibitors SLP1 (STH ) and ST12 of S.lividans 1326 or 66 and S.longisporus. respectiv ely (Fig. 4). The VSI signal peptide contains 3 positive charges compared to +2 for the others. In addition, the hydrophobic region of VSI signal peptide has a very high number of Ala residues.

The precursor thus consists of 146 amino acids and the mature Ibim is composed of 1 18 amino acids. The size of VSI is 12.03kDa. closely to the size estimated by SDS-PAGE and its isoelectric point is 7.02. Downstream from the stop codon is a 15bp inverted repeat followed by a A/T-rich sequence which may serve as a transcriptional terminator ( Fig. 1 ).

3 Primer extension studies and promoter identification

To localize the promoter sequence preceding the vsi ORF. the start point of the vsi transcript was detennined by primer extension experiments using mRNA extracted from 23h-old cultures of S.venezuelae and the ³²P-labeled antisense oligonucleotide 93G3 14. The size of the extended product was determined by performing DNA sequence analysis using the same primer. The results indicated the presence of 2 transcripts starting at nucleotide -37 and -36 upstream the initiation codon. Upstream the mRNA transcription start site, the DNA sequence showed at the -10 but not at the -35 region a high similarity with the sequence of the ssiP2 promoter of the S. albogriseolus s.si gene (Taguchi et al.. 1989) ( Fig. 5). Furthermore, the putative v.s/ promoter did not show any obvious similarity with the SEP-like promoters ( Strohl. 1992) which is consistent with the observation that the vsi promoter failed to function in Ecoli. Direct repeats in the neighbourhood of the promoter sequence, a common phenomenon was seen in regulated promoters (Delic et al. 1992). also observed for vsi (Fig. 1 ). However, so far there are no indications that the vsi promoter is subjected to regulation. The primer extension studies did not reveal any other transcript, indicating that the vsi ORF was preceded by a single promoter sequence in contrast to e.g S.albogriseolus s.si.

Example 3: Comparison of the VSI-sequence to other Streptomyces subtilisin inhibitor protein sequences

The secondary structure of Streptomyces subtilisin inhibitors ( Fig. 6) shows 5β-strands and 2α-helices. with the reactive site located on a flexible loop between the αl -helix and the β4-strand as elucidated from crystallographic studies (Takeuchi et al., 1992).

The most conserved region resides in the core of the β-sheet. the β1. β2 and IM-strand and to some extent in β3. which all are of structural importance. The β4- strand contains most of the residues in contact with the proteolytic enzyme. Replacement of amino acids in these regions may result in less stable proteinase inhibitors, which might be subjected to degradation by various proteases. The flexible loop contains the reactive site. It shows surprisingly high variability between the different inhibitors. As a consequence, the amino acid sequence seems not to be important, as long as flexibility is prov ided. However, the amino acid residue at P1 determines the specificity of the inhibitory activity A mutation of this amino acid residue results in an alteration of the specificity ( Kojima et al. , 1990).

When comparing the VSI amino acid sequence with those of other inhibitors (Fig. 6). the alignment indicates that VSI has an amino acid identity with other proteinase inhibitors from 48.7% (for STT2) to 54.0% (for SIL2). Based on the similarity, the reactive site of VSI is supposed to be Lys-Glu. This is consistent with its trypsin inhibitory activity since other Streptomyces subtilisin inhibitors with trypsin inhibitory activ ity contain at the P1 site Lys or Arg (Kojima et al., 1990). VSI shows a relatively high identity of amino acids in the conserved domains, including the presence of 4 Cys. which are considered to form two intrachain disulfide bridges. On the other hand VSI shows also some striking differences. Several amino acid residues that are completely conserved in the already published protease inhibitors are changed to amino acids of low or no similarity, e.g.. Val₂₉, Ser ₄₇ or Arg₆₄ instead of Pro. Ala or Asp, respectively. These amino acids aie situated outside the β-strands and αl-h-lix. Therefore, of more fundamental importance might be the difference of amino acids in the β5-strand for which the conserved sequence Ser-Tyr-Glu-Arg is replaced by the non-similar amino acids Asn₉₅-Trp₉₆-Thr₉₇-Ser₉₈. The substitution of these highly conserved amino acid residues might suggest that VSI belongs to a class of Streptomyces subtilisin inhibitors, different from the 3 classes mentioned by Taguchi et al. ( 1994) as also suggested by phylogenetic tree analysis (Fig. 9). For this phylogenetic analvsis. a 1 10 amino acid alignment was made, starting with the conserved Tyr at the beginning of the β1 strand and ending with the carboxyterminal Phe.

Example 4: Secretion of mTNF in S.lividans mediated by the regulatory sequences and signal sequence of vsi

S.lividans TK24[pCBS2mTNF] transformants were cultured in NM medium (Van Mellaert et al., 1994) for 32 and 48 h. Culture filfrates were assayed for mTNF production by SDS-PAGE, immunoblot analysis and bioassay. On Coomassie stained SDS-PAGE gels

3 abundantly secreted proteins were detected in 48 h cultures two of which were about 17kDa and a third of 13kDa (Fig. 8). N-terminal sequence analysis of the VSI proteins in band 1 showed that the VSI-mTNF fusion protein was precisely processed by the S.lividans signal peptidase. However. mTNF with heterogeneous N-termini (+3, +4, +5 forms) were detected in band 2 (Fig. 7). This heterogeneity was the result of aminopeptidase activity of S.lividans, as already mentioned previously (Van Mellaert et al., 1994). N-terminal analysis of the proteins in band 3 indicated that this band consisted of the S.lividans subtilisin inhibitor. Immunoblot analysis of S.lividans TK24[pCBS2mTNF] showed that the protein bands 1 and 2 were immunoreactive, but also band 3 showed some cross-reactivity. Biological assays gave values of 2.5 - 8 × 10¹⁰ U mTNF/L culture corresponding to 1 10 - 300 mg mTNF/1. Upon culturing TK24 [pCBSmTNF] transformants in NM medium, biological mTNF activities up to 2.5×10¹⁰ U/l were obtained in the culture filtrate, indicating that the immediate fusion between the VSI signal peptide and the mTNF protein also resulted in efficient secretion of mTNF. These values are about 10 to 100 times higher than when using the α-amylase regulatory sequences and signal peptide (Table 5) and equalize the highest values obtained for eukaryotic proteins secreted by S.lividans.

Example 5: Secretion of hIL 10 in S. lividans mediated by the regulatoly sequences and signal sequence of vsi.

S. lividans TK24 [pCBS2hIL10] transformants were cultured in NNI medium for 30 and 40h. Culture filtrates were checked for hIL 10 production by SDS-PAGE. immunoblot analysis and bioassay. After SDS-PAGE of proteins of the culture filtrate, the concentration of hIL10 obtained was below the detection limit of Coomassie staining. On the other hand the immunoblot with anti-IL10 antibodies revealed 2 immunoreactive bands of about 18 and 24 kDa (see Fig. 10). Biological hIL10 activities measured in the culture filtrates were up to 1600 U/l after 30 h of growth.

Example 6: Secretion of hIL 2 by S.lividans mediated by the regulatory sequences and signal sequence of vsi.

S.lividans TK24 [pCBS2hIL2] and [pCBShIL2] transfonnants were grown in NM-medium during 28h to 48h on a rotary shaker (280rpm. 27°C). Proteins piesent in the culture filtrates were separated by SDS-PAGE ( 12.5%) and subsequently, blotted on a nitrocellulose membrane (BioRad). Immunoblot analysis of the blotted proteins revealed the presence of two immunoreactive bands. The proteins of about 15kDa corresponded to the processed hIL2 (with or without 2 additional amino acids of mature vsi). the protein with the size of 18kDa was likely to be unprocessed vsi-hIL2 fusion preprotein. By comparison of the intensity of bands on immunoblots containing a known amount of mature recombinant hIL2 and the immunoreactive proteins, the amount of hIL2 secreted by S.lividans TK24 was estimated to be 20 to 40μg hIL2/1 culture. Example 7: Secretion of hGM-CSF by S.lividans mediated by the regulatory sequences and signal sequence of vsi.

Transformants of S.lividans TK24 containing pCBS2hGM-CSF were grown in NM-medium for 24h to 48h on a shaker (280 rpm. 27°C) Proteins present in the filtrates of these cultures were analysed by SDS-PAGE and subsequent immunoblot assay . Binding of anti-hGM-CSF antibodies occurred on several proteins with a molecular weight varying from 14 to 18 kDa The protein with the lowest M, corresponded to mature hGM-CSF, the origin of the larger immunoreactive proteins has still to be investigated but were supposed to be aggregates of hGM-CSF. By comparison on immunoblot with a known amount of recombinant hGM-CSF standard the amount of hGM-CSF protein secreted by S.lividans TK24 was estimated to be up to 125μg hGM-CSF/1 culture. Example 8: Secretion of BLA by S.lividans mediated by the regulatoiy sequences and signal sequence of vsi.

After transformation of S.lividans TK24 protocols with pCBS2sBLA and pCBSsBLA, respective transformants were grown in phage medium (Anne et al, 1984) or in 1/2 YPD (0.5% bacto yeast extract, 1% bacto peptone and 1% glucose) on a shaker (280 rpm. 27°C). At different time intervals, samples of culture filtrates were taken and analysed for the presence of BLA by immuno radiometric assays. BLA levels of 2 to 2.5mg/l culture could be obtained.

Example 9: Influence of positive charges in the signal peptide on protein secretion

To investigate the importance of the three positive charges in the N-tenninal region of the signal peptide for effecting secretion and to analyse whether the level of secreted mTNF could still be improved several mutants of pCBS2mTNF were made by in vitro mutagenesis as indicated in Figure 1 1. As arginine residues aie more commonly used than lysines in streptomycete signal peptides, extra positive charges were added by introducing CGT or CGC. The three positive charges of the WT signal peptide were gradually replaced by glutamine residues. In this way, we obtained signal peptide mutants with a positiv e charge ranging from 0 to 5. After transformation of S. lividans with the different constructs. mTNF was measured both by a biological activity test and by Western blotting. For each of the mutants 8 different transformants were sampled for the presence of intracellular and extracellular mTNF after 30 and 48 hours of growth. The average results after 48 hours of growth aie shown in Figure

12A. It is clear that charge variation does not inhibit export of mature biologically active mTNF. Nevertheless, the differences in amounts of secreted mTNF are significant. Taking the amount of mTNF secreted under the influence of the wild type signal peptide as a reference (100 %). we found a change in the amounts of secreted mTNF to values of 16 % for zero charge, 80 % for +1, 320 % for +2. 80 % for -t4 and 50 % for -5. This variation was also visualised on SDS-PAGE followed by Coomassie staining (Figure 12B). We can conclude from this that the presence of positive charges in the N-region is not absolutely required to obtain export.

Immunoblot analysis of the recombinant mTNF produced by S. lividans fransfomiants each carrying a different mutant VSI signal peptide showed the mature fonn and a few degradation products in the culture fluid. Cell lysates contained both the processed and unprocessed form of mTNF together with a protein of intermediate size which could be a degradation product of the precursor protein (Figure 13). Whether the mature mTNF present in the cell extracts is originating from the cytoplasm or is attached to the cell wall or cell membrane was not further investigated.

Calculating the real secretion efficiencies (Figure 12C) reveals that, although the wild type is most efficient in directing export of mTNF, the +2 mutant is able to secrete more in absolute terms. Respectively 76 % and 60 % of the total synthesized mTNF is secreted by the wild type and the +2 mutant.

Example 10: Positive secretion effect correlates with increased mRNA level

To analyse a possible effect of the introduced mutations at the level of tianscription, total RNA was prepared from S. lividans cultures harbouring the different mutant constructs including the wild type. Results obtained after slot-blotting and hybridisation with a mTNF cDNA specific probe are shown in Figure 14 together with an internal control to ensure that the cultures were approximately at the same position on the growth curve. Therefore a chromosomal stil probe was used As can be concluded from this figure in combination with

Figure 3 A. for each mutant, the level of mTNF mRNA is different and seems to be tightly correlated with the total amount (intracellular and extracellular) of synthesized mTNF.

REFERENCES

Anne, J., W. Wohlleben. H.J. Burkardt, R Springer and A. Pühler ( 1984). Morphological and molecular characterization of several actinophages isolated from soil which lyse Streptomyces cattley a or S.venezuelae. J. Gen. Microbiol. 130:2639-2649.

Anne. J. and Van Mellaert, L. (1993). Streptomyces lividans as host for heterologous protein production. FEMS Microbiol. Lett. 114, 121-128.

Asseline U, Delarue M Lancelot G, Toulme F, Thuong N ( 1984) Nucleic acid-binding molecules with high affinity and base sequence specificity : intercalating agents covalently linked to oligodeoxynucleotides. Proc. Natl. Acad. Sci. USA 81(1 1):3297-301. Bibb. MJ. Findlay, P R. and Johnson MW. (1984). The relationship between base composition and codon usage in bacterial genes and its use for the simple and reliable identification of protein-coding sequences. Gene 30, 157-166.

Brawner, M., Poste, G., Rosenberg, M and Westpheling, J. (1991 ). Streptomyces: a host for heterologous gene expression. Curr. Op. Biotechnol. 2. 674-681.

Chang. A.C.Y. and Cohen, S.V. (1978). Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the p15A cryptic miniplasmid. J. Bacteriol. 134, 1 141-1156.

Delic, I.. Robbins, P. and Westpheling, J. (1992). Direct repeat sequences aie implicated in the regulation of two Streptomyces chitinase promoters that aie subject to carbon catabolite control. Proc. Natl. Acad. Sci. USA, 89. 1885-1889.

Fornwald J.A., Donovan, M.J., Guber, R, Keller, J.. Taylor. D P.. Arcuri. E.J. and Brawner, M.E. (1993). Soluble forms of the human T cell receptor CD4 aie efficiently expressed by Streptomyces lividans. Biotechnology 11, 1031-1036. Hopwood DA, Bibb, M.J., Chater, K.F., Kieser. T.J., Brutαn. C.J.. Kieser. H.M. Lydiate. D.J., Smith, C.P., Ward J M and Schrempf H. ( 1985). Genetic Manipulation of Streptomyces: a laboratory manual. John Innes Foundation. Norwich. U.K.

Jacobs et al. (1988). Nucleic Acids Research, 16: 4637-4650.

Hopwood D.A., Bibb, M.J., Chater , K.F. and Kieser, T. ( 1987). Plasmid and phage vectors for gene cloning and analysis in Streptomyces. Methods Enzymol. 153: 1 16-166.

Kojima, S., Obata, S.. Kumagai, I. and Miura. K.-i. ( 1990). Alteration of the specificity of the Streptomyces subtilisin inhibitor by gene engineering. Biotechnology 8. 449-452.

Kojima, S., Terabe M., Taguchi, S., Momose. H., and Miura. K.-i Primary structure and inhibitory properties of a proteinase inhibitor produced by Stivpiomyces cacaoi. Biochem.

Biophys. Acta 1207: 120-125.

Kumazaki, T., Kajiwara, K.. Kojima, S.. Nliura. K.-i. and lshii. S -i. ( 1993). Interaction of Streptomyces subtilisin inhibitor (SSI) with Str epiomyces grisens metal lo-endopeptidase II (SGMPII). J. Biochem. 114. 570-575.

Kwok et al. Nucleic Acids Research 18:999-1005. 1990.

Laemmli, U.K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685.

Laskowski, Jr., M. and Kato, I. (1980). Protein inhibitors of proteinases Ann. Rev . Biochem.

49. 593-626. Matsukura M Shinozuka K, Zon G, Tvlitsuya H. Reitz M. Cohen J. Broder S ( 1987) Phosphorothioate analogs of oligodeoxynucleotides : inhibitors of replication and cytopathic effects of human immunodeficiency virus. Proc. Natl. Acad. Sci. USA 84(21 ): 7706- 10. Miller P, Yano J, Yano E, Carroll C, Jayaram K, Ts'o P ( 1979) Nonionic nucleic acid analogues. Synthesis and characterization of dideoxyribonucleoside methylphosphonates. Biochemistry 18(23):5134-43.

Miller, J.H. (1972). Experiments in molecular genetics. Cold Spring Harbor Laboratory, New York. Mitsui, Y., Satow, Y., Watanabe, Y, Hirono, S. and litaka Y. ( 1979). Crystal structure of a bacterial protein protease inhibitor and its complex with subtilisin BPN'. Nature 277, 447-452.

Nielsen P, Egholm M. Berg R, Buchardt O ( 1991) Sequence-selective recognition of DNA by strand displacement with a thymine-substituted polyamide. Science 254(5037): 1497-500.

Nielsen P. Egholm M. Berg R, Buchardt O ( 1993) Sequence specific inhibition of DNA restriction enzyme cleavage by PNA. Nucleic- Acids- Res. 21(2): 197-200.

Norrander, J., Kempe. J. and Messing, J. (1983). Construction of improved M13 vectors using oligonucleotide-directed mutagenesis. Gene 26. 101-106.

Obata, S.. Taguchi. S.. Kumagai. I. and Miura. K -i. ( 1989). Molecular cloning and nucleotide sequence determination of gene encoding Streptomyces subtilisin inhibitor (SSI). J. Biochem. 105, 367-371.

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

Strickler, J.E., Berka. T.R, Gorniak, J., Fornwald J.. Keys. R. Rowland J.J.. Rosenberg. M. and Taylor, D.P. (1992). Two novel Streptomyces protein protease inhibitors. J. Biol. Chem 26-.

3236-3241. Strohl, W.R (1992). Compilation and analysis of DNA sequences associated with apparent Streptomycete promoters. Nucl. Acids Res. 20. 961-974

Sugino, H., Kahinuma. A. and Iwanaga, S. ( 1978). Plasminostreptin. a protein proteinase inhibitor produced by Streptomyces antifibrinolyticus. J. Biol. Chem. 253. 1546-1555.

Suzuki, K , Uyeda, M and Shibata, M. ( 1981). Partial amino acid sequence of an alkaline protease inhibitor, API -2 (b and c). Agric. Biol. Chem. 45. 629-634 Taguchi, S. (1992). Streptomyces subtilisin inhibitor: general characteiization and its application.

Actinomycetol. 6, 9-20.

Taguchi, S., Nishiyama, K-i., Kumagai, I. and Miura, K -i. ( 1989). Analysis of transcriptional control regions in the Streptomyces subtilisin-inhibitor-encoding gene Gene H4. 279-286.

Taguchi, S., Kumagai, I. and Miura, K -i. (1990). Comparison of secretory expression in Escherichia coli and Streptomyces of Streptomyces subtilisin inhibitor (SSI) gene. Biochim. Biophys. Acta 1049. 278-285. Taguchi, S., Maeno, M, and Momose, H. ( 1992). Extiacellulai production system of heterologous peptide driven by a secretory protease inhibitor of Streptomyces. Appl. Microbiol. Biotechnol. 36, 749-753.

Taguchi, S., Kikuchi. H., Suzuki, M, Kojima, S., Terabe. M. Nliura K-i., Nakase. T. and Momose, H. (1993). Streptomyces subtilisin inhibitor-like proteins are distributed widely in streptomycetes. Appl. Environm. Microbiol 59. 4338-4341.

Taguchi, S., Kojima S.. Terabe. M., Miura K -i. and Momose. H ( 1 994) Comparative studies on the primary structures and inhibitory properties of subtilisin-trypsin inhibitors from Streptomyces. Eur. J . Biochem. 220, 91 1-918. Takeuchi, Y., Noguchi, S., Satow, Y., Kojima S., Kumagai, I.. Miura K-i., Nakamuia K T. and Mitsui, Y. (1991). Molecular recognition and the activ e site of subtilisin BPN: crystallographic studies using genetically engineered proteinaceous inhibitor SSI (Streptomyces subtilisin inhibitor). Protein Engineering 4. 501 -508.

Takeuchi, Y., Nonaka T., Nakamura K.T., Kojima S.. Miura K-i and Mitsui, Y. ( 1992). Crystal structure of an engineered subtilisin inhibitor complexed with bovine nypsin. Proc. Natl. Acad Sci. USA 89. 4407-4411. Terabe M., Kojima, S., Taguchi, S., Momose, H., and Miura K-i. ( 1994a) Primary structure and inhibitory properties of a subtilisin-chymotiypsin inhibitor from Streptomyces virginae. Eur. J. Biochem 266: 627-632.

Terabe M, Kojima S., Taguchi, S., Momose, H., and Miura K-i. ( 1994b) Three novel subtilisin-trypsin inhibitors from Streptomyces: primary structures and inhibitory properties.

J. Biochem 116: 1 156-1 163

Terabe M.. Kojima S.. Taguchi, S., Momose. R, and Miura K.-i. ( 1995) A subtilisin inhibitor produced by Streptomyces bikini ens is possesses a glutamine residue at reactive site Pl . J. Biochem. 117: 609-613.

Thompson-Snipes, L.. Dhar, V., Bond M., Mosmann. T. Moore. K. and Rennick. D. ( 1991). lnterleukin 10 : a novel stimulatory factor for mast cells and their progenitors. Vögtli. M. and Hütter, R (1987). Characterisation of the hydroxysfreptomycin phosphotransferase gene (sph) of Streptomyces glaiicescens: nucleotide sequence and promoter analysis. Mol. Gen. Genet. 208, 195-203.

Van Mellaert, L., Dillen. C. Proost, P., Sabloa E, DeLeys. R. Van Broekhoven. A.. Heremans. H. Van Damme, J.. Eyssen. H. and Anne, J. (1994). Efficient secretion of biologically active mouse tumor necrosis factor α by Streptomyces lividans. Gene 150. 153-158. Yiaene. A., G. Volckaert, M Joniau, A. De Baetselier and F. Van Cauwelaert ( 1991 ). Efficient expression of bovine alpha-lactalbumin in Sacchctromyces cerevisiae. Eur. J. Biochem, 202:

471-477. von Heijne, G. and Abrahamsen, L. (1989). Species-specific variation in signal peptide design.

FEBS Lett. 244, 439-446.

Ward J.M, Janssen, G.R, Kieser, T., Bibb. M.J.. Buttner. M .l. and Bibb. M.J. ( 1986). Construction and characterization of a series of multicopy promoter-probe plasmid vectors for Streptomyces using the arranoglycosiclase phosphofransferase gene from Tn5 as indicator.

Mol. Gen. Genet. 203, 486-478.

Wbmble, D.D., Taylor. D.P. and Roward, R.M. ( 1977). Method for obtaining more-accurate covalently closed circular plasmid- to -chromosome ratios from bacterial lysates by dye- boyant density centrifugation. J. Bacteriol. 130. 148-153.

Wright, F. and Bibb, M.J. (1992). Codon usage in the G +C-rich Strepionn ccs genome. Gene 113. 55-65.

Claims

1. A polynucleic acid in substantially isolated form comprising a sequence of at least 10 contiguous nucleotides selected from:

(a) the polynucleic acid sequence as shown in Fig. 1 (SEQ ID NO 1 ) or

(b) the polynucleic acid sequences which are degenerate as a result of the genetic code to the polynucleic acid sequence as defined in (a) and which code for a polypeptide showing a subtilisin inhibitory activity, or

(c) the polynucleic acid sequences which hybridize to the polynucleic acid sequences as defined in (a) or (b).

2. A recombinant DNA molecule comprising a polynucleic acid according to claim 1.

3. A recombinant DNA molecule according to claim 2. comprising the gene expression unit of any of the polynucleic acid sequences as defined in claim 1. or part of it. said gene expression unit comprising the promoter sequence, the ribosome binding site, the sequence coding for the secretion signal, the sequence coding for the mature protein, and the termination sequence.

4. A recombinant DNA molecule according to any of claims 2 to 3. comprising the regulatory sequence of any of the polynucleic acid sequences as defined in claim 1. or part of it, operably linked to a heterologous coding sequence, said regulatory sequence comprising the promoter sequence, the ribosome binding site and the nucleotide sequence encoding the secretion signal.

5. A recombinant DNA molecule according to any of claims 2 to 4. comprising the regulatory sequence of the polynucleic acid sequence represented by SEQ ID NO 1. or part of it, operably linked to a heterologous coding sequence, said regulatory sequence comprising - the promoter sequence located in the region extending from pos (-686) to pos (-38) in Fig.1 (SEQ ID NO 1), more specifically, located in the region extending from nucleotide pos (-81 ) to pos (-38) in Fig. 1, and/or,

- the ribosome binding site located in the region extending from pos (-37) to pos (- 1 ) in Fig. 1 (SEQ ID NO 1), more specifically, located in the region extending from pos (-15) to (-1) in Fig. 1, and/or ,

- the nucleotide sequence encoding the secretion signal extending from position (+1 ) to pos (+84) in Fig. 1 (SEQ ID NO 1) .

6. A recombinant DNA molecule according to any of claims 2 to 5. comprising the regulatory sequence of the polynucleic acid sequence represented by SEQ ID NO 1. or part of it, operably linked to the coding sequence: of SEQ ID NO 1. or part of it. wherein said coding sequence is fused in frame with a second heterologous coding sequence, said coding sequence of SEQ ID NO 1 extending from pos (+85) to pos (+439) in Fig. 1.

7. A recombinant DNA molecule according to any of claims 2 to 6. wherein the sequence encoding the signal peptide has been specifically mutated resulting in one at least of the following amino acid substitutions in the signal peptide:

R substituted by Q at amino acid position 2 of Fig 1. and/or

R substituted by Q at amino acid position 3 of Fig. 1. and/or

T substituted by R at amino acid position 4 of Fig. 1. and/or

L substituted by R at amino acid position 5 of Fig. 1. and/or

K substitued by Q at amino acid position 6 of Fig. 1.

8. A recombinant vector capable of transforming a host cell, and capable of replication and/or directing expression in said host cell, wherein said recombinant v ector comprises a recombinant DNA molecule according to any one of claims 2 to 7

9. A host cell transformed by a recombinant vector according to claim 8. said host cell being chosen from among eucaryotic organisms, such as plant cells, animal cells, or yeast cells, or chosen from among procaryotic organisms, such as bacteria.

10. A host cell according to claim 9, said host cell belonging to the genus Streptomyces.

1 1. Process for producing a heterologous protein from a host cell, comprising the steps of:

- growing a culture of a host cell according to any of claims 9 to 10. under conditions enabling expression and possibly secretion of the heterologous protein .

- recovering said heterologous protein from said culture.

12. Polypeptide in substantially pure form comprising a sequence of at least 9 contiguous amino acids selected from

(a) me polypeptide sequence encoded by any of the polynucleic acid sequences as defined in claim 1, or

(b) the polypeptide sequence of 146 amino acid residues as represented in Fig. 1 (SEQ ID NO 2), or

(c) polypeptide sequences resulting from the modification of one or several amino acids in the polypeptide sequence represented by SEQ ID NO 2. said modification comprising substitution and/or deletion and/or addition of one or several amino acids, prov ided that said modification does not abolish the subtilisin inhibitory activity of the polypeptide encoded by SEQ ID NO 2.

13. A fusion protein comprising a polypeptide according to claim 12 fused to a heterologous protein.

14. A polypeptide according to claim 12 for use as a medicament.

15. Use of a polypeptide according to claim 12 for treating medical distresses.

16. Use of a polypeptide according to claim 12 in the non-medical field more specifically in food and feed processing.

17. A probe hybridizing specifically with any of the polynucleic acid sequences as defined in claim 1.

18. A primer amplifying specifically me polynucleic acid sequences as defined in claim 1