Background of the invention Translocation of proteins across membranes is an intricate process involving a large number of accessory cellular functions ranging from necessary enzymatic activities and structurally important proteins to energetic requirements.

Over the past two decades the so called "secretion pathway"is slowly being unravelled, starting at the molecular level with the recognition that secreted proteins are translated as a precursor polypeptide carrying an N-terminal extension, the secretion signal peptide, which upon translocation across the membrane is cleaved off by an enzyme, the signal peptidase, to yield the mature protein (J. M. Gennity, M. Inouye. 1991.

Curr. Opin. Biotechnol. 2: 661-667).

In more recent years it has become evident that for this simple scheme to become true, in different classes of organisms a variety of additional factors is necessary. Thus in eukaryotic organisms during translation

of the protein to be secreted a complex of proteins and RNA, the signal recognition particle, mediates targeting to the membrane where a number of other proteins takes care of the actual membrane crossing. In prokaryotes no firm evidence has been gathered until now for a similar signal recognition particle. However, at least as many proteins appear to be involved in both gram-positive and gram-negative bacteria to effect secretion; some functions attributed to accessory proteins are membrane targeting, (timing of) protein folding, actual membrane translocation and signal peptide cleavage, and final protein folding.

Minor differences exist between the gram-positive and gram-negative bacteria: e. g. not all secretion associated proteins from gram-positive bacteria have a counterpart in gram-negative bacteria, and vice versa. Moreover, gram- negative bacteria have an additional secretion pathway for translocation of protein across the outer membrane (K. L. Bieker, T. Silhavy, 1990, Trends in Genetics 6: 329- 334; C. Wandersman, 1992, Trends in Genetics 8: 317-322; M. Simonen, I. Palva, 1993, Microbiol. Rev. 57: 109-137).

The mechanism of protein secretion has received very much attention over the past two decades because of its anticipated use for extracellular production of proteins of industrial and pharmaceutical interest. The ways in which secreted enzymes are provided are not always as efficient and economically feasible as desired.

For the purpose of high level production of the Penicillin G acylase enzyme there are several major hurdles to take. In wild-type Escherichia coli strains (E. coli strains), transcription of the acylase gene needs to be induced by the addition of aryl fatty acids (e. g.

phenylacetic acid) and is sensitive to catabolite repression (S. Scherrer ibid.). Translation of Penicillin G acylase appears to be inhibited at temperatures above 30°C, resulting in the unfavourable situation that the temperature by which both host cell growth is optimal and efficient synthesis of the desired periplasmic protein occurs, do not coincide (Keilman, C., Wanner, G., Boeck, A., 1993, Biol. Chem. Hoppe Seyler 374,983-992). Folding of penicillin G acylase occurs properly at 22°C, whereas at 26°C only 50% of the enzyme is still active, and at 30°C most of the enzyme occurs in an inactive form.

Although modern recombinant DNA technology in principle offers numerous methods to bypass or alleviate this type of mechanism, none of the approaches followed in various research laboratories has provided efficient solutions.

For example, when trying to overproduce the protein under a strong promoter, this resulted in an increased level of protein, but both secretion across the cytoplasmic membrane and correct processing into the subunits was considerably reduced (S. Scherrer, N. Robas, H. Zouheiry, G. Branlant, C. Branlant. 1994. Appl. Microbiol.

Biotechnol. 42: 85-91). A similar observation is made when the 5 nucleotide spacer causing temperature dependent translation is removed (C. Keilman et al. ibid.).

An alternative approach, aimed at the production of Penicillin G acylase intracellularly by removal of the secretion signal peptide led to the expected accumulation of protein inside the cell, but this protein was in an inactive, unprocessed form (G.

Schumacher, D. Sizmann, H. Haug, P. Buckel, A. Boeck.

1986. Nucl. Acids Res. 14: 5713-5727; K. S. Choi, J. A. Kim, H. S. Kang. 1992. J. Bacteriol. 174: 6270-6276). Production

of both the subunits separately to circumvent the latter problem in its turn resulted in inactive subunit aggregates in the cytoplasm.

Both induction and catabolite repression phenomena of Penicillin G acylase are linked with the DNA region upstream of the structural gene, whereas inhibition of mRNA translation is mediated by a 5 nucleotide spacer region between the Shine-Delgarno sequence and the ATG startcodon (C. Keilman et al. ibid.).

The specific requirements of the production organism with respect to the recognition of the secretion signal peptide have led to the generally accepted conclusion that, independent of the mature protein origin, for efficient secretion of this protein to occur, it should be equipped with a secretion signal originating from the production host (i. e. homologous secretion signal). This conclusion is of particular interest for the secretion of heterologous proteins, i. e. proteins not normally produced by the production organisms, to ensure that the host secretion machinery will recognise and correctly process the primary translation product. (G. von Heijne, L. Abrahmsen, 1989, FEBS Lett. 244: 439-446).

Contrary to this general knowledge, in the underlying invention, we have now surprisingly found that the use of a heterologous secretion signal, originating from a species different from the production host species, allows more efficient secretion of proteins such as ß- lactam acylases, than the use of a homologous secretion signal, even at high expression level conditions and at elevated temperatures.

Description of the Figures Figure 1: Physical and functional maps of plasmids pBRK, pMCtrpEC and pKECtrp. Km (R): kanamycin resistance gene; Tc (R): tetracyclin resistance gene; Cm (R): chloramphenicol resistance gene; ori327: origin of replication from plasmid pBR327; trpP: modified trp promoter; ECpga: E. coli PenG acylase.

Figure 2: Physical and functional maps of plasmids pKAFssECtrp and pKAFssECaro. Km (R): kanamycin resistance gene; Tc (R): tetracyclin resistance gene; Cm (R): chloramphenicol resistance gene; ori327: origin of replication from plasmid pBR327; trpP: modified trp promoter; ECpga: E. coli PenG acylase. AFss: A. faecalis secretion signal sequence; aroP: aro promoter.

Summary of the invention The present invention provides an expression cassette comprising a promoter sequence, a secretion signal and a DNA sequence encoding a protein, wherein said secretion signal is heterologous. This secretion signal originates from Alcaligenes faecalis. The expression cassette comprises a DNA sequence encoding the mature part of a naturally secreted protein, which DNA sequence is heterologous with respect to Alcaligenes faecalis, i. e. the DNA sequence does not originate form Alcaligenes faecalis. The DNA sequence encoding a protein preferably is a P-lactam acylase, e. g. Penicillin G acylase of E. coli origin, or originating from Escherichia coli,

Kluyvera citrophila, Providencia rettgeri, Arthrobacter viscosus or Bacillus megaterium.

Furthermore, the expression cassette comprises a promoter sequence preferably selected from the group consisting of aro, lac, trp and tac promoter. The expression cassette may be integrated in a host chromosome, or may be present on a plasmid which is capable of propagation in a suitable host. Various hosts or host cell strains may be used, for example E. coli or Pseudomonas species are suitable hosts. Preferably, an E. coli host cell strain is used which is capable of production and secretion of the protein.

In another embodiment of the invention, a process for the preparation of the protein encoded by the expression cassette is provided characterised by growing the host cell strain in a suitable nutrient medium allowing initiation of expression of said protein in the host cell strain, whereby that host cell strain produces said protein, followed by harvesting said protein from that medium or host cell. In this process the host cell strain is preferably grown at a temperature of 20-30°C, more preferably 22-28°C.

In another embodiment of the invention, a process for the preparation of an amino ß-lactam compound is provided, by the application of the protein of present invention. This process is optionally followed by a process for the preparation of a semisynthetic a-lactam antibiotic, wherein the corresponding amino P-lactam compound, is subsequently reacted with a suitable side chain derivative by applying a suitable a-amino acid hydrolase. In this latter process said semisynthetic ß- lactam antibiotic is preferably selected from the group

consisting of Amoxicillin, Ampicillin, Cloxacillin, Dicloxacillin, Flucloxacillin, Cefotaxim, Cephalexin, Cefadroxil, Cephradine, Cefetamet, Cefroxadine, Cefatrizine, Cefoperazon, Cefprozil, Cefaclor, Loracarbef, Cefazolin, Cefotiam and Cefamandole.

Description of the invention Many proteins are secreted across the cytoplasmic membrane. For example, E. coli Penicillin G acylase is a heterodimeric protein located in the periplasmic space and thus has to be translocated across the cytoplasmic membrane. In addition to being secreted into the periplasm, the single peptide chain (preproprotein) resulting from translation of the messenger RNA not only loses the signal peptide yielding the proprotein, but subsequently requires proteolytic modification by removal of a central portion of 54 amino acids resulting in the mature form of two closely associated subunits, i. e. a small a-subunit and a largep- subunit (G. Schumacher et al. ibid.).

Hydrolases and specifically acylases are involved in the enzymatic preparation of ß-lactams. P- Lactam acylases, such as penicillin G acylases (E. C.

3.5.1.11, benzylpenicillin amidohydrolase or penicillin amidase), glutaryl acylases, adipyl acylases and dicarboxylate acylases can be obtained from various microorganisms and can be used in enzymatic steps for semi-synthetic preparation of P-lactam antibiotics such as penicillins, cephalosporins and their derivatives.

Penicillin acylases are efficiently able to cleave the benzyl side chain off penicillin G and penicillin V yielding 6-APA. Penicillin G acylase is a

member of a family of evolutionary related Penicillin G acylases from different Gram positive (e. g. Arthrobacter viscosus and Bacillus megaterium) and Gram negative bacteria (e. g. E. coli, Kluyvera citrophila and Providencia rettgeri). In addition to the latter Penicillin G acylase family, other enzymes have been identified displaying different properties like e. g. altered substrate specificities or different stability under application conditions. One example is Alcaligenes faecalis Penicillin G acylase (R. M. D. Verhaert, A. M. Riemens, J.-M. van der Laan, J. van Duin, W. J. Quax. 1997.

Appl. Environm. Microbiol. 63: 3412-3418).

Glutaryl, adipyl or benzyl acylases are capable of specifically hydrolyzing the glutaryl, adipyl or benzyl side chain of the respective Cephalosporin C derivative, yielding 7-ACA and 7-ADCA.

6-APA, 7-ACA and 7-ADCA are important precursors for semi-synthetic ß-lactam antibiotics such as Amoxicillin, Ampicillin, Cloxacillin, Dicloxacillin, Flucloxacillin, Cefotaxim, Cephalexin, Cefadroxil, Cephradine, Cefetamet, Cefroxadine, Cefatrizine, Cefoperazon, Cefprozil, Cefaclor, Loracarbef, Cefazolin, Cefotiam and Cefamandole.

In addition to the a-lactam acylases, there are a-amino acid ester hydrolases. It has been shown that these hydrolysing enzymes can be used for the reverse reaction too, i. e. are capable of attachment of a desired side chain to a ß-lactam structure: such as the condensation of activated side chain derivatives such as D- (-)-phenylglycine (PG), D- (-)-2,5-dihydrophenylglycine, D- (-)-4-hydroxyphenylglycine (HPG), 1H-tetrazoleacetic acid, (2-aminothiazol-4-yl) acetic acid and D- (-)-mandelic

acid and amides or methyl esters thereof, with amino ß- lactams such as 6-amino-penicillanic acid (6-APA), 7- aminocephalosporanic acid (7-ACA), and 7- aminodesacetoxycephalosporanic acid (7-ADCA).

This may even be the preferred application because in this way the diversity of semi-synthetic penicillins and cephalosporins can actually be enlarged.

Moreover, besides their intrinsically better substrate selectivity and specific activity, the use of enzymes may replace existing complex chemical treatments and thus contribute to more environmentally attractive routes.

As described above, acylases are involved in many enzymatic reactions in pharmaceutical processes.

However, efficient synthesis of the desired acylase proteins in the periplasmic compartment remains a major problem since the temperatures at which both host cell growth and protein synthesis is optimal do not always coincide. It is known that expression of Penicillin G acylase needs to be tightly controlled by the cell, because both the secretion pathway and the folding process are very sensitive to high level expression of the protein. The latter phenomenon is not uncommon in recent biotechnological research involving E. coli as a production organism: several (classes of) proteins of interest have been observed to be blocked in either efficient secretion, resulting in accumulation of intracellular inactive complexes, or after secretion are incorrectly folded in the periplasm. Examples can be found e. g. in the expression and secretion of single-chain antibody fragments (J. E. Somerville Jr, S. C. Goshorn, H. P Fell, R. P. Darveau. 1994. Appl. Microbiol. Biotechnol. 42: 595- 603), and even in E. coli homologous gene fusions (A.

Guigueno, P. Belin, P. L. Boquet. 1997. J. Bacteriol. 179: 3260-3269).

The basic element to realise protein expression is called the expression cassette. This cassette comprises: a promoter with accessory regulatory regions for induction or repression involved in directing the transcription of the downstream DNA into messenger RNA followed by a region, which in the messenger RNA is responsible for the recognition and correct positioning of the starting point of translation, the actual protein assembly mechanism. From this starting point onwards the nucleotide sequence of the messenger RNA determines the amino acid sequence of the encoded protein. Embedded in this amino acid sequence, a secretion signal is present targeting the protein to a cellular compartment, e. g. the cell membrane, periplasmic space (gram-negative bacteria only), peptidoglycan layer, the outer membrane (gram- negative bacteria only), or completely out of the cell. In the underlying invention, secretion is defined as at least one membrane crossing effectuated by the secretion signal.

Finally, the expression cassette may contain signals mediating efficient transcription termination, translation termination, as well as sequences increasing messenger RNA stability.

In the so-called expression cassettes, a strong promoter ensuring high level messenger RNA availability must be present. The promotor used in the expression cassette according to the invention may be selected from the well-known set of inducible promoters for highly expressed operons/genes like the lactose operon (lac, lacW5), the arabinose operon (ara), the tryptophan operon (trp), and the operon encoding enzymes common to the biosynthesis of all aromatic amino acids (aro), or

functional hybrids of these, e. g. the tac promoter, which is a fusion of the trp and the lac promoter (E. Amann, J. Brosius, M. Ptashne. 1983. Gene 25: 161-178).

Alternatively, constitutive promoters can be used providing for a constant supply of messenger RNA throughout the cell's life. Similar strategies can be used to improve translation of the messenger RNA pool by introducing, via recDNA methodology, 5'untranslated leader regions from efficiently translated messenger RNA's like those obtainable from the tuf gene encoding the highly expressed Elongation Factor Tu protein, or modified or synthetic variants of the tryptophan operon.

Transcription terminators, possibly also contributing to messenger RNA stability can be selected either from the native gene to be expressed or from sources like rRNA genes or viral operons, e. g. the ribosomal RNA terminator, or the fd terminator (J. Sambrook, E. F. Fritsch, T.

Maniatis. 1989. Molecular Cloning 2nd edition. CSH Press).

The extrachromosomal elements in which the expression cassettes can be inserted may be plasmid or virus derived, and preferably but not necessarily comprise a marker enabling selection of cells harbouring the element, as well as DNA sequences responsible for autonomous propagation and/or equal distribution of the element within the host cell and its daughter cells.

Alternatively, chromosomal integration can be envisaged.

In an embodiment of the present invention the expression cassette comprises a DNA fragment encoding the mature part of a protein, preferably a secreted protein (crossing at least one membrane), more preferably an acylase, most preferably one or both subunits of a ß- lactam acylase, in combination with a DNA fragment encoding a secretion signal peptide of a heterologous

protein i. e. originating from a different species, i. c.

A. faecalis, which is subsequently introduced into a production host and which yield high expression levels of said proteins.

In another embodiment of the present invention, the embodiment of the expression cassette last mentioned is additionally modified by replacement of the original promoter by preferably the trp promoter or the aro promoter. In order to fully exploit the basic improvement of secretion efficiency, additional modifications relating to the increased gene expression, messenger RNA translation and plasmid stability may be applied to the recDNA constructs used to create the actual production strain e. g. addition of the Transcription terminator of phage fd, or the introduction of the partitioning function par from plasmid pSC101 (G. Churchward, P. Linder, L. Caro, 1983. Nucleic Acid Res.

11: 5645-5659).

To even further increase production of the desired protein it is possible to insert the expression cassette on extrachromosomal elements, for example plasmids ColEl, ColD, R1162, RK2 or derivatives which may be present in either predetermined low copy numbers or, often dynamic, high copy numbers and which are capable of propagation or autonomous replication in e. g. E. coli strains HB101, B7, RV308, DH1, HMS174, W3110, BL21.

For the enzymatic production of semisynthetic P-lactam antibiotics and amides or esters thereof, the enzyme obtained in the present invention may be used as isolated free enzymes, or preferably immobilised on various types of water-insoluble carrier materials known in the art (WO 97/04086 and W099/01566). Of course recovery of 6-APA, 7-ACA and 7-ADCA and their conversion

into compounds such as Amoxicillin, Ampicillin, Cloxacillin, Dicloxacillin, Flucloxacillin, Cefotaxim, Cephalexin, Cefadroxil, Cephradine, Cefetamet, Cefroxadine, Cefatrizine, Cefoperazon, Cefprozil, Cefaclor, Loracarbef, Cefazolin, Cefotiam and Cefamandole and their use as pharmaceutically active compounds do also form an aspect of the underlying invention.

The following examples are only to be considered as illustration of the present invention.

Example 1 Construction of an expression cassette with the native E. coli Penicillin G acylase gene.

All molecular biological techniques employed were essentially performed according to Maniatis (J. Sambrook, E. F. Fritsch, T. Maniatis. 1989. Molecular Cloning 2nd edition. CSH Press).

To be able to express and analyse a 6-lactam modifying enzyme plasmid vectors should be free of ß- lactamase activity (which would destroy the substrate for the acylase). Initially for this purpose the plasmid pMC5 was used (P. Stanssens, H.-J. Fritz. 1989. Nucleic Acids Res. 17: 4441-4454).

For directing transcription and translation in the expression cassette, a modified trp promoter was designed which lacked the attenuator region, contained an adapted ribosome binding site, and within the translation start codon incorporated a NdeI restriction site. A DNA fragment according to this design was assembled synthetically using an Applied Biosystems DNA synthesizer.

The complete nucleotide sequence of this fragment is in SEQ ID nr 1.

The native gene encoding E. coli Penicillin G acylase (G. Schumacher et al. ibid.) was obtained in two parts: a. the N-terminal coding region was obtained by PCR using the following two oligonucleotides: 5'- CTGCCAGAGGATCATATGAAAAATAG-3' (SEQ ID nr 2) introducing an NdeI site at the Penicillin G acylase start codon, and 5'-GGGCATAAATATGCGGCATGCCG-3' (SEQ ID nr 3). The resulting fragment was treated with T4 DNA polymerase to ensure blunt ends at both sides. b. The fragment encoding the Penicillin G acylase C- terminal coding region was isolated directly from chromosomal DNA from E. coli strain ATCC 11105 by complete digestion with SphI and SmaI and insertion of this fragment in the general cloning vector pUC18 (Yanisch- Perron, J. Vieira, J. Messing. 1985. Gene 33: 102-119).

Clones harbouring the correct Sphl-Smal fragment were identified by colony hybridization techniques using synthetic oligonucleotides designed on the published sequence.

Construction of the expression cassette with the native gene for E. coli Penicillin G acylase was finalised by first a three fragment ligation involving pMC5 digested with EcoRI and SmaI, the modified synthetic trp promoter fragment digested with EcoRI and NdeI, and the N-terminal coding region of the E. coli Penicillin G acylase digested with NdeI. The resulting plasmid was subsequently digested with SphI and SmaI, ligated with the C-terminal SphI-SmaI Penicillin G acylase fragment, and transformed to E. coli strain DH1 (ATCC33849), to yield plasmid pMCtrpEC (Figure 1).

Example 2 Insertion of the expression cassette according to Example 1 on an alternative plasmid For expression studies and production at 10 litre scale a new plasmid, pBRK (Figure 1), was constructed from the standard cloning vector pBR327 (X, Soberon, L. Covarrubias, F. Bolivar. 1980. Gene 9: 287-305) by exchanging the ampicillin resistance gene (a ß- lactamase) by a kanamycin resistance gene from plasmid pBGS18+ (B. G. Spratt, P. J. Hedge, S. te Heesen, A. Edelman, J. K. Broome-Smith. 1986. Gene 41: 337-342) as follows.

Plasmid pBR327 was digested with Dral and AatII, treated with T4 DNA polymerase to obtain blunt ends on both sides of the fragment, and treated with T4 DNA ligase in the presence of a purified HaeII fragment from pBGS18+. The ligation mixture was transformed to E. coli strain DH1 and cells resistant to tetracyclin and kanamycin were analysed for the presence of recombinant plasmid. Plasmid was identified by restriction enzyme mapping and a new plasmid named pBRK (Figure 1) was found as being the expected combination of fragments.

Plasmid pBRK was digested with StyI, treated with T4 DNA polymerase to create a blunt end, then digested with EcoRI, ligated to the EcoRI-SmaI fragment containing the expression cassette for the Penicillin G acylase from pMCtrpEC, and transformed to E. coli strain HB101 (ATCC33694) to yield plasmid pKECtrp (Figure 1).

Example 3 Construction of an expression cassette with an Alcaligenes faecalis-E. coli fusion gene under the direction of the trp promotor In this example, the secretion signal present in the expression cassette according to example 2, is replaced by the A. faecalis secretion signal.

The Alcaligenes faecalis secretion signal coding region was linked with the coding region for the mature E. coli Penicillin G acylase coding region employing the fusion PCR technique. Two oligonucleotides 5'- GAAACCATTATTATCATGACA-3' (SEQ ID nr 4) and 5'- ACGACTGCTCCGCGTGGGTCGGTGC-3' (SEQ ID nr 5) were used to amplify the Alcaligenes faecalis secretion signal coding region including the modified trp promoter from a plasmid essentially identical to pKECtrp except that the expression cassette contained the Alcaligenes faecalis Penicillin G acylase (R. M. D. Verhaert, ibid.). Two other oligonucleotides 5'-GACCCACGCGGAGCAGTCGTCAAGTG-3' (SEQ ID nr 6) and 5'-CAGTAGTTACGACGGATATCT-3' (SEQ ID nr 7) were used to amplify the N-terminal coding region for the mature E. coli Penicillin G acylase. The PCR fragments thus obtained were purified and used in subsequent cycles as templates for oligonucleotides SEQ ID nrs 4 and 7 yielding a 523 basepair fragment. This fragment was digested with EcoRI and SphI, and exchanged for the corresponding EcoRI- SphI fragment of pKECtrp by restriction enzyme digestion, ligation and transformation to E. coli strain HB101. The resulting plasmid was named pKAFssECtrp (Figure 2). The sequences of this fusion gene are given in SEQ ID nrs 11 and 12.

Example 4 Construction of a new expression cassette: the A. faecalis- E. coli fusion gene under the direction of the aro promoter The aro promoter (G. R. Zurawski, R. P. Gunsalus, K. D. Brown, C. Yanofsky. 1981. J. Mol. Biol. 145: 47-73) was linked to the A. faecalis secretion signal sequence fused to the E. coli Penicillin G acylase mature protein coding region (AFssEC Penicillin G acylase) in an analogous way with the fusion PCR methodology as described in Example 3.

Two oligonucleotides 5'-TCGACTGAATTCTCGATATCATGGGCCTTAGT- 3' (SEQ ID nr 8) and 5'-TGAACTTGCGTAGCATGATAACAAA-3' (SEQ ID nr 9) were designed to obtain the aro promoter by PCR on chromosomal DNA from E. coli strain RR1. Two other oligonucleotides 5'-TATCATGCTACGCAAGTTCACGTAAAAAGGAGG-3' (SEQ ID nr 10) and SEQ ID nr 7 (Example 3) were used to amplify the N-terminal coding region of the AFssEC Penicillin G acylase. Both resulting fragments were subsequently used as templates for oligonucleotides SEQ ID nrs 7 and 8 only, yielding a new DNA fragment of 417 basepairs.

This 417 basepairs fragment was digested with EcoRI and SphI and used to replace the corresponding fragment in pKAFssECtrp by ligation and transformation to E. coli strain HB101, yielding plasmid pKAFssECaro (Figure 2).

Example 5 Production of E. coli Penicillin G acylase: effect of the A. faecalis signal sequence Culturing of E. coli Penicillin G acylase producing strains The following medium was prepared: Yeast extract (total nitrogen content of 2.1 g/1) 19.0 g Na2HP04.2H20 8.9 g

KH2PO4 6.8 g NH4C1 2.4 g Distilled water to 1000 g final medium The pH was adjusted to 6.8 and 100 ml aliquots were placed in 500 ml Erlenmeyer flasks and sterilised (30 min. 121 °C). To each flask glucose and filter sterilised Neomycin stock solutions were added aseptically to a final concentration of 2 g/1 and 10 mg/1 respectively.

The shake flasks were then seeded with a 1.0 ml vegetative suspension of said E. coli strains and incubated at 27 °C under constant orbital agitation till a late logphase culture was obtained.

Two of above cultures were employed to inoculate 5 liter of production medium in a 10 1 glass fermenter. The production medium had the following composition: per kg medium Low salt yeast extract with a free L-tryptophan contents of about 40 mmol/kg (e. g. Gist brocades LS paste'9) 30.0 g Bactopeptone 20mmol/kg (e. g. Difco) 12.5 g (NH4) 2S04 5. 0 g Citric acid 9.0 g CaCl2.2H20 1.25 g FeS04.7H20 0.63 g MnS04. H20 0.025 g Antifoam agent (e. g Basildon) 1.0 g Tapwater to 1000 g final medium After adjusting the pH to 5.5 with NaOH the medium was sterilised at 121 °C for 30 minutes. Then

dextrose, K2HPO4, MgS04.7H20, neomycin and necessary nutritional supplements were added aseptically from sterile stocksolutions to a final concentration of: Dextrose 11.0 g K2HP04 7.5 g MgS04.7H20 2.0 g Neomycin 0.01 g Necessary nutritional supplements (e. g. vitamin B1, amino acids like leucine and proline) were added aseptically to batch medium and/or together with the carbon feed in such amounts that during the fermentation no shortages of these components occurred. Heat labile components were filter sterilised.

As carbon feed 400 g/kg glucose or an equivalent concentration of another carbon source suitable to the host as high DE glucose syrup, glycerol etc. can be applied.

After adjustment of the pH to 4-5 the feed was sterilised for 30 min at 121°C. Heat labelled nutritional supplements were filter sterilised separately and added aseptically after the heat sterilised feed had cooled down sufficiently.

After inoculation the fermentation was started under optimal aeration. Dissolved oxygen concentration in the broth was maintained above 20 % saturation by adjusting stirrerspeed and air flow. The pH was controlled between 6.9 and 7.8 using ammonia and H2SO4. In case of ammonia shortage during the fermentation additional (NH4) 2SO4 was added. The fermentations were conducted at different temperatures between 22 °C and 30 °C.

After all glucose from the batchmedium was consumed a carbon feed was started according to an exponential feed profile. The feed started with lg/kg initial broth per hour. The exponential feed profile was chosen such that at a given temperature no carbon source accumulation or acetate accumulation occurred.

After the maximum feed rate equivalent to 6 g glucose/kg initial broth/hr was reached, this feed rate was maintained till the end of the fermentation.

Fermentation was stopped when acylase titer started to decrease. This happened between 90 and 140 hours after seeding.

Table 1 : Production of E. coli Penicillin G acylase Strain Signal Fermentation Relative Sequence temperature units per g broth HB101 pKECtrp E. coli 23.5 °C 1 from example 2 27.0 °C 0.1-1 HB101 pKAFssECtrp A. 23.5 °C 17 from example 3 faecalis 27.0 °C 74 HB101 pKAFssECaro A. 27.0 °C 55 from example 4 faecalis The assay of E. coli Penicillin G acylase was based on kinetic, photometric measurement at 405 nM of 3- amino-6-nitrobenzoicacid formed from 6-Nitro- 3 (PhenylAcetamido) benzoic acid. Reaction temperature was

37°C. Cell culture suspension were first sonicated.

Recovery samples in which the Penicillin G acylase was liberated from the biomass, were measured directly.

Example 6 Recovery of E. coli Penicillin G acylase After fermentation the recombinant strains expressing E. coli Penicillin G acylase (as produced in Example 5) were killed by adding 1-octanol to a final concentration of 4 g/l. The mixture was incubated for 4 hours and after this cooled to 10-15°C. The cells were disrupted by homogenisation, using two passes through a high pressure slit (600-700 bars). Temperature was maintained at 15°C by cooling. Alternatively, cell mass removal is possible by microfiltration followed by diafiltration.

The mixture was collected in a vessel and the pH was adjusted to a pH of 7. Flocculant was added in a concentration of 4-8 g/1 depending on filterability and stirred for 1.5 hours. After this 10 wt% of dicalite 4108 was added. The solids were filtered off by means of a membrane filter press. After filtration the cake was washed with 2.5 cake volumes of water.

The octanol was removed by adding active carbon CA1 at 2-6 g/1 of filtrate and 1-3% Dicalite 4108.

After 1.5 hours of stirring the solids were filtered off using a membrane filter press. After filtration washing was performed with 2,5 cake volumes of water. The filtrate was filtered over a K700 filter followed by an EKS filter to gain a resulting filtrate low in germs.

This filtrate was adjusted to pH 6-8 by using 25% ammonia, and was concentrated by ultrafiltration using Polysulfonic membranes of 50 kD. After sufficient

concentration (6 times) diafiltration was performed to lower the conductivity to < 2 mS/cm. The resulting filtrate was germ filtrated over K700 and EKS filtration plates.

The filtrate was purified by adsorption onto a Sepharose S gel at pH 5.5 (2.5 mS/cm). The column was rinsed with 20 mM NaAc/HAc buffer at pH 5.0 (1.5 mS/cm).

Elution was performed using 10 mM NaAc/30 mM NaCl at pH 6.0 (4.4 mS/cm), followed by conditioning of the resin with 10 mM NaAc/2.0 mM NaCl at pH 5.5 (150 mS/cm) and washing with 10 mM NaAc at pH 5.5 (0.8 mS/cm).

This elution material was again concentrated by ultrafiltration until the required concentration of product for application was reached. The product was formulated with propylene glycol till a concentration of 30 wt/wt % and again germ filtrated over K700 and EKS.

After this the enzyme was immobilised onto a carrier according to procedures described in the art. Overall yield for acylase was 30%, which means that 30% of input Acylase activity was recovered in immobilized form.

Sequence Listing <110> DSM N. V.

<120> Novel expression cassette for efficient production of a protein <130> EP-2885P <140> <141> <160> 12 <170> PatentIn Ver. 2.1 <210> 1 <211> 151 <212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA <400> 1 gaattcaagg cgcactcccg ttctggataa tgttttttgc gccgacatca taacggttct 60 ggcaaatatt ctgaaatgag ctgttgacaa ttaatcatcg aactagttaa ctagtacgca 120 agttcacgta aaaaggaggt atcgacatat g 151 <210> 2 <211> 26 <212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA

<400> 2 ctgccagagg atcatatgaa aaatag 26 <210> 3 <211> 23 <212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA <400> 3 gggcataaat atgcggcatg ccg 23 <210> 4 <211> 21 <212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA <400> 4 gaaaccatta ttatcatgac a 21 <210> 5 <211> 25 <212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA <400> 5 acgactgctc cgcgtgggtc ggtgc 25

<210> 6 <211> 26 <212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA <400> 6 gacccacgcg gagcagtcgt caagtg 26 <210> 7 <211> 21 <212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA <400> 7 cagtagttac gacggatatc t 21 <210>8 <211> 32 <212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA <400> 8 tcgactgaat tctcgatatc atgggcctta gt 32 <210> 9 <211> 25

<212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA <400> 9 tgaacttgcg tagcatgata acaaa 25 <210> 10 <211> 33 <212> DNA <213> Artificial Sequence <220> <223> Description of Artificial Sequence: Synthetic DNA <400> 10 tatcatgcta cgcaagttca cgtaaaaagg agg 33 <210> 11 <211> 2541 <212> DNA <213> Genomic DNA <220> <221> CDS <222> (1).. (2541) <223> A. faecalis secretion signal coding sequence linked with mature E. coli Peng acylase coding sequence <400> 11 atg cag aaa ggg ctt gtt cgt acc ggg ctt gtg gcc gct ggt ttg atc 48 Met Gln Lys Gly Leu Val Arg Thr Gly Leu Val Ala Ala Gly Leu Ile 1 5 10 15

ttg ggt tgg gcg ggg gca ccg acc cac gcg gag cag tcg tca agt gag 96 Leu Gly Trp Ala Gly Ala Pro Thr His Ala Glu Gln Ser Ser Ser Glu 20 25 30 ata aag att gtt cgc gat gaa tac ggc atg ccg cat att tat gcc aat 144 Ile Lys Ile Val Arg Asp Glu Tyr Gly Met Pro His Ile Tyr Ala Asn 35 40 45 gat aca tgg cac cta ttt tat ggc tat ggc tat gta gta gca caa gat 192 Asp Thr Trp His Leu Phe Tyr Gly Tyr Gly Tyr Val Val Ala Gln Asp 50 55 60 cgc ctt ttt cag atg gaa atg gca cgt cgc agt act caa ggg act gtc 240 Arg Leu Phe Gln Met Glu Met Ala Arg Arg Ser Thr Gln Gly Thr Val 65 70 75 80 gcg gaa gtg ctt ggc aaa gat ttt gtg aaa ttt gat aaa gat atc cgt 288 Ala Glu Val Leu Gly Lys Asp Phe Val Lys Phe Asp Lys Asp Ile Arg 85 90 95 cgt aac tac tgg ccg gat gct atc cgg gcg caa att gct gcc ctt tcc 336 Arg Asn Tyr Trp Pro Asp Ala Ile Arg Ala Gln Ile Ala Ala Leu Ser 100 105 110 cca gag gat atg tcc att ctg caa ggc tac gct gat gga atg aat gcc 384 Pro Glu Asp Met Ser Ile Leu Gln Gly Tyr Ala Asp Gly Met Asn Ala 115 120 125 tgg att gat aag gta aat acc aat cca gag acg ctc tta cca aaa cag 432 Trp Ile Asp Lys Val Asn Thr Asn Pro Glu Thr Leu Leu Pro Lys Gln 130 135 140 ttt aat aca ttt ggc ttt act cct aag cgc tgg gaa ccg ttt gat gtc 480 Phe Asn Thr Phe Gly Phe Thr Pro Lys Arg Trp Glu Pro Phe Asp Val 145 150 155 160 gcg atg ata ttt gtg ggc acc atg gca aac cgc ttc tct gat agc act 528 Ala Met Ile Phe Val Gly Thr Met Ala Asn Arg Phe Ser Asp Ser Thr 165 170 175

agc gaa att gat aat ctg gca ctg cta acg gct tta aaa gat aaa tat 576 Ser Glu Ile Asp Asn Leu Ala Leu Leu Thr Ala Leu Lys Asp Lys Tyr 180 185 190 ggt gta tca caa ggc atg gcg gta ttt aat cag ttg aaa tgg ctg gta 624 Gly Val Ser Gln Gly Met Ala Val Phe Asn Gln Leu Lys Trp Leu Val 195 200 205 aac cca tca gcg cca acc act att gcc gta caa gag agt aac tac cca 672 Asn Pro Ser Ala Pro Thr Thr Ile Ala Val Gln Glu Ser Asn Tyr Pro 210 215 220 ctt aaa ttt aat cag caa aac tcg caa aca gca gct ctg ttg cca cgc 720 Leu Lys Phe Asn Gln Gln Asn Ser Gln Thr Ala Ala Leu Leu Pro Arg 225 230 235 240 tac gat tta cct gca cca atg ctt gac cga cca gca aaa ggg gcg gat 768 Tyr Asp Leu Pro Ala Pro Met Leu Asp Arg Pro Ala Lys Gly Ala Asp 245 250 255 ggc gca ctg ctg gcg tta aca gca ggg aag aac cgg gaa act att gct 816 Gly Ala Leu Leu Ala Leu Thr Ala Gly Lys Asn Arg Glu Thr Ile Ala 260 265 270 gca caa ttt gca cag ggt ggt gcc aat ggt ctg gcg ggg tat cca acg 864 Ala Gln Phe Ala Gln Gly Gly Ala Asn Gly Leu Ala Gly Tyr Pro Thr 275 280 285 acc agc aat atg tgg gtg atc ggc aaa agc aaa gcc cag gat gcg aaa 912 Thr Ser Asn Met Trp Val Ile Gly Lys Ser Lys Ala Gln Asp Ala Lys 290 295 300 gca atc atg gta aat ggt ccg cag ttt ggc tgg tat gcg cct gcg tat 960 Ala Ile Met Val Asn Gly Pro Gln Phe Gly Trp Tyr Ala Pro Ala Tyr 305 310 315 320

act tat ggt att ggt ctg cac ggt gct ggt tat gat gtc act ggc aat 1008 Thr Tyr Gly Ile Gly Leu His Gly Ala Gly Tyr Asp Val Thr Gly Asn 325 330 335 aca cca ttt gcc tat cct ggg ctg gtt ttt ggt cat aat ggt gtg att 1056 Thr Pro Phe Ala Tyr Pro Gly Leu Val Phe Gly His Asn Gly Val Ile 340 345 350 tcc tgg gga tca acg gca ggt ttc ggc gat gat gtc gat att ttt gct 1104 Ser Trp Gly Ser Thr Ala Gly Phe Gly Asp Asp Val Asp Ile Phe Ala 355 360 365 gaa cgg ctg tcg gca gag aaa cca ggc tac tac ttg cat aat ggt aag 1152 Glu Arg Leu Ser Ala Glu Lys Pro Gly Tyr Tyr Leu His Asn Gly Lys 370 375 380 tgg gtg aaa atg tta agc cgt gag gaa acc att acg gtg aaa aat ggt 1200 Trp Val Lys Met Leu Ser Arg Glu Glu Thr Ile Thr Val Lys Asn Gly 385 390 395 400 cag gca gag acc ttt act gtc tgg cgt acg gtg cat ggc aac att ctc 1248 Gln Ala Glu Thr Phe Thr Val Trp Arg Thr Val His Gly Asn Ile Leu 405 410 415 caa act gac cag acg aca caa acg gct tac gct aaa tcc cgc gca tgg 1296 Gln Thr Asp Gln Thr Thr Gln Thr Ala Tyr Ala Lys Ser Arg Ala Trp 420 425 430 gat ggt aaa gag gtg gcg tct ttg ctg gcc tgg act cat cag atg aag 1344 Asp Gly Lys Glu Val Ala Ser Leu Leu Ala Trp Thr His Gln Met Lys 435 440 445 gcc aaa aat tgg cag gag tgg aca cag cag gca gcg aaa caa gca ctg 1392 Ala Lys Asn Trp Gln Glu Trp Thr Gln Gln Ala Ala Lys Gln Ala Leu 450 455 460 acc atc aac tgg tac tat gct gat gta aac ggc aat att ggt tat gtt 1440 Thr Ile Asn Trp Tyr Tyr Ala Asp Val Asn Gly Asn Ile Gly Tyr Val 465 470 475 480

cat act ggt gct tat cca gat cgt caa tca ggc cat gat ccg cga tta 1488 His Thr Gly Ala Tyr Pro Asp Arg Gln Ser Gly His Asp Pro Arg Leu 485 490 495 ccc gtt cct ggt acg gga aaa tgg gac tgg aaa ggg cta ttg cct ttt 1536 Pro Val Pro Gly Thr Gly Lys Trp Asp Trp Lys Gly Leu Leu Pro Phe 500 505 510 gaa atg aac cct aag gtg tat aac ccc cag tcg gga tat att gct aac 1584 Glu Met Asn Pro Lys Val Tyr Asn Pro Gln Ser Gly Tyr Ile Ala Asn 515 520 525 tgg aac aat tct ccc caa aaa gat tat ccc gct tca gat ctg ttt gcc 1632 Trp Asn Asn Ser Pro Gln Lys Asp Tyr Pro Ala Ser Asp Leu Phe Ala 530 535 540 ttt ttg tgg ggt ggt gca gat cgc gtt acg gag atc gac cga ctg ctt 1680 Phe Leu Trp Gly Gly Ala Asp Arg Val Thr Glu Ile Asp Arg Leu Leu 545 550 555 560 gag caa aag cca cgc tta act gct gat cag gca tgg gat gtt att cgc 1728 Glu Gln Lys Pro Arg Leu Thr Ala Asp Gln Ala Trp Asp Val Ile Arg 565 570 575 caa acc agt cgt cag gat ctt aac ctg agg ctt ttt tta cct act ctg 1776 Gln Thr Ser Arg Gln Asp Leu Asn Leu Arg Leu Phe Leu Pro Thr Leu 580 585 590 caa gca gcg aca tct ggt ttg aca cag agc gat ccg cgt cgt cag ttg 1824 Gln Ala Ala Thr Ser Gly Leu Thr Gln Ser Asp Pro Arg Arg Gln Leu 595 600 605 gta gaa aca tta aca cgt tgg gat ggc atc aat ttg ctt aat gat gat 1872 Val Glu Thr Leu Thr Arg Trp Asp Gly Ile Asn Leu Leu Asn Asp Asp 610 615 620

ggt aaa acc tgg cag cag cca ggc tct gcc atc ctg aac gtt tgg ctg 1920 Gly Lys Thr Trp Gln Gln Pro Gly Ser Ala Ile Leu Asn Val Trp Leu 625 630 635 640 acc agt atg ttg aag cgt acc gta gtg gct gcc gta cct atg cca ttt 1968 Thr Ser Met Leu Lys Arg Thr Val Val Ala Ala Val Pro Met Pro Phe 645 650 655 gat aag tgg tac agc gcc agt ggc tac gaa aca acc cag gac ggc cca 2016 Asp Lys Trp Tyr Ser Ala Ser Gly Tyr Glu Thr Thr Gln Asp Gly Pro 660 665 670 act ggt tcg ctg aat ata agt gtt gga gca aaa att ttg tat gag gcg 2064 Thr Gly Ser Leu Asn Ile Ser Val Gly Ala Lys Ile Leu Tyr Glu Ala 675 680 685 gtg cag gga gac aaa tca cca atc cca cag gcg gtt gat ctg ttt gct 2112 Val Gln Gly Asp Lys Ser Pro Ile Pro Gln Ala Val Asp Leu Phe Ala 690 695 700 ggg aaa cca cag cag gag gtt gtg ttg gct gcg ctg gaa gat acc tgg 2160 Gly Lys Pro Gln Gln Glu Val Val Leu Ala Ala Leu Glu Asp Thr Trp 705 710 715 720 gag act ctt tcc aaa cgc tat ggc aat aat gtg agt aac tgg aaa aca 2208 Glu Thr Leu Ser Lys Arg Tyr Gly Asn Asn Val Ser Asn Trp Lys Thr 725 730 735 cct gca atg gcc tta acg ttc cgg gca aat aat ttc ttt ggt gta ccg 2256 Pro Ala Met Ala Leu Thr Phe Arg Ala Asn Asn Phe Phe Gly Val Pro 740 745 750 cag gcc gca gcg gaa gaa acg cgt cat cag gcg gag tat caa aac cgt 2304 Gln Ala Ala Ala Glu Glu Thr Arg His Gln Ala Glu Tyr Gln Asn Arg 755 760 765 gga aca gaa aac gat atg att gtt ttc tca cca acg aca agc gat cgt 2352 Gly Thr Glu Asn Asp Met Ile Val Phe Ser Pro Thr Thr Ser Asp Arg 770 775 780

cct gtg ctt gcc tgg gat gtg gtc gca ccc ggt cag agt ggg ttt att 2400 Pro Val Leu Ala Trp Asp Val Val Ala Pro Gly Gln Ser Gly Phe Ile 785 790 795 800 gct ccc gat gga aca gtt gat aag cac tat gaa gat cag ctg aaa atg 2448 Ala Pro Asp Gly Thr Val Asp Lys His Tyr Glu Asp Gln Leu Lys Met 805 810 815 tac gaa aat ttt ggc cgt aag tcg ctc tgg tta acg aag cag gat gtg 2496 Tyr Glu Asn Phe Gly Arg Lys Ser Leu Trp Leu Thr Lys Gln Asp Val 820 825 830 gag gcg cat aag gag tcg cag gaa gtg ttg cac gtt cag aga taa 2541 Glu Ala His Lys Glu Ser Gln Glu Val Leu His Val Gln Arg 835 840 845 <210> 12 <211> 847 <212> PRT <213> Protein <400> 12 Met Gln Lys Gly Leu Val Arg Thr Gly Leu Val Ala Ala Gly Leu Ile 1 5 10 15 Leu Gly Trp Ala Gly Ala Pro Thr His Ala Glu Gln Ser Ser Ser Glu 20 25 30 Ile Lys Ile Val Arg Asp Glu Tyr Gly Met Pro His Ile Tyr Ala Asn 35 40 45 Asp Thr Trp His Leu Phe Tyr Gly Tyr Gly Tyr Val Val Ala Gln Asp 50 55 60 Arg Leu Phe Gln Met Glu Met Ala Arg Arg Ser Thr Gln Gly Thr Val 65 70 75 80

Ala Glu Val Leu Gly Lys Asp Phe Val Lys Phe Asp Lys Asp Ile Arg 85 90 95 Arg Asn Tyr Trp Pro Asp Ala Ile Arg Ala Gln Ile Ala Ala Leu Ser 100 105 110 Pro Glu Asp Met Ser Ile Leu Gln Gly Tyr Ala Asp Gly Met Asn Ala 115 120 125 Trp Ile Asp Lys Val Asn Thr Asn Pro Glu Thr Leu Leu Pro Lys Gln 130 135 140 Phe Asn Thr Phe Gly Phe Thr Pro Lys Arg Trp Glu Pro Phe Asp Val 145 150 155 160 Ala Met Ile Phe Val Gly Thr Met Ala Asn Arg Phe Ser Asp Ser Thr 165 170 175 Ser Glu Ile Asp Asn Leu Ala Leu Leu Thr Ala Leu Lys Asp Lys Tyr 180 185 190 Gly Val Ser Gln Gly Met Ala Val Phe Asn Gln Leu Lys Trp Leu Val 195 200 205 Asn Pro Ser Ala Pro Thr Thr Ile Ala Val Gln Glu Ser Asn Tyr Pro 210 215 220 Leu Lys Phe Asn Gln Gln Asn Ser Gln Thr Ala Ala Leu Leu Pro Arg 225 230 235 240 Tyr Asp Leu Pro Ala Pro Met Leu Asp Arg Pro Ala Lys Gly Ala Asp 245 250 255 Gly Ala Leu Leu Ala Leu Thr Ala Gly Lys Asn Arg Glu Thr Ile Ala 260 265 270 Ala Gln Phe Ala Gln Gly Gly Ala Asn Gly Leu Ala Gly Tyr Pro Thr 275 280 285

Thr Ser Asn Met Trp Val Ile Gly Lys Ser Lys Ala Gln Asp Ala Lys 290 295 300 Ala Ile Met Val Asn Gly Pro Gln Phe Gly Trp Tyr Ala Pro Ala Tyr 305 310 315 320 Thr Tyr Gly Ile Gly Leu His Gly Ala Gly Tyr Asp Val Thr Gly Asn 325 330 335 Thr Pro Phe Ala Tyr Pro Gly Leu Val Phe Gly His Asn Gly Val Ile 340 345 350 Ser Trp Gly Ser Thr Ala Gly Phe Gly Asp Asp Val Asp Ile Phe Ala 355 360 365 Glu Arg Leu Ser Ala Glu Lys Pro Gly Tyr Tyr Leu His Asn Gly Lys 370 375 380 Trp Val Lys Met Leu Ser Arg Glu Glu Thr Ile Thr Val Lys Asn Gly 385 390 395 400 Gln Ala Glu Thr Phe Thr Val Trp Arg Thr Val His Gly Asn Ile Leu 405 410 415 Gln Thr Asp Gln Thr Thr Gln Thr Ala Tyr Ala Lys Ser Arg Ala Trp 420 425 430 Asp Gly Lys Glu Val Ala Ser Leu Leu Ala Trp Thr His Gln Met Lys 435 440 445 Ala Lys Asn Trp Gln Glu Trp Thr Gln Gln Ala Ala Lys Gln Ala Leu 450 455 460 Thr Ile Asn Trp Tyr Tyr Ala Asp Val Asn Gly Asn Ile Gly Tyr Val 465 470 475 480 His Thr Gly Ala Tyr Pro Asp Arg Gln Ser Gly His Asp Pro Arg Leu 485 490 495

Pro Val Pro Gly Thr Gly Lys Trp Asp Trp Lys Gly Leu Leu Pro Phe 500 505 510 Glu Met Asn Pro Lys Val Tyr Asn Pro Gln Ser Gly Tyr Ile Ala Asn 515 520 525 Trp Asn Asn Ser Pro Gln Lys Asp Tyr Pro Ala Ser Asp Leu Phe Ala 530 535 540 Phe Leu Trp Gly Gly Ala Asp Arg Val Thr Glu Ile Asp Arg Leu Leu 545 550 555 560 Glu Gln Lys Pro Arg Leu Thr Ala Asp Gln Ala Trp Asp Val Ile Arg 565 570 575 Gln Thr Ser Arg Gln Asp Leu Asn Leu Arg Leu Phe Leu Pro Thr Leu 580 585 590 Gln Ala Ala Thr Ser Gly Leu Thr Gln Ser Asp Pro Arg Arg Gln Leu 595 600 605 Val Glu Thr Leu Thr Arg Trp Asp Gly Ile Asn Leu Leu Asn Asp Asp 610 615 620 Gly Lys Thr Trp Gln Gln Pro Gly Ser Ala Ile Leu Asn Val Trp Leu 625 630 635 640 Thr Ser Met Leu Lys Arg Thr Val Val Ala Ala Val Pro Met Pro Phe 645 650 655 Asp Lys Trp Tyr Ser Ala Ser Gly Tyr Glu Thr Thr Gln Asp Gly Pro 660 665 670 Thr Gly Ser Leu Asn Ile Ser Val Gly Ala Lys Ile Leu Tyr Glu Ala 675 680 685 Val Gln Gly Asp Lys Ser Pro Ile Pro Gln Ala Val Asp Leu Phe Ala 690 695 700

Gly Lys Pro Gln Gln Glu Val Val Leu Ala Ala Leu Glu Asp Thr Trp 705 710 715 720 Glu Thr Leu Ser Lys Arg Tyr Gly Asn Asn Val Ser Asn Trp Lys Thr 725 730 735 Pro Ala Met Ala Leu Thr Phe Arg Ala Asn Asn Phe Phe Gly Val Pro 740 745 750 Gln Ala Ala Ala Glu Glu Thr Arg His Gln Ala Glu Tyr Gln Asn Arg 755 760 765 Gly Thr Glu Asn Asp Met Ile Val Phe Ser Pro Thr Thr Ser Asp Arg 770 775 780 Pro Val Leu Ala Trp Asp Val Val Ala Pro Gly Gln Ser Gly Phe Ile 785 790 795 800 Ala Pro Asp Gly Thr Val Asp Lys His Tyr Glu Asp Gln Leu Lys Met 805 810 815 Tyr Glu Asn Phe Gly Arg Lys Ser Leu Trp Leu Thr Lys Gln Asp Val 820 825 830 Glu Ala His Lys Glu Ser Gln Glu Val Leu His Val Gln Arg 835 840 845