AU697156B2

AU697156B2 - Cloning and expression of the chondroitinase I and II genes from (p. vulgaris)

Info

Publication number: AU697156B2
Application number: AU68183/94A
Authority: AU
Inventors: Kiran Manohar Khandke; Jason Arnold Lotvin; Michael Joseph Ryan; Bruce Clifford Tilley
Original assignee: American Cyanamid Co
Current assignee: Wyeth Holdings LLC
Priority date: 1993-04-23
Filing date: 1994-04-22
Publication date: 1998-10-01
Anticipated expiration: 2014-04-22
Also published as: EP0702715A1; AU6818394A; WO1994025567A1; JPH09500011A; EP0702715A4; CA2161125A1

Description

vp,-- ,00 i: 11_: I I I WO 94/25567 PCT/US94/04495 CLONING AND EXPRESSION OF THE CHONDROITINASE I AND II GENES FROM P. VULGARIS Field of the Invention This invention relates to the DNA sequence encoding the major protein component of chondroitinase ABC, which is referred to as "chondroitinase from Proteus vulqaris vulgaris). This invention further relates to the DNA sequence encoding a second protein component of chondroitinase ABC, which is referred to as "chondroitinase II", from P. vulgaris.

This invention also relates to the cloning and expression of the genes containing these DNA sequences and to the amino acid sequences of the recombinant chondroitinase I and II enzymes encoded by these DNA sequences.

This invention additionally relates to methods for the isolation and purification of the recombinantly expressed major protein component of chondroitinase ABC, which is referred to as "chondroitinase from Protaus vulgaris (P.

vulgaris). This invention further relates to methods for the isolation and purification of the recombinantly expressed second protein component of chondroitinase ABC, which is referred to as "chondroitinase II", from P. vulcaris. These methods provide significantly higher yields and purity than those obtained by adapting for the recombinant enzymes the method previously used for isolating and purifying the native chondroitinase I enzyme from P. vulqaris.

i i/ uI

~~I

r 1 WO 94/25567 PCT/US94/04495 WO 94/25567 PCT/US94/04495 2 Background of the Invention Chondroitinases are enzymes of bacterial origin which have been described as having value in dissolving the cartilage of herniated discs without disturbing the stabilizing collagen components of those discs.

Examples of chondroitinase enzymes are chondroitinase ABC, which is produced by the bacterium P. vulgaris, and chondroitinase AC, which is produced by A. aurescens. The chondroitinases function by degrading polysaccharide side chains in proteinpolysaccharide complexes, without degrading the protein core.

Yamagata et al. describes the purification of the enzyme chondroitinase ABC from extracts of P.

vulraris (Bibliography entry The enzyme selectively degrades the glycosaminoglycans chondroitin-4-sulfate, dermatan sulfate and chondroitin-6-sulfate (also referred to -espectively as chondroitin sulfates A, B and C) at pH 8 at higher rates than chondroitin or hyaluronic acid. However, the enzyme did not attack keratosulfate, heparin or heparitin sulfate.

Kikuchi et al. describes the purification of glycosaminoglycan degrading enzymes, such as chondroitinase ABC, by fractionating the enzymes by adsorbing a solution containing the enzymes onto an insoluble sulfated polysaccharide carrier and then desorbing the individual enzymes from the carrier Brown describes a method for treating intervertebral disc displacement in mammals, including humans, by injecting into the intervertebral disc space effective amounts of a solution containing chondroitinase ABC The chondroitinase ABC was

(C

I

c;,.1:3 j

-U,

WO 94/25567 PCT/US94/04495 3 isolated and purified from extracts of P. vulgaris.

This native enzyme material functioned to dissolve cartilage, such as herniated spinal discs.

.Specifically, the enzyme causes the selective chemonucleolysis of the nucleus pulposus which contains proteoglycans and randomly dispersed collagen fibers.

Hageman describes an ophthalmic vitrectomy method for selectively and completely disinserting the ocular vitreous body, epiretinal membranes or fibrocellular membranes from the neural retina, ciliary epithelium and posterior lens surface of the mammalian eye as an adjunct to vitrectomy, by administering to the eye an effective amount of an enzyme which disrupts or degrades chondroitin sulfate proteoglycan localized specifically to sites of vitreoretinal adhesion and thereby permit complete disinsertion of said vitreous body and/or epiretinal membranes The enzyme can be a protease-free glycosaminoglycanase, such as chondroitinase ABC.

Hageman utilized chondroitinase ABC obtained from Seikagaku Kogyo Co., Ltd., Tokyo, Japan.

In isolating and purifying the chondroitinase ABC enzyme from the Seikagaku Kogyo material, it was noted that there was a correlation between effective preparations of the chondroitinase in vitrectomy procedures and the presence of a second protein having an apparent molecular weight (by SDS- PAGE) slightly greater than that of the major protein component of chondroitinase ABC. The second protein is now designated "chondroitinase II", while the major protein component of chondroitinase ABC is referred to as "chondroitinase The chondroitinase I and II proteins are /,sic proteins at neutral pH, with similar isoelectric points of 8.30-8.45. Separate

I'\

i,

A;

WO 94/25567 PCTIUS94/04495

I

4 purification of the chondroitinase I and II forms of the native enzyme revealed that it was the combination of the two proteins that was active in the surgical vitrectomy rather than either of the proteins individually.

Use of the chondroitinase I and II forms of the native enzyme to date has been limited by the small amounts of enzymes obtained from native sources.

The production and purification of the native forms of the enzyme has been carried out using fermentations of P. vulqaris in which its substrate has been used as the inducer to initiate production of these forms of the enzyme. A combination of factors, including low levels of synthesis, the cost and availability of the inducer (chondroitin sulfate), and the opportunistically pathogenic nature of P. vulgaris, has resulted in the requirement for a more efficient method of production. In addition, the native forms of the enzyme produced by conventional techniques are subject to degradation by proteases present in the bacterial extract. Therefore, there is a need for a reliable supply of pure material free of contaminants in order for the medical applications of the two forms of this enzyme to be evaluated properly and exploited. There is also a need for methods to isolate and purify a reliable supply of the chondroitinase I and II enzymes free of contaminants.

k Summary of the Invention Accordingly, the present invention is directed to the production of chondroitinase I and chondroitinase II in quantities not readily achievable using present non-recombinant bacterial fermentation and extraction techniques.

WO 94/25567 PCTIUS94/04495 1 a This invention is also directed to the production of chondroitinase I and chondroitinase II, each in a form substantially free of proteases which would otherwise degrade the enzyme and cause a loss of its activity.

These aspects of the present invention are realised through the use of an alternative approach to the problems presented by large scale bacterial fermentation of these two forms of enzyme.

Separately for chondroitinase I and chondroitinase II, the gene that encodes the enzyme is cloned and the enzyme is expressed at high levels in a heterologous host. In a preferred embodiment, this invention is directed to the cloning of 'the P. vulraria gene for chondroitinase I and the high level expression of that enzyme in E. coli, as well as the cloning of the P. vulqaris gene for chondroitinase I1 and the high level expression of that enzyme in E. coli.

This invention provides a purified isolated DNA fragment of P. vulqaris which comprises a sequence encoding for chondroitinase I. This invention further provides a purified isolated DNA fragment of P.

vulqaris which hybridizes with a nucleic acid sequence encoding for amino acids as follows: the chondroitinase I enzyme with its signal peptide (SEQ ID NO:2, amino acids 1-1021) or a biological equivalent thereof (encoded for example by: nucleotides numbered 119-3181 of SEQ ID NO:1, and nucleotides numbered 119-3181 of SEQ ID NO:3, where the three nucleotides immediately upstream of the initiation codon are changed (SEQ ID NO:3, nucleotides 116- 118)) I i ii i i i, r, iii iii i I,,

I

i

I

i i i i s

I

i I i r- WO 94/25567 PCT/US94/04495 567 PCTUS9404495 6 the mature chondroitinase I enzyme (SEQ ID NO:2, amino acids 25-1021) or a biological equivalent thereof (encoded for example by: nucleotides numbered 191-3181 of SEQ ID NO:1, and nucleotides numbered 191-3181 of SEQ ID NO:3, where the three nucleotides immediately upstream of the initiation codon are changed (SEQ ID NO:3, nucleotides 116-118)); and the mature chondroitinase I enzyme where the sequence encoding the signal peptide has been replaced with a sequence which adds a methionine residue to the amino terminus of the enzyme (SEQ ID NO:5, amino acids 24- 1021) or a biological equivalent thereof (encoded for example by nucleotides numbered 188-3181 of SEQ ID NO:4).

The recombinant chondroitinase I is produced by transforming a host cell with a plasmid containing a purified isolated DNA fragment of P. vulgaris which contains one of the above-described sequences, and culturing the host cell under conditions which permit expression of the enzyme by the host cell.

This invention also provides a purified isolated DNA fragment of P. vulaaris which comprises a sequence encoding for chondroitinase II. This invention further provides a purified isolated DNA fragment from P. vulgaris which hybridizes with a nucleic acid sequence encoding for amino acids as follows: the chondroitinase II enzyme with its signal peptide (SEQ ID NO:40, amino

I

fWi Li d WO 94/25567 PCTUS94/044 9 r 7 acids 1-1013) or a biological equivalent thereof (encoded for example by nucleotides numbered 3238-6276 of SEQ ID NO:39); and the mature chondroitinauie II enzyme (SEQ ID NO:40, amino acids 24-1013) or a biological equivalent thereof (encoded for example by nucleotides numbered 3307-6276 of SEQ ID NO:39).

The recombinant chondroitinase II is produced by transforming a host cell with a plasmid containing a purified isolated DNA fragment of P.

vulgaris which contains one of the above-described sequences, and culturing the host cell under conditions which permit expression of the enzyme by the host cell.

This invention is also directed to providing methods for the isolation and purification of the recombinantly expressed chondroitinase I enzyme of P. vulcaris.

In particular, this invention is directed to providing methods which result in significantly higher yields and purity of the recombinant chondroitinase I enzyme than those obtained by adapting for the recombinant enzyme the method previously used for isolating and purifying the native chondroitinase I enzyme from P. vulaaris.

These aspects of the present invention are realised through either of two methods described and claimed herein for the chondroitinase I enzyme. The first method comprises the steps of: lysing by homogenization the host cells which express the recombinant chondroitinase I enzyme to release the enzyme into the supernatant;

I

aa,a a a a a I' i,-r 1-i m r lri i id\U' -~*i'irL-Y n WO 94/25567 PCTIUS94/04495 subjecting the supernatant to diafiltration to remove salts and other small molecules; passing the supernatant through an anion exchange resin-containing column; loading the eluate from step to a cation exchange resin- containing column so that the enzyme in the eluate binds to the cation exchange column; and eluting the enzyme bound to the *cation exchange column with a solvent capable of releasing the enzyme from the column.

In the second method, prior to step of the first method just described, the following two steps are performed: treating the supernatant with an acidic solution to precipitate out the enzyme; and recovering the pellet and then dissolving it in an alkali solution to again place the enzyme in a basic environiment.

The present invention is also directed to providing methods for the isolation and purification of the recombinantly expressed chondroiti-nase II enzyme of P.vlars Additionally; this invention provides methods which results in significantly higher yields and purity of the recombinant chondroitinase II enzyme than those obtained by adapting for the recombinant enzyme the method previously used for isolating and purifying the native chondroitinase I enzyme from P.

vulgarig.

These aspects of the present invention are also realised through either of

II

I

5 I I I

V

WO 94125567 PTU9149 PCTIUS94/04495 s- -r -4~ WO 94/25567 PCT/US94/04495 9 25 30 35K1 two methods described and claimed herein for the chondroitinase II enzyme. The first method comprises the steps of: lysing by homogenization the host cells which express the recombinant chondroitinase I enzyme to release the enzyme into the supernatant; subjecting the supernatant to diafiltration to remove salts and other small molecules; passing the supernatant through an anion exchange resin-containing column; loading the eluate from step to a cation exchange resin-containing column so that the enzyme in the eluate binds to the cation exchange column; obtaining by affinity elution the enzyme bound to the cation exchange column with a solution of chondroitin sulfate, such that the enzyme is coeluted with the chondroitin sulfate; loading the eluate from step to an anion exchange resin-containing column and eluting the enzyme with a solvent such that the chondroitin sulfate binds to the column; and concentrating the eluate from step (f) and crystallizing out the enzyme from the supernatant which contains an approximately 37 kD contaminant.

In the second method, prior to step of the first method just described, the following two steps are performed: treating the supernatant with an acidic solution to precipitate out the enzyme;

Q

iri i 'il r j i i

II

:I

E

-li; 1 51

I~)

-r I WO 94/25567 PCT/US94/04495 I I 21 i I I I Ii WO 94/25567 PCTUS94/04495 10 and recovering the pellet and then dissolving it in an alkali solution to again place the enzyme in a basic environment.

Use of the methods of this invention results in significantly higher yields and purity of each recombinant enzyme than those obtained by adapting for each recombinant enzyme the method previously used for isolating and purifying the native chondroitinase I enzyme from P. vul.ris.

Brief Description of the Figures Figure 1 depicts a preliminary restriction map for the subcloned approximately 10 kilobase Nsi fragment in pIBI24. The Nsi fragment contains the complete gene encoding chondroitinase I and a portion of the gene encoding chondroitinase II. The restriction sites are shown in their approximate positions. The restriction sites are useful in the constructions described below; other restriction sites present are not shown in this Figure; some are set forth in Example 13 below.

Figure 2 depicts the elution of the recombinant chondroitinase I enzyme from a cation exchange chromatography column using a sodium chloride gradient. The method useer to purify the native enzyme is used here to attempt to purify the recombinant enzyme. The initial fractions at the left do not bind to the column. They contain the majority of the chondroitinase I enzyme activity. The fractions at right containing the enzyme Lre mjrked "eluted actv ity". The radient is from 0.0, to 250 mM NaCl.

Figure 3 ,epicts the elution of the E I J ii 1 ;i.i.*~-lil WO 94/25567 PCTIUS94/04495 11 recombinant chondroitinase I enzyme from a cation exchan7e column, after first passing the supernatant through an anion exchange column, in accordance with a method of this invention. The initial fractions at the left do not bind to the column, and contain only traces of chondroitinase I activity. The fractions at right containing the enzyme are marked "eluted activity". The gradient is from 0.0 to 250 mM NaCl.

Figure 4 depicts sodium dodecy- sulfatepolyacrylamide gel chromatography (SDS-PAGE) of the recombinant chondroitinase I enzyme before and after the purification methods of this invention are used.

In the SDS-PAGE gel photograph, Lane 1 is the enzyme purified using the method of the first embodiment of the invention; Lane 2 is the enzyme purified using the method of the second embodiment of the invention; Lane 3 represents the supernatant from the host cell prior to purification many ot~er proteins are present; Lane 4 represents the following molecular weight standards: 14.4 kD lysozyme; 21.5 kD trypsin inhibitor; 31 kD carbonic anhydrase; 42.7 kD ovalbumin; 66.2 kD bovine serum albumin; 97.4 kD phosphorylase B; 116 kD betagalactosidase; 200 kD myosin. A single sharp band is seen in Lanes 1 and 2.

Figure 5 depicts SDS-PAGE chromatography of the recombinant chondroitinase II enzyme during various stages of purification using a method of this invention. In the SDS-PAGE gel photograph, Lane 1 is the crude supernatant after diafiltration; Lane 2 the eluate after passage of the supernatant through an anion exchange resin-containing column; Lane 3 is the enzyme after elution through a cation exchange resincontaining column; Lane 4 is the enzyme after elution through a second anion exchange resin-containing

AI

I

CS1--~

~C

j-~ i I WO 94/25567 PCTUS94/04495 12 column; Lane 5 represents the same molecular weight standards as described for Figure 4, plus 6.5 kD aprotinin; Lane 6 is the same as Lane 4, except it is overloaded to show the approximately 37 kD contaminant; Lane 7 is the 37 kD contaminant in the supernatant after crystallization of the chondroitinase II enzyme; Lane 8 is first wash of the crystals; Lane 9 is the second wash of the crystals; Lane 10 is the enzyme in the washed crystals after redissolving in water.

Detailed Description of the Invention Preliminary experiments indicated that E.

coli could not use the hydrolysis products yielded by chondroitinase I as a sole carbon source, suggesting that this gene could not be cloned by selecting for its expression in E. coli. Another approach, followed in this application, is to use a physical method to identify DNA fragments that encode the chondroitinase I enzyme. This is accomplished using an appropriately labeled probe for hybridization with individual clones that, together, make up a gene bank comprising the complete genome of P. vulgaris. The probe itself is generated using Polymerase Chain Reaction (PCR) In this procedure, the genomic DNA of P. vulgaris is denatured and oligonucleotides (designed to bracket part of the chondroitinase I gene) are annealed and DNA synthesis is carried bu't in vitro. This cycle of denaturation, annealing and DNA synthesis using the Soligonucleotides as primers is repeated many times 30), with the yield of the desired product (the DNA fragment that lies between the two oligonucleotides) increasing exponentially with each cycle.

A putative nucleotide sequence of the ~r )r c -I_1~I-~4 WO 94/25567 PCT/US94/04495 I: WO 94/25567 PCT/US94/04495 13 appropriate oligonucleotides is constructed from available amino acid sequence information derived from the protein purified from P. vularis bacteria. Once this is done, the DNA fragment produced by PCR is cloned and its DNA sequence determined to verify that it is part of the chondroitinase I gene. It is then labeled and used as a probe to indicate which members of the gene bank actually contain the chondroitinase I gene. Subsequent restriction mapping and Southern hybridization narrows the location to a piece of DNA of approximately four thousand base-pairs This is then sequenced using the Sanger dideoxy chain termination method to reveal the exact position of the gene and guide the subsequent manipulations used to place the gene into a high-level expression system in E. coli. A fermentation at a 10 liter scale carried out with this E. coli strain containing a recombinant plasmid expressing the P. vulgaris chondroitinase I gene yields a maximum chondroitinase I titer of approximately 600 units/ml (which is the same as 1.2 mg/ml). This yield far exceeds that of the native P. vulqcaris fermentation process which had not achieved a titer of more than 2 units/ml.

The process of cloning and expression of the chondroitinase I gene is summarized by the following series of stages: 1) The isolation of P. vulgaris genomic DNA and the construction of a cosmid gene bank.

2) PCR experimentation designed to yield an authentic piece of the chondroitinase I gene for use as-a hybridization probe.

3) Colony hybridization studies to identify at least a portion of the chondroitinase I gene.

4) Restriction mapping, Southern hybridij i: ii

W

f:i

I

ai~c.ui* rl ;L1, L -i i: I r a o i j

I

:L k -1 r

Q

WO 94/25567 PCT/US94/04495 ;a ;e WO 94/25567 PCT/US94/04495 14 zation, DNA sequencing, and chondroitinase I enzyme assays that, collectively, serve to place the location of the chondroitinase I gene more precisely within the cloned DNA.

DNA sequence analysis to reveal the exact coding region and location of the chondroitinase I gene.

6) Site-specific mutagenesis, related manipulations, and genetic engineering leading to the regulated, high-level expression of the P. vulgaris gene in E. coli.

These six stages are described in specific detail in Examples 1-7 below. The rationale for the stages is as follows. In the first stage, genomic DNA is obtained. DNA is separated from protein and other material contained in a P. vulgaris fermentation.

Study of the genomic DNA is facilitated by the insertion of fragments of the DNA into cosmid vectors.

The genomic DNA is digested with an appropriate restriction endonuclease, such as Sau3A, and then ligated into a cosmid vector. The packaged recombinant cosmids containing the P. vulgaris DNA fragments are introduced into an appropriate bacterial host strain, such as an E. coli strain, and the resulting culture is grown to allow gene expression.

The gene banks are engineered to contain a marker, such as ampicillin or kanamycin resistance, to assist in the screening of the gene banks for the presence of the chondroitinase I gene.

Applicants have conducted some amino acid sequencing of the native chondroitinase I enzyme.

Samples of the enzyme are generated by fermentation of P. vulgaris. Samples may also be obtained from Seikagaku Kogyo Co., Ltd., Tokyo, Japan. The amino acid sequence information is used to design p; ii 1 f* i cc, 'i! i _L .a PCT/US94/04495 WO 94/25567

D^'

WO 94/25567 PCT/US94/04495 15 oligonucleotides for use in screening for the chondroitinase I gene.

In the second stage, oligonucleotides are designed for use in PCR. A first set of oligonucleotides is designed so as to encode a heptapeptide that has minimal degeneracy of its genetic code. Seven amino acids near the amino terminus of the chondroitinase I enzyme (SEQ ID NO:2, amino acids 19-25) are potentially encoded by 512 different nucleotide sequences (SEQ ID NO:6; see Example The number of potential sequences is reduced to 32 by selecting specific nucleotides at the end, because of the observation that mismatched nucleotides in PCR primers are of less consequence at the 5' end than at the 3' end of the primer The sequences of the pool of 32 primers are set out at SEQ ID NOS:7-14.

Applicants have discovered that the approximately 110 kD chondroitinase I enzyme is cleaved proteolytically into an 18,000 MW ("18 kD") fragment and an approximately 90,000 MW ("90 kD") fragment. Furthermore, the 18 kD fragment is further fragmented by treatment with cyanogen bromide and trypsin. The various fragments are then used to design additional sets of oligonucleotide primers for

PCR.

Seven amino acids within the 18 kD fragment (SEQ ID NO:2, amino acids 114 120) are potentially encoded by 512 different nucleotide sequences (SEQ ID NO:15; see Example The complementary strand has the same nrumber of potential sequences (SEQ ID NO:16; see Example Using the criteria described above for the first set of cligonucleotides, the number of potential sequences is reduced to 128, whose sequences are set out at SEQ ID NOS:17-24.

s* Illilil i WO 94/2 35 25567 PCT/US94/04495 -16 Six amino acids located near the aminoterminus of the "90 kD" fragment (SEQ ID NO:2, amino acids 165-170) are potentially encoded by a large number of different nucleotide sequences (SEQ ID and 26; see Example The complementary strand has the same number of potential sequences (SEQ ID NOS:27 and 28; see Example Using the criteria described above for the first set of oligonucleotides, the number of potential sequences is reduced to the sequences set out at SEQ ID NOS:29-36.

PCR amplifications are conducted using these 24 mixtures of oligonucleotides. The most effective amplifications are observed as discrete bands on electrophoretic gels. Products approximately 500 and 350 base pairs (bp) in size are obtained. The approximately 350 bp product is a subfragment of the approximately 500 bp product. The approximately 500 bp product is isolated and, following successive cloning procedures described in Example 2, is isolated as a 455 bp PCR product.

This 455 bp fragment is sequenced and translated into an amino acid sequence which is in virtual agreement with the sequence available from the native chondroitinase I enzyme. The sequences differ by one amino acid; subsequent experiments reveal that the nucleotide and amino acid sequences of the 455 bp fragment are correct, while the native amino acid sequence identification is in eror.

In the third stage, the PCR amplification fragment is used as a probe to identify the cosmid gene banks prepared in the first stage which contain the chondroitinase I gene. The PCR fragment is denatured and labelled with, for example, digoxigeninlabelled dUTP (Boehringer-Mannheim, Indianapolis, IN).

The cosmid gene banks are then used to infect a

I

i-*I

I

i

L

a P WO 94/25567 prT/TTCOiA/AAOC

A:

I. WO 94/25567 PCT[US94/04495 17 bacterial strain. The resulting colonies are lysed and their DNA subjected to colony hybridization with the labelled probe, followed by exposure to an alkaline phosphatase-conjugated antibody to the digoxigenin-labelled material. Positive clones are visualized and then picked to be grown in selective media.

In the fourth stage, Southern hybridization and restriction mapping are used to localize the position of the chondroitinase I gene within individual clones. The PCR-generated fragment described above is used as a Southern hybridization probe against P. vulqaris genomic DNA that is first digested by restriction enzymes and fractionated. In a second PCR amplification, several of the oligonucleotides described above are used as primers.

The results indicate that the portion of the chondroitinase I gene that hybridizes to the probe is carried on several large DNA fragments.

These large DNA fragments are digested to yield individual fragments which are isolated, tested for the presence of chondroitinase I sequences by Southern hybridization, and then subcloned into appropriate vectors. Example 3 details the cloning strategy used. Restriction maps are generated to assist in the identification of the portions of the fragments carrying the desired sequences. In addition, in vitro chondroitinase I assays in which the activity of the enzyme based on measuring the release of unsaturated disaccharide from chondroitin sulfate C at 232 nm are conducted on several samples to assist in the placement and orientation of the chondroitinase I gene. The results of these procedures suggest that a 4.2 kb EcoRV-EcoRI fragment of a larger 10 kb NsiI fragment could contain the -1 4 i 7 a i 7 PCTJUS94/O 44 9 WO 94/25567 to the production of chondroitinase I and chondroitinase II in quantities not readily achievable using present non-recombinant bacterial fermentation and extraction techniques.

,i 1< a WO 94/25567 PCT1US94104495 18 entire chondroitinase I gene.

In the fifth stage, the above-mentioned 4.2 kb fragment is subjected to DNA sequence analysis.

The resulting DNA sequence is 3980 nucleotides in length (SEQ ID NO:1). Translation of the DNA sequence into the putative amino acid sequence reveals a continuous open reading frame (SEQ ID NO:1, nucleotides 1l9-3lI.1) e ncoding 1021 amino acids (SEQ ID NO:2).

In turn, analysis of the amino acid sequence reveals a 24 residue signal sequence (SEQ ID NO:2, aminio acids 1-24), followed by a 997 residue mature (processed) chondroitinase I enzyme (SEQ ID NO:2, amino acids 25-1021) Signal sequences are required for a complex series of post-translational processing steps which result in secretion of a protein from a host cell.

The signal sequence constitutes the amino-terminal end of the protein to be secreted. In most cases, the signal sequence is cleaved off by a specific protease, called a signal peptidase, The "18 kf" and "90 lcD" fragments are found to be.adjacent to each other, with the "18 kD" fragment constituting the first 157 amino acids of the mature protein (SEQ ID NO:2, amino acids 25-181), and the 1190 kD" fragment constituting the remaining 840 .Yiino acids of the mature protein (SEQ ID NO:2, amino acids 182-1021).

The chondroitinase I enzyme of this invention is expressed using established recombinant DNA methods. Suitable host organisms include bacteria, viruses, yeast, insect or mammalian cell lines, as well as other conventionZxl organisms. The host cell is transformed with a plasmid containing a purified isolated Dk A fragment encoding for

F;

rT'* A WO 94/25567 PCTIUS94/04495 the three nucleotide. immediately upstream of the initiation codon are RA changed (SEQ ID NO:3, nucleotides 116- I 3 118)) WO 94/25567 PCT/US94/04495 -19 chondroitinase I enzyme. The host cell is then cultured under conditions which permit expression of the enzyme by the host cell.

In the sixth stage, the gene is subjected to site-directed mutagenesis to introduce unique restriction sites. These permit the gene to be moved, in the correct reading frame, into an expression system which results in expression of chondroitinase I enzyme at high levels. Such an appropriate host cell is the bacterium E. coli.

As detailed in Example 6 below, two different constructs are prepared. In the first, the three nucleotides immediately upstream of the initiation codon are changed (SEQ ID NO:3, nucleotides 116-118) through the use of a mutagenic oligonucleotide (SEQ ID NO:37). The coding region and amino acid sequence encoded by the resulting construct are not changed, and the signal sequence is preserved (SEQ ID NO:3, nucleotides 119-3181; SEQ ID NO:2).

In a preferred embodiment of this invention, the second construct is used. In the second construct, the site-directed mutagenesis is carried out at the junction of the signal sequence and the start of the mature protein. A mutagenic oligonucleotide (SEQ ID NO:38) is used which differs at six nucleotides from those of the native sequence (SEQ ID NO:1, nucleotides 185-190). The sequence Si differences result in the deletion of the signal S- sequence, and the addition of a methionine residue at the amino-terminus, resulting in a 998 amino acid protein (SEQ ID NO:4, nucleotides 188-3181; SEQ ID NO: In the absence of a signal sequence, the enzyme is not secreted. Fortunately, it is not retained within the cell in the form of insoluble WO 94/25567 PCTUS94/04495 r WO 94/25567 PCT/US94/04495 20 inclusion bodies. Instead, at least some of the enzyme is produced intracellularly as a soluble active enzyme. The enzyme is extracted by homogenization, which serves to lyse the cells and thereby release the enzyme into the supernatant. Even with the signal sequence present, much of the enzyme is not secreted, because it is thought that this expression system provides such high yields of enzyme that it exceeds the capacity of the host cell to secrete that much enzyme.

As described in Example 7 below, the gene lacking the signal sequence is inserted into an appropriate expression vector. One such vector is pET-9A Novagen, Madison, WI), which is derived from elements of the E. coli bacteriophage T7. The resulting recombinant plasmid is designated pTM49-6.

The plasmid is then used to transform an appropriate expression host cell, such as the E. coli B strain BL21/(DE3)/pLysS (10; Novagen).

Samples of this E. coli B strain BL21(DE3)/pLysS carrying the recombinant plasmid pTM49-6 were deposited by Applicants on February 4, 1993, with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852, and have been assigned ATCC accession number 69234.

Expression of the chondroitinase I enzyme using the deposited host cell yields approximately 300 times the amount of the enzyme as was possible using a ae size fermentation vessel with native (nonilcombinant) P. vulcaris.

After Qxpression of the chondroitinase I enzyme, the supernatant from the host cells is treated to isolate6and urify the enzyme. Initial attempts to isolate and purify the recombinant chondroitinase I enzyme do not result in high yields of purified "l 2,' Av (ii 2' 2 WO 94!255' PCTIUS94/04495 -i i lysing by homogenization the host cells which express the recombinant chondroitinase I enzyme to release the enzyme into the supernatant; i 2 ~nl ii WO 94/25567 PCTIUS94/04495 21 protein. Tie previous method for isolating and purifying native chondroitinase I from fermentation cultures of P. vulgaris is found to be inappropriate for the recombinant material.

The native enzyme is produced by fermentation of a culture of P. vulcaris. The bacterial cells are first recovered from the medium and resuspended in buffer. The cell suspension is then homogenized to lyse the bacterial cells. Then a charged particulate such as Bioacryl (Toso Haas, Philadelphia, PA), is added to remove DNA, aggregates and debris from the homogenization step. Next, the solution is brought to 40% saturation of anmonium sulfate to precipitate out undesired proteins. The chondroitinase I remains in solution.

The solution is hen filtered and the retentate is washed to recover most of the enzyme.

The filtrate is concentrated and subjected to diafiltration with a phosphate to remove the salt.

The filtrate containing the chondroitinase I is subjected to cation exchange chromatography using a cellulose sulfate column. At pH 7.2, 20 mM sodium phosphate, more than 98% of the chondroitinase I binds to the column. The native chondroitinase I is then eluted from the column using a sodium chloride gradient.

The eluted enzyvic' is then subjected to additional chromatography steps, such as anion "'exchange and hydrophobic interaction column chromatography. As a result of all of these procedures, chondroitinase I is obtained at a purity of 90-97%. The level of purity is measured by first performing ,SDS-PAGE. The proteins are stained using Coomassie blue, destained, and the lane on the gel is scanned using a laser beam of wavelength 600 nm. The ft ~i ~J? v(TflCOdmCAAOC 22 "2E /3 purity is expressed as the percentage of the total absorbance accounted for by that band.

However, the yield of the native protein is only 25-35%. The yield is measured as the remaining activity in the final purified product, expressed as a percentage of the activity at the start (which is taken as 100%). In turn, the activity of the enzyme is based on measuring the release of unsaturated disaccharide from chondroitin sulfate C at 232 nm.

This purification method also results in the extensive cleavage of the approximately 110,000 dalton (110 kD) chondroitinase I protein into a 90 kD and an 18 kD fragment. Nonetheless, the two fragments remain non-covalently bound and exhibit chondroitinase I activity.

When this procedure is repeated with homogenate from lysed host cells carrying a recombinant plasmid encoding chondroitinase I, significantly poorer results are obtained. Less than 10% of the chondroitinase I binds to the cation exchange column at standard stringent conditions of pH 7.2, 20 mM sodium phosphate.

Under less stringent binding conditions of pH 6.8 aid 5 mM phosphate, an improvement of binding with one batch of material to 60-90% is observed.

However, elution of the recombinant protein with the NaCI gradient gives a broad activity peak, rather than a sharp peak (see Figure This indicates the product is heterogeneous. Furthermore, in subsequent fermentation batches, the recombinant enzyme binds poorly even using the less stringent bindirg conditions. Most of these batches are not processed to the end, as there is poor binding. Therefore, their overall recovery is not quantified.

Based on these results, it is concluded that i -i i: i i I WO 94/25567 PCTUS94/04495 23 the recombinant chondroitinase I enzyme has a reduced basicity compared to the native enzyme, and that the basicity also varies between batches, as well as within the same batch.

It is evident that the method used to isolate and purify the native enzyme is not appropriate for the recombinant enzyme. The method produces low yields of protein at high cost.

Furthermore, for large batches, large amounts of solvent waste are produced containing large amounts of a nitrogen-containing compound (ammonium sulfate).

This is undesirable from an environmental point of view.

A hypothesis is then developed to explain these poor results and to provide a basis for developing improved isolation and purification methods. It is known that the native chondroitinase I enzyme is basic at neutral pH. It is therefore assumed that the surface of the enzyme has a net excess of positive charges.

Without being bound by this hypothesis, it is believed that, in recombinant expression of the enzyme, the -host.cell contains or produces, small, negatively charged molecules. These negatively charged molecules bind to the enzyme, thereby reducing the number of positive charges on the enzyme. If these negatively charged molecules bind with high enough affinity to copurify with the enzyme, they can cause an alteration of the behavior of the enzyme on the ion exchange column.

Support for this hypothesis is provided by the data described below. In general, cation exchange Sresins b.ind to proteins better at lower pH's than higher pH's. Thus, a protein which is not very basic, and hence does not bind at a high pH, can be made to i ti /il II i II '_14 m rraITTLcnA IAAnE to the column. Theycontain the majority of the "'chondroitinase I enzyme activity. The fractions at riaht containing the enzyme r.re marked "eluted ii acivity". .The radient is from 0.0, to 250 mM NaCI.

Figure 3 /lepIcts the elution of the f lr Vi if Ci o.

WO 94/25567 PCT/US94/04498 24 bind to the cation exchanger by carrying out the operation at a lower pH. At pH 7.2, the native enzyme binds completely to a cation exchange resin. However, the recombinant.-derived enzyme, due to the lowered basicity as a result of binding of the negatively charged molecules, does not bind very well (less than This enzyme can be made to bind up to 70% by using a pH of 6.8 and a lower phosphate concentration mM rather than 20 mM), but heterogeneity and low yield remain great problems. Indeed, only one fermentation results in a 70% binding level; typically, it is much less (less than 10%) even at pH 6.8. This level of bind.ng varies dramatically between different fermentation batches.

This hypothesis and a possible solution to the problem are then tested. If negatively charged molocules are attao'hing non-covalently to chondkoitinase 1, thus decreasing its basicity, it should be possible to remove these undesired molecules by using a strong, high capacity anion exchange resin.

Removal of the negatively charged molecules should then restore the basicity of the enzyme. The enzyme could then be bound to a cation exchange resin and eluted therefrom in pure form at higher yields.

Experiments demonstrate that this approach indeed provides a solution-to the problem encountered with the isolation and purification of the recombinantly expressed chondroitinase I enzyme.

As is discussed below, chondroitinase I is recombinantly expressed in two forms. The enzyme is ekpressed with a signal peptide, which is then cleaved to produce the mature enzyme. The enzyme is also expressed without a,,signal peptide, to produce directly the mature enzyme. The twob embodiments of thisinvention 'which will now be discussed are ii e

II

'i

B

r; i ;i ;n L 1- i ;a a -1 .1 j:1 WO 94/25567 PCTUS94/04495 25 suitable for use in purifying either of these forms of the enzyme.

In the first embodiment of this aspect of the invention, the host cells which express the recombinant chondroitinase I enzyme are lysed by homogenization to release the enzyme into the supernatant. The supernatant is then subjected to diafiltration to remove salts and other small molecules. However, this step only removes the free, but not the bound form of the negatively charged molecules. The bound form of these charged species is next removed by passing the supernatant through a strong, high capacity anion exchange resin-containing column. An example of such a resin is the Macro-Prep

M

High Q resin (Bio-Rad, Melville, Other strong, high capacity anion exchange columns are also suitable. Weak anion exchangers containing a diethylaminoethyl (DEAE) ligand also are suitable, although they are not as effective. Similarly, low capacity resins are also suitable, although they too are not as effective. The negatively charged molecules bind to the column, while the enzyme passes through the column. It is also, found that some unrelated, undesirable proteins also bind to the column.

Next, the eluate from the anion exchange column is tiredtly loaded to a cation exchange resincontaining column. Examples of such resins are the S- Sepharose M (Pharmacia, Piscataway, and the Macro-Prep TM High S (Bio-Rad). Each of these two, resin-containing columns has SO 3 ligands bound thereto in order to facilitate the exchange of cations. Other cation exchange columns are also suitable. The enzyme binds to the column and is then eluted with a solvent capable of releasing the enzyme from the column.

N,

p.

I*

R- C rt i I I I i" WO 94/25567 PCT/US94/04495 -26 Any salt which increases the conductivity of the solution is suitable for elution. Examples of such salts include sodium salts, as well as potassium salts and ammonium salts. An aqueous sodium chloride solution of appropriate concentration is suitable. A gradient, such as 0 to 250 mM sodium chloride is acceptable, as is a step elution using 200 mM sodium chloride.

A sharp peak is seen in the sodium chloride gradient elution (Figure The improvement in enzyme yield over the prior method is striking. The recombinant chondroitinase I enzyme is recovered at a purity of 99% at a yield of 80-90%.

The purity of the protein is measured by scanning the bands in SDS-PAGE gels. A 4-20% gradient of acrylamide is used in the development of the gels.

The band(s) in each lane of the gel is scanned using the procedure described above.

These improvements are related directly to the increase in binding of the enzyme to the cation exchange column which results from first using the anion exchange column. In comparative experiments, when only the cation exchange column is used, only 1% of the enzyme binds to the column. However, when the anion exchange column is used first, over 95% of the enzyme binds to the column.

The high purity and yield obtained with the first embodiment of this invention make it more feasible to manufacture the recombinant chondroitinase I enzyme on a large scale.

In a second embodiment of this aspect of the invention, two additional steps are inserted in the method before the diafiltration step of the first embodiment. The supernatant is treated with an acidic solution to precipitate out the desired enzyme. The 1 i A1 WO 94/25567 PCT/US94/04495 27 pellet is recovered and then dissolved in an alkali solution to again place the enzyme in a basic environment. The solution is then subjected to the diafiltration and subsequent steps of the first embodiment of this invention.

In comparative experiments with the second embodiment of this invention, when only the cation exchange column is used, only 5% of the enzyme binds to tha column. However, when the anion exchange column is used first, essentially 100% of the enzyme binds to the column. The second embodiment provides comparable enzyme purity and yield to the first embodiment of the invention.

Acid precipitation removes proteins that remain soluble; however, these proteins are removed anyway by the cation and anion exchange steps that follow (although smaller columns may be used). An advantage of the acid precipitation step is that the sample volume is decreased to about 20% of the original volume after dissolution, and hence can be handled more easily on a large scale. However, the additional acid precipitation and alkali dissolution steps of the second embodiment mean that the second embodiment is more time consuming than the first embodiment. On a manufacturing scale, the marginal improvements in purity and yield provided by the second embodiment may be outweighed by the simpler procedure of the first embodiment, which still provides highly pure chondroitinase I enzyme at high yields. An additional benefit of the two embodiments of the invention is that cleavage of the enzyme into kD &and 18 kD-lfragments is avoided.

The high purity of the enzyme produced by the two embodiments of this invention is depicted in Figure 4. A single sharp band is seen in the SDS-PAGE

LI

W 94 5 PCT/US94/044 9 'WO 94/25567 WO 94/25567 PCT/US94/04495 28 gel photograph: Lane 1 is the enzyme using the method of the first embodiment; Lane 2 is the enzyme using the method of the second embodiment (Lane 3 represents the supernatant from the host cell prior to purification many other proteins are present; Lane 4 represents molecular weight standards).

The material deposited with the ATCC can also be used in conjunction with the sequences disclosed herein to regenerate the native chondroitinase I gene sequence (SEQ ID NO:1) or the modified chondroitinase I gene sequence which includes the signal sequence (SEQ ID NO:3) using conventional genetic engineering technology.

Production of native chondroitinase I enzyme in P. vulqaris after induction with chondroitin sulfate does not provide a high yield of enzyme; the enzyme represents approximately 0.1% of total protein present. When the recombinant construct with the signal sequence deleted is used in E. coli, approximately 15% of the total protein is the chondroitinase I enzyme.

In addition to the three DNA sequences just described for the chondroitinase I gene (SEQ ID NOS:1, 3 and the present invention further comprises DNA sequences which, by virtue of the redundancy of the genetic code, are biologically equivalent to the sequences which encode for the enzyme, that is, these other DNA sequences are characterized by nucleotide sequences which differ from those set forth herein, but which encode ah enzyme having the same amino acid sequences as those encoded by the DNA sequences set forth herein.

In particular, the invention contemplates those DNA sequences which are sufficiently duplicative of the sequences of SEQ ID NOS:1, 3 or 4 so as to

UL

4~

I

ii i I ri:

:'I

WO 94/25567 PCT/US94/04495 .4 WO 94/25567 PCTUS94/04495 29 permit hybridization therewith under standard high stringency Southern hybridization conditions, such as those described in Sambrook et al. as well as the biologically active enzymes produced thereby.

This invention also comprises DNA sequences which encode amino acid sequences which differ from those of the chondroitinase I enzyme, but which are the biological equivalent to those described for the enzyme (SEQ ID NOS:2 and Such amino acid sequences may be said to be biologically equivalent to those of the enzyme if their sequences differ only by minor deletions from or conservative substitutions to the enzyme sequence, such that the tertiary configurations of the sequences are essentially unchanged from those of the enzyme.

For example, a codon for the amino acid alanine, a hydrophobic amino acid, may be substituted by a codon encoding another less hydrophobic residue, such as glycine, or a more hydrophobic residue, such as valine, leucine, or isoleucine. Similarly, changes which result in substitution of one negatively charged residue for another, such as aspartic acid for glutamic acid, or one positively charged residue for another, such as lysine for arginine, as well as changes based on similarities of residues in their hydropathic index, can also be expected to produce a biologically equivalent product. Nucleotide changes which result in alteration of the N-terminal or Cterminal portions of the protein molecule would also not be expected to alter the activity of the protein.

Each of the proposed modifications is well within the routine skill in the art, as is determination of retention of biological activity of the encoded pr ducts. Therefore, where the terms "chondroitinase I gene" or "chondroitinase I enzyme" are used in i i

U

r) r r 1 I ~mM I Ivski i h B I^

I

WO 94/25567 PCTJS94/04495 30 either the specification or the claims, each will be understood to encompass all such modifications and variations which result in the production of a biologically equivalent protein.

The starting point for the cloning and expression of the chondroitinase II gene is partial amino acid sequencing of the mature native chondroitinase II protein obtained from P. vulgaris.

The N-terminal sequence of the mature native chondroitinase II protein is found to include the following 22 amino acids: Leu-Pro-Thr-Leu-Ser-His-Glu-Ala-Phe-Gly-Asp-Ile-Tyr- Leu-Phe-Glu-Gly-Glu-Leu-Pro-Asn-Thr (SEQ ID NO: amino acids 1-22) The nucleotide sequence determined above for the region encoding the chondroitinase I gene includes an additional approximately 800 base pairs beyond the translation termination codon (SEQ ID NOS:1 and 39, nucleotides 3185-3980). An inspection of this region reveals that the sequence between nucleotides 3307 and 3372 (SEQ ID NOS:1 and 39) encodes the identical 22 amino acids in the same order as the first 22 amino acids of native chondroitinase II.

Furthermore, an ATG initiation codon (SEQ ID NOS:1 and 39, nucleotides 3238-3240) is found upstream of this region and in-frame, indicating that this gene is expressed with a 23 amino acid signal peptide sequence for the export of chondroitinase II' (SEQ ID amino acids 1-23). Although a Shine-Dalgarno sequence (AGGA; SEQ ID NOS:1 and 39, nucleotides 3225- 3228) is found upstream of the initiation codon, there is no apparent promoter sequence, suggesting that both the 110 kD and 112 kD forms of the P. vulqaris ZII l--_C

;:I

i 1 4 i; i WO 94/25567 PCTUS94/04495 31 chondroitinase enzyme are expressed as part of a single messenger RNA.

The coding sequence that starts with this ATG was originally not found to be continuous in SEQ ID NO:1, since a termination codon (TAA) was thought to be present in-frame at base-pairs identified as 3607-3609. Re-examination of the sequencing data, however, revealed that a residue was overlooked and that a T should be inserted between nucleotides originally identified as 3593 and 3594. This change restores the open reading frame which then extends through the end of SEQ ID NO: 39 (SEQ ID NOS:1 and 39 include the inserted T as nucleotide 3594). (Thus, the three bases TAA at base-pairs 3608-3610, properly numbered, do not constitute a termination codon.) With this information available, the cloning and expression of the P. vulgaris chondroitinase II gene is performed in three stages. In the first stage, because the N-terminal sequences are known, a site-specific mutagenesis is carried out. This is necessary in order for this gene to be placed, eventually, directly into the desired T7-based expression vector pET9A that is used (as described above) for the chondroitinase I gene. The mutagenized bases are upstream of the coding region (an AT sequence (SEQ ID NOS:1 and 39, base pairs 3235 and 3236) is replaced by a CA sequence).

The second stage, which can be carried out in parallel with the first, involves the identification, isolation and DNA sequencing of an appropriate DNA fragment which will include the Cterminal coding region of the chondroitinase II gene.

The available DNA sequence information is adequate to account for approximately 220 amino acids of an estimated 1000 for the entire chondroitinase II dii

I

'1 ;i n J

I

I,

WO 94/25567 PCT/US94/04495 32 protein. The missing coding sequences, therefore, would extend for another 2400 base pairs beyond the end of SEQ ID NO: 1.

The third stage involves the assembly of an intact gene for chondroitinase II that has been modified to include the initiation codon as part of an NdeI site and to be followed by a BamHI site downstream of the coding region. This allows a directed insertion of this gene into the pET9A expression vector (Novagen, Madison, WI) without further modification.

Sequencing of the entire assembled gene confirms the presence of the initiation codon at nucleotides 3238-3240, where this codon represents the start of the region coding for the signal peptides at nucleotides 3238-3306, the region coding for the mature protein at nucleotides 3307-6276, and a termination codon at nucleotides 6277-6279 (SEQ ID NO:39). Thr translation of this sequence results in 1013 amino acids, of which the first 23 amino acids are the signal peptide and 990 amino acids constitute the mature chondroitinass II protein at residues numbered 24-1013 (SEQ ID In this construction, the signal peptide is retained, such that the expressed gene is processed and secreted to yield the mature native enzyme structure that has a leucine residue at the N-terminus.

As described in Example 13 below, the gene encoding the chondroitinase II protein is inserted into pET9A and the resulting recombinant plasmid is designated LP 2 1359. The plasmid is then used to transform an appropriate expression host cell, such as the E. coli B strain BL21(DE3)/pLysS (which is also used for the expression of the chondroitinase I gene.

Samples of this E. coli B strain designated I:lp rr: ct C n -Thi i WO 94/25567 PCTIUS94/04495 33 TD112, which is BL21(DE3)/pLysS carrying the recombinant plasmid LP 2 1359, were deposited by Applicants on April 6, 1994, with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852, and have been assigned ATCC accession number 69598.

Expression of the chondroitinase II enzyme using the deposited host cell yields approximately times the amount of the enzyme as was possible using a same size fermentation vessel with native (nonrecombinant) P. vulcaris.

After expression of the enzyme, the supernatant from the host cells is treated to isolate and purify the enzyme. Because of the virtually identical isoelectric points and similar molecular weights for the two proteins, the first method described above for isolating and purifying the recombinant chondroitinase I protein is adapted for isolating and purifying the recombinant chondroitinase II protein, and then modified as will now be descr'ibed.

The need for the modification of the method is based on the fact that the recombinant chondroitinase II protein is expressed at levels approximately several-fold lower than the recombinant chondroitinase I protein; therefore, a more powerful and selective solution is necessary in order to obtain a final chondroitinase II product of a purity equivalent to that obtained for the chondroitinase I protein.

The first several steps of the method for \[the chondroitinase II protein are the same as those used to isolate and purify the chondroitinase I protein. Initially, the host cells which express the recombinant chondroitinase II enzyme are lysed by 1; y' I I jl W I; I I -r CssLT I IB i 4: 'Wn GIAI799'7, PrTII.qOd/9dd; r 1. 1 fl J- 'WO 94/2556'/ PCT/US94/04495 34 homogenization to release the enzyme into the supernatant. The supernatant is then subjected to diafiltration to remove salts and other small molecules. However, this step only removes the free, but not the bound form of the negatively charged molecules. The bound form of these charged species is next removed by passing the supernatant through a strong, high capacity anion exchange resin-containing column. An example of such a resin is the Macro-Prep" High Q resin (Bio-Rad, Melville, Other strong, high capacity anion exchange columns are also suitable. Weak anion exchangers containing a diethylaminoethyl (DEAE) ligand also are suitable, although they are not as effective. Similarly, low capacity resins are also suitable, although they too are not as effective. The negatively charged molecules bind to the column, while the enzyme passes through the column. It is also found that some unrelated, undesirable proteins also bind to the column.

Next, the eluate from the anion exchange column is directly loaded to a cation exchange resincontaining column. Examples of such resins are the S- Sepharose" (Pharmacia, Piscataway, and the Macro-Prep™ High S (Bio-Rad). Each of these two resin-containing columns has SO03 ligands bound thereto in order to facilitate the exchange of cations. Other cation exchange columns are also suitable. The enzyme binds to the column, while a significant portion of contaminating proteins elute unbound.

At, this point, the method diverges from that used for the chondroitinase I protein. Instead of eluting the protein with a a non-specific salt solution capable of releasing the enzyme from the cation exchange column, a specific elution using a jt~ i

J

WO 94/25567 PCT/US94/04495 35 solution containing chondroitin sulfate is used.

This procedure utilizes the affinity the positively charged chondroitinase II protein has for the negatively charged chondroitin sulfate. The affinity is larger than that accounted for by a simple positive and negative interaction alone. It is an enzyme-substrate interaction, which is similar to other specific biological interactions of high affinity, such as antigen-antibody, ligand-receptor, co-factor-protein and inhibitor/activator-protein.

Hence, the chondroitin sulfate is able to elute the enzyme from the negatively charged resin. In contrast, the resin-enzyme interaction is a simple positive and negative interaction.

Although affinity elution chromatography is as easy to practice as ion-exchange chromatography, the elution is specific, unlike salt elution. Thus, it has the advantages of both affinity chromatography (rpecificity), as well as ion-exchange chromatography (low cost, ease of operation, reusability).

Another advantage is the low conductivity of the eluent (approximately 5% of that of the salt eluent), which allows for further ion-exchange chromatography without a diafiltration/dialysis step, which is required when a salt is used. Note, that this is not a consideration in the method for the chondroitinase I protein, because no further ionexchange chromatography is needed in order to obtain the purified chondroitinase I protein.

There is another reason for not using the method for purifying recombinant chondroitinase I.

Chondroitinase II obtained usiiig the chondroitinase I salt elution purification method has poor stability; there is extensive degradation at 4 C within one week.

In contrast, chondroitinase II obtained by affinity F I i -1 1-1 1 i WO 94/25567 PCT/US94/04495 36 elution is stable. The reason for this difference in stability is not known. It is to be noted that chondroitinase I obtained by salt elution is stable.

The cation exchange column is next washed with a phosphate buffer to elute unbound proteins, followed by washing with borate buffer to elute loosely bound contaminating proteins and to increase the pH of the resn to that required for the optimal elution of the chondroitinase II protein using the substrate, chondroitin sulfate.

Next, a solution of chondroitin sulfate in water, adjusted to pH 9.0, is used to elute the chondroitinase II protein, as a. sharp peak (recovery and at a high purity of approximately 95%. A 1% concentration of chondroitin sulfate is used. A gradient of this solvent is also acceptable.

Because the chondroitin sulfate has an affinity for the chondroitinase II protein which is stronger than its affinity for the resin of the column, the chondroitin sulfate co-elutes with the protein. This ensures that only protein which recognizes chondroitin sulfate is eluted, which is desirable, but also means that an additional process step is necessary to separate the chondroitin sulfate from the chondroitinase II protein.

In this separation step, the eluate is adjusted to a neutral pH and is loaded as is onto an anion exchange resin-containing column, such as the Macro-Prep M High Q resin. The column is washed with a phosphate buffer. The chondroitin sulfate binds to the column,.while the chondroitinase II protein flows through in the unbound pool with greater than recovery. At this point, the protein is pure, except for the presence of a single minor contaminant of approximately 37 kD. The contaminant may e a

SI

4i r, j I r-r i I~

I

4 1 1 1 j 1 a

J-^

SWO 94/25567 PCT/US94/04495 37 breakdown product of the chondroitinase II protein.

This contaminant is effectively removed by a crytallization step. The eluate from the anion exchange column is concentrated and the solution is maintained at a reduced temperature, such as 4°C, for several days to crystallize out the pure chondroitinase II protein. The supernatant contains the 37 kD contaminant. Centrifugation causes the crystals to form a pellet, while the supernatant with the 37 kD contaminant is removed by pipetting. The crystals are then washed with water. The washed crystals are composed of the chondroitinase II protein at a purity of greater than 99%.

In a second embodiment of this aspect of the invention for the chondroitinase II protein, two additional steps are inserted in the method before the diafiltration step of the first embodiment. The supernatant is treated with an acidic solution to precipitate out the desired enzyme. The pellet is recovered and then dissolved in an alkali solution to again place the enzyme in a basic environment. The solution is then subjected to the diafiltration and subsequent sa ps of the first embodiment of this invention.

Acid precipitation removes proteins that remain soluble; however, these proteins are removed anyway by the cation and anion exchange steps that follow (although smaller columns may be used). An advantage of the acid precipitation step is that the sample volume is decreased compared to the original volume after dissolution, and hence can be handled more easily on a large scale. However, the additional acid precipitation and alkali dissolution steps of the second embodiment mean that the second embodiment is morer e consumiing than the first embodiment. On a I Z! in

I

expressed without a signal peptide, to produce directly the mature enzyme. The two embodiments of this invention which will now be discussed are Yi j 'WO 94/25567 PCT/US94/04495 38 manufacturing scale, tne marginal improvements in purity and yield provided by the second embodiment may be outweighed by the simpler procedure of the first *embodiment, which still provides highly pure chondroitinase II enzyme at high yields.

Production of native chondroitinase II enzyme in P. vulgaria after induction with chondroitin sulfate does not provide a high yield of enzyme; the enzyme represents approximately 0.1% of total protein present. When the recombinant construct is used in E.

coli, approximately 2.5% of the total protein is the chondroitinase II enzyme.

In addition to the DNA sequence just described for the chondroitinase II gene (SEQ ID NO:39), the present invention further comprises DNA sequences which, by virtue of the redundancy of the genetic code, are biologically equivalent to the sequences which encode for the enzyme, that is, these other DNA sequences are characterized by nucleotide sequences which differ from those set torth herein, but which encode an enzyme having the same amino acid sequences as those encoded by the DNA sequences set forth herein.

In particular, the invention contemplates those DNA sequences which are sufficiently duplicative of the sequence of SEQ ID NO:39 so as to permit hybridization therewith under standard high stringency Southern hybridization conditions, such as those described in Sambrook et al. as well as the biologically active enzymes produced thereby.

This invention also comprises DNA sequences Swhich encode amino acid sequences which differ from those of the chondroitinase II enzyme, but which are the biological equivalent to these described for the enzyme (SEQ ID NO:40). Such amino acj sequences may i e iIt

'I,

i T -r 0 I ,1

C

WO 9425567 'WO 94/25567 PCT/US94/04495 39 be said to be biologically equivalent to those of the enzyme if their sequences differ only by minor deletions from or conservative substitutions to the enzyme sequence, 3uch that the tertiary configurations of the sequences are essentially unchanged from those of the enzyme.

Bach of the proposed modifications is well within the routine skill in the art, as is determination of retention of biological activity of the encoded products. Therefire, where the terms "chondroitinase II gene" or "chondroitinase II enzyme" are used in either the specification or the claims, each will be understood to encompass all such modifications and variations which result in the production of a biologically equivalent protein.

If desired, one of ordinary skill in the art can ligate together the two pieces of DNA from the two deposits, for example, at the HindIII site at nucleotide 3326, so as to express both the chondroitinase I and chondroitinase II proteins under PT i .4

I

'WO 94/25567 PCT/US94/04495 40 the control of the T7 promoter upstream of the coding sequence for chondroitinase I.

In order that this invention may be better understood, the following examples are set forth. The examples are for the purpose of illustration only and are not to be construed as limiting the scope of the invention.

Examples Standard molecular biology techniques are utilized according to the protocols described in Sambrook et al. (11).

Example 1 Isolation Of P. vulgaris Genomic DNA And Construction Of A Cosmid Bank In E. coli Two 35 ml aliquots (designated A and B) of a P. vulqaris large-scale (1000 liter) fermentation are obtained and centrifuged. Both pellets are resuspended with 7 ml of 0.05M glucose-0.025M Tris- HC1-0.01M EDTA (pH 8) containing 4 mg/ml of egg-white lysozyme. After 30 minutes of incubation at 37 0 7 ml of 1% SDS-0.16M EDTA-0.02M NaC1 (pH 8) are added to sample and incubation is continued at 37 0 C for another hour.

After the initial lysozyme treatment, sample is centrifuged and the cell pellet taken up with 7 ml of 0.05M glucose-0.025M Tris-HC1-0.01M EDTA (pH 8) containing 40 Ag/ml of DNAase-free RNAase and then 7 ml of 1% SDS-0.16M EDTA-0.02M NaCl (pH 8) are added to this resuspended material. Finally, proteinase K (Boehringer Mannheim, In&danapolis, IN) is added to both samples to a final concentration of 100 Ag/ml and '0 in'

I

h WO 94/25567 PCT/US94/04495 41 f C3 incubation is continued overnight a.t 37°C.

The next day, the samples are extracted once with an equal volume (14 ml) of equilibrated phenol followed by two further extractions in which the samples are extracted with 7 ml of phenol followed by the addition of 7 ml of chloroform, continued shaking and finally, centrifugation to separate the two phases. The DNA is precipitated by adding one-quarter volume of 5M ammonium acetate and 0.6 volumes of isopropanol followed by centrifugation. The pelleted DNA is rinsed once with 70% ethanol, dried under vacuum and then resuspended with 1 ml of TE (0.01M Tris-HCl-0.0OlM EDTA, pH The nucleic acid concentration for sample is 1.2 mg/ml while that for sample is 1.3 mg/ml, as determined by their ultraviolet absorption at 260 nm.

Fragmentation of the genomic DNA to yield pieces of a size suitable for insertion into cosmid vectors (approximately 25-35 kilobases is accomplished by partial digestion with the restriction endonuclease Sau3A. Duplicate 0.2 ml reactions are set up (one with preparation and the other with DNA from preparation each containing .100 Ag of the P. vulgaris genomic DNA, 0.1M NaCl, 0.01M MgC.

2 0.01M Tris-HCl (pH 7.5) and 80 units of the enzyme Sau3A.

Incubation is carried out at 37°C and 25 Al aliquots are rer at appropriate time points (5,6,7,8,9,10,11 and 20 minutes) and added to 25 Al of 0.2M EDTA (pH The individual samples are heated to 70°C and then 10 Al are removed for a sizedistribution analysis on an agarose gel. The sample obtained after five minutes of Sau3A digestion of preparation and that obtained after 6 minutes with preparation are chosen for further use.

I,

WO 94/25567 PCT/US94/04495 i r- i ,r f i WO 94/25567 PCT/US94/04495 42 In each case, an aliquot (4 gl, which is approximately equal to 2 gg) of the chosen partial digest is ligated to the appropriate "left" and "right" arms of the cosmid vector DNA using approximately 1 pg and 2 Ag of each, respectively, in Al reactions containing 0.066M Tris-HCl (pH 7.4), 0.01M MgC12, 0.001M ATP, and 400 units (as defined by the manufacturer (New England Biolabs, Beverly, MA)) of T4 DNA ligase. Incubation is carried out at 11 0

C

overnight. The "left" and "right" arms of the cosmids are DNA fragments which, when ligated to an appropriately sized piece of P. vulqaris DNA, comprise a recombinant molecule of approximately 35-50 kb.

Both arms contain "cos" sites which are recognized by the packaging enzymes in the next step. In addition, these arms carry the origin of replication and ampicillin-resistance functions of pIBI24 (International Biochemical Inc., New Haven, CT).

Each of the above ligase reactions is added to one tube of a X packaging extract (Packagene", a trademark of Promega Corp., Madison, WI) and the reaction is allowed to proceed at room temperature for two hours, at which point 0.5 ml of PDB (0.1M NaCl- 0.01M Tris-HCl (pH 7.9)-0.01M MgSO 4 is added followed by approximately 0.05 ml of chloroform. Each tube of packaged DNA is, therefore, a gene bank of the P.

vulQaris genome.

Because this method of construction creates a pool of infectious particles X phage heads filled with the cosmid vector joined to approximately to 35 kb of P. vulqaris DNA), the number of potential clones is quantitated by adsorbing an aliquot of the packaged material to an appropriate, sensitive E. coli host strain, and then after outgrowth, plating the mixture on selective media.

i r 0

I?

-w -L L P O r" a

I'

SWO 94/25567 PCT/US94/04495 WO 94/25567 PCT/US94/04495 43 For example, an overnight culture of the E.

coli strain ER1562 (New England Biolabs, Beverly, MA) grown in 20-10-5 medium is diluted 1:20 into fresh media (20-10-5 supplemented with 1% maltose) and grown for three hours at 37 0 C. The cells (1 ml) are then centrifuged, resuspended with PDB (0.2 ml) and 0.02 ml of the appropriate gene bank added. After adsorption for twenty minutes at 37 0 C, the samples are diluted to 2 ml with 20-10-5 medium and grown at 37 0 C for minutes. The culture is then spread on 20-10-5 plates containing 100 Ag/ml of ampicillin and colonies scored after overnight incubation at 37 0 C. The results indicate that there are approximately 68,000 and 95,000 infectious particles (potential cosmid clones) present in the two samples, designated PV1-GB and PV2- GB, corresponding to the and preparation of P.

vulgaris genomic DNA, respectively.

In addition, four other P. vulgaris gene banks are prepared, as above, using two different cosmid vectors. These two cosmids differ from the above-mentioned vectors in that a kanamycin resistance determinant is used in one case rather than the ampicillin resistance, while in the other, the replication functions of paR322 (New England Biolabs, Beverly, MA) are used instead of those of pIBI24.

These four "libraries," designated L1974, L1975, L1976, and L1977, contain, respectively, approximately 18,000 34,000 (ampr), 13,000 (kan') and 15,000 (kanrt members. Aliquots of each of these six gene banks are screened for the presence of the P. vulqaris chondroitinase I gene (see below).

I i i a.

'WO 94/25567 PCT/US94/04495 44 Example 2 PCR Experimentation Designed To Yield An Authentic Piece Of The Chondroitinase I Gene For Use As A Hybridization Probe The Polymerase Chain Reaction (PCR) allows the geometric amplification of a DNA sequence that lies between oligonucleotide primers that can be extended by a DNA polymerase in vitro. The enzyme used in these experiments is the Tag DNA polymerase (isolated originally from Thermus acuaticus), which is preferred because of its thermotolerance which allows it to survive the repeated DNA denaturation steps that are carried out at 94 0

C.

In order for this method to be employed successfully, the oligonucleotides used must have sequences that are as close as possible to those of the target sequence the P. vulgaris chondroitinase I gene. An approximation of that sequence can be derived from the limited available amino acid sequence data. To minimize uncertainty in the sequence presented by the degeneracy of the genetic code (a given amino acid can be encoded by up to six codons), the first approximation involves choosing an amino acid sequence that has the least degeneracy. For example, in the amino-terminal sequence of the P.

vulgaris chondroitinase I gene, there are the following consecutive amino acids: His.-Phe-Ala-Gln- Asn-Asn-Pro (SEQ ID NO:2, amino acids 43-49).

This amino acid sequence could be encoded by any one of 512 different nuclEotide sequences, represented as 5'-CAY-TTY-GCN-CAR-AAY-AAY-CCN-3' (SEQ ID NO: f)-i where R stands for purine (A or Y for pyrimidine (C or and N indicates that any one of the four nucleotides (A T, G, or C) at this position 1; v i i 1 I; i- -I l-:ii -e i

V

'WO 94/25567 pCTUS9/044 9

I

1 'WO 94/25567 PCT/US94/04495 45 will constitute a nucleotide sequence that could encode the indicated amino acid sequence. One possible approach would be to synthesize an ,oligonucleotide mixture containing a total of 512 different olignucleotides, represented as: (GATC)-3' (SEQ ID NO:6).

Although use of such mixtures in PCR has been successful, another approach is to use a number of oligonucleotide mixtures, each of which is made up of a relatively smaller set of nucleotide sequences.

In order to simplify this further, advantage is taken of the observation that mismatched nucleotides in PCP primers are of less consequence at the 5'-end of the primer than they are at the 3'-end. Using these criteria, a set of eight oligonucleotides (each made up of four unique sequences) is designed, where the individual sets of oligonucleotides have the following sequences: 5'-CAC-TTC-GC(GATC)-CAA-AAT-AAT-CC-3' 5'-CAC-TTC-rC(GATC)-CAA-AAC-AAC-CC-3' 5'-CAC-TTC-G"C(GATC)-CAA-AAC-AAT-CC-3' C(GATC)-CAA-AAT-AAC-CC-3' 5'-CAC-TTC-GC(GATC)-CAG-AAT-AAT-CC-3' 5'-CAC-TTC-GC(GATC)-CAG-AAC-AAC-CC-3' 5'-CAC-TTC-GC(GATC)-CAG-AAC-AAT-CC-3' 5'-CAC-TTC-GC(GATC)-CAG-AAT-AAC-CC-3'

(SEQ

NO: 7) NO: 8) NO:9) NO: NO: 11) NO:12) NO:13) NO:14) One of these pools is perfectly matched for the first eleven nucleotides (counting from the 3end), and, furthermore, within this pool of four oligonucleotides, one is a perfect match for the first i i; rt

I

i ~T 1 I--L _1_1-I f FROs7I wo 94125567 PCTIUS94/04495 the E. coli B strain BL21(DE3)/pLysS (which is also j used for the expression of the chondroitinase I gene.

Samples of this Samples of this E. coli B strain designated WO 94/25567 PCTUS94/04495 46 fourteen nucleotides. This is important because it permits stringent annealing conditions to be used that discriminate against imperfect matches that give rise to PCR products that are unrelated to the chondroitinase I gene.

A further aid in the design of oligonucleotides to be used in these PCR experiments is derived from the observation that the P. vulqaris 110 kD chodroitinase enzyme appears to have a structure that leaves one particular region hypersensitive to proteolytic cleavage. The result of this hydrolysis is that the normally approximately 110 kD protein is split into two predominant species of 18 kD and approximately 90 kD. The amino-terminal sequences of the "110 kD" protein and the "18 kD" fragment are the same, while that for the "90 kD" has been found to be different.

The "18 kD" peptide is further fragmented by treatment with cyanogen bromide and trypsin and the resulting oligopeptides sequenced, affording still more information with which to design oligonucleotides for PCR. This information from the "18 kD" and kD"' regions is also valuable because the locations of these amino acidsequences relative to each other and the N-terminal sequences of the intact protein are well defined. In fact, the nucleotide distance between the regions encoding the N-termini of the "110 SkD" and "90 kD" entities can be predicted to be approximately 400-500 bp.

Two further sets of oligonucleotide pools are then designed with one further consideration: The i first eight oligonucleotides hybridize to one strand of the DNA and, during the in vitro DNA synthesis, they are extended toward the "90 kD" N-terminal coding

S

0 35 sequences. Consequently, the oligonucleotides 'WO 94/25567 PCT/US94/04495 47 corresponding to amino acid sequences from within the "18 kD" peptide and at the N-terminus of the "90 kD" peptide must be designed so that they anneal to the complementary DNA strand of the P. vulqaris genome, so that they extend, in vitro, toward the region encoding the N-terminus of the intact protein.

In this way, the oligonucleotides effectively "bracket" the region of the P. vulqaris chromosome that encodes the N-terminal region of the chondroitinase I gene. It is worth noting that the PCR methodology offers an extremely large potential amplification of this bracketed region. Thirty PCR cycles, in theory, increase the number of copies of this DNA segment by a factor of one billion. This allows the use of very small quantities of P. vulqaris genomic DNA as a template which will yield, potentially, microgram amounts of synthesized product which can be readily visualized, isolated and cloned.

Using the above logic, oligonucleotide mixtures are designed based on the following amino acid sequence that is found within the "18 kD" peptide: Glu-Ala-Gln-Ala-Gly-Phe-Lys (SEQ ID NO:2, amino acids 138-144). This heptapeptide is encoded by the following nucleotide sequences: S 5'-GAR-GCN-CAR-GCN-GGN-TTY-AAR-3' (SEQ ID The complementary strand, therefore, has the following A sequences: 5'-YTT-RAA-CC-NGC-YTG-NGC-YTC-3' which is the same as A 5' -(CT)TT- (AG)AA- (GATC)CC- (GATC)GC-(CT)TG-(GATC)GC- (CT)TC-3' (SEQ ID NO:16) I 35 Using the same criteria as described above 1 •!j 770 WO 94/25567 PCTIUS94/04495 -48for the first set of eight oligonucleotides, a further set of eight oligonucleotides (each made up of 16 unique sequences) is designed, where the individual sets of oligonucleotides have the foillowing sequences: 9. 5'-TT-GAA- (AG)CC-(GATC)GC- (CT)TG-GGC-TTC-3' (SEQ ID NO:17) 5'-TT-GAA-(AG)CC-(GATC)GC-(CT)TG-AGC-TTC-3' (SEQ ID NO:18) 11. 5'-TT-GAA- (AG)CC- (GATC)GC- (CT)TG-TGC-TTC-3' (SEQ ID NO:19) 12. 5'-TT-GAA- (AG) CC- (GATC)GC- (CT)TG-CGC-TTC-3' (SEQ ID 13. 5'-TT-GAA- (AG)CC-(GATC)GC- (CT)TG-GGC-CTC-3' (SEQ ID NO:21) 14. 5'-TT-GAA- (AG)CC- (GATC)GC- (CT)TG-AGC-CTC-3' (SEQ ID NO:22) 5'-TT-GAA- (AG)CC- (GATC)GC- (CT)TG-TGC-CTC-3' (SEQ ID NO:23) 16. 5' -TT-GAA- (AG)CC-(GATC)GC- (CT)TG-CGC-CTC-3' (SEQ ID NO:24) Unlike oligonucleotides 1-8 above, one base is deleted from the 5' end of oligonucleotides 9-16 in order to reduce the number of sequence permutations.

in this case, one pool has a perfect matchfor the first eight nucleotides at the 3'-end, while of this same pool has an eleven-nucleotide perfect match with the genomic DNA of P. vu1garis encoding' chondroitinase I.

For a third set of oligonucleotide mixtures, the following amino acid sequence, obt; ined as part of the N-terminal amino acid sequence of the "190 lcD" peptide, is used: Gl-l-y-a-s-*r(SEQ ID NO:2, amino acikds 189-194). This hexapeptide can-be Wfl QdI74:gA7 49 encoded by the following nucleotide sequences: -GGN-GCN-AAR-GTN-GAY-TCN-3' (SEQ ID or -GG -N-GCN-AAR-GTN-GAY-AGY-3' (SEQ ID NO:26) The complement of this sequence is: -NGA-RTC-NAC-YTT-NGC-NCC-3' (SEQ ID NO:27) or 5'-RCT-RTC-NAC-YTT-NGC-NCC-3' (SEQ ID NO:28) These possible sequences are represented using the following oligonucleotide mixtures: 17. 5'-GA-GTC-(GATC)AC-(TC)TT- (AG)GC-GCC-3' (SEQ ID NO:29) 18. 5'-GA-GTC-(GATC)AC- (TC)TT- (AG)GC-AcC-3' (SEQ ID 19. 5'-GA-GTC- (GATC)AC- (TC)TT- (AG)GC-TCC-3' (SEQ ID NO:31) 5'-GA-GTC- (GATC)AC- (TC)TT- (AG)GC-CCC-3' (SEQ ID NO:32) 21. 5'-GA-GTC- (GATC)AC- (TC)TT- (TC)GC-GCC-3' (SEQ ID NO:33) 22. 5'-GA-GTC-(GATC)AC-(TC)TT-(TC)GC-ACC-3' (SEQ ID NO:34) 23. 5'-GA-GTC-(GATC)AC-(TC)TT- (TC)GC-TCC-3' (SEQ ID 24. 5' -GA-GTC-(GATC)AC-(TC)TT-(TC)GC-CCC-3' (SEQ ID NO:36) Unlike ol igonucl eo tides 1-8 above, one base is deleted from the 5' end of oligonucleotides 17-24 in order to reduce the number of sequence hi 1.

i WO 94/25567 PCTUJS94/04495 'WO 94/25567 PCTUS94/04495 50 permutations.

In this case, one oligonucleotide mixture has half of its members perfectly matched for the first eight nucleotides at the 3'-end, and one quarter of the oligonucleotides in the pool are perfectly matched for eleven nucleotides at the 3'-end.

These twenty-four oligonucleotide mixtures are purchased from Biosynthesis, Inc. (Denton, TX), and are provided as fully deprotected, purified and lyophilized samples. In each case (except oligonucleotide 5 O.D. units of synthetic DNA are obtained. This is resuspended in 0.5 ml of water to yield a solution that contains approximately 50-60 pmoles of oligonucleotide per microliter. The remaining sample (oligonucleotide #20) contains O.D. and is resuspended with one ml of water to give a solution with approximately 90 pmole/pl.

A typical 50 tl PCR reaction contains approximately 20 ng of P. vulgaris genomic DNA as template; 200 AM each of dATP, dGTP, dCTP, dTTP; KC1; 10mM Tris-HCl (pH 1.5 mM MgC12; 0.01% gelatin; 2.5 units of Ampli-Taq" DNA polymerase (PerkizElmer/Cetus, Norwalk, CT); and 50 pmoles of each oligonucleotide pool to be tested. The reactions are overlaid with mineral oil (Plough) and incubated in a Perkin-Elmer/Cetus Thermalcycler.

SFor each cycle, the inst'rment is programmed to denature the template DNA at 94 0 C f£ 1.25 minutes, anneal the oligonucleotide primers to ta- denatured template at 60 0 C or 62 0 C for one minute, and to extend i these primers via DNA synthesis at 72°t for 2.25 minutes. Thirty such cycles are carri d out in an experimental amplification. The prodults are analyzed !j 35 by running an aliquot on a 4% NuSieveT 0 (FMC Biochemicals, Rockland, ME) GTG gel calitaining if J' I. 1 WO 94/25567 PCT/US94/04495 51 approximately 0.5 g/ml ethidium bromide using either Tris-borate or Tris-acetate buffers at either full or half strength. These gels are usually run overnight at approximately 1V/cm and photographed on a long wavelength UV transilluminator using a red filter and Polaroid Type 57 film.

PCR experiments are run testing the pairwise combinations between oligonucleotide pools #1-8 (derived from the "110 kD" amino-terminal sequence of chondroitinase pools #9-16 (derived from a peptide sequence contained within the "18 kD" fragment), and pools #17-24 (derived from the amino-terminal sequence of the "90 kD" fragment). The most effective amplifications observed (based on the visual yield of a discrete DNA band detected on gel electrophoretic analysis of the reaction products) are between oligonucleotide pools #4 and #18, and pools, #4 and #9,10,11, or 12. In general, the other pools, which differ by one nucleotide from these pools, also yield some amplification. A difference of two nucleotides results, essentially, in no observed product. It is important to note, however, that the annealing temperatures are deliberately kept at 60-62 0 C to enhance such discrimination.

PCR amplifications using oligonucletide pools #4 and #18 yield a product of approximately 500 bp as estimated relative to size standards (pBR322 digested with MSP-1 (New England Biolabs, Beverly, MA) ranging from 30 to 700 bp on NuSieve T agarose gels.

The product from the use of oligonucleotide pool #4 combined with pools 10, 11, or 12 is approximately 350 bp in length. Furthermore, the larger product could be isolated from an agarose gel, diluted a thousand-fold, and then used as the template in a second PCR reaction employing oligonucleotide pools #4 'I 'i

;A

41- WO 94/25567 2CT/US94/04495 52 and #9 as primers, which yield a product of approximately 350 bp. That is, the smaller PCR product is synthesized from the larger one in .agreement with what would be expected if these sequences were all derived from the P. vulqaris chondroitinase I gene. This indicates that the desired region of the genome is amplified.

The larger PCR product is isolated from an agarose gel using a QiaexM extraction procedure according to the manufacturer's instructions (Qiagen, Chatsworth, CA). The isolated DNA is then subjected to a "fill-in" reaction (11) to remove the extra, protruding adenine residue that Tag DNA polymerase tends to add to the 3'-end of DNA in a teniplateindependent reaction The isolated DNA is then treated with T 4 polynucleotide kinase to add a phosphate moiety to the 5'-ends of the PCR products to allow them to be joined to the vector DNA. After these treatments, the PCR product is ligated to pIBI24, a high copy vector containing a polylinker (IBI, New Haven, CT), that is first sequentially digested with PstI, "filled-in" and then treated with calf intestinal alkaline phosphatase (Boehringer- Mannheim).

Once the PCR product is cloned into pIBI24, it is removed as an EcoRI-HindIII fragment by virtue of the restriction sites within the polylinker carried by the plasmid. This fragment is then cloned into *both M13mpl8 and M13mpl9 (13; New England Biolabs, Beverly, MA) after cleavage with both EcoRI and HindIII and then phosphatased. Single stranded DNA corresponding to these constructions is then isolated and subjected to DNA sequence analysis usino aan A 1 oster .Cty, CA) n n Applied Biosytems poster City, CA) instriint an S35 Tg sequencing kit. The results indicate that thil .9

I

I___A

WO 94125567 P T PCT/US94/04495 53 larger PCR product is 455 bp in length. As expected, the ends of the fragment are derived from the oligonucleotide pools used as primers.

The DNA sequence is translated into an uninterrupted amino acid sequence that is in agreement (with one exception described below) with the available data obtained by amino acid sequence analysis on the native chondroitinase I protein itself, including, for example, a twelve residue oligopeptie (SEQ ID N:O2, amino acids 133-144). An eight residue oligopeptide derived from the DNA sequence (SEQ ID NO:2, amino acids 71-78) also matches a previously sequenced oligopeptide derived by a combination of trypsin digestion and cyanogen bromide treatment of the native protein. The only discrepancy between the two sequences is at amino acid residue #162 of the mature protein (SEQ ID NO:2, amino acid 186), where the DNA sequence codes for an arginine, while the native protein sequence indicates a leucine.

Since a single nucleotide alteration would change a leucine codon (CTT) to an arginine codon (CGT), an initial interpretation suggests that this may be caused ,y a lack of perfect incorporation fidelity by the Tag DNA polymerase during the i. vitro amplification process. However (see below) later results indicate that the DNA sequence is correct, rather than the amino acid sequence obtained by analyzing the native enzyme. These results also indicate that the "18 kD" and the "90 kD" fragments are, in fact, contiguous pieces of the chondroitinase I protein that has been cleaved (pr"umably by a contaminating protease), predominately between residues #157"(Gln) and #158 (Asp) of the mature protein (SEQ ID NO:2, between amino acids 181 and 182). All of the above information supports the 1 ri iti 11 *w' S.J r- 7~ WO 94/25567 PCT/US94/04495 WO 94/25567 PCTUS94/04495 54 interpretation that the cloned DNA (at least that portion that is bracketed by the oligonucleotide primers) generated by PCR amplification represents "part of the authentic chondroitinase I gene of P.

vulqaris and, therefore, can be used as a probe to identify cosmid clones that carry the intact gene.

Although it is possible to isolate the entire gene coding for a protein of interest using PCR amplification (thereby avoiding construction of a gene bank and many of the other steps described below) by employing oligonucleotide primers derived from the amino-terminus of the protein coupled with primers derived from the carboxyl-terminal amino acid sequence, there are several potential problems in this approach. In the case of the P. vulqaris chondroitinase I, the problems include: the assumption that the protein being sequenced has not been processed at either end (not likely to be true, for example, with a secreted protein), the occasional lack of fidelity exhibited by Tag DNA polymerase during PCR reactions, and the rather large size of the bracketed region of the DNA that is to be amplified which was expected to be approximately r 3000 bp (deduced from the apparent molecular weight of approximately 110 kD). Consequently, the approach of Sconstructing a gene bank is selected.

Si. "Generation Of A Labeled Probe, Colony Hybridization And Identification Of Positive Cosmid Clones From The P. vulqaris Gene Bank The cloned PCR product correspaoding to the 455 bp near the amino-terminal coding portion of the P. vulqaris chondroitinase I gene is released from the 1 II "1 55 plasmid DNA into which it had been cloned by digestion with the restriction enconuclease Sall. This is a consequence of the presence of one Sal site within the polylinker sequence and a second Sall site within the cloned PCR amplification product (this is fortuitous in that the latter Sall site is derived from the nucleotide sequence of oligonucleotide pool #18 near its 5'-end; in ;act, there is no recognition site for Sall within the P. vulqaris chondroitinase I gene itself). A total of approximately 260 gg of plasmid DNA is digested with SalI and the products separated by electrophoresis on a NuSieve

T

M GTG agarose gel. The desired approximately 450 bp fragment is isolated using a Qiaex' extraction protocol. The fragment is then denatured by heating at 95-100 0 C for 5-15 minutes, followed by rapid cooling. The denatured fragment is then labelled with digoxigeninlabelled dUTP (Boehringer-Mannheim, Indianapolis, IN) in two 200 jl reactions.

Aliquots of the six P. vulgaris cosmid gene banks described in Example 1 above are used to infect the E. coli strain ER1562 described above and a total of approximately 10,000 colonies are obtained on the ii appropriate selective plates. These colonies (on a total of 50 plates) are replica plated onto two nylon membranes on selective agar as well as to a third selective plate. After overnight incubation, the colonies on the filters are lysed by sequentially treating with 10% sodium dodecyl sulfate (SDS) and M NaOH for 5-30 minutes each. The cells from the lysed colonies are neutralized by being placed on sheets saturated with 1,M Tris-HCl (pH 7.4) (twice) and then on paper saturated with 2X standard saline citrate priori vacuum drying at 80°C. The DNA from the lysed colonies is then fixed to the membranes.

1 1 o 'WO 94/25567 PCT/US94/04495 i SWO 94/25567 PCT/US94/04495 56 The filters are then washed by incubation of the filters at 42*C with agitation for 1-3 hours, using at least 10 ml/filter of 0.05 M Tris HC1, 0.5-1 M NaCl/0.001 M EDTA, pH 8, 0.1% SDS and 0.05 mg/ml proteinase K. The filters are then rinsed with 2 X SSC and pre-hybridized by incubation with a hybridization buffer at 65°C for 1-3 hours. The filters are then hybridized overnight at 65-68 0 C using the digoxigenin-labeled probe described above (0.5-50 ng/ml in a hybridization solution). The hybridized filters are washed with SSC and SDS, re-blocked with a blocking reagent (Component #11 of DNA Labelling and Detection Kit, Nonradioactive, Boehringer Mannheim, Indianapolis, IN) and exposed .o polyclonal sheep anti-digoxigenin Fab fragments conjugated to alkaline phosphatase.

The positive clones are visualized by incubation of the antibody-labeled filters in the presence of BCIP (bromo-chloro-indolyl-phosphate) and NBT (nitro-blue tetrazolium). The presence of the desired DNA fragment within a colony will result in a dark brownish-purple spot in the filter after this hybridization procedure. After approximately four hours, the developed filters are used as templates to guide the selection of a total of 117 clones which are then picked to selective media. A small-volume ml) culture ("Miniprep") of each of these clones is grown in selective media and plasmid DNA is then isolated using materials and protocols supplied by Qiagen.

A _i W IO 94/25567 PCT/US94/04495 35 57 Example 4 Restriction Mapping And Southern Hybridization Used To Localize The Position Of The Chondroitinase I Gene Within Individual Clones A number of approaches are used to guide the selection of particular cosmid clones for further study. One is to carry out Southern hybridization (8) using the same PCR-generated fragment as a probe against P. vulgaris genomic DNA that had been digested by a number of restriction enzymes and then fractionated on an agarose gel prior to transfer to a nylon membrane. The probe is labeled with digoxigenin-dUTP by including this nucleotide analogue in a PCR amplification. In this reaction, the gelpurified product of a previous PCR amplification (that using P. vulqaris genomic DNA as template) is diluted 10,000-fold and serves as the template in a second PCR amplification.

This latter reaction is made up as a 0.5 ml mixture, which is then divided into ten individual tubes and amplified as described above for 25 cycles using oligonucleotide pools #2 and #10 (see above) as the primers. The normal complement of deoxyribonucleoside triphosphates is replaced with a digoxigenin-dUTP labeling mixture from the manufacturer (Boehringer-Mannheim, Indianapolis, IN), which yields a final concentration of 100 MM each of dATP, dCTP and dGTP, 65 pM dTTP and 35 Al digoxigenindUTP. The reactions are pooled and precipitated according to the manufacturer's recommendations. An aliquot of the resuspended product is examined by gel electrophoresis and exhibits,, a single band between approximately 300 and approximately 400 bp in length as expected for the "smaller" PCR product described

I~

I

1 t ie r rr I I r; $I j a 'WO 94125567 PCT/US94/04495 58 above.

To avoid problems encountered with the highly viscous P. vulgaris genomic DNA preparation, the DNA (approximately 5 pl) is diluted into large (0.35 ml) volumes for digestion with the various restriction enzymes. The DNA is then concentrated by ethanol precipitation prior to fractionation on agarose gels and transfer to nylon membranes. The data obtained in these experiments indicates that the chondroitinase I gene (at least that portion that hybridizes to the N-terminal coding region represented by the probe described above) is carried on a BstYI fragment of approximately 2800 bp, an EcoRV fragment of 5400 bp, and on large (equal to or greater than approximately 10kb) DNA fragments generated by Nsil, BglII, HindIII, and Stvl.

Large scele cultures (500 ml) of a number of hybridizing cosmid clones are grown and plasmid DNA is isolated from these cultures for use in mapping the location of the chondroitinase I gene. The DNA of the gene is expected to represent only approximately per cent of the P. vulgaris DNA carried within each cosmid. A number of these clones are digested with BstYI and Nsil and the products are fractionated on an agarose gel. Individual fragments are then isolated, a portion tested for the presence of chondroitinase I sequences by Southern hybridization, and then subcloned into appropriate vectors.

Two of these fragments are of special interest. The first, a BstYI fragment of approximately 2800 bp, is observed in a number of cosmid clones, including those designated #2 and The DNA isolated from these two cosmid clones is designated LP 2 751 and LP 2 760. With LP 2 760, the approximately 2800 bp BstYI fragment is well separated i

I

g -i

;I

:i i I g o i 59 i i 3 5 from the other BstYI fragments and is therefore more readily subcloned into another vector designated pT660-3. The plasmid designated pT660-3 is a .derivative of pBR322 in which the DNA from a point immediately downstream of the promoter for tetracycline resistance (approximately bp 80) as far as the PvuII site (approximately bp 2070) is deleted and replaced with a BamHI linker. Similarly, the approximately 10 kb NsiI fragment (which hybridizes with the chondroitinase probe described above) is readily isolated from a digest performed on LP 2 751.

These two fragments are referred to as the "2800 bp BstYI" fragment and the "10 kb Nsil" fragment.

The 2800 bp BstYI fragment is small enough to permit a second restriction enzyme digestion on this piece of DNA in order to obtain a fragment suitable for DNA sequence analysis. This is important because the hybridization experiments serve to identify the N-terminal coding region of the chondroitinase I gene, due to how the probe i, derived. This procedure does not, however, indicate to which side the rest of the gene is located. Given the relative size of the probe (less than 500 bp) compared to the predicted size of the intact gene (greater than 3000 bp), this is not a trivial consideration. The nucleotlde sequence, however, clearly indicates in which direction the gene would be "read" and therefore, which restriction fragments should be cloned in order to obtain the entire gene.

The subcloned 2800 bp BstYI fragment contains two internal EcoRV sites, which suggests that the resulting fragments might be small enough for DNA sequencing-. However, the EcoRV sites are symmetricaily placed within the 2800 bp BstYI fragment; \ach EcoRV site is approximately 1200 bp i;

E

m saiiiir ii i, 1- WO 94/25567 PCT/US94/04495 of the DNA and, during the in vitro DNA synthesis, they are extended toward the "90 kD" N-terminal coding sequences. Consequently, the oligonucleotides Iii.' WO 94/25567 PCT/US94/04495 60 from one end, with the space between them equal to approximately 400 bp. The subcloned fragment is digested asymmetrically by taking advantage of unique restriction sites present within the vector. In this manner, the "halves" of the 2800 bp BstYI fragment are distinguished physically and, by Southern hybridization, the "end" that contains the chondroitinase I N-terminal coding region is ascertained. Once this is done, the appropriate piece, which is a HindIII-EcoRV fragment of approximately 1200 bp, is subcloned into both M13mpl8 and M13mpl9 vectors which are first digested with both HindIII and Smal and subsequently treated with calf intestinal alkaline phosphatase.

The DNA sequence derived from these subclones reveals a number of features that clearly establish the location of the chondroitinase I gene, as well as the direction in which it is read.

Starting with nucleotide #183 in this sequence (SEQ ID NO:1, nucleotide 191), a coding region is observed which matches the first thirty previously-identified amino acids of the P. vulqaris chondroitinase I enzyme. Preceding this sequence, it is possible to discern a number of other features by their analogy to corresponding sequence motifs from previously analyzed E. coli genes. These features include: nucleotides 32-37 (SEQ ID NO:1, nucleotides 40-45) which match in three of six positions with the consensus region of a promoter and, after a 17 nucleotide space, a region of a promoter (matching in six of seven positions with the consensus region); a putative "Shine-Dalgarno" sequence can be noted between nucleotides 98-103 (SEQ ID NO:1, nucleotides 106-111); and there is an in-frame ATG initiation codon at nucleotides 111-113 (SEQ ID NO:1, nucleotides 0 i

W

r ;i 1 j ;t i

I

i; n i o

I

,ic e i

I

WO 94/25567 PCT/US94/04495 61 119-121), which indicates that the P. vulgaris chondroitinase I enzyme is synthesized with a 24 amino acid signal sequence which is, presumably, removed as the protein is transported across the inner membrane.

The second fragment that is subcloned (into a pIBI24 derivative that is first modified to include an NsiI restricticn site in place of the PstI site normally present in the polylinker of this vector) is the approximately 10 kb Nsil fragment. Digestion of this approximately 14 kb recombinant molecule (the approximately 10 kb Nsil fragment in pIBI24) with EcoRV yields four fragments of approximately 9 kb, 2.3 kb, 2.1 kb, and 0.4 kb. Southern hybridization analysis using the probe derived from the N-terminal amino acid sequence indicates that the related, chondroitinase gene sequences are contained within the largest fragment (the approximately 9 kb EcoRV fragment).

Since there is no other fragment larger than 2.9 kb (the size of pIBI24 which has no internal EcoRV recognition sites), this approximately 9 kb EcoRV fragment must contain the vector as well as P.

vulaaris DNA. A double digestion of this recombinant molecule with Nsil and EcoRV releases the pIBI24 vector as a 2.9 kb fragment; it also yields fragments of approximately 4.5 kb, 2.3 kb, 2.1 kb, 1.0 kb and 0.4 kb. Taken together (along with the information presented above on the 2.8 kb BstYI fragment which has two internal EcoRV sites separated by approximately 0.4 kb), an initial restriction map is constructed.

A double digestion with EcoRV and HindIII releases fragments of approximately 4.1 kb, 2.3 kb, 2.1 kb, 2.0 kb, 1.3 kb, 1.1 kb and 0.4 kb. Three of these fragments (2.3 kb, 2.1 kb, and 0.4 kb) are apparently EcoRV fragments that have not been cut by .4

JI

'V

I

zne ro±Jowing amino acia sequence, UUI--- the N-terminal amino acid sequence of the "90 kD" peptide, is used: Gly-Ala-Lys-Val-Asp-Ser (SEQ ID NO:2, amino acids 189-194). This hexapeptide can be Ii 1 1 i~ )r:a kc

I

j)y 'WO 94/2; 5567 PCT/US94/04495 62 HindIII. Again, the only fragment larger than the vector (4.1 kb) indicates that this fragment includes pIBI24 (2.9 kb). The approximately 2.0 kb fragment hybridizes with the chondroitinase probe, thereby serving to place one of the HindIII sites. Since there is a HindIII site in the polylinker, it too can be placed, leaving the last HindIII site to be placed by deduction.

Double digestion of the cloned approximately kb Nsil fragment with EcoRV and EcoRI yields six fragments (of approximately 4.2 kb, 3.5 kb, 2.3 kb, 2.1 kb, 1 kb, and 0.4 kb), indicating the presence of two EcoRI sites one in the polylinker and one in the cloned P. vulqaris DNA. Southern hybridization reveals that the approximately 4.2 kb band in this double digest contains the chondroitinase I N-terminal coding sequence. Adding this information to the above data yields a preliminary restriction map for the subcloned approximately 10 kb Nsil fragment in pIBI24 (Figure 1).

It should be noted that, in further support of the placement and orientation of the chondroitinase I gene, in vitro chondroitinase I assays in which the activity of the enzyme based on measuring the release of unsaturated disaccharide from chondroitin sulfate C at 232 nm are carried out on a small number of samples. In one case, an aliquot of an overnight culture used to prepare LP 2 751 (ER1562 carrying cosmid DNA selected from the colony hybridizations) is found to exprss 0.12 units/ml of chondroitinase. In addition, one of the EcoRV-deletion constructions (to be described below) is grown overnight in the presence of ampicilli This culture is then inoculated into fresh selective media either with or without isopropyl-beta D-thiogalactopyranoside (IPTG) which is i i"n -t r 'in oraer to reduce the number of sequence WI 'O 94/25567 PCT/US94/04495 63 expected to increase the level of transcription from the lac promoter present in pIBI24. The assay results of 0.29 units/ml of chondroitinase without and 0.36 units/ml with IPTG induction indicate that, even after the EcoRV deletion, the gene is still intact and possibly oriented in the same direction as that of the lac promoter.

Although the sizes of the fragments ir. the above discussion are approximate (especially the approximately 1 kb region between the EcoRI/NsiI in the polylinker and the nearest EcoRV site; in aidition, there also might be another small EcoRV fragment that is still unmapped), overall they suggest that the approximately 4.2 kb EcoRV-EcoRI fragment contains the entire chondroitinase I gene. In order to facilitate the restriction mapping, an EcoRV deletion is constructed using the approximately 10 kb Nsil fragment cloned into pIBI24 (LP 2 776). This DNA is digested with EcoRV, treated with calf intestinal alkaline phosphatase, and fractionated on an agarose gel. The largest (approximately 10kb) fragment is extracted from the gel and ligated together in the presence of a phosphorylated EcoRI linker. The resulting construction (LP 2 786) is next digested with EcoRI to yield three fragments. Although it is not completely separated from the pIBI24-containing, somewhat smaller fragment, an approximately homogenous, approximately' 4.2 kb EcoRI fragment is obtained after extraction from the gel. This EcoRI fragment is then used for DNA sequence analysis.

7 WO 94/25567 PCT/US94/04495 64 Example DNA Sequence Analysis Of The Approximately 4.2 kb EcoRI Fragment The approximately 10 kb Nsil fragment, cloned into pIBI24, is digested with EcoRV (as described above) and ligated together in the presence of EcoRI linkers. The net result of this construction is the deletion of approximately 5 kb of P. vulgaris DNA from this subcloned piece of DNA and the simultaneous introduction of another EcoRI site into the molecule. One hundred micrograms of this "EcoRV deletion" construction (LP 2 786) is digested with EcoRI and fractionated on an agarose gel. The desired approximately 4.2 kb fragment is eluted from the gel, precipitated and resuspended in 150 Al TE described above. One-third of this material is then ligated to itself (polymerized) and, after destruction of the DNA ligase by heating, the DNA is sonicated to generate random, small pieces suited to DNA sequence analysis.

The ends are rendered flush in a "fill-in" reaction mediated by the "Klenow fragment" (10; New England Biolabs, Beverly, MA) of the DNA polymerase I of E. coli, and then ligated into Smal-cut and phosphatased M13mpl9. This recombinant DNA is used to transform the male E. 'Ai strain MV1190 and 500 of the phage plaques obt& d are picked into SM buffer (NaCl, 100 mM, 8 mM, Tris-HCl, pH 7.4, 50 mM and i 0.01% gelatin) ?r ve as stocks for the infection of i 30 small (leLs th i equal to 10 ml) cultures that are then used for the isolation of single stranded templa te DNA.

DNA sequencing is carried out at elevated temperatures using Tag DNA polymerase and 35 fluorescently-labeled oligonucleotide primers. The i 350 bp in length. Furthermore, the larger product could be isolated from an agarose gel, diluted a thousand-fold, and then used as the template in a second PCR reaction employing oligonucleotide pools #4 ii c:3 WO 94/25567 PCT/US94/04495 65 data are collected using a Model 370A DNA sequencing system (Applied Biosystems, Foster City, CA).

Sequence editing, overlap determinations and derivation of a consensus sequence are performed using a collection of computer programs obtained from the Genetics Computer Group at the University of Wisconsin The resulting DNA sequence of this EcoRI fragment is 3980 nucleotides in length (SEQ ID NO:1).

It is to be noted that the EcoRI site near the Nterminal coding sequence is derived from the linker ligated into this site; it is not present in the P.

vulgaris chromosome. This position actually is an EcoRV site in the cloned cosmid DNA.

Translation of the DNA sequence into the putative amino acid sequence reveals a continuous open reading frame encoding of 1021 amino acids (SEQ ID NO:2), with a 24 residue signal sequence (SEQ ID NO:2, amino acids 1-24), followed by a 997 residue coding sequence for the mature (processed) chondroitinase I protein (SEQ ID NO:2, amino acids 25-1021). Computer analysis using the programs described above of this sequence predicts a molecular weight of 115,090.94 for the unprocessed protein, a molecular weight of 112,507.82 for the mature "110 kD" (transported) protein, 17,503.43 for the first 157 amino acids (the "18 kD" fragment) (SEQ ID NO:2, amino acids 25-181) and 95,022.40 for the remaining 840 amino acids (the kD" fragment) (SEQ ID NO:2, amino acids 182-1021) and a molecular-weight of 2601.14 for the 24-residue signal sequence. One notable feature of the amino acid composition is the absence of cysteine which could be important if the protein has to be re-folded at any point.

In the nucleotide sequence, it was noted above that there is a unique SphI restriction site

U

V;

.l, r icorresponding to these constructions is then isolatea and subjected to DNA sequence analysis usingaan Applied Biosy.tems ("oster City, CA) instruMnt ana gTag sequencing kit. The results indicate that th4 11 Ui j z 4 8- 'WO 94/25567 PCTIUS94/04495 66

I-K

located approximately 230 bp beyond the end of the gene (SEQ ID NO:1, nucleotides 3414-3419), which presents a unique target site that can be manipulated .to allow the facile movement of the gene to achieve the overall goal of expressing chondroitinase at high levels in E. coli. Although there are two recognition sites for Clal (ATCGAT), one of them (SEQ ID NO:1, nucleotides 2702-2707) is embedded within the E. coli dam recognition sequence (GATC) (SEQ ID NO:1, nucleotides 2701-2704). The resulting adenine methylation by the dam-encoded enzyme blocks cleavage of this site by Clal; therefore, there is, in effect, a "unique" Clal site (SEQ ID NO:1, nucleotides 497- 502) which is used, as described below, to reconstruct the chondroitinase I gene after the appropriate sitespecific mutageneses are carried out.

Example 6 Site-specific Mutagenesis Of The Cloned P. vul~aris Chondroitinase I Gene The site-specific mutagenesis method employed is based on that of Kunkel using materials purchased from Bio-Rad, Melville, N.Y.

(Muta-Gene M In Vitro Mutagenesis Kit). In this procedure, the target DNA to be mutagenized is first cloned into an appropriate Ml3-derived vector. In this case, the recombinant molecule used (M13mpl9 into which is clone e t2e approximately 1200 bp EcoRV- HindIti fragment as described above) encompasses the N-terminal coding region of the chondroitinase I gene.

This recombinant phage is replicated in the E. coli host strain CJ236 (Bio-Rad), a male strain that carries the dut and ung alleles. The combination of these two mutations, dut (dUTPase) and un (uracil-N- O0 residues #157 (Gln) and #158 (Asp) of the mature protein (SEQ ID NO:2, between amino acids 181 and 182). All of the above information supports the 1

I-)

94/25567 WO 94/25567 PCT/US94/04495 67 glycosylase), results in the incorporation of some uracil, rather than thymine, residues into the DNA synthesized in this organism. Single stranded template is then isolated after propagation on CJ236 and an appropriate, mutagenic, synthetic oligonucleotide is annealed to this DNA.

This oligonucleotide serves as a primer for T7 DNA polymerase which copies the entire recombinant molecule. T4 DNA ligase is then used to seal the nick between the first residue of the mutagenic oligonucleotide and the last residue added in vitro.

The newly synthesized DNA (containing the desired base changes) therefore does not contain uracil, while the template DNA does. Transformation of a non-mutant (with respect to the dut and un alleles) male E. coli strain yields phage progeny that are primarily derived from the mutagenized strand synthesized in vitro as a result of the inactivation of the uracil-containing template strand.

In this specific case, four resuspended plaques (aliquots of which had been used for DNA sequencing which established the N-terminal coding region of the chondroitinase I gene and included another 110 bp "upstream" of the presumed translation initiation site (see above)) are used to infect the male host strain CJ236 (dut ung). Individual plaques are picked to 0.5 ml of phage dilution buffer (FB).

One picked plaque from each transformation is adsorbed to log phase CJ236 and the infected cultura grown for hours. The cells are pelleted by centrifugation, and the supernatant heated to 55°C for 30 minutes and then stored at 4 0 C. Single stranded DNA is isolated from 100 ml of each supernatant and resuspended in a total volume of 0.1 ml of TE.

The goal of the site-specific mutagenesis is (1:

I'

K if -C 'i r WO 94/25567 PCT/US94/04495 68 to modify the "ends" of this gene to allow it to be moved, precisely, into an appropriate high-level E.

coli expression system. The target vector chosen (pET9-A; see above) is one derived from genetic regulatory elements present in the bacteriophage T7.

In this sytem, there is a unique NdeI site (CATATG) that includes the translation initiation codon as well as a downstream BamHI site that, together, allow the direct, unidirectional, insertion of a gene encoding the protein that is to be expressed. These two sites are preceeded by a T7-specifc promoter sequence and trailed by a transcription terminator that functions with the T7 RNA polymerase. Accordingly, these two restriction sites (Ndel and BamHI) are introduced into the cloned gene for P. vulqaris chondroitinase I.

In order to introduce the NdeI site (containing the ATG initiation codon) both before the signal sequence as well as, in a second construction, before the coding sequence for the mature protein (thereby deleting the signal sequence), two synthetic oligonucleotides are designed and synthesized (purchased from BioSynthesis, Inc., Denton, TX). The first, designated oligonucleotide 25 (SEQ ID NO37), retains the signal sequence while the second, oligonucleotide #26 (SEQ ID NO:38), deletes the signal sequence and allows the direct expression of the mature chondroitinase I protein (which can have an additional methionine residue at the N-terminus (SEQ ID NO:5, amino acid number The native sequence, including the predicted initiation codon, is presented on line 1 below while the mutagenic oligonucleotide #25 (which differs in the three nucleotides immediately upstream of the initiation codon) is presented on line 2: "3 0 I, I I I II I I r ir: 1 I, S SI 'WO 94/25567 PCT/US94/04495 69 1) TCGTTTTACTGC-3' (SEQ ID NO:1, nucleotides 94-141) 2) TCGTTTTACTGC-3' (SEQ ID NO:37) For the construction in which the signal sequence is deleted, the site-specific mutagenesis is carried out at the junction of the signal sequence and the start of the mature protein (line 3) using the mutagenic oligonucleotide 26 (line 4) (which differs by six nucleotides, including the location of the initiation codon): 5'-GCGCCTTATAACGCGATGGCAGCCACCAGCAATCCTG-3' (SEQ ID NO:1, nucleotides 170-206) 5'-GCGCC.'"ATAACGCGCATATGGCCACCAGCAATCCTG-3' (SEQ ID NO:38) d '1 The underlined GCC in line 3 corresponds to the codon for alanine which is the N-terminal amino acid for the mature, processed form of the P. vulgaris chondroitinase I.

In order for these oligonucleotides to be used, their 5'-ends need to be phosphorylated. Therefore, oligonucleotide 25 (5 O.D. units) is resuspended with 0.5 ml of TE, while oligonucleotide 26 (also 5 O.D. units) is resuspended in 0.65 ml TE to yield stocks that are approximately 20 nM, 20 pmole/Al. Three nanomoles (150 Al of stock solution) of each oligonucleotide are kinased in separate (0.35 ml) reactions containing 35 Al 10x ligase salts (New England Biolabs, Beverly, MA): 0.5 M Tris-HCl (pH

I

i if" t r i i i; oi~ 'WO 94/25567 PCT/US94/04495 70 0.1 M MgC12, 0.2 M dithiothreitol, 10 mM ATP, mg/ml bovine serum albumin), 35 yl 0.1 M dithiothreitol, 10 Al (100 units) T4 polynucleotide kinase (New England Biolabs) and made up to volume with 120 Al TE. The reactions are incubated at 37 0

C

for 40 minutes and the enzyme inactivated at 70°C for minutes.

Template DNA (5 yl of the preparation described above) and phosphorylated mutagenic primer (approximately 2 pmole) are annealed in a 20 Al volume containing 20 mM Tris-HCl (pH 2 mM MgC12, and mM NaCl. The sample is heated at 70°C for 45 minutes in a Perkin-Elmer/Cetus Thermalcycler'. The sample is then gradually cooled from 70 0 C to 25 0 C over a minute period. The annealed mixture is placed on ice and the following components added: 2 Al of 10 X synthesis buffer (Bio-Rad): 5mM each of dATP, dGTP, dCTP, dTTP; 10 mM ATP; 100mM Tris-HCl (pH 50 mM MgC 2 1; 20 mM dithiothreitol), 2 Il of T4 DNA ligase (6 units) and 1 gl of T7 DNA polymerase (1 unit). These reactions are incubated for 5 minutes each at 0 C (on ice), 11 0 C, 25°C, and finally for 30 minutes at 37 0

C.

The reactions are stopped by the addition of 75 Al of mM Tris-HCl-10 mM EDTA (pH 8.0) and placed at -20 0

C.

After the mutagenized DNA is thawed, it is used to transform the male E. coli strain MV1190 (dut* mung*). Individual plaques obtained are picked and single-stranded DNA is isolated and sequenced. For those cases in which the desired sequence changes are introduced, another aliquot of the resuspended plaque is used to infect strain MV1190, but in this case the intracellular, double-stranded replicative form of the recombinant DNA is isolated from the infected cell pellets using the Mini-Prep procedure referenced above.

c lr' i: i ,j

L

WO 94/25567 PCT[US9404495 71 Example 7 Reconstruction Of The Site-Specifically Mutagenized Chondroitinase I Gene And Its High-Level Expression In E. coli 4 Example 6 described the site-specific mutageneses that created an NdeI site immediately preceeding the signal sequence, as well as a second construction which placed the Ndel site adjacent to the triplet which codes for the N-terminal alanine found on the mature, processed P. vulqaris chondroitinase I gene. In each case, the ATG sequence of the NdeI recognition site (CATATG) can function as the translation initiation codon for the protein (either with or without the signal sequence).

In order to transfer these alterations from the M13 vector in which they were constructed, to t;he full chondroitinase I gene, the isolated replicative form is digested with KpnI and Clal. The KpnI site is part of the M13mpl9 polylinker, while the Clal site is found approximately 490 bp from the end of the cloned fragment of the chondroitinase I gene. The restriction digestion products obtained are fractionated on a 4% NuSieveT GTG agarose gel run in 1/2 X Tris-Acetate buffer (TAE). The appropriate approximately 500 bp band is extracted from the gel using Qiaex

M

Similarly, plasmid DNA (LP 2 786) carrying the chondroitinase I gene is also digested with KpnI and Clal and then fractionated on a 0.8% agarose gel run in 1/2 X TAE. In this case, the KpnI site is part of the polylinker of pIBI24, while the Clal site corresponds to the one described above. (As stated above, there is a second Clal site in the chondroitinase I gene, but it is not cleaved by Clal because this site is apparently blocked by dam I ~I i: :9h T1 Ft Id~ PCT/US94/04495 'WO 94/25567 72 methylation. The site-specific mutagenesis and reconstruction of the chondroitinase I gene were carried out before the entire nucleotide sequence was ascertained).

The approximately 7 kb fragment containing the pIBI24 vector and the large fragment of the chondroitinase I gene are isolated from the agarose gel by electroelution followed by ethanol precipitation. This 7 kb fragment is then treated with calf intestinal alkaline phosphatase, extracted first with phenol-chloroform, then with chloroform, and then precipitated twice with ethanol and finally resuspended with 0.1 ml TE. The two isolated Nterminal encoding fragments (the two approximately 500 bp KpnI-ClaI pieces containing the two sitespecifically mutagenized sequences, one with and one without the signal sequence) are each ligated to the approximately 7 kb fragment encompassing the remainder of the chondroitinase I gene and the pIBI24 vector.

The ligase reaction is then used to transform the E.

coli strain 294 and ampicillin resistant derivatives obtained. DNA is isolated from small (10 ml) cultures and digested with NdeI to verify the presence of this restriction site within the reconstructed DNA.

In order to remove the (apparent) P.

vulqaris promoter and ribosome binding site, the modified chondroitinase I genes are isolated as approximately 4.5 kb NdeI-NsiI fragments and subcloned into a pBR322 variant in which the EcoRI site is first filled-in, then dephosphorylated, and finally a phosphorylated Nsil linker (New England Biolabs) inserted. The sequence of the linker used (TGCATGCATGCA) to place the Nsil site (ATGCAT) into pBR322 also includes an SphI site (GCATGC). In order to trim extra, non-coding DNA from the subcloned Ndel- Ir*C Ii rr WO 94/25567 PCT/US94/04495 73 Nsil fragments, as well as to introduce a unique restriction site to be used later, plasmids (representing two clones each with the signal sequenc retained [LP 2 861 and LP 2 863] and two with the signal sequence deleted [LP 2 865 and LP 2 867]) containing the approximately 4500 bp Ndel-NsiI segments including the chondroitinase I gene are first digested with SphI, the ends "filled-in" with the "Klenow" fragment (11) of the E. coli DNA polymerase I and the resulting DNA fragments fractionated on an agarose gel in 1/2 X TAE). The appropriate bands (approximately 5200 bp) are eluted from the gel using Qiaex T and then treated with calf alkaline phosphatase. After the removal of this enzyme by phenol-chloroform and chloroform extractions, the DNA is precipitated twice and finally resuspended with 0.1 ml TE.

This DNA is then ligated in the presence of a phosphorylated BamHI linker and the mixture used to transform the E. coli strain 294. Six representative, ampicillin resistant colonies from each of the four constructions are grown in small (10 ml) cultures and plasmid DNA is isolated. Digestion of the DNA from the 24 clones examined with the enzymes NdeI and BamHI indicates which contain the BamHI site and, simultaneously, releases the approximately 3400 bp Ndel-BamHI fragment which contains the chondroitinase I gene. Seventeen clones (eight with and nine without the signal sequence) yield the desired fragment which is extracted from the agarose gel with QiaexM.

These approximately 3.4 kb NdeI-Bam-HI chondroitinase I gene-containing fragments (both with and without the signal sequence) are then used to construct a high-level expression system. The expression vector used, pET-9A Novagen), is derived from elements of the E. coli bacteriophage T7.

I

la i 1 i i I 1 j .a 'WO 94/25567 PCT/US94/04495 74 It contains an origin of replication derived from the Col El plasmid, a kanamycin resistance determinant, and the transcription and translation initiation determinants of the T7 gene 10. The naturallyoccurring translation initiation codon for this gene is part of an NdeI site. This region is followed by a unique BamHI site and a T7 transcription terminator.

A sample of this expression vector is digested with the restriction enzymes Ndel and BamHI, dephosphorylated with calf intestinal alkaline phosphatase, and purified by agarose gel electrophoresis. Each of the chondroitinase I gene fragments (both with and without the signal sequence) is ligated to the expression vector fragment. The resulting recombinant DNA mixture is used to transform the E. coli K-12 host, HMS174 (Novagen). Kanamycinresistant colonies obtained are grown in small scale ml) and plasmid DNA is isolated and examined to confirm the predicted structure.

Samples of these constructions are then used to transform the expression host BL21(DE3)/pLysS This E. coli B strain carries the T7 RNA polymerase gene under lac control (and is therefore inducible by either lactose or IPTG) on a lambda phage integrated within the E. coli chromosome, as well as the Col Elcompatible plasmid pLysS. This latter replicon specifies resistance to chloramphenicol and contains the T7 lysozyme gene inserted into the tetracyclineresistance determinant of pACYC184 (ATCC 37033, American Type Culture Collection, Rockville, MD) in the "silent" orientation (read in the opposite direction relative to the tetracycline resistance gene). The T7 lysozyme is expressed at a relatively low level in this construction and serves as an inhibitor of the T7 RNA Polymerase thereby

I

3T 1 I I f 'WO 94/25567 PCT/US94/04495 75 minimizing the basal-level expression of the gene to be overexpressed.

Derivatives of BL21(DE3)/pLysS carrying the chondroitinase I gene (with the signal sequence retained and which have been subjected to the sitedirected mutagenesis described in Example 6 (SEQ ID NO:3)) in pET9-A are designated LL2084, LL2085, LL2086 and LL2087. They are not tested for expression of the chondroitinase I enzyme. The native chondroitinase I gene (with the signal sequence retained) (SEQ ID NO:l), which has not been subjected to site-directed mutagenesis, is inserted into a different expression host.- Expression of the chondroitinase I enzyme is achieved.

One of the derivatives of BL21(DE3)/pLysS carrying the signal-less chondroitinase I gene which has been subjected to the site-directed mutagenesis described in Example 6 (SEQ ID NO:4) inserted into pET9-A, is designated LL2088, tested and used to establish a master cell bank. The insertion of the gene into pET9-A yields the plasmid designated pTM49- 6. Samples of the E. coli B strain BL21(DE3)/pLysS carrying the plasmid pTM49-6 constitute the deposited strain ATCC 69234.

An overnight culture of this deposited strain is grown at 30 0 C in the presence of 40 g/ml of kanamycin and 25 gg/ml of chloramphenicol. A 0.5 ml aliquot of this culture is used to inoculate 100 ml of a rich "expression" medium containing M9 salts (17) supplemented with 20 g/l tryptone, 10 g/l yeast extract, and 10 g/l dextrose in addition to the same level of kanamycin and chloramphenicol.

The culture is grown at 30°C to an appropriate density (a value of 1 at A 6 and then chondroitinase I expression is induced by the addition ~i i i 1 'WO 94/25567 PCT/US94/04495 76 of IPTG to a final concentration of 1 mM. After three hours, samples are taken, centrifuged, and the cell pellets frozen on dry ice prior to assay. The frozen pellets are thawed, resuspended in buffer and sonicated. A value of 56 units/ml is obtained (relative to the original culture volume), which indicates that this expression system is functional.

A subsequent 10 liter fermentation under controlled conditions at a higher cell density yields a maximum value of approximately 600 units/ml of chondroitinase I. This represents a substantial improvement over fermentation of the original native P. vulqaris, which had not expressed chondroitinase I at a level above 2 units/ml.

Method For The Isolation AJ4 Purfic:tion Of The Native Chondroitx;, 3 En~ As Adapted To The Riaoibina;nt Enzyme The native enzyme is ploduced by fermentation of a culture of P. ylari. The bacterial cells are first recoveped from the medium and resuspended in buffer. The cell suspension is then homogenized to lyse the bacterial cells. Then a charged particulate such as 50 ppm Bioacryl (Toso Haas, Philadelphia, PA), is added to remove DNA, aggregates and debris from the homogenization step.

Next, the solution is brought to 40% saturation of ammonium sulfate to precipitate out undesired proteins. The chondroitinase I remains in solution.

The solution is then filtered using a 0.22 micron SP240 filter (Amicon, Beverly, MA), and the retentate is washed using nine volumes of 40% ammonium sulfate solution to recover 7ost of the enzyme. The

I

i i-I i I i I j c

I'

WO 94I/5567 PCT/US94/04495 77 filtrate is concentrated and subjected to diafiltration with a sodium phosphate buffer using a kD filter to remove salts and small molecules.

The filtrate containing chondroitinase I is subjected to cation exchange chromatography using a Cellufine' cellulose sulfate column (Chisso Corporation, distributed by Amicon) At pH 7.2, 20 mM sodium phosphate, more than 98% of the chondroitinase I binds to the column. The native chondroitinase I is then eluted from the column using a 0 to 250 mM sodium chloride gradient, in 20 mM sodium phosphate buffer.

The eluted enzyme is then subjected to additional chromatography steps, such as anion exchange and hydrophobic interaction column chromatography. As a result of all of these procedures, chondroitinase I is obtained at a purity of 90--97% as measured by SDS-PAGE scanning (see above) However, the yield of the native protein is only 25-35%, determined as described above. This method also results in the cleavage of the approximately 110 kD chondroitinase I protein into a kD and an 18 kD fragment. Nonetheless, the two fragments remain non-ionically bound and exhibit chondroitinase I activity.

When this procedure is repeated with lysed host cells carrying a recombinant plasmid encoding chondroitinase I, significantly poorer results are obtained. Less than 10% of the chondroitinase I binds to the cation exchange column at standard stringent conditions of pH 7.2, 20 mM sodium phosphate.

,Under less stringent binding conditions of pH 6.8 aL 5 mM phosphate, an improvement of binding with one batch of material to 60-90% is observed.

However, elution of the recombinant protein with the NaC1 gradient gives a broad activity peak, rather than ffc WO 94/25567 PCTUS94/04495 78 a sharp peak (see Figure This indicates the product is heterogeneous. Furthermore, in subsequent fermentation batches, the recombinant enzyme binds poorly even using the less stringent binding conditions. Batches that bind poorly are not completely processed, so their overall recovery is not quantified.

Example 9 First Method For The Isolation And Purification Of Recombinant Chondroitinase I According To This Invention As a first step, the host cells which express the recombinant chondroitinase I enzyme are homogenized to lyse the cells. This releases the enzyme into the supernatant.

In one embodiment of this invention, the supernatant is first subjected to diafiltration to remove salts and other small molecules. An example of a suitable filter is a spiral wound 30 kD filter made by Amicon (Beverly, MA). However, this step only removes the free, but not the bound form of the negatively charged molecules. The bound form of these charged species is removed by passing the supernatant through a strong, high capacity anion exchange resincontaining column. An example of such a resin is the Macro-Prep M High Q resin (Bio-Rad, Melville, Other strong, high capacity anion exchange columns are also suitable. The negatively charged molecules bind to the column, while the enzyme passes through the column. It is also found that some unrelated, undesirable proteins also bind to the column.

Next, the eluate from the anion exchange column is directly loaded to a cation exchange resinp Ii

K--

I) r V.

WO 94/25567 PCT/US94/04495 79 containing column. Examples of such resins are the S- Sepharose M (Pharmacia, Piscataway, and the Macro-Prep M High S (Bio-Rad). Each of these two resin-containing columns has S03O ligands bound thereto in order to facilitate the exchange of cations. Other cation exchange columns are also suitable. The enzyme binds to the column and is then eluted with a solvent capable of releasing the enzyme from the column.

Any salt which increases the conductivity of the solution is suitable for elution. Examples of such salts include sodium salts, as well as potassium salts and ammonium salts. An aqueous sodium chloride solution of appropriate concentration is suitable. A gradient, such as 0 to 250 mM sodium chloride is acceptable, as is a step elution using 200 -M sodium chloride.

A sharp peak is seen in the s, _hloride gradient elution (Figure The improveient in enzyme yield over the prior method is striking. The recombinant chondroitinase I enzyme is recovered at a purity of 99% at a yield of 80-90%.

The purity of the protein iis measured by scanning the bands in SDS-PAGE gels. A 4-20% gradient of acrylamide is used in the development of the gels. The band(s) in each lane of the gel is scanned using the procedure described above.

[1 ft WO 94/25567 PCT/US94/04495 79 containing column. Examples of such resins are the S- SepharoseT (Pharmacia, Piacataway, and the Macro-Prep M High S (Bio-Rad). Each of these two resin-containing columns has SO 3 ligands bound thereto in order to facilitate the exchange of cations. Other cation exchange columns are also suitable. The enzyme binds to the column and is then eluted with a solvent capable of releasing the enzyme from the column.

Any salt which increases the conductivity of the solution is suitable for elution. Examples of such salts include sodium salts, as well as potassium salts and ammonium salts. An aqueous sodium chloride solution of appropriate concentration is suitable. A gradient, such as 0 to 250 mM sodium chloride is acceptable, as is a step elution using 200 mM sodium chloride.

The purity of the protein is measured by scanning the bands in SDS-PAGE gels. A 4-20% gradient of acrylamide is used in the development of the gels. The band(s) in each lane of the gel is scanned using the procedure described above.

These improvements are related directly to the increase in binding of the enzyme to the cation exchange column which results from first using the S 30 anion exchange column. In comparative experiments, I when tly the cation exchange column is used, only 1% of the enzyme binds to the column. However, when the anion exchange column is used first, over 95% of the j enzyme binds to the column.

4r LrJ-;LCIL W1 94 WO 94/25567 PCTIUS94/04495

I

r WO 94/25567 PCT/US94/04495 Example Second Method For The Isolation And Purification Of Recombinant Chondroitinase I According To This Invention In the second embodiment of this aspect of the invention, two additional steps are inserted in the method before the diafiltration step of the first embodiment. The supernatant is treated with an acidic solution, such as 1 M acetic acid, bringing the supernatant to a final pH of 4.5, to precipitate out the desired enzyme. The pellet is obtained by centrifugation at 5,000 x g for 20 minutes. The pellet is then dissolved in an alkali solution, such as 20-30 mM NaOH, bringing it to a final pH of 9.8.

The solution is then subjected to the diafiltration and subsequent steps of the first embodiment of this invention.

In comparative experiments with the second embodiment of this invention, when only the cation exchange column is used, only 5% of the enzyme binds to the column. However, when the anion exchange column is used first, essentially 100% of the enzyme binds to the column. The second embodiment provides comparable enzyme purity and yield to the first embodiment of the invention.

Acid precipitation removes proteins that remain soluble; however, these proteins are removed anyway by the cation and anion exchange steps that follow (although smaller columns may be used). An advantage of the acid precipitation step is that the sample volume is decreased to about 20% of the original volume after dissolution, and hence can be handled more easily on a large scale. However, the additional acid precipitation and alkali dissolution ,r WO 94/25567 PCT/US94/04495 WO 94/25567 PCT/US94/04495 -81 steps of the second embodiment mean that the second embodiment is more time consuming than the first embodiment. On a manufacturing scale, the marginal improvements in purity and yield provided by the second embodiment may be outweighed by the simpler procedure of the first embodiment, which still provides highly pure enzyme at high yields.

The high purity of the recombinant enzyme obtained by the two embodiments of this invention is depicted in Figure 4. A single sharp band is seen in the SDS-PAGE gel photograph: Lane 1 is the enzyme using the method of the first embodiment; Lane 2 is the enzyme using the method of the second embodiment; Lane 3 represents the supernatant from the host cell prior to purification many other proteins are present; and Lane 4 represents molecular weight standards.

Example 11 Site-Specific Mutagenesis Of A Fragment Encoding The N-Terminal Recion Of Chondroitinase II The approach taken in the case of the chondroitinase II gene is to modify the naturallyoccurring ATG initiation codon to embed it within an NdeI site. This results in a construction in which the signal peptide is retained, such that the i expressed gene is processed and secreted to yield the l mature native enzyme structure that has a leucine S 30 residue at the N-terminus. The mutagenized bases are upstream of the coding region.

The method used for this site-specific alteration is that described above for the expression i of the chondroitinase I gene and is based on the work of Kunkel (15) using the Muta-Gene tm In Vitro w1 w 'WO 94/25567 PCT/US94/04495 4E

I

WO 94/25567 PCT/US94/04495 82 Mutagenesis Kit Version 2 (Bio-Rad, Melville, In this procedure, the target DNA to be mutagenized is first cloned into a suitable M13-derived vector to generate single-stranded DNA. This recombinant phage is replicated in the E. coli host strain CJ236 (Bio- Rad), a male strain that carries the dut and ung alleles. The combination of these two mutations, dut (duTPase) and ung (uracil-N-glycosylase), results in the incorporation of some uracil, rather than thymine, residues into the DNA synthesized in this organism.

Single-stranded template is then isolated after propagation on CJ236 and the appropriate mutagenic, synthetic oligonucleotide (SEQ ID NO:41) is annealed to this DNA.

The newly synthesized DNA (containing the desired base changes) therefore does not contain uracil while the template DNA (with the native sequences) does.

Transformation of a non-mutant (with respect to the ung and dut alleles) male E. coli strain yields phage progeny that are primarily derived from the mutagenized strand synthesized in vitro as a result of the inactivation of the uracil-containing template strand.

In this specific case, the fragment to be cloned for the mutagenesis is a MunI-EcoRI fragment that spans the region between nucleotides 2943 to 3980 (SEQ ID NOS:1 and 39). The DNA digested to obtain this fragment is designated LP 2 783. This plasmid is construcad in the same way as LP 2 786 (described in Example except that a HindIII linker is inserted i: i:! I i i.; i:' 'i

V-

i: ji i i: i, i i p:

I

1::f i k:; %i *WO 94/25567 PCT/US94/04495 "1 WO 94/25567 PCT/US94/04495 83 into the EcoRV deletion of LP 2 776 rather than the EcoRI linker. This MunI-EcoRI fragment is ligated into the unique EcoRI site of LP 2 941, an M13mpl9 derivative in which the normal polylinker is replaced with that found in the plasmid vector pNEB193 (New England Biolabs, Beverly MA). The four base overhang produced by MunI digestion can be ligated to an EcoRI site, but the resulting recombinant sequence cannot be digested by either enzyme. The EcoRI digested LP 2 941 is also dephosphorylated with calf intestinal alkaline phosphatase (Boehringer Mannheim, Indianapolis IN) prior to gel purification and use.

The ligated DNA mixture is used to infect the male E. coli strain MV1190 and the plaques obtained are picked to 0.5 ml. of SM buffer and the phage allowed to elute by diffusion. These are then used to infect 10 ml. cultures of MV1190 and grown overnight. The cultures are centrifuged and the pellets used for the isolation of the double-stranded replicative forms of the recombinant virus. The supernatants, which contai the corresponding phage particles, are stored under refrigeration until needed. The orientation of the cloned fragment is determined by digestion of the replicative form DNA and HindIII, because there is one site within the polylinker and a second, aymmetrically placed site (SEQ ID NOS:1 and 39, nucleotides 3326-3331) within the above MunI-EcoRI fragment.

Once the desired orientation is identified, the corresponding phage-containing supernatant is serially diluted, used to infect the E. coli strain CJ236, and then plated to obtain single plaques which are picked and eluted as above. One of these is then used (o infect CJ236 and another 10 ml culture grown and the single-stranded DNA is isolated from the '1* '-7 _iL, WO 94/25567 PCT/US94/04495 1 WO 94/25567 PCTI[U94/04495 84 phage-containing supernatant using QiaexM columns and materials and methods recommended by the manufacturer (Qiagen, Chatsworth, CA) and finally resuspended in a volume of 0.01 ml. The recombinant phage are grown on CJ236 (dut" uncr) for two rounds in order to maximize the accumulation of uracil residues in the template and strand prior to the actual site-specific mutagenesis.

The mutagenic oligonucleotide used is obtained from Bio-Synthesis (Denton, TX) and has the following sequence: 5'-ATT-TGC-AGG-AAA-TCT-GCA-TAT-GCT-AAT-AAA-AAA-CCC-3' (SEQ ID NO:41) This sequence differs from the corresponding region of SEQ ID NOS:1 and 39 in that an AT sequence (base pairs 3235 and 3236) is replaced by a CA sequence which creates the desired NdeI sequence (CATATG) at the start of the presumed leader sequence for the chondroitinase II gene. One optical density unit of this oligonucleotide is dissolved in 0.46 ml.

of TE 7.4 (0.01M TrisHCl, pH 7.8-0.001M EDTA, pH yielding an oligonucleotide concentration of approximately 6 pmol/Al. Three hundred picomoles of this oligonucleotide are phosphorylated in a 0.1 ml reaction containing 0.05 M TrisHCl, pH 7.8, 0.01 M MgCI 2 0.02M dithiothreitol, 0.001 M ATP, 25 14g/ml bovine serum albumin and 100 units of T4 polynucleotide kinase (New England Biolabs) at 37 0

C

for 30 minutes, followed by incubation at 750 for minutes to inactivate the enzyme. The phosphorylated oligonucleotide is then stored frozen at -200 at a concentration of approximately 3 pmoles/Al.

For the site-spi.-fic mutagenesis, 1 gl (3 itt 'i a iS WO 94/25567 PCT/US94/04495 85 pmole) of the mutagenic oligonucleotide is mixed with 6 1 l of the single-stranded DNA prepared above in a Al volume of 0.02 M TrisHCl, pH 7.4, 0.002 M MgCl 2 0.05 M NaCl. The oligonucleotide is annealed to this template by first incubating the sample at 70 0 C for minutes and then cooling this sample at 25°C over a minute period in a DNA Thermal Cycler T (Perkin-Elmer Cetus/Norwalk, CT). The sample is maintained at for another 5 minutes before being cooled to 20 0 C and finally transferred to an ice bath.

The annealed primer is then extended after the addition of 1 Al of 10X synthesis buffer (Bio-Rad; containing 0.005 M of each of the dNTP's, 0.01 M ATP, 0,1 M TrisHCl, pH 7.4, 0.05 M MgC12, 0.02 M DTT). One Al of T4 DNA ligase (3 units/gl Bio-Rad) and 1 Ml of T7 DNA polymerase (0.5 units/gl Bio-Rad). The in vitro DNA synthesis is carried out on ice for minutes, at 11 0 C for ten minutes, and at 37 0 C for minutes prior to transfer to ice.

This sample is used directly to transform the male E. coli host MV1190 (dut' unq) and the resulting plaques, containing the site-specifically mutagenized phage, are obtained, picked and eluted as described above. Aliquots of these phage stocks are used in infect 10 ml. cultures of MV1190 and allowed to grow overnight. The cultures are centrifuged and the replicative forms of the recombinant phage are isolated using Qiaex" columns and methods recommended by the manufacturer (Qiagen, Chatsworth CA). The DNA isolated is resuspended in 0.1 ml of TE 7.4. Initial digestions of a portion of each of these DNA samples with NdeI reveals that at least four appeared to have acquired a new Ndel site, indicate that the sitespecific mutagenesis is successful. Consequently, larger samples of these four clones (0.04 ml each) are fir 'WO 94/25567 PCT1US94104495 'WO 94/25567 PCT/US94/04495 -86 digested with NdeI and EcoRI and fractionated on a 1.4% agarose gel run in a Tris-acetate-EDTA buffer system.

The desired approximately 740 base pair fragment is observed in each case and this band is excised from each pattern. The four samples are then combined and the DNA extracted from the gel using a QiaexM resin and buffers according to the manufacturer's reconmmendations (Qiagen, Chatsworth CA) and resuspended in 0.05 ml. of TE, pH 7.4. This isolated, site-specifically mutagenized N-terminal coding region of the cloned P.'vulgaris gene for the chondroitinase II gene is then subcloned into the plasmid pNEB193 (New England Biolabs, Beverly MA) between the (dephosphorylated) unique NdeI and EcoRI sites present in this plasmid. After transformation of the E. coli host strain 294, 10 ml cultures derived from the individual transforma.ts are grown and the recombinant plasmid DNA isolated as above. The DNA sample from one of the positive clones is designated m#15-5712. This sample represents the modified Nterminal region that is to be joined to the C-terminal coding region for the chondroitinase II gene, which is described in Example 12.

Example 12 Isolation, Characterization And DNA Sequence Analysis Of A Fragment Encoding The C-terminal Region Of i Chondroitinase II The DNA sequence contained in SEQ ID NO: 39 j indicates that chondroitinase II is encoded by a 1 region that is downstream of that for chondroitinase I. This information is derived from a portion of a 1 35 kilobase NsiI fragment of P. vulqaris that is .r I r r Jb

I

77

I

WO 94/25567 PCT/US94/04495 87 subcloned originally from a cosmid clone designated

LP

2 751. The combination of the DNA sequencing and the restriction map in Figure 1 reveals that the chondroitinase II coding region initiates to the "left" of the EcoRI site that lies within the P.

vulgaris derived DNA and proceeds toward the NsiI site at the "right" end of the fragment depicted in Figure 1. Therefore, this restriction map should be expanded to the "right" to find a suitable fragment that will include the C-terminal coding region for the chondroitinase II gene.

Digestion of LP 2 751 reveals three EcoRI fragments of approximately 20 kb, 13 kb, and 10 kb, and indicates that there are three EcoRI sites within

LP

2 751. Because there are two EcoRI sites that bracket the cloning site, the conclusion is that there is one EcoRI site within the cloned P. vulgaris DNA in this clone. Furthermore, since the approximately 13 kb fragment corresponds to the size of the cosmid vector Rer se, this unique EcoRI site lies between the approximately 20 kb and the approximately 10 kb fragments noted above. Since it is known that the entire coding region for chondroitinase I, as well as the N-terminal coding region for chondroitinase II, are both contained within the approximately 10 kb NsiI fragment, restriction digestions that compare the patterns obtained among the cloned 10 kb NsiI (present in the recombinant plasmid designated LP 2 770) and gelpurified samples of the above approximately 20 kb EcoRI and approximately 10 kb EcoRI fragments indicate which of these EcoRI fragments contain the chondroitinase I coding sequence and, therefore by deduction, which will carry the C-terminal coding region for chondroitinase II. Consequently, digestions are carried using the restriction enzymes i i iii

IUCI

i WO 94/25567 PCTIUS94/0449a -I i Wh 4 WO 94/25567 PCT/US94/04495 88 AflIII, ClaI, EcoRV, and HindIII each of which has been noted by Applicants to yield eight to ten fragments upon digestion of the original cosmid clone designated LP 2 751.

The recombinant molecule carrying the subcloned approximately 10 kb NsiI fragment (LP 2 770) and the individually gel-purified approximately 20 kb EcoRI and approximately 10 kb EcoRI fragments are digested with each of these enzymes to yield patterns of fragments that are compared. These digestions reveal that the approximately 20 kb EcoRI and the

LP

2 770 patterns have a number of fragments in common.

This indicates that the chondroitinase I gene and the N-terminal coding region of the chondroitinase II gene are contained within the larger EcoRI fragment and, therefore, the C-terminal coding region for the chondroitinase II gene is on the approximately 10 kb EcoRI fragment.

The approximately 10 kb 7coRI fragment is cloned into the unique EcoRI site of the derivative of pNEB193 (New England Biolabs, Beverly MA) that is designated lacpoA pNEB193. This vector carries two deletions relative to the parental molecule pNEB193.

The first removes the sequences between the unique NdeI and EcoRI sites, retaining the EcoRI site but removing the NdeI site (and one of the two PvuII sites). The second deletion removes the region between the HindIII site at the other end of the polylinker and the (now unique) PvuII site, maintaining the HindIII site, while removing the PvuII site. The recombinant DNA molecule carrying the subcloned approximately 10 kb EcoRI fragment in the vector lacdpo pNEB193 is designated LP 2 1263. The orientation of the 112 kD C-terminal coding region within LP 2 1263 is determined by restriction enzyme i

I

;r

I

i D~I.IC1-T1).- ~t c" il_ iT,::c- 'WO 94/25567 PCT/US94/04495 i w WO 94/25567 PCT/US94/04495 89 mapping. The results indicate that this region is positioned so as to proceed from the EcoRI site (defined as the "left" end) toward the HindIII site at the other end of the polylinker. Similarly, unique restriction sites for SmaI, XhoI, NocI and NdeI are found approximately 2.6, 4.6, 5.8 and 8.5 kb from the "left" end of the approximately 10 kb EcoRI fragment.

Digestion of LP 2 1263 with SmaI, therefore, deletes a downstream region of approximately 7.4 kb from the site within the cloned P. vulqaris DNA to the second site within the polylinker region, leaving approximately 2.6 kb which should be enough to encode the missing region of the chondroitinase II gene.

This construction also "places" a BamHI site (present in the polylinker) downstream of the coding region for the chondroitinase II gene. This recombinant DNA molecule which carries the chondroitinase II gene from the EcoRI site to (and presumably just beyond) the termination codon for this gene has been designated m#25-5712.

DNA sequence analysis is initiated on the approximately 10 kb EcoRI fragment derived from LP 2 1263 and is completed after the assembly of the intact gene for chondroitinase II. The materials and methods for the DNA sequencing of this fragment are essentially the same as those used for the approximately 4 kb fragment containing the gene for chondroitinase I.

Random fragments are derived from this approximately kb EcoRI fragment by self-ligating the DNA and then fragmenting the polymerized DNA by sonication as well as by partial digestion with the restriction enzymes Sau3A or MseI. These pieces are then eventually cloned into M13 derived vectors and the singlestranded recombinant molecules sequenced using the standard protocols described above.

i I- t WO 94/25567 PCTIUS94/04495 -rnI

L

i r i I WO 94/25567 PCTIUS94/04495 90 Finally, with the two set of sequence data available, an approximately 300 base-pair BclI fragment is identified that is predicted to contain the EcoRI site that is the junction between the two P.

vulgaris fragments of approximately 20 kb and approximately 10 kb obtained by digestion with EcoRI.

This small fragment is sequenced in both directions to verify the nucleotide sequence through this junction point used in the constructions described below.

Example 13 Assembly Of The Entire Site-Specifically Mutagenized Gene For Chondroitinase II During the DNA sequencing, the molecule designated m#25-5712 is digested with EcoRI and BamHI.

This releases a DNA fragment of approximately 2.6 kb.

Similarly, the construction designated m#15-5712 is digested with EcoRI and BamHI and then dephosphorylated prior to purification by gel electrophoresis. The latter molecule therefore carries the N-terminal coding region of the chondroitinase II gene from the ATG initiation codon (now present as part of an NdeI site from the sitespecific mutagenesis) to the EcoRI site.

These two fragments are ligated and then the mixture used to transform the E. coli strain 294.

Plasmid DNA is isolated from the transformants and positive clones identified. Restriction digestion with Ndel and BamHI releases the desired fragment encoding the chondroitinase II gene (SEQ ID NO:39, nucleotides 3235-6518, followed by 14 nucleotides derived from the polylinker, which includes a BamHI site). This fragment is then ligated to the expression vector pET9A (Novagen, Madison, WI) jj WO 94/25567 PCT/US94/04495 i i 1 0, 4 I-I I II -I -1 WO 94/25567 PCT/US94/04495 91 described in the expression of the chondroitinase I gene.

The coding region of the chondroitinase II gene includes nucleotides 3238-6276 of the SEQ ID NO: 39, which encodes 1013 amino acids (SEQ ID NO:40). Of this region, nucleotides 3238-3306 encode the 23 amino acid signal peptide (SLQ ID NO:40, amino acids 1-23), while nucleotides 3307-6276 encode the mature 990 amino acid chondroitinase II protein (SEQ ID amino acids 24-1013).

Restriction analysis with four enzymes of the region spanning both chondroitinase genes and flanking sequences thereof reveals the following restriction sites: Enzyme Nucleotide Enzyme Nucleotide EcoRI 2 MunI 4510 HindIII 2046 HindIII 4530 MunI 2904 MunI 5176 MunI 2943 HindIII 5427 HindIII 3326 Smal 6515 EcoRI 3974 In addition, restriction analysis with Sau3AI reveals a multiplicity of sites, including those at SEQ ID NO:39, nucleotides 212, 602, 890, 1042, 1181, 1241, 1442, 1505, 1746, 2330, 2363, 2701, 2705, 2920, 3697, 3708, 3745, 3868, 4087, 4800, 4872, 5565, 5635, 5860, 6058 and 6467.

One of the recombinant molecules (the chondroitinase II gene inserted into pET9A) obtained in this experiment is grown in large scale (0.5 liter) and the expression system containing the chondroitinase II gene isolated and designated LP 2 1359.

An aliquot of this DNA is used to transform the expression host BL21(DE3)/pLysS described in the i"/ b~i

I,

A"

WO 94/25567 PCTIUS9404495 I .1 WO 94/25567 PCT/US94/04495 92 expression of the chondroitinase I gene. The resulting strain is designated TD112 and is used for large-scale fermentation and isolation of the chondroitinase II enzyme.

A fermentation at a 10 liter scale carried out with this E. coli strain containing the plasmid expressing the chondroitinase II protein, provides a maximum chondroitinase II titer of approximately 0.3 mg/ml, which is approximately 25 times that of the approximately 0.012 mg/ml obtained from the native P.

vulqaris fermentation process for chondroitinase II.

Example 14 First Method For The Isolation And Purification Of Recombinant Chondroitinase II According To This Invention The initial part of this method is the same as that used for the recombinant chondroitinase I enzyme. As a first step, the host cells which express the recombinant chondroitinase II enzyme are homogenized to lyse the cells. This releases the enzyme into the supernatant.

In one embodiment of this invention, the supernatant is first subjected to diafiltration to remove salts and other small molecules. An example of a suitable filter is a spiral wound 30 kD filter made by Amicon (Beverly, MA). However, this step only removes the free, but aot the bound form of the negatively charged molecules. The bound form of these charged species is removed by passing the supernatant (see the SDS-PAGE gel depicted in Figure 5, lane 1) through a strong, high capacity anion exchange resincontaining column. An example of such a resin is the Macro-Prepm High Q resin (Bio-Rad, Melville,

K

i ,lj i:..i 1 WO 94/25567 PCT/US94/04495 93 Other strong, high capacity anion exchange columns are also suitable. The negatively charged molecules bind to the column, while the enzyme passes through the column with approximately 90% recovery of the enzyme.

It is also found that some unrelated, undesirable proteins also bind to the column.

Next, the eluate from the anion exchange column (Figure 5, lane 2) is directly loaded to a cation exchange resin-containing column. Examples of such resins are the S-Sepharose (Pharmacia, Piscataway, and the Macro-Prep M High S (Bio- Rad). Each of these two resin-containing columns has S03- ligands bound thereto in order to facilitate the exchange of cations. Other cation exchange columns are also suitable. The enzyme binds to the column, while a significant portion of contaminating proteins elute unbound.

At this point, the method diverges from that used for the chondroitinase I protein. Instead of eluting the protein with a non-specific salt solution capable of releasing the enzyme from the cation exchange column, a specific elution using a solution containing chondroitin sulfate is used. A 1% concentration of chondroitin sulfate is used; however, a gradient of this solvent is also acceptable. The specific chondroitin sulfate solution is preferred to the non-specific salt solution because the recombinant chondroitinase II protein is expressed at levels approximately several-fold lower than the reconmwinant chondroitinase I protein; therefore, a more powerful and selective solution is necessary in order to obtain a final chondroitinase II product of a purity equivalent to that obtained for the chondroitinase I protein.

The cation exchange column is next"washed 4i9a

I

t

I:

i j ji ii

I

B

I:

n*" i c; r i 7 Lii

I'

WO 94/25567 PCT[US94/04495 94 with a phosphate buffer, pH 7.0, to elute unbound proteins, followed by washing with borate buffer, pH to elute loosely bound contaminating proteins and to increase the pH of the resin to that required for the optimal elution of the chondroitinase II protein using the substrate, chondroitin sulfate.

Next, a 1% solution of chondroitin sulfate in water, adjusted to pH 9.0, is used to elute the chondroitinase II protein, as a sharp peak (recovery and at a high purity of approximately 95% (Figure lane However, the chondroitin sulfate has an affinity for the chondroitinase II protein which is stronger than its affinity for the resin of the column, and therefore the chondroitin sulfate coelutes with the protein. This ensures that only protein which recognizes chondroitin sulfate is eluted, which is desirable, but also means that an additional process step is necessary to separate the chondroitin sulfate from the chondroitinase II protein.

In this separation step, the eluate is adjusted to pH 7.0 and is loaded as is onto an anion exchange resin-containing column, such as the Macro- Prep T High Q resin. The column is washed with a 20 mM phosphate buffer, pH 6.8. The chondroitin sulfate binds to the column, while the chondroitinase II protein flows through in the unbound pool with greater than 95% recovery. At this point, the protein is pure, except for the presence of a single minor contaminant of approximately 37 kD (Figure 5, lanes 4 and The contaminant may be a breakdown product of the chondroitinase II protein.

This contaminant is effectively removed by a crytallization step. The eluate from the anion exchaage column is concentrated to 15 mg/ml protein i

'D

'E

r.i

I:

WO 94/25567 PCTfUS94/04495 4, 4

I

'WO 94/25567 PCT/US94/04495 95 using an Amicon stirred cell with a 30 kD cutoff. The solution is maintained at 4 0 C for several days to crystallize out the pure chondroitinase II protein.

The supernatant contains the 37 kD contaminant (Figure lane Centrifugation causes the crystals to form a pellet, while the supernatant with the 37 kD contaminant is removed by pipetting, and the crystals washed twice with water. After the first wash, some of the contaminant remains (Figure 5, lane but after the second wash, only the chondroitinase II protein is visible (Figure 5, lane The washed crystals are redissolved in water and exhibit a single protein band on SDS-PAGE, with a purity of greater than 99% (Figure 5, lane Example Second Method For The Isolation And Purification Of Recombinant Chondroitinase II According To This Invention In the second embodiment of this aspect of the invention, two additional steps are inserted in the method for purifying the chondroitinase II enzyme before the diafiltration step of the first embodiment.

The supernatant is treated with an acidic solution, such as 1 M acetic acid, bringing the supernatant to a final pH of 4.5, to precipitate out the desired enzyme. The pellet is obtained by centrifugation at 5,000 x g for 20 minutes. The pellet is then dissolved in an alkali solution, such as 20-30 mM NaOH, bringing it to a final pH of 9.8. The solution is then subjected to the diafiltration and subsequent steps of the first embodiment of this aspect of the invention.

:o i: t R1' 1 this fragment is designated Lp 2 783. This plasmid is constructed in the same way as Lp 2 786 (described in Example except that a HindIII linker is inserted WO 94/25567 PCTIUS94/04495 96 1 243, 1523-1535 2.

Bib).iocrraph Yamagata, et J. Biol. Chem., (1968).

Kikuchi; et U.S. Patent Numnber 5,198,355.

3. Brown, M. U.S. Patent Num~ber 4,696,816.

4. Hageman, G. U.S. Patent Numiber 5,292,509.

Malitschek, and Scharti, M., Biotechniques, 7, 177-178 (1991).

6. Sanger, et Proc. Natl.. Acad.

Sci. USA, 74, 5463-5467 (1977).

7. Inmis, M. and Gelfand, D. H., "Optimization of PC~sI', pages 3-12 in PCR Protocols. A Guide to Methods and Applications, Academic Press, New York, N.Y. (1990).

8. Southern, J. 98, 503- 517 (1975).

9. Studier, F. et Methods in EnzvMo).og, 185 60-89 (1990).

Studier, F. and Moffatt, B. J.

Mol. Biol., 189, i13-130 (1986); Moffatt, B. and Studier, F. Cell, 49, 221-227 (1987).

11. Sambrook, et Molecular Clonincy: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

(1989).

12. Clark, Nucleic Acids Research, 1- 9677 (1988).

13. Yanisch-Perron, et Gene, 3, 103-119 (1985).

14. Devereaux, et Nucleic Acids Research, 12, 387-3 95 (1964).

Kunkel., T. Proc. Nat).. Acad. Sci., I. "I WO 94/25567 PCT/US94/04495 A k WVO 94/25567 PCTIUS94/04495 .97 USA, L2 488-492 (1985).

16. Studier, F. J. Mol. Biol., 219, 1 37- 44 (1991).

17. Miller, J. Experiments in Molecular Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. (1972).

L

A

WO 94125567PCUS4O 9

I

WO 94/25567 PCT/US94/04495 98 SEQUENCE LISTING GENERAL INFORMATION: APPLICANT: American Cyanamid Company (ii) TITLE OF INVENTION: Cloning And Expression Of The Chondroitinase I and II Genes From P. Vulgaris (iii) NUMBER OF SEQUENCES: 41 (iv) CORRESPONDENCE ADDRESS: ADDRESSEE: American Cyanamid Company STREET: One Cyanamid Plaza CITY: Wayne STATE: New Jersey COUNTRY: U.S.A.

ZIP: 07470-8426 COMPUTER READABLE FORM: MEDIUM TYPE: Floppy disk COMPUTER: IBM PC compatible OPERATING SYSTEM: PC-DOS/MS-DOS SOFTWARE: PatentIn Release Version #1.25 (vi) CURRENT APPLICATION DATA: APPLICATION NUMBER: PCT/US94/ FILING DATE:

CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION: NAME: Gordon, Alan M.

REGISTRATION NUMBER: 30,637 REFERENCE/DOCKET NUMBER: 31,726-00/PCT (ix) TELECOMMUNICATION INFORMATION: TELEPHONE: 201-831-3244 TELEFAX: 201-831-3305 INFORMATION FOR SEQ ID NO:1: SEQUENCE CHARACTERISTICS: LENGTH: 3980 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (ix) FEATURE: NAME/KEY: CDS LOCATION: 119..3181 (xi) SEQUENCE DESCRIPTION: SEQ ID NOl: SSUBSTITUTE SHEET (RULE 26)

A

'WO 94/25567 -99 GGAATTCCAT CACTCAATCA TTAAATTTAG GCACAACGAT AATTTAATGA AGGACGCATT GGTTTCACTG TTAGCCAGCG ATG CCG ATA TTT CGT TTT ACT GCA CTT GCA ATG Ment Pro Ile Phe Arg Phe Thr Ala Leu Ala Met PCT[US94/04495 GGGCTATCAG CGTTATGACA TTTCTAAGGA, GAAAA.ATA ACA TTG GGG CTA TTA Thr Leu Gly Lou Leu TCA GCG Ber Ala CCT AAA Pro Lys CCA TTA Pro Lou GAT AAA Asp Lys GGT GOT Oly Oly AAA GAA Lys Glu TTT TG Ph. Trp TTC GGA Phe Gly 130 GTA AU~ Val Lyen 145 AAC GA7 Aen An; TCT GMI Ser Asp GAT AGI Asp Sex TAT ATC Tyr Ili 21C CCT TAT AAC Pro Tyr Ann 20 AAT CTO ATG Ann Lou Met 35 GCA GAC TTC Ala Asp Phe CGT AGC ATT Arg Sar Ile AGT AGC TTT Bar Bar Ph.

GCA TCT AAA Ala Bar Lye 100 CTT TAC AAT Lou Tyr Ann 115 GAA AAA CTC Giu Lys Lou TTA GAT TTC Lou Asp Ph.

CTT GAA AAT Lou Giu An 165 IGOT ACT CAA Gly Thr Gin 160 ATT CGT TTT Ile Arg Ph.

195 GAC CGT ATT kAsp Arg Ile GCG ATG Ala Met CAG TCA Gin Ser TCA TCA Bar Ber 55 ATG GGA Met Gly 70 ACT TTA Thr Lou GCA TGG Ala Trp, GAA AA Giu Lys ATT TCA Ile Bar 135 ACT GGC Thr Gly I50 CGA GAG Arg Giu GAC AGC Asp Bar AAA GCG Lys Ala ATG TTT Not Ph.

215

GCA

Ala

GAA

Giu 40

GAT

Asp

AAC

An

CAT

His

GGA

Gly

CCG

Pro 120

ACC

Thr

TGG

Trp

ATG

met

CCT

Pro 200

TCT

Bar

GCC

Ala 25

ATT

Ile

AAA

Lys

CAA

Ulin

AA

Lye

CGC

Arg 105

ATT

Ile

AGT

Ber

CGT

Arg

ACC

Thr

G

Gly 185

TCT

Bar

GTC

Val

ACC

Thr

TAC

Tyr

AAC

An

TCT

Bar

AAA

Lys 90

TCA

Bar

OAT

Asp

GAG

Glu

GCT

Aia

TTA

Lou 170

CGT

Arg

AAT

An

GAT

Asp

AGC

Ser

CAT

Hisn

TCA

Ser

CTT

Lou 75

CTG

Lou

TCT

Bar

GT

Gly

GCT

Ala

GTG

Val 155

AAT

An

TCT

Bar

GTG

Val

GAT

Asp

AAT

An

TTT

Phe

ATA

Ile

TTA

Lau

ATT

Ile

ACC

Thr

TAT

Tyr

CAG

Gin 140 00K Gly

GCA

Ala

TTA

Lou

AGT

Ser

OCT

Ala 220 CCT OCA Pro Ala 30 GCA CAA Ala ;l r Lou Thl.

TGG AA Trp Lys GTC CCC Val Pro CCC GTT Pro Val 110 CTT ACT Lau Thr 125 GCA GGC Ala Gly OTC TCT Val Bar ACC AAT Thr An GGT GCT Gly Ala 190 CAG GOT Gin Gly 205 CGC TAC Arg Tyr TTT GAT Phe Asp AAT AAC Ann Ann .TA TCT Lou Bar TOG AA Trp Lys ACC GAT Thr Asp TTC TCA Ph. Bar ATC GET Ile Asp TTT AAA Ph. Lys TTA AAT Lou Ann 160 ACC TCC Thr Bar 175 AAA GTC Lye Val GAA ATC Giu le CAA TGG Gin Trp CAA TTTZ Gin Phe 358 TCT GAT TAT CAA GTA AAA Sar Asp Tyr Gin Val Lys ACT CGC TTA TCA GAA CCT GAA ATT Thr Arg Lou Bar Giu Pro Gu. Ile SUBSTITUTE SHEET (RULE 26) WO 94/25567 WO 9425567PCT!{ ,94IO4495I region that is downstream of that for chondroitinase 1. This information is derived from a portion of a kilobase NsiI fragment of P. vulgaris that is

I

1 mk I. WO 94/25567 PCT1US94/04495 100 225 V CAC His

ATT

Ile

AAA

Lys

GAT

Asp

GGC

Gly 305

AAT

Ann

GGT

Gly

GAA

Glu

ATG

Met

GTG

Val 385

ACG

Thr

GTT

Val

GAT

Asp

TTA

Lou

AAG

Lys 465 GTA AAG Vai Lys CTT ATT Lou Ile 260 ACA AAC Thr Asn 275 GAT GCT Asp Ala CAT CTG His Lou AAC TCC Ann Sor TAC ACG Tyr Thr 340 GAT CCC Asp Pro 355 AAG CAT Lys His ACC CAT Thr His TTA ATG Lou Met GAT TCA Asp Ser 420 AAA GTA Lye Vai 435 CGC CAA Arg Gin ATC AAC Ile Aan

CCA

Pro 245

CGC

Arg

CTC

Lou

CTT

Lou

ATC

Ile

CAA

Gin 325

ACA

Thr

ACA

Thr

TTA

Lou

CAC

His

TCT

Ser 405

TTA

Lou

AGT

Ser

CAT

His

TTA

Lou 230 CAA CTA Gin Lou CAA CGT Gin Arg GCA TTA Ala Lu AAT ATT An Io 295 ACT GAT Thr Asp 310 GAT AAA Asp Lys TTA ATO Lou Met CAA AAG Gin Lys TTA GAT Lou Asp 375 TGG GGA Trp Gly 390 GAT GCA Asp Ala CTG TGG Lou Trp GCT GAT Ala Asp TTA GCC Lou Ala 455 GTT AAT Val Ann1 470 235 GTA ACA CCT Val Thr Pro 250 ATT AAT GAA Ilie Ann Giu 265 GAG AAT ATC GiU AnnIlie ACT TTA GCA Thr Lou Ala CAA ATC ATT Gin Iie Ile 315 CTA TTT GAT Lou Ph. Asp 330 AAT ATT AGC Ann Ilie Ser 345 CAA CTA AAG Gin Lou Lyse GGC TTT GTTJ Giy Ph. ValI AGT TCT CGT Ser Ser Arg 395 AAA GAP. GCGJ Lys Giu Alaj 410 TCA CGT GAG! Ser Arg Glu1 425 TCT CAT CTA Sor ks5p Louj TTA TTA CTA( Lou Lou Lou( TTC AGC CAT Ph. Ser Hist 475

GAP.

Giu

TTT

Ph.

AGC

Sor

AAT

Ann 300

ATT

Ile

AAT

An

CGT

Arg

CAG

Gin

AAA

Lyrs 380

TGG

Trp

AAC

An

TTT

Ph.

CAT

Asp

GAG

Giu 460

TAT

Tyr

TTA

Lou

GGA

Giy 270

TTA

Lou

GGA

Gly

CAA.

Gin

GTT

Val

TAT

Tyr 350

TAC

Tyr

AGT

Ser

TAT

Tyr

CAP.

Gin

AGT

Ser 430

TTC

Ph.

GAT

Asp

ACT

Thr 240 GCG GCC Ala Ala 255 GGT GAP.

Gly Giu AAA AGT Lye Ser ACG CAP.

Thr Gin CCA GAG Pro Giu 320 ATT TTA Ile Lou 335 GTG CTG Vai Lou TTA TTA Lou Lou GCT TTA Ala Lou ATT TCC Ile Sor 400 ACT CAA Thr Gin 415 AGT TTT Ser Ph.

AP.T ACC Aen Thr CAT CAP.

Asp Gin GGC GCA Gly Ala 480 1030 1078 1126 1174 1222 1270 1318 1366 1414 1462 1510 1558 .4 SUBSTITUTE SHEET (RULE 26) m

I

'WO 94125567 PCT/US94/04495 deduction, which will carry the C-terminal coding region for chondroitinase ii. Consequently, digestions are carried using the restriction enzymes i it

R

St r *r i 1,, 4 S 5.

WO 94/25567 PCTIUS94/449S 101

TTA

Lou

ACA

Thr

TTT

Ph.

TCA

Ser

GCG

Ala 545

CAC

His

TGG

Trp

ATT

Ii.

TTT

Ph.

TTT

Ph.

625

ACA

Thr

AAA

Lys

GTG

Val

GAT

Asp

GAC

Asp 705

OGA

Gly ACG CAA Thr Gin GCA TGG Ala Trp AAA AAT Lys Ann 515 GTG GGT Val Gly 530 TOG ATC Trp Ile CCT TTT Pro Ph.

CTT GCC Lou Ala TAT CTT Tyr Lou 595 OGA GAA Oly Olu 610 AAT GGC Aen Gly CTO AAA Lou Lys GA AAC Asp Ann AGT AAT Ser Ann 675 TOG AAT Trp Ann 690 TTA GAC Lou Asp TTT AGC Ph. Ser

GTG

Val

CGA

Arg 0CC Ala

OAA

Olu

TAC

Tyr

AAC

Ann

ATG

met 580 oco Ala

ACT

Thr

GGT

Oly

OCT

Ala

CGT

Arg 560

GGC

Gly

AGA

Arg

AGT

Ser

OGA

Gly

CCA

Pro 485

CAT

His

TCT

Ser

AGT

Ser

AGT

Ser

TCA

Ser 565

TCT

Ser

ATT

I1

ATT

Ile

GCT

Ala

TAT

Tyr 645

ITAT

Tyr

TCG

Sor

ATG

met

C'M

ACA

Thi

CCG

Pro

GAA

Olu

CAG

Gin

GGT

Oly

AAT

Asn 550

CCT

Pro

OCA

Ala

AGT

Ser

ACA

Thr

TTT

Ph.

630

AAC

Ann

GGC

Gly

C.AG

Gin

CA&

Gln

AAA

Lys 710

TCA

Ser GGT GGT Gly Gly GGC AAC Gly Ann CTT ATT Lou lie 520 TGG AAT Trp Ann 535 CCA OAA Pro Glu TCG TTA Bar Lou AAA TCA Lys Ser OAT AA.A Asp Lys 600 CCA OCO Pro Ala 615 GT ATT Gly 1i ACC AAT Thr Ann COT TAC Arg Tyr CTT TCA Lou Ser 680 G0G OCA (Ala CCT CAT Pro His TCC CTT Ser Lou

AAL

Lys

TAT

Tyr 505

TAT

Tyr

AAC

Asn

OTT

Val

AAA

Lye

TCG

Ser 585

ACA

Thr

T!CT

Ser

CAT

His

OTT

Val

CAA

Gln 665

CAG

Gin

ACC

Thr

ACC

Thr

GAA

Glu

TTA

Lou

TAC

Tyr

CGC

Arg

ILAA

Lys 540

CCG

Pro

OCT

Ala

AAA

Lys

GAA

Glu

CAA

Gln 620

CAA

Gin

TCT

Ser

GOT

Gly

CAG

Gin

CAC

His 700

CAA

Gin

TAT

Tyr

CCT

Pro

TTC

Ph.

510

ACA

Thr

ATG

Met

OCA

Ala

GGC

Oly

OTT

Lau 590

ACT

Thr

TTC

Ph.

AAA

Lys

ATT

Ile

OCT

Ala 670

OAA

Glu

CCT

Pro

GGA

Gly

ATO

Met 1606 1654 1702 1750 1798 1846 1894 1942 1990 2038 2086 2134 2182 2230 2278 2326 L /i SUBSTITUTE SHEET (RULE 26) vector lacpoA pNEB193 is designated Lp 2 1263. The orientation of the 112 kD C-terminal coding region within LP'1263 is determined by restriction enzyme WO 94/25567 PCTIUS94/04495 102 TTC GAT Ph. Asp ACT GCG Thr Ala GGT AGC Gly Ser 770 TTA TTC Lou Ph.

785 OGA CAA Gly Gin GAT TOO Asp Trp OAA AA Giu Lys AAT COC Ann Arg 850 AOC ACT Bar Thr 865 OCO ACA.

Ala Thr AAT 000 Ann Oly CTC OAT Lau Asp TCA ATT Bar Ile 930 ATO ACT Not Thr, 945 OAT TTA

CTT

Lau

AAA

Lys 755

AAT

Ann

CAA.

Gin

AAO

Lys

TTA

Lau

OTA

Val 835

CAA.

Oin

CGC

Arg

CCT

Pro

TTA

Lou

AAA

Lys 91is

GA?.

Oiu

CAT

Z!is

AAT

725 ATT TAT le Tyr 740 AAG AGT Lys Bar ATA AAT Ile Ann CAT 0CC His Ala AT?. OAA Ile Oiu 805 ATT OAT le Asp 820 AAT OTA Ann Val CCO ACA Pro Thr CCC AAA Pro Lys GA. AAA? Oiu Lys 885 TAT CAG Tyr Oin 900 CTC AOC, Lou, Ser OAC AA Asp Lys CO?. CA?.

Arg Glz ATG ACT

CCC

Pro

OTA

Val

AGT

Ser

ATT

Ile 790

AAC

Ann

AGC

Bar

AOT

Bar

OAA

Oiu

OAT

Asp 870

ATG

Met

OTT

Val

,AAT

'kn

TG

Trp

AA

Ly8 950

CGC

GCC AAT Ala Ann TTA 0CC Lou Aia 760 AGT GAT AAA Ser Asp Lys 775 ACT CCA ACA Thr Pro Thr ATG CCT TAT Met Pro Tyr AIAT GGC AAT Ann Oly Ann 825 CGC CAA CAT Arg Oin His 840 GGA AAC TTT Oly Ann Ph.

855 0CC AOT TAT Ala Ser Tyr 00?. GAO ATO Gly Oiu Met CTT COT AAO Lou Arg Lys 905 OTA ACO GO?.

Val Thr Oly 920 ATC YAA AAO Ile Lys Lye 935 GAC ACT CTT Asp Thr Lou CAA. AAA OCA Gin Lys Ala 730 GAG COT Giu Arg OAT AAT Asp Ann AAT AAA Ann Lys TTA AAT Lou An 795 CAA ACA Gin Thr 810 GOT TAC Oly Tyr CAG OTT Gin Vai AOC TCO Bar 8cr GAG TAT Oiu Tyr 875 OCA CAA Ala Gin 890 OAT AAA Asp Lye TAT 0CC Tyr Ala OTT AAT Val Ann ATT OTC Ile Val 955 OCA AC~T Ala Thr 970 735 TTT OAT CCT AAT Ph. Asp Pro Ann 750 CAC TTA ATT TTT His Lou Ile Phe 765 AAT OTT OAA ACO Ann Val Oiu Thr 780 ACC CTT TOO ATT Thr Lou Trp Ile ACA CTT CA?. CAA Thr Lou Gin Gin 815 TTA ATT ACT CAA.

Lau Ilo Tir Gin 830 TCA OCO OAA AAT Ser Ala Oiu Ann 845 OCA TOO ATC tAAT Ala Trp 119 Anp 860 ATG OTC TTT TTA Met Vai Ph, Lou ALA? TTC COT GAA Lye Phe Arg Giu 895 GAC OTT CAT ATT Asp Val His Ile 910 TTT TAT CAO CCA Phe Tyr Gin Pro 925 AAA? CCT OCA ATT Lys Pro Ala Ile 940 AOT OCA OTT ACA Sar Ala Val Thr CCT OTC ACC ATC Pro Val Thr Ile 975

TTC

Phe

ATT

le

ACC

Thr

AAT

Ann 800

GGT

Oly

OCA

Ala Lys

CAC

His

OAT

Asp 880

AAT

Ann

ATT

le

GCA

GTG

Val

CCT

Pro 960

A?.T

Ann 2374 2422 2470 2518 2566 2614 2662 2710 2758 2806 2854 2902 2950 2998 3046

ON

p.

A

Asp Lou Ann Met Thr Arg 965 SUB STIITUTE SHEET (RULE 26) iia It WO 94/25567 PCTIUS94/04495 103 GTC ACG ATT Val Thr Ile AAM TAT CAG L~ys Tyr Gin 995 TTT GGT ATT Ph. (fly Ile 1010 AAT GGC AAA TGG CAA TCT GCT GAT Asn Gly Lyn Trp, Gin Ser Ala Asp 980 985 GTT TCT GGT GAT AAC ACT GAA CTG Val Scr Gly Asp Asn Thr Giu Leu 1000 CCA CAA OAA ATC MAA CTC TCG CCA Pro Gin Giu Ile L~ys Lou Ser Pro 1015 AAA MAT AGT GMA GTG Lys Asn Ser Glu Val 990 ACG TTT ACG AGT TAC Thr Phe Thr Ser Tyr 1.005 CTC CCT TGATTTAATC Leu Pro 1020 CTGATTATGC TAATMAAAA AAAAGAACGC

CCCTTTAGCC

CACTCTGTCT

CCTTACCACT

ACAATCACTC

TTACCAAGAT

AAAACCTCAA

TTTTAATGCT

GCAAGGCTCT

AACACTCTTT

TGACTATCAA

TGCATTATTG

TGAATTTCGC

AGrGAATTCC

TCTTGCGTTC

CACGCGGTTA

CATGAAGCTT

TCAAATAATA

AAATGGCAAT

GATAAAAATA

TCTTCCCCAT

GAACTTAATT

GCGACAGGTC

TTTGATCAAA

ACACCTTACG

ATGTACGATC

GATGACCAAA

CTTTTTTATT

CATTAAGCCT

TCGG1CGATAT

ATCAATTATC

ATCAACCACA

CAGCCACACC

TAACGTTAGC

TTACGGGGTG

AACTTGATCA

TCATCATGAG

TAAATAACGC

AG;ATGTTTCK

CAGAAATGGC

TGCAGGAAAT

CTGTTTATCA

TTATCTT'-TT

GCTAAGCAAA

AGCAACATTA

ACTCACTTTT

ATTTAAACAA

GCGAGGTATT

ATTAGTGATC

TGTACCGTTA

AGTAAACACG

AGCCCATTAC

TTCGATTTAT

TTACCCGCAC

GMGGTGAAT

CAGCATGCTA

ACACTAAATA

ATGATGTGGA

AATAATAAAA

GCTGTTCCTT

ACCGCTCCAA

GACP.ATCGTT

ATGGTTAGTA

CCTACTTTAA

CAGCGCTTTG

MGCATTACC

TACCCAATAC

AGATGOTQA

ATATTGTTAA

TTTATAATGA

TTGCACTAAG

TTCGTGATAT

ACCAAGCCGG

GGGCAGTACC

AAAACTGGAG

ACTTCGATAC

AATATTATCA

3094 3142 3191 3251 3311 33' i 3431 3491 3551 3611 3671 3731 3791 3851 3911 3971 3980 INFORMATION FOR SEQ ID NO:2:

'I

Wi SEQUENCE CHARACTERISTICS:- LENGTH: 1021 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION* SEQ ID NO:2: Met Pro Ile Ph. Arg Ph. Thr Ala Leu Ala Met Thr 1 510 Ser Ala Pro Tyr Ann A1,; Met Ala Ala Thr Bar An h'25 Leu Gly Leu Lou Pro Ala Ph. Asp

I,

171

I

Gin Asn An Pro Lys An Pro Lau Ala Met Gin Bar Glu Ile Tyr His Ph. Ala 40 Asp Ph. 8cr Ser Asp Lys Aun Bar le Lou Thr Lou Bar SUR1TITI~ Ii It pCT1US94/O 449 'WO 9415567

I

WO 94/25567 WO 9425567PCT[US94104495 104 so Asp Lys Arg Gly Gly Ser Lys Glu Ala Phe Trp Lou 115 Ph. Gly Giu 130 Val Lys Leu 145 Ann Asp Leu Ser Asp Gly Amp Ser Ile 195 Tyr Ile Asp 210 Sor Asp Tyr 225 His Ann Val Ile Asp Lau Lys Giu Thr 275 Amp Ph. Asp 290 Gly Arg His 305 Ann Lou An Gly Ann Tyr Glu Lys Asp 355 Met Thr Lys 370 Ser Ile Sor Phe Sor Lys 100 Tyr Ann Lys Lou Asp Phe Giu Ann 165 Thzr Gin Arg Ph.

Arg Ile Gin Val Lys Pro 245 Ile Arg 260 Ann Lou Ala Lou Lou Ile Ser Cdln 325 Thr Thr 340 Pro Thr Met 70 Thr Ala Giu Ile Thr 150 Arg Asp Lys Met Lye 230 Gin Gin Ala An Thr 310 Asp Lau Gin 55 Gly Ann Lau His Trp Gly Lye Pro 120 Ser Thr 135 Gly Trp Giu Met Sor Ile Ala Pro 200 Ph. Sor 215 Thr Arg Lou Pro Arg Lou Lou Giu 280 Ile His 295 Asp Lye Lye Gin Net Ph.

Lys Ala 360 Gin Sor Leu 75 Lye Lys Lou 90 Arg Sor Ser 105 Ile Asp Gly Sor Glu Ala Arg Ala Val 155 Thr Lou Ann 170 Gly Arg Ser 185 Ser Ann Val k I AL my

I

I>

Val Lou Val Ile 265 Giu Thr Gin Lou An 345 Gin Lou Trp Ile Val Thr Pro Tyr Lau 125 Gin Ala 140 Gly Val Ala Thr Lou Giy Sor Gin 205 Ala Arg 220 Pro Giu Giu Ann Pho Val Ser Lye 285 Ann Giy 300 Ile Tyr Aen Tyr Arg Ala Gin Mot 365 Lye Gly 380 Lye Trp, Pro Thr Val Ph.

110 Thr Ile Gly Phe Ser Lou Ann Thr 175 Ala Lys 190 cly Giu Tyr Gin 11o Gin Lou Ala 255 Gly G;iy 270 Lou Lye Gly Thr Gin Pro Val Ile 335 Tyr Val 350 Tyr Lou Sor Ala Lys Asp Sor Asp Lys Ann 160 Sor Val le Trp, Ph.

240 Ala Glu Bor Gin Giu 320 Lou Lau Lou Lou Him8 Lou Lou Asp Gin 375 SUBSTiTUTE SHEET (RULE 26) 4 711' .01 0, 4~ WO 94/25567 PCT[US94/04495 105 Val 385 Thr Val Asp Lou Lys 465 Lou Thr Phe Ser Ala 545 His Trp Ile Ph.

Ph.

625 Thr Lys Vail Asp Asp 705 Thr Leu Asp Lys 435 Arg Ile Gin Trp Ann 515 Giy Ile Ph.

Ala Lou 595 Giu Gly An Ann 675 Kn Asp His Met Sor 420 Val Gin Ann Val Arg 500 Ala Glu Tyr Ann Met 580 Ala Thr Arg 660 Gly Arg Ser Gly Tyr Ser Leu Lou Arg Lou Gly Sor 570 Pro Gin Lou Arg Trp 650 Sor Gly Thr Lou Asp 525 Ala Lou Gin Thr Bar 605 Gly Asp Giu Val Gln 685 Lou Arg Tyr G2,n Sor 430 Ph.

Asp Thr Pro Ph.

510 Thr met Ala Giy Lou 590 Thr Pho ILys Ile Ala 670 Glu Pro Gly Ile Ser 400 Thr Gin 415 Ser Ph.

Ann Thr Asp Gin Gly Ala 480 Asp Giy 495 Pro Ala Pro Ph.

Val Ser Gly Arg 560 Tyr Tyr 575 Ala Sor Ala Ile Tyr Ala M~c Val.

640 Tyr Ann Gin Ile Gly Trp Lou Lye Giu Arg 720 SUBSTI1TUTE SHEET (RULE 26' -w WO 94/25567 PCTIUS94/04495 106 Gly Pho Ser Ph. Asp Lou Thr Ala Lys 755 Gly Sor Ann 770 Lou Phe Gin 785 Gly Gin Lys *Asp Trp Lou Gl, Lys Val 835 Ann Arg Gin 850 Bar Thr Arg 865 Ala Thr Pro Ann Gly Lou Lou Asp Lys 915 Ser Ile Glu 930 Met Thr His 945 Asp Lou, Ann Val Thr Ilo Lys Tyr din.

995 Ph. MlY Ile 1010 Gly Ile 740 Lys Ile His Ile Ile 820 Ann Pro Pro Giu Tyr 900 Lou Asp Arg Met 980 Val Pro Thr Ber 725 Tyr Pro Bar Val Asn Sor Ala Ile 790 Giu Ann 805 Asp Bar Val Ser Thr Glu Lys Asp 870 Lys met 885 Gin Val Bar Amn Lys Trp Gin Lys 950 Thr Arg 5 1G2.1y Lys Bar Gly Glr-. lU Ser Ala Lou Soe- 775 Thr Met Ann Arg Gly 855 Ala Gly Lou Val le 935 Asp Gin Trp Asp Ile 1015 Gly Gin Tyr Gly Met 730 Giu Arg Pho Asp Pro 750 Asp Asn His Lou Ile 765 Asn Lys Ann Val Glu 780 Lou Ann Thr Lou Trp 795 Gin Thr Thr Lou Gin 810 Gly Tyr Lou Ile Thr 830 Gin Val Ser Ala Giu -845 Bar Bar Ala Trp Ile 860 Giu Tyr Met Val Phe 875 Ala Gin Lys Pho Arg 890 Asp Lys Asp Val His 910 Tyr Ala Ph. Tyr Gin 925 Val Ann Lys Pro Ala 940 Ile Val Bar Ala Val 955 Ala T-hr Prt. Val Thr 970 Ala Asp Lys Ann Bar 990 Glu Lou Thr Ph. Thr 1005 Met 735 Ann Pho Thr Ile Gin 815 Gin Ann Asp Lou Glu 895 Ile Pro 11e Thr le 975 Glu Ber Ala Phe Ile Thr Asnn 800 Gly Ala Lys His Asp 880 Ann le Ala Val Pro 960 Ann Val Tyr Arg Lys 1905 Thr Gly 920 Lys Lys Thr Lou Lys Ala Gln Ber 985 Ann Thr 1000 Lys Lou Bor Pro Lou Pro 1020 INFORMATION FOR SEQ ID NO: 3: SEQUENCE CHARACTERISTICS: LENGTH: 3980 base pairs TYPE: nucleic acid Q1 4z r I- r% ~L~M~iLW ILLI+ L QWJ1f F- 412 WO 94/25567 WO 9425567PCTIUS94/04495 WO 94/25567 PTU9149 PCT[US94/04495 -107 STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HrYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xci) SEQUENCE DESCRIPTION: SEQ ID NO:3:

GGAATTCCAT

AATTTAATGA.

GCCGATATTT

ircGCGATGGCA

TTACCA.TTTT

AACGTTATCT

TGGTAGTAGC

AGCATGGGvGA

TGATGGTTAT

AGGCTTTAAA.

CGATCTTGAA

AGACAGCATT

TAATGTGAGT

CTACCkAATGG

CAACGTAAAG

CCAACGTCTA

GAATATCAGC

TGGAACGCA.A

TCTTAACTCC

ATTAATGTTT

ACTAAAGCAG

GAGTGCTTTA

GTTATTAATG

ACTGTGGTAT

TGATCTAGAT

CACTCAATCA

AGGACGCATT

CGTTTTACTG

GCCACCAGCA

GCACAAAATA

GATAAACGTA

TTTACTTTAC

CGCTCATCTA

CTTACTATCG

GTAAAATTAG

AATCGAGAGA

GGGCGTTCTT

CAGGGTGAAA

TCTGATTATC

CCACAACTAC

ATTAATGAAT

A)JOTTAAAAA

GGCAGACATC

CAAGATAAAC

AATATTAGCC

ATGTACTTAT

GTGACAACC

TCTGATGCAC

TCACGTGAGT

TATTTCAATA.

TTAAATTTAG

GGTTTCACTG

CACTTGCAAT

ATCCTGCATT

ACCCATTAGC

GCATTATGGG

ATAAAAAPICT

CCCcCGTTTT

ATTTCGGAGA

ATTTCACTGG

TGACCTTAAA

TAGGTGCTAA

TCTATATCGA

AAGTAAAAAC

CTGTAAC1ACC

TTGTCGGAGG

GTGATTTCGA

TGATCACTGA

AACTATTTGA

GTGCTTATGT

TAATGACAA

ATCACTGGGG

TAAAAGAAGC

TTAAAAGTAG

CCTTATCTCG

GCACAACGAT

TTAGCCAGCG

GACATTGGGG

TGATCCTAAA

A%;ACTTCTCA

AAACCAATCT

GATTGTCCCC

CTCATTTTGG

AAAACTCATT

CTGGCGTGCT

TGCAACCAAT

AGTCGATAGT

CCGTATTATG

TCGCTTATCA

TGAAAATTTA

TGAAAAAQAG

TGCTCTTAAT

TAAACAAATC

TAATTATGTT

GCTGGAAAAA

GCATTTATTA

ATACAGTTCT

GAACCTACAA

TTTTGATATG

GGGCTATCAG

TTTCTAAGGA

CTATTATCAG

AATCTGATGC

TCAGATAAAA

CTTTTATGGA

ACCGATAAAG

CTTTACAATG

TCLILCCAGTG

GTGGGAGTCT

ACCTCCTCTG

ATTCGTTTTA.

TTTTCTGTCG

GAACCTGAAA

GCGGCCATTG

ACA.AACCTCG

ATTCACACTT

ATTATTTATC

ATTTTAGGTA

GATCCCACAC

GATCAAGGCT

CGTTGGTGGT

ACTCAAGTTT

AAAGTAAGTG

CGTTATGACA

GAAAACATAT

CGCCTTATAA.

AGTCAGAAAT

ACTCAATACT

AATGGAAAGG

AAGCATCTAA

AAAAACCGAT

AGGCTCAGGC

CTTTAAZTAA

ATGGTACTCA

AAGCGCCTTC

ATGATGCTCG

TTCAATTTCA

ATCTTATTCG

CATTAGAAGA

TAGCAAATGG

AACCAGAGAA

ATTACACGAC

AAAAGGCGCA.

TTGTTAAAGG

ATATTTCCAC

ATGATTCATT

CTGATAGCTC

120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 1140 1200 1260 1320 1380 14,40 1~500 CCAACATTTA GCCTTATTAT TACTAGAGCC SUBSTITUTE SHEET (RULE 26) WO 94/25567 PCT/US94/04495 108 TGATGATCAA AAGCGTATCA AACGCAAGTG CCACCGGGTG TGAAGGCAAC TATCCGGGCT TTTATTAPGC GATACACCAT GATGGTTTCA GCGTGGATCT CCCTTTTAAC TCACCTTCGT TGCAALAATCA TCGCCTGATA ACAAAATGI4A TCAACTGCTA TTTCTATGCC TTTAkTGGCG ACTGAAAGCT TATP.ACACCA TGGCCGTTAC CAAAGTCATG GGGCTATCAG CAAGAAGGTT TCCTCTTAA. GACTTAGACA ATTTAGCGGA ACATCATCCC TCCCGCCAAT CTTGAGCGTT TGATAATCAC TTAATTTTTA TGAAACGACC TTATTCCAAC ACAAAAQATA GAAAACATGC TAGCAATGGC AATGGTTACT TCAGGTTTCA GCGGAAAATA GATCGATCAC AGCACTCGCC GACACCTGAA AAAATGGGAG GGTTCTTCGT AAGGATAAAG ATATGCCTTT TATCAGCCAG TGCAATTGTG ATGACTCATC TT'AA4ATATG ACTCGCCAAA CAAATGGCAA TCTGCTGATA TGAACTGACG TTTACGAGTT *.NTTGATTTAAT CAAAAGAACG CTWNTAAA ACCCTTTAGC

ACTTAGTTAA

GTAAAGATGG

ACTCTTTCCC

TTTCAGTGGG

ACAGTAATCC

TAAPJATCAGT

AAA.CACTTGC

TTTTTGGAGA

GTGCTTTTGG

ATGTTTGGTC

GTGTCGCTCA

GGGATTGQPJA

GTCCTAAACC

TTGAAGGTCA

TTGATCCTAA

TTGGTAC-CAA

ATGCCATTAC

CTTATCAAAC

TACTTTCAGC

TTTACGCCCT

AGCCTTTAAA

TGAAAGTGGT

AGAAGTTGGA

CGCTCAAGGC

ATCTATTTAT

AACTATTACA

T1 TTCATCGT

ATCTGAAATT

AATAGTGAGT

TAG.AATG;CAAj

CATTATATCA

GATGGTACAG

AATGCCTCTC

TGGAATAACC

TTACCGCTTG

TATTACTGGC

CTTGCGATTA

CCAGCGTCTT

CTGGCGCATT

CATGGCGACA

AGCTTATTTA

TGAAAA.AAGC

CAGGAAGACA

TTGCCATGTC

GTGATAAAAC

TACCTCAAGG

TGGCI

1 AGATA AAATGGTGAC TATAACAAAG ATAACCGTTA

AATGGCTCGC

GGGGCAACCA

TCATACCTTA ATGCAACGTG ATATGGCATG ATGGCATTCG TTTCACTGCG APJJ.AGAGTG

TATAAATAGT

TCCAACATTA

AACACTTCAA

TAATTACTCA AGCAGAJ~AA AAAATCGCCA ACCGACAGAA

CCAAAGATGC

AGATGGCACA

ACGTTCATAT

CATCAATTGA

GACAAAAAGA

AAGCAGCAAC

AAAATAGTGA

ACTTTGGTAT

CTCTTGCGTT

CCACGCGGTT

CAGTTATGAG

AAAATTCCGT

TATTCTCGAT

AGACAAATGG

CACTCTTATT

TCCTGTCACC

AGTGAAATAT

TCCACAAGAA

CCTTTTTTAT

ACIATTAAGCC

AGTGATAAAA

AATACCCTTT

CAAGGTGATT

GTAAATGTAA

GGAAACTTTA

TATATGGTCT

GAAAATAATG

AAACTCAGCA

ATCAAAAAGG

GTCAGTGCAG

ATCAATGTCA

CAGGTTTCTG

ATCAPJLCTCT

TTGCAGGAAA

TC.'GTTTATC

AGCT"TTCACA

CTATTCACCT

GAGAGCGTGG

ATCTTATTTA

TATTAGCCGC

ATAAAAATGT

GGATTAATGG

GGTTAATTGA

GTCGCCAACA

GCTCGGCATG

TTTTAGATGC

GGTTATATCA

ATGTAACGGG

TTAATAAACC

TTACACCTGA

CGATTAATGG

GTGATAACAC

CGCCACTCCC

TCTGATTATG

ATTACCCGCA

1560 1620 1680 1740 1800 1860 1920 1980 2040 2100 2160 2220 2280 2340 2400 2460 2520 2580 2640 2700 2760 2820 2880 2940 3000 3060 3120 3180 3240 3300 3360 CAAGCATTAC CCACTCTGTC TCP 4 TGAAGCT. TTCGGCGATA TTTATCTTTT TGAAGGTGAA SUB ISTITUT E SHIEET (RULE 26) WVO 94/25567 PCTIUS94/04495

TTACCCAATA

AAAGATGGTG

AATATTGTTA

ATTTATAATG

ATTGCACTAA

TTTCGTGATA

AACCAAGCCG

TGGGCAGTAC

AAAAACTGGA

AACTTCGATA

CCCTTACCAC

AACAATCAtC,

ATTACCAAGA

AAAAACCTCA

GTTTTAATGC

TCCAAGGCTC

GAACACTCTT

CTGACTATCA

GTGCATTATT

CTGAATTTCG

TTCAAATAAT

CAATGGCAA

TGATAAAAAT

ATCTTCCCCA

TGAACTTAAT

TGCGACAGGT

TTTTGATCAA

AACACCTTAC

GATGTACGAT

CGATGACCAPI

AATCAATTAT

TATCAACCAC

ACAGCCACAC

TTAACGTTAG

TTTACGGGGT

CAACTTGATC

ATCATCATGA

GTAAATAACG

CAGATGTTTC

ACAGAhATGG

CGCTAAGCAA

AAGCAACATT

CACTCACTTT

CATTTAAACA

GGCGAGGTAT

AATTAGTGAT

GTGTACCGTT

CAGTAAACAC

AAGCCCATTA

CTTCGATTTA

ACAGCATGCT

AACACTAAAT

TATGATGTGG

AZ ,ATA&tTAAA

TGCTGTTCCT

CACCGCTCCA

AGACAATCGT

GATGGTTAGT

CCCTACTTTA

TCAGCGCTTT

3420 3480 3540 3600 3660 3720 3780 3840 3900 3960 3980 GAATATTATC AAGGAATTCC INFORMATION FOR SEQ ID NO:4: Mi SEQUENCE CHAR.ACTERISTICS: LENGTH: 3980 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: No (iv) ANTI-SENSE: NO (ix) FEATURE: NAME/KEY: CDS LOCATION: 188. .3181 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4: GGAATTCCAT CACTCAATCA ?'TAAATTTAG GCACAACGAT AATTTAATGA AGGACGCATT GGTTTCACTG TTAGCCAGCG GCCGATATTT CGTTTTACTG CACTTGCAAT GACATTGGGG CGCGGAT ATG GCC ACC AGC AAT CCT GCA TTT GAT Met Ala-Thr Ser Asn Pro Ala Phe Asp 1 5 GGGCTATCAG CGTTATGACA TTTCTAAGGA GAAAAATAAT CTATTATCAG CGCCTTATAA CCT AAA AAT CTG ATG Pro Lys Aen Leu Met CCA TTA GCA QAC TTC Pro Leu Ala Asp Phe GAT AAA CGT AGC ATT Asp Lys Arg Ser Ile 120 180 229 277 325 CAG TCA GAA ATT TAC Gln Ser Glu Ile Tyr TCA TCA CAT AMA AAC Ser Ser Asp Lys Asn CAT TTT GCA CAA AAT AAC His Phe Ala Gln Asn Asn 20 25 TCA ATA CTA ACG TTA TCT Ser Ile Leu Thr Leu Ser 40 SUBSTITUTE SHEET (RULE 26) r~,t ,~2 f 'WO 9425567PCTIUS94/0 4 4 9 nA r .1 11 9 'WO 94/25567 PCTIUS9404495 110

ATG

Met

ACT

Thr

GCA

Ala

GAA

Glu

ATT

Ile

ACT

Thr

CGA

Arg

GAC

Asp

AAA

Lye 175

ATG

Met

AAA

Lye

CAA

Gln

CAA

Gln

GCI

Ala 255

AAT

Aen

ACT

Thr GGA AAC CAA TCT Gly Asn Gin Ser TTA CAT AAA AAA Lou His Lye Lye TGG GGA (GC TCA Trp Gly Arg Ser AAA CCG ATT GAT Lye Pro Ile Asp TCA ACC AGT GAG Sor Thr Sor Glu 115 GGC TGG CGT GCT Gly Trp Arg Ala 130 GAG ATG ACC TTA Glu Met Thr Leu 145 AGC ATT GGG CGT Ser Ile Gly Arg 160 GCG CCT TCT AAT Ala Pro Sor Asn TTT TCT GTC GAT Ph. Sor Val Asp 195 ACT CGC TTA TCA Thr Arg Lou Ser 210 CTA CCT GTA ACA Lou Pro Val Thr 225 CGT CTA ATT AAT Arg Lou Ile Aen 240 TTA GAA GAG AlT Lou Giu Giu Asn ITT CAC ACT TTA lie His Thr Lou 275 GAT AAA CAA ATC Asp Lye Gln Ile CTT TTA Lou Leu CTG ATT Lou Ile TCT ACC Ser Thr 85 GGT TAT Gly Tyr 100 GCT CAG Ala Gin GTG GGA Val Gly AAT GCA Asn Ala TCT TTA Ser Lou 165 GTG AGT Val Ser 180 GAT GCT Asp Ala GAA CCT Glu Pro CCT GAl Pro Glu GAl TTT Glu Ph.

245 ATC AGC Ile Ser 260 GCA AAT Ala Asn ATT ATT lie Ile TGG AAA TGG Trp Lye Trp 55 GTC CCC ACC Val Pro Thr 70 CCC GTT TTC Pro Val Phe CTT ACT ATC Lou Thr Ile GCA GGC TTT Ala Gly Ph.

120 GTC TCT TTA Val Sor Lou 135 ACC AlT ACC Thr Aen Thr 150 GGT GCT AAA Gly Ala Lye CAG GGT GAl Gin Gly Glu CGC TAC CAA Arg Tyr Gin 200 GAA ATT CAA Giu Ii Gin 215 AlT TTA GCG Aen Lou Ala 230 GTC GGA GGT Val Gly Gly AAA TTA AAA Lye Lou Lye GGT GGA ACG Gly Gly Thr 280 TAT/ CAA CCI Tyr Gin Pro AlAA GGT Lye Gly GAT AAA Asp Lye TCA TTT Ser Ph.

CAT TTC Asp Ph.

105 AAA CTA Lye Val AAT AAC Asn len TCC TCT Sor Ser GTC GAT Val Asp 170 ATC TAT Ile Tyr 185 TGG TCT Trp Ser TTT CAC Ph. His GCC ATT Ala Ile GAA AAA Glu Lye 250 ACT GAT Ser Asp 265 CAA GGC Gln Gly GIG AAT Giu Asn

GGT

Cly

GAA

Glu

TGG

Trp

GGA

Gly

AAA

Lye

GAT

Asp

GAT

Asp 155

AGT

Ser

ATC

Ile

GAT

Asp

AAC

Asn

CAT

Asp 235

GAG

Glu

TTC

Ph.

AGA

Arg

CTT

Lou

AGT

Ser

GCA

Ala

CTT

Lou

AA

Glu

TTA

Lou

CTT

Lou 140

GGT

Gly

ATT

Ile

GAC

Asp

TAT

Tyr

GTA

Val 220

CTT

Lou

ACA

Thr

GAT

Asp

CAT

His

AAC

Aen

AGC

Ser

TCT

Ser

TAC

Tyr Al Lye

GAT

Asp 125 GAl Glu

ACT

Thr

CGT

Arg

CGT

Arg

CAA

Gln 205

AAG

Lye

ATT

I1

AAC

Asn

GCT

Ala

CTG

Lou 285

TCC

Ser

TTT

Phe

AAA

Lye

AAT

Asn

CTC

Lou 110

TTC

Ph.

AAT

len

C&A

Gln

TTT

Ph.

ITT

lie 190

GTA

Val

CCA

Pro

CGC

Arg

CTC

Lou

CTT

Lou 270

ATC

Ile

CAA

Gln 901 1045 1093 i ur x i.

1 SUBSTITUTE SHEET (RULE 26) WO 94/25567 PCTIUS94/04495 MNEEImbwr--- w -md

I

WO 94/25567 PCTIUS94/04495 ill

GAT

Asp

TTA

Lou

CAA

Gin 335

TTA

Lou

TOO

Trp

OAT

Asp

CTG

Lou

OCT

Ala 415

TTA

Lou

GTT

Val

CCG

Pro

GAA

Olu

CAG

Gin 495 Oly AAA CAP.

Lys Gin 305 ATO TTT Mot Pho 320 AAG OCO Lys Ala OAT CA Asp Gin GGA TAC Oly Tyr OCA CTA Ala Lou 385 TOO TAT Trp Tyr 400 OAT AGC Asp Bor 0CC TTA Ala Lou AAT ACT Ann Thr GOT GOT Oly Oly 465 GOC AAC Oly An 480 CTiT ATT Lou Ilo TOO AAT Trp An 290

CTA

Lou

AAT

An

CAA.

Gin

GGC

Oly

AGT

Sor 370

AA

Lys

TCA

Ser

TCT

Sor

TTA

Lou

TTC

Pho 450

AAA

Lys

TAT

Tyr

TAT

Tyr

AAC

An

OTT

Val 530 TTT GAT Pho Asp ATT AGC Ilo Sor CTA AAG Lou Lys 340 TTT OTT Pho Vai 355 TCT COT Sor Arg GAIL OCO Oiu Ala COT GAG Arg Giu OAT CTA Asp Lou 420 TTA CTA Lou Lou 435 AOC CAT Sor His OAT GOT Asp Oly CCO GOC Pro Oly TTA TTA Lou Lou 500 CTG AAA Lou Lye 515 295 ILAT TAT OTT Ann Tyr Val 310 CGT OCT TAT Arg Ala Tyr 325 CAG ATO TAC Gin Mot Tyr AAA GOG ILOT Lye Oly Sor TOO TOG TAT Trp'Trp Tyr 375 AAC CTA CAA Ann Lou Gin 390 TTT AAA AGT Pho Lys Ser 405 OAT TAT TTC Asp Tyr Pho GAG CCT OAT Oiu Pro Asp TAT ATC ACT Tyr Ile Thr 455 TTA COC CCT Lou Arg Pro 470 TAC TCT TTC Tyr Sor Phe 485 CGC OAT ACA Arg Asp Thr AAA OCO ATO Lys Ala Hot

ILAT

An 315

AAA

Lys

ACA

Thr

ACA

Thr

TTA

Lou

TAT

Tyr 395

ATO

Mot

TCT

Sor

COT

Arg

ACG

Thr

GCA

Ala 475

AAA

Lys

OTO

Val

TOO

Trp, 300 TAC ACG Tyr Thr GAT CCC Asp Pro ILAG CAT Lys His ACC CAT Thr His 365 TTA ATG Lou Hot 380 OAT TCA Asp Sor AAA OTA Lys Val COC CAP.

Arg Oin ATC AAC le Ann 445 CALA OTG Gin Val 460 TOG COA Trp Arg FLAT 0CC Ann Ala GOT GA Gly Oiu ATC TAC le Tyr 525 TTT ILAC Pho Ann 540 1141 1189 1237 1285 1333 1381 1429 1477 1525 1573 1621 1669 1717 1765 1813 1,AT CCA GAP.

Ann Pro Giu GOP. TTA CCG CTT Gly Lou Pro Lou

OCA

Ala 535 GOIL AGA CAC CCT Gly Arg His Pro 1 (SUBSTITUTE SHEET (RULE 26) WO 94/25567 PCTIUS94/04495 112 CCT TCG TTA AAA TCA Pro Ser Lau Lys Ser *1 545 GCA AAA TCA TCG CCT Ala Lys Bar Ser Pro 560 AGT GAT AAA ACA CAA Sar Asp Lys Thr Gin 575 ACA CCA GCG TCT TTA Thr Pro Ala Ser Len 595 TTT GGT ATT CAT COT Ph. Gly Ile His Arg 610 AAC ACC AAT OTT TG Ann Thr Ann Val Trp 625 OGC CGT TAC CAA AGT Gly Arg Tyr Gin Bar 640 CGO CTT TCA CG GOC Gin Lou Sor Gin Gly 655 CAA 000 GCA. ACC ACT Gin Gly Ala Thr Thr 675 AAA CCT CAT ACC TTA Lys Pro His Thr Lou 690 TCA TCC CTT OAA GOT Ser Bar Lou Gin Gly 705 CCC 0CC AAT CTT GAG Pro Ala Ann Lou Giu 720 GTA TTA GCC OCT GAT Val Lou Ala Ala Asp 735 AOT AGT OAT AAA AAT Sor Sor Asp Lys Amn 755 ATT ACT CCA. ACA TTA le Thr Pro Thr Lau 770 AAC ATG CCT TAT CAK Ann Met Pro Tyr Gin OCT CAA GGC Ala Gin Gly 550 AAA ACA CTT Lye Thr Lan 565 GAA TCA ACT Gin Ser Thr CAA GOT TTC Gin Gly Phe CAA CAT AAA Gin Asp Lys 615 TCT OAk ATT Ser Gin Ile 630 GOT OTC OCT Gly Vai Ala 645 CG CAA GAk Gin Gin Gin CAC CTT CCT His Len Pro CAA COT 00k Gin Arg Gly 695 TAT GGC ATG Tyr Gly Met 710 TTT GAT CCT Ph., Asp Pro 725 CAC TTA ATT Hi's Len Ile AAT GTT OAA Ann Val Gin ACC CTT TG Thr Len Trp 775 ACA CTT CAA TOG CTT GCC Trp Lau Ala 555 ATT TAT CTT Ile Tyr Lau 570 TTT GGA OAk Ph. Giy Gin TTT AAT GGC Ph. Ann Oly ACA CTG AA Thr Lau Lys 620 AAA OAT AAC Lys Asp Ann 635 GTG AGT AAT Val Bar Ann 650 GAT TOG ?4AT Asp Trp Ann CAC TTA GAC Asp Lou Asp GGA TTT kOC Gly Ph. Sar 700 TTC GAT CTT Ph. Asp Leu 715 ACT OCG AAA Thr Ala Lye 730 GOT AOC AAT Gly Bar Ann TTA TTC CAA Len Ph. Gin GG0I'CAA AAG Gly Gin Lye 780 GAT TOG TTA ATG TCT Met Bar GCG ATT Ala Ile ACT ATT Thr Ile 590 GOT GCT Oly Ala 605 OCT TAT Ala Tyr CGT TAT Arg Tyr GOC TCG Gly Ber AGA ATO Arg Not 670 AOT CCT Bar Pro 685 G0A ACA Gly Thr ATT TAT Ile Tyr AAG AGT Lys Bar ATA KAT le Ann 750 CAT 0CC Kin Ala 765 ATA OAk Ile Gin ATT GAT 1861 1909 1957 2005 2053 2101 2149 2197 2245 2293 2341 2389 2437 2485 2533 2581

L

r Thr Len Gin Gin Giy Asp Trp, Len 119 Asp SUBSTITUTE SHEET (RULE 26) 2 PCTIUS94l04 9 wo 94125567 4 WO 94/25567 PCT!( ,94/04495 113

AGC

Ser

AGT

Sor 815

GAA

Giu

GAT

Asp

ATG

met

GTT

Val

AAT

An 895

TGG

Trp Lys

CGC

Arg Lys

GGT

Gly 975

GA

785 AAT GGC Ann Gly Soo CGC CAA Arg Gin GGA AAC Gly An GCC AGT Ala Ser GGA GAG Gly Glu 865 CTT CGT Leu. Arg 880 GTA ACG Val Thr ATC AAA Ile Lys GAC ACT Asp Thr CAA AA Gin Lys 945 TGG CAA Trp Gin 960 GAT AAC Asp Ann ATC AAA 790 TTA ATT ACT CAA GCA Lou Ile Thr Gin Ala 805 TCA GCG GAA AAT AAA Ser Ala Giu Aen Lys 825 GCA TGG ATC GAT CAC Ala Trp le. Asp His 840 ATG GTC TTT TTA GAT Met Val Phe Lou Asp 855 AAA TTC CGT GAA AAT Lys Phe Arg Giu Aen 870 GAC GTT CAT ATT ATT Asp Val His Ile Ile 885 TTT TAT CAG CCA GCA Ph. Tyr Gin Pro Ala 905 AAA CCT GCA ATT GTG Lys Pro Ala Ile Val 920 AGT GCA GTT ACA CCT Ser Ala Val Thr Prt 935 CCT GTC ACC ATC AAT Pro Val Thr Ile An 950 AAA AAT AGT GAtA GTG Lys Ann Ser Giu Val 965 ACG TTT ACG AGT TAC Thr Phe Thr Ser Tyr 985 CTC CCT TGATTTAATC 795 GAA AAA Giu Lys 810 AAT CGC Ann Arg AGC ACT Ser Thr GCG ACA Ala Thr AAT GG0 Ann Gly 875 CTC GAT Lou Asp 890 TCA ATT Sor Ile ATG ACT Met Thr GAT TTA Asp Lou GTC ACG Val Thr 955 AAA TAT Lys Tyr 970 TTT GGT

GTA

Val

CAA

Gin

CGC

Arg

CCT

Pro 860

TTA

Lou

AA

Lys

GAA

Giu

CAT

Hia

AAT

Asn 940

ATT

Ile

CAG

Gin

ATT

APLT

An

CCG

Pro

CCC

Pro 8

GAA

Giu

TAT

Tyr

CTC

Lou

GAC

Asp

CGA

Arg 925

ATG

met

AAT

Ann

GTT

Val

CCA

GTA

Val

ACA

Thr 830 Lys

AAA

Lys

CAG

Gin

AGC

Bar

AAA

910

CAA

GIn

ACT

Thr

GGC

Gly

TCT

Ser

CAA

2629 2677 2725 2773 2821 2869 2917 2965 3013 3061 3109 3157 3211 3271 3331 3391 3451 Phe Gly Ile Pro Gin 990 hAAAGAACGC TCTTGCGTTC kk Glu Ile Lye Lou Ser Pro Lou Pro CTTTTTTATT TGCAGGAAATCTGATTATGC TAATAAAAAA CCCTTTAGCC CACGCGGTTA CATTAAGCCT CTGTTTATCA TTACCCGCAC AAGCATTACC CACTCTGTCT CATGAAGCTT TCGGCGATAT TTATCTTTTT GAAGGTGAAT TACCCAATAC CCTTACCACT TCAAATAATA ATCAATTATC GCTAAGCAAA CAGCATGCTA AAGATGGTGA ACAATCACTC AAATGGCAAT I SUBSTITUTE SHEET (RULE 26) 'WO 94/25567 PCTIUS94/0449 'WO 94/25567 W0 9425567PCT1US94/04495 114

ATCAACCACA

CAGCCACACC

TAACGTTAGC

TTACGGGGTG

AACTTGATCA

TCATCATGAG

TAAATAACGC

AGATGTTTCA.

CAGA.AATGGC

AGCAACATTA ACACTAAATA ATATTGTTAA ACTCACTTTT ATGATGTGGA TTTATAATGA ATTTAAACAA AATAATAAAA TTGCACTAAG GCGAGGTATT GCTGTTCCTT TTCGTGATAT ATTAGTGATC ACCGCTCCAA ACCAAGCCGG TGTACCGTTA GACAATCGTT GGGCAGTACC AGTAAACACG ATGGTTAGTA AAAACTGGAG AGCCCATTAC CCTACTTTAA ACTTCGATAC TTCGATTTAT CAGCGCTTTG AATATTATCA

TTACCAAGAT

AAAACCTCAA

TTTTAATGCT

GCAAGGCTCT

AACACTCTTT

TGACTATCAA

TGCATTATTG

TGAATTTCGC

AGGAATTCC

GATAAAAATA

TCTTCCCCAT

GAACTTAATT

GCGACAGGTC

TTTGATCAAA

ACACCTTACG

ATGTACGATC

GATGACCAAA

3511 3571 3631 3691 3751 3811 3871 3931 3980 INFORMATION FOR SEQ ID Wi SEQUENCE CHARACTERISTICS: LENGTH: 998 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE (xi) SEQUENCE Met Ala Thr Bar Asn 1 5 Glu Ile Tyr His Phe Asp Lys Asn Bar Ile Aun Gin Bar Lou Leu His Lys Lys Lou Ile TYPE: protein DESCRIPTION: SEQ ID Pro Ala Phe Asp Pro Lys Asn Lau Met 10 Gin Bar is Ala Gin Asn Asn Pro Leu Ala 25 Lou Thr Leu Ser 40 Trp Lys Trp Lys Val Pro Thr Asp 70 Pro Val Phe Bar Asp Lys Arg Gly Giy Bar Lys Giu Ala Asp Ph. Bar Bar Ser 1ie Met Gly Sar Ph. Thr Lou Bar Lys Ala Gly Arg Bar Bar Thr Trp Lou Tyr &sn Glu Lys le Bar Pro Ile Asp Thr Bar Giu 115 Trp, Arg Ala 130 Met Thr Lou 145 Ile Gly Arg Giy 100 Ala Val Tyr-Lou Thr Ile Asp 105 Gin Ala Gly Phe Lys 120 Gly Val Ser Lou Asn 13F-, Ph. Gly Giu Lys Val Lys Leu Asp Phe Thr Gly 125 Asn Asp Leu Giu Asn Arg Giu 140 Asn Ala Thr Asn Thr Ser Bar 150 Bar Lou Gly Ala Lye Val Asp 165 170 Asp Gly Thr Gin 155 Bar le Arg Ph.

Asp Ser 160 Lys Ala 175

T

SUBSTITUTE SHEET (RULE 26) p' PCTIUS94/04495 WO 94/25567

V

0 WO 94125567 PCTIUS94/04495 115 Pro Ser Arg Pro 225 Lou Giu His Lys Gin 305 Ph.

Ala Gin Tyr Lou 385 Tyr Ser Lou Thr Gly 465 Asn Ser Asn Val Asp 195 Lou.* Ser 210 Val Thr Ile An Giu An Thr Lou 275 Gin Ile 290 Lou Ph.

Ann XIe Gin Lou Gly Ph.

355 Sor Sew.

370 Lys Glu Sor Arg Sor Asp Lou Lou 435 Ph. Sor 450 Lys Asp Tyr Pro Val 180 Asp Giu Pro Giu Ile 260 Ala Ile Asp Sor Lys 340 Val Arg Ala Giu Lou 420 Lou His Giy Gly Sor Ala Pro Giu Ph.

245 Ser An Ile An Aurg 3:25 Gin Lys Trp An Phe 405 Asp Giu Tyr Lou Tyr 485 Gin Arg Giu An 230 Val Lys Gly Tyr Tyr 310 Ala Met Gly Trp Lou 390 Lys Tyr Pro Ile Arg 470 Sor Gly Tyr Ile 215 Lou Giy Lou Gly Gin 295 Val Tyr Tyr Sor Tyr 375 Gin Sor Ph.

Asp Thr 455 Pro Ph.

Giu Gin 200 Gin Ala Gly Lys Thr 280 Pro Ile Val Lou Ala 360 Ile Thr Ser An Asp 440 Gly Asp Pro Ile Tyr 185 Trp Sor Pho His Ala Ile Giu Lys 250 Sor Asp 265 Gin Giy Giu An Lou Gly Lou Giu 330 Lou Mot 345 Lou Val Sor Thr Gin Vai Pha Asp 410 Thr Lou 425 Gin Lys Ala Lou Gly Thr Ala Ph.

490 Ile Asp Asp Tyr Ann Vai 220 Asp Lou 235 Giu Thr Ph. Asp Arg His Le1i An 300 Ann Tyr 315 Lys Asp Thr Lys Thr Thr Lou Lou 380 Tyr Asp 395 Mot Lys Sor Arg Arg Ile Thr Gin 460 Ale, Trp 475 Ly's An Arg Ilo 190 Gin Vai 205 Lys Pro Ile Arg Ann Lou Ala Lou 270 Lou Ile 285 Ser Gin Thr Thr Pro Thr His Lou 350 His His 365 Hot Ser Sor Lou Val Sor Gin His 430 Ann Lou 445 Val Pro Arg His Ala Sor Giu Sor 510 A&1

NP

Asp Lou Gin 335 Lou Trp Asp Lou Ala 415 Lou Val Pro Glu Gin 495 Gly Pho Thr Lou Arg 240 Lou Ile Asp Lys mot 320 Lys Asp Gly Ala Trp 400 Asp Ala An Gly Gly 480 Lou Trp Ile Tyr Lou Lou Arg 'Asp Thr Pro Ph. Sor Val Gly 500L 505 SUBSTITUTE SHEET (RULE 26) *WO 94125567 PCT[US94/04495 'WO 94/25567 PCT1US94/04495 -116- Ann Ann Len Lys Lys Ala Met Val Ser Ala Trp lle Tyr Ser Asn Pro 515520 525 Giu Val Gly Leu Pro Leu Ala Gly Arg His Pro Phe Ann Ser Pro Ser 530 535 540 Leu Lys Ser Val Ala Gin Gly Tyr Tyr Trp Lau Ala Met Ser Ala Lys 545 550 555 560 Ser Ser Pro Asp Lys Thr Leu Ala Ser Ile Tyr Leu Ala Ile Ser Asp 565 570 575 Lys Thr Gin Asn Giu Ser Thr Ala Ile Phe Gly Glu Thr Ile Thr Pro 580 585 590 Ala Ser Leu Pro Gin Gly Phe Tyr Ala Phe Ann Gly Gly Ala Phe Gly 595 600 605 le Hiis Ai-g Trp Gin Asp Lys Met Val Thr Leu Lys Ala Tyr Asn Thr 610 615 620 Ann Val Trp Ser Ser Glu Ile Tyr Asn Lys Asp Ann Arg Tyr Gly Arg 625 630 635 640 Tyr Gin Bar His Gly Val Ala Gin Ile Val Ser Ann Gly Ser Gin Leu 645 650 655 Bar Gin Gly Tyr Gin Gin Glu Gly Trp Asp Trp Asn Arg Met Gin Gly 660 665 670 AaTrThr Ile His Leu Pro Leu Lys Asp Leu Asp Ser Pro Lys Pro AaTr675 680 685 His Thr Leu Met Gin Arg Gly Glu Arg Gly Phe Ser Gly Thr Ser Bar 690 695 700 Lau Glu Gly Gin Tyr Gly Met Met Ala Phe Asp Leu Ile Tyr Pro Ala 705 710 715 720 Ann Lou Giu Arg Phe Asp Pro Asn Phe Thr Ala Lys Lys Ser Val Leu 725 730 735 Ala Aia Asp Ann His Leu Ile Phe Ile Gly Ser Asn Ile Asn Ser Bar 740 745 750 Asp Lys Ann Lys Asn Val Glu Thr Thr Leu Phe Gin His Ala Ile Thr 755 760 765 Pro Thr Lou Ann Thr Leu Trp Ile Asn Gly Gin Lys Ile Giu Ann Met 770 775 780 Pro ,Tvr Gin Thr Thr Lou Gin Gin Gly Asp Trp Leu le Asp Bar Ann 78~ 790 795 800 Gly Asn Gly Tyr Lau Ile Thr Gin Ala Giu Lys Val Asn Val Ser Arg 805 810 815 Gin His Gin Val Bar Ala Giu Ann Lys Ann Pxg Gin Pro Thr Glu Gly 820 825 830 BnaheSr Bar Ala Trp Ile Asp His Ser Thr Arg Pro Lys Asp Ala AnP.835 840 845 SUBSTITUTE: SHEET (RULE 2 WO 9425567PCT/US94/04495 S .5 .WO 94/125567 PCT1US94/04495 117 Ser Tyr 850 Glu Tyr Met Val Phe Leu Asp Ala 855 Thr Pro Glu Lys Met Gly 860 Glu 865 Arg Thr Lys Thr Lys 945 Gin An Lys Met Ala Gln Lys Phe Lys Gly Lys Lou 930 Ala Ser Thr Lou Asp Tyr Val 915 Ile Ala Ala Giu Ser 995 Lys Asp 885 Ala Phe 900 Ann Lys Vai Ser Thr Pro Asp Lys .965 Lou Thr 980 Pro Lou 870 Val Tyr Pro Ala Val 950 An Pho Pro Arg His Gin Ala Val 935 Thr Sor Thr Giu Ile Pro Ile 920 Thr le Giu Ser Asn Ile Ala 905 Val Pro Aen Val Tyr 985 An Lou 890 Ser Met Asp Val Lys 970 Ph.

Gly 875 Asp 119 Thr Lou Thr 955 Tyr Gly Lou Lys Giu His An 940 Ile Gin Ile Tyr Lou Asp Arg 925 Met An Val Pro Gin Ser Lys 910 Gin Thr Gly Sor Gin 9 9l Val Lou 980 Ann Val 895 Trp Ile Lys Asp Arg Gin Lys Trp 960 Gly Asp 975 Giu Ile INFORMATION FOR SEQ ID NO:6: SEQUENCE CHARACTERISTICS: LENGTH: 21 base pairs TYPE: nucleic acid CC) STRAZIDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: No (xi) SEQUENCE DESCRIPTION: SEQ ID NO:6: CAYTTYGCNC ARAAYKAYCC N INFORMATION FOR SEQ ID NO:7: Ci) SEQUENCE CHARACTERISTICS: CA) LENGTH: 20 base pairs CEB) TYPE: nucleic acid CC) STRANqDEDNESS: single D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA Cgenomic) -(U11i) HYPOTHETICAL: NO '1~ SUBSTITUTE SHEET (RULE 26) V/O 94/25567 PCT/U$94O4495 118 (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7: CACTTCGCNC AAAATAATCC INFORMATION FOR SEQ ID NO:8: Wi SEQUENCE CHARACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: No (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8: CACTTCGCNC AA.AACAACCC INFORMATION FOR SEQ ID No:9: SEQUENCE CHARACTERISTICS: CA) LENGTH: 20 base pair.

(B TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICZI: NO Civ) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:9: CACTTCGCNC AAAACA.ATCC INFORMATION FOR SEQ ID NO:l0: Ci SEQUENCE CHARACTERISTICS: CA) LENGTH: 20 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO SUBSTITUTE SHEET (RULE 26) a a ~a.

Asp Trp Ann Arg Met Gin Gly Ala Thr Thr Ile His Lou Pro Leu Lys 690 695 700 Asp Leu Asp Ser Pro Lys Pro His Thx Lou Met Gin Arg Gly Glu, Arg 705 710 715 720 'SUBSTITUTE SHEET (RULE 26)77

AV

WO 94125567 PCTIUS94/04495 -119- (xi) SEQUENCE DESCRIPTION: SEQ ID NO:l0: CACTTCGCNC AAAATAACCC INFORMATION FOR SEQ ID NO:11: SEQUENCE CHARACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYP'E: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:ll: C.ACTTCGCNC AGAATAATCC INFORMATION FOR SEQ ID NO:12: Wi SEQUENCE CHARACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii)4 AYPOTHETICAL: NO (iv) ANTI-SENSE: No (xi) SEQUENCE DESCRIPTION: SEQ ID NO:12: CaU.'TCGCNC AGAACAACCC INFORMATION FOR SEQ ID NO%13: Ci) SEQM1WCE CHAR.ACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid TOPOLOGY: linear (ii) MOLECULE TYPE. DNA (genomic) -j(iii) HYPOTHETICAL: NO (iv) ANITI-SENSE, O (xi) SEQUENCE DESCRIPTION: SEQ ID NO:13,- SUBSTITUTE SHEET (RULE 26)_ VWO 94/25567 P~I~A1t~l Th-9rjTTOnA IAAAn WvO 94/25567 PCTIUS94/04495 -120fjCACTTCGCNC AGAAcAATCC 2 (2 IOR1iiTIoN FOR SEQ ID NO:14: 2 Wi SEQUENCE CHARACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid MC STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: No (iv) ANTI-SENSE: No (xi) SEQUENCE DESCRIPTION: SEQ ID NO:14: CACTTCGCNC 4GAATAACCC INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: 21 base paj.rs TYPE: nucleic acid STRANqDEDNESS: single TOPOLOGY: linear (i)MOLEcULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO ir~ (xi) SEQUENCE DESCRIPTIONi SEQ ID %3ARGCNCARG CNGGNTTYAA R 21 INFOPM!ATION FOR SEQ ID NO:16: i)SEQUENCE CHARACTERISTICS: LENGTH: 21 base pairs TYPE: nuclific acid STRANDEDNESS: single TOPOLOGY: linear (ii) 14OLECULE TYPE: DNA (genomic) j (iii) HYPOTHETICAL: NO (iv) ANTI-SNSE:, YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 16: YTTRAANCCN GCYTGNGCYT C 21 SBSTITUTE SHEET (RULE 26)

K)

WO 94/25567 PCTIUS94/04495 121 INFORMATION FOR SEQ ID NO:17: Wi SEQUENCE CHARACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:1?: TTGA.ARCCNG CYTGGGCTTC INFORMATION FOR SEQ ID NO:lB: Ci) SEQUENCE CHARACTERISTICS: LENGTH: 20 bae pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:lB: TTGAARCCNG CY2GAGCTTC INFORMATION FOR SEQ ID NO:19: Ci) SEQUENCE CHARtACTERISTICS:.

CA) LENGTH: 20 base pairs TYPE: nucleic acid C)STRANDEDNESS: single CD) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SMSE: YES Cxi) SEQUENCE DESCRIPTION: SEQ ID NO:19: TTQAARCCNG CYTGTGCTTC INFORMATION FOR SEQ ID NO:20: U SUBSTITUTE SHEET (RULE 26)

U.

V. PCT[US94/04495 W0 94125567 -122- SEQUENCE CHARACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear t (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID TTGA.\RCCNrG CYTGCGCTTC INFORMATION FOR SEQ ID NO:21: SEQUENCE CHARACTERISTICS:I LENGTH: 20 base pairs TYPE: nucleic acid STRANDEDNESS: single CD) TOPOLOGY: linear i (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO Civ) ANTI-SENSE: YES Cxi) SEQUENCE DESCRIPTION: SEQ ID NO:21: TTGAARCCNG CYTGGGCCTC INFORMATION FOR SEQ ID NO:22: Wi SEQUENCE CHARACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid STR.ANDEDNESS: single TOPOLOGY: linear "K Ui) MOLECULE TYPE: DNA (genomic) iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:22: SKEETNC (RULECT26)TI20 LENGTH: 20 base pairs C 4 WO 94/25567 PTU9149 -123- TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:23: TTGAARCCNG CYTGTGCCTC INFORMATION FOR SEQ ID NO:24: Ci) SEQUENCE CHARACTERISTICS: LENGTH: 20 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (gonomic) iii) HYPOTHETICAL: No (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:24: TTGAARCCNG CYTGCGCCTC INFORMATION FOR SEQ ID SEQUENCE CHARACTERISTICS: LENGTH: IS base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (i)MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: No (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID GGNGCNAARG TNGAYTCN 18 INFORMATION FOR SEQ ID NO:26: Ci SEQUENCE CHARACTERISTICS: LENGTH: 18 base pairsF ()TYPE: nucleic acid CC) STRANDEDNESS: single SUBSTITUTE SHEET (RULE 26) PO<4 '44 WO 94125567 PCT[US94/04495 T- 41 a I I I I i -11- II rXI~ l WO 94/25567 PCT/US94/04495 124 TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:26: GGNGCNAARG TNGAYAGY INFORMATION FOR SEQ ID NO:27: SEQUENCE CHARACTERISTICS: LENGTH: 18 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:27: NGARTCNACY TTNGCNCC INFORMATION FOR SEQ ID NO:28: SEQUENCE CHARACTERISTICS: LENGTH: 18 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:28: RCTRTCNACY TTNGCNCC INFORMATION FOR SEQ ID NO:29: SEQUENCE CHARACTERISTICS: LENGTH: 17 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear SUBSTITUTE SHEET (RULE 26) I r WM~ s~ ~.L~tA .XL L I %UA UliA AWjA CAL; LUX. 'LrW*X .jLU-Lc .tsj Auri Pro Glu Val Gly Lou Pro Lou Ala Gly Arg His Pro Ph. Asn Ser 530 535 540 SUBSTITUTE SHEET (RULE 26) W0 9425567PCTJ[US94/04495 -125- (ii) MOLECULE TYPE: DNA (genomic) iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:29: GAGTCNACYT TRGCGCC 17 INFORMATION FOR SEQ ID Ci SEQUENCE CHARACTERISTICS: LENGTH: 17 base pairs TYPE: nucleic acid STRANDEDNESS: single CD) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID GAGTCNA"YT TRGCACC 17 INFORMATION FOR SEQ ID NO:31: Ci) SEQUENCE CHARACTERISTICS: CA) LENGTH: 17 base pairs TYPE: nucleic acid STRANDEDNESS: single CD) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (gonomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES Cxi) SEQUENCE DESCRIPTION: SEQ ID NO:31: GAGTCKACYT TRGCTCC 17 INFORMATION FOR SEQ ID NO:32: Ci) SEQUENCE CHARACTRISTICS: CA) LENGTH: 17 base pairs CB) TYPE: nucleic acid CC) STR&NDEDNESS: single CD) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA ,(k-enoMic) SUBSTITUTE SHEET (RULE 26) WO 94/25567 WO 9425567PCT1US94/04495 126 (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:32: GAGTCNACYT TRGCCCC INFORMATION FOR SEQ ID NO:33: Wi SEQUENCE CHARACTERISTICS: LENGTH: 17 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:33: GAGTCNACYT TYGCGCC INFORMATION FOR SEQ ID NO:34: Ci) SEQUENCE CHARACTERISTICS,.r LENGTH: 17 baue pairs TYPE: nucleic acid CCY STRANDEDNESS: single CD) iTOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomtic) iii) HYPOTHETICAL: NO Civ) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:34: GAGTCNACYT TYGCACC INFORMATION FOR SEQ ID Ci) SEQUENCE CHARACTERISTICS~: CA) LENGTH: 17 base pairs TYPE: nucleic acid CC) STRANDEDNESS: single CD) TOPOLOGY: linear Ci)MOLECULE TYPE: DNA (genomic) Ci)HYPOTHETICAL: NO

I-

1w SUBSTITUTE SHEET (RULE 26) 4 I -if.- WO 94125567 PCT/US94104495 127 (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID GAGTCNACYT TYGCTCC INFORMATION FOR SEQ ID NO:36: SEQUENCE CHARACTERISTICS: LENGTH: 17 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: YES (xi) SEQUENCE DESCRIPTION: SEQ ID NO:36: GAGTCNACYT TYGCCCC INFORMATION FOR SEQ ID NO:37: SEQUENCE CH RACTERISTICS: LENGTH: 48 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (xi) SEQUENCE DESCRIPTION: SEQ ID NO:37: GCCAGCGTTT CTAAGGAGAA AACATATGCC GATATTTCGT TTTACTGC INFORMATION FOR SEQ ID NO:38: SEQUENCE CHARACTERISTICS: LENGTH: 37 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO 48

L

i V- B '0 SUBSTITUTE SHEET (RULE 26) ~-~FriT~ 4 WO 94/25567 PCT/US94/04495 659 IS394- WO 94/25567 WO 9425567PCT1US94/04495 128 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:38: GCGCCTTATA ACGCGCATAT GGCCACCAGC AATCCTG INFORMATION FOR SEQ ID NO:39: Wi SEQUENCE CHARACTERISTICS: LENGTH: 6519 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO (iv) ANTI-SENSE: NO (ix) FEATURE: NAME/KEY: CDS LOCATION: 3238. .6276 (Xi) SEQUENCE DESCRIPTION: SEQ ID NO:39: G3AATTCCAT

AATTTAATGA

GCCGATATTT

CGCGATGGCA

TTACCATTTT

CACTCAATCA TTAAATTTAG AGGACGCATT GGTTTCACTG CGTTTTACTG, CACTTGCAAT GCCACCKGCA ATCCTGCATT GCACAAAATA ACCCATTAGC GCACAACGAT GGGCTATCAG CGTTATGACA TTAGCCAGCG TTTCTAAGGA GAAAAATAAT GACATTGGGG CTATTATCAG CGCCTTATAA TGATCCTAAA AATCTGATGC AGTCAGAAAT AGACTTCTCA TCAGATAAAA ACTCAATACT AAACCAATCT CTTTTATGGA AATGGAAAGG GATTGTCCCC ACCGATAAAG'AAGCATCTAA AACGTTATCT GATAAACGTA TGGTAGTAGC TTTACTTTAC AGCATGGGGA CGCTCATCTA TGATGGTTAT CTTACTATCG AGGCTTTAAA. GTAWATTAG CQATCTTQAA AATCGAGAGA AQACAGCLTT GGGCrITTCTT TAATtvTGAGT CAGGGTGAAA "FTACCA.&TGG TCTGATTATC (ICAACGTAAAG CCACAACTAC 'CCAACGTCTA )7"AATG&AT GAATATC N fZAA TGGAACGCAA GGCAGACATC

GCATTATGGG

ATAAAAAACT

CCCCCGTTTT

ATTTCGGAGA

ATTTCACTGG

TGACCTTAAA

TAGGTGCTAA

TCTATATCGA

AAGTAAAAAC

CTGTAACACC

TTGTCGGAGG

GTGZ.TT.TCGA

TGATCACTGA

CTCATTTTGG

AAAACTCATT

CTGGCGTGCT

TGCAkCCAAT

AGTCGATAGT

CCGTATTATG

TCGCTTAT"A

TGAAAATTTA

TGAAAAAGAG

TGCTCTTAAT

TAAACAAATC

CTTTACAATG AAA&ACCGAT

TCAACCAGTG

GTGGGAGTCT

ACCTCCTCTG

ATTCGTTTTA

TTTTCTGTCG

GAACCTGAAA%

GCGGCCATTG

ACAAACCTCG

ATTCACACTT

ATTATTTATC

AGGCTCAGGC

CTTTAAATAA

ATGGTACTCA

A.AGCGCCTTC

ATGATGCTCG

TTCAATTTCA

;k'-CTATTCG

CATTAGAAGA,

TAGCAAATGG

AACCAGAGAA

120 180 240 300 360 420 480 540 600 660 720 780 840 900 960 1020 1080 C4 w SUBSTITUTE SHEET (RULE 26) TCTTAACTCC CAAGATAAAC AACTATTTGA TAATTATGTT ATTTTAGGTA ATTAATGTTT AATATTAGCC ACTAAAGCAG ATGTACTTAT GAGTGCT~rTA GTGACAACCC GTTATTAATG TCTGATGCAC ACTGTGGTAT TCACGTGAGT TGATCTAGAT TATTTCAATA TGATGATCAA AAGCGTATCA AACGCAAGTG CCACCQGGTG TGAAGGCAAC TATCCGGGCT TTTATTACGC Q1.TACACCAT GATGGTTTCA GCGTGQATCT CCCTTTTAAC TCACCTTCGT TGCAAAATCA TCGCCTGzATP.

ACAAA1ATQPJA TCAACTGCTA TTTCTATGCC TTTAkLTGGCG ACTGA1JAGCT TATAACACCA TGGCCGTTAC CAAAGTCATG GGGCTATCA'G CAAGAAQGTT TCCTCTTAAA GACTTAGACA ATTTAGCGIGA ACATCATCCC TCCCGCCAAT CTTGAGCGTT TGATAATCAC TTAATTTTTA TQAAACGACC TTATTccAAc ACAPJJ.GATA GAAAACATGc TAGCA&TGGC AATGGTTACT TCAGGTTTMZ GCGGAJJTA GATCGATCAC AGCACTCGCC GACACCTGAA AMAATGGGAG GGTZCTCGT AAGGATAAAG ATATGCCTTT TATCAGCCAG

GTGCTTATGT

TAATGACAAA

ATCACTGGGG

TAAAA~GAAGC

TTAAAAGTAG

CCTTATCTCG

ACT.TAGTTAA

GTAAAGATGG

ACTCTTTCCC

TTTCAGTGGG

ACAGTAATCC

TAAAATCAGT

AAACACTTGC

TTTTTGGAGA

GTGCTTTTG/3

ATGTTTGGTC

GTGTCGCTCA

GGGATTGGAA

GTCCTAP*ACC

TTGAAGGTCA

TTGxATCCTAA

TTGGTAGCAA

ATGCCATTAC

CTTATCAAAC

TAATTACTCA

APA"TCGCCA

CCAAAGATGC

AGATGGCILCA

ACGTTCATAT

CATCAATTGA

GCTGGAAAAA

GCATTTATTA

ATACAGTTCT

GAACCTACAA

TTTTQATATG

CCAACATTTA

TACTTTCAGC

TTTACGCCCT

AGCCTTTAAA

TGAAAGTGGT

AGAhGTTGGA

CGCTCAAGGC

ATCTATTTAT

AACTATTACA

TATTCATCGT

ATCTGAAATT

AATAGTGA(IT

TAGAkTGCAA

TCATACCTTA

ATATGGCATG

TTTCACTGCG

TATAAATAGT

TCCAACATTA

AACACTTCAA

AGCAGAAAAA

ACCGACAGAA

CAGTTATGAG

AAAATTCCGT

TATTCTCQPLT

AGACAA6ATGG

GATCCCACAC

GATCAAGGCT

CGTTGGTGGT

ACTCkLAGTTT

AAAGTAAGTG

GCCTTATTAT

CATTATATCA

GATGGTACAG

AATGCCTCTC

TGGAATAACC

TTACCGCTTG

TATTACTGGC

ATTACACGAC

AAAAGGCGCA

TTGTTAAAGG

ATATTTCCAC

ATGATTCATT

CTGATAGCTC

TACTAGAGCC

CTGGCGCATT

CATGGCGACA

AGCTTATTTA

TGAAAAGC

CAGGAAGACA

TTGCCATGTC

CTTGCGATTA GTQATAAAAC CCAGCGTCTT TACCTCAkGG TGGCAAGATA AAATGGTGAC 1140 1200 1260 1320 1s 1440 1500 1560 1620 1680 1740 1800 1860 1920 1980 2040 2100 2160 2220 2280 2340 2400 2460 2520 2580 2640 2700 2760 2820 2880 2940 4 1 ~Ii

'H

TATAACAAAG

AATGGCTCGC

GGGGCAkCCA

ATGCAACGTG

ATGGCATTCG

AAAAAGAGTG

AGTGATAAAA

AATACCCTTT

CAAGGTGATT

GTAL&TGTAA

GGAACTTTA

ATAACCGTTA

AGCTTTCACA

CTATTCACCT

GAGAGCGTGG

ATCTTATTTA

TATTAGCCGC

ATAAAAATGT

GGPITTAATGG

GGTTAATTGA

GTCGCCAACA

GCTCGGCATG

TATATGGTCT TTTTAQPITGC GAAAATAATG GGTTATATcA AAACTCAGCA ATGTAACGGG ATCAAAAAGG TTAATAAACC SUBSTITUTE SHEET (RULE 26) iVO 94/25567 WO 9412567 TiCT1TJS94/0449S WO 94/25567 PCT1US,94/04495 -130- TGCAATTGTG ATGACTCLTC GACAAAAAGA CACTCTTATT TTTAAATATG ACTCGCCLLA AAGCAGCAAC TCCTGTCACC CAAATGGCAA TCTGCTGATA AA~a.TAGTGA AGTGAAATAT TGAACTGACG TTTACGAGTT ACTTTGGTAT TCCACAAGAA TTGATTTAAT CAAAAGAACG CTCTTGCGTT CCTTTTTTAT GTCAGTGCAG TTACACCTGA ATCAATGTCA. CGATTAATGG CAGGTTTCTG GTGATAACAC ATCLLACTCT CGCCACTCCC TTGCAGGAAA TCTGATT ACA TTA AGC CTC TGT ATG CTA ATA Met Lou Ile AAA AAC CCT TTA GCC CAC Lys Ann Pro Lou Ala His GCG OTT Ala Val Thr Lou Ser Lou Cys

I.

TTA

Lou

GOC

Gly

TCA

Bar

GAL

Giu

AAT

Ann

ACT

Thr

ACG

Thr

GAL

Olu

ATO

Not 145

CCA

Pro

CCG

Pro

ALT

Ann TCA TTA Ser Lou GAT ATT Asp Ile 35 ALT ALT Ann Ann 50 CAA TCA Gin Bar AAT ATT Ann Ile TTT ATG Ph. Mot TTA GCA Lou Ala 115 CT? ALT Lou Ann 130 CAA GOC Gin Glly AAC CAA Ann Gin TTA GAC Lou Asp AAC GCA Ann Ala 195 5 CCC GCA Pro Aia 20 TAT CTT Tyr Lou ALT CAA Ann Gin CTC AA Lou Lys GTT AT Vai Ann ATG TOG Met Trp 100 TTT AA Ph. Lys TTT ACG Phe Thr TCT OCG Sar Ala GCC GGA Ala Gly 165 ALT COT Ann Arg 180 OTA AAC Val Ann CAA OCA TTA Gin Ala Lou TTT GAL GOT Phe Giu Gly 40 TTA TCO CTA Lou Bar Lou 55 TOO CAA TAT Trp Gin Tyr 70 TAC CAA OAT Tyr Gin Asp ATT TAT ALT Ile Tyr Ann CAA AAT ALT Gin Ann Ann 120 G00 TOG COA Gly Trp Arg 135 ACA GOT CAA Thr Gly Gin 150 ACA CTC TTT Thr Lou Ph.

TGG GCL GTA Trp Ala Vai ACO ATG OTT Thr Z4et Val 200 10 CCC ACT CTO Pro Thr Lou 25 OAA TTA CCC Olu Lou Pro AOC AAA CGO Sor Lys Gin CAA CCA CAA Gin Pro Gin 75 OAT AAA ALT Asp Lys Ann 90 GAL AAA CCT Oiu Lys Pro 105 AAA ATT OCA Lys Ile Ala GOT ATT OCT Oly Ile Ala CTT OAT CAA Lou Asp Gin 155 TTT OAT CAA Ph. Asp Gin 170 CCT GAC TAT Pro Asp Tyr 185 LOT AAA AAC Bar Lys Ann TCT CAT Sor His ALT ACC Ann Thr CAT OCT His Ala OCA ACA Ala Thr ACA 0CC Thr Ala CAL TCT Gin Bar CTL AGT Lou Ser 125 OTT CCT Val Pro 140 TTA OTO Lou Val ATC ATC 110 Ile CAA ACA Gin Thr TOG LOT Trp Bar 205 GAL OCT TTC Giu Ala Phe CTT ACC ACT Lou Thr Thr ALA OAT GOT Lys Asp Gly TTA ACA CTL Lou Thr Lou ACA CCA CTC Thr Pro Lou TCC CCA TTA Bar Pro Lou 110 TTT ALT OCT Pho Ann Ala TTT COT OAT Ph. Lrg Asp ATC ACC OCT Ile Thr Ala 160 ATO LOT OTA Met Bar Val 175 CCT TAC GTA Pro Tyr Val, 190 OCA TTA TTO Ala Lou Lou 3000 3060 3120 3180 3237 3285 3333 3381 3429 3477 3525 3573 3621 3669 3717 3765 3813 3861 4

I.

L

SUBSTITUTE SHEET (RULE 26)

I

A ,aflY _4 1 'WO 94/25567 PCT1US94/04495 131

ATG

Met

ACT

Thr 225

TTT

Phe

ATG

Met

CAC

His

CMA

Gin Ala 305

CAA

Gin

AAA

Lys

CMA

Gin

TAC

Tyr

CTT

Lou 385

TAC

Tyr

GCA

Ala

TTA

Lou

CTG

Lou TAC GAT CAG ATG Tyr Asp Gin Met 210 GMA TTT CGC GAT Glu Phe Arg Asp GMA TAT TAT CMA Glu Tyr Tyr Gin 245 CTA GAT MAA CAT Lou Asp Lys His 260 GCT GAT GGC TCA Ala Asp Gly Ser 275 CAT TTT ATG MA His Ph. Met Lys 290 TTA CTT GAT GCC Lou Lou Asp Ala ACT GCT ATT TAC Thr Ala Ile Tyr 325 MAA TTA GMA GAG Lys Lou Glu Glu 340 GGT TTT ACA CGA Gly Ph. Thr Arg 355 CMA ACC AGA GMA Gin Thr Arg dlu 370 OCA MAA MAT MC Ala Lyn Asn An MAC GCC ACA GGA Asn Ala Thr Gly 405 M&T GTC GAT ATT Asn Val Asp Ile 420 TTG ATG CTA CCG Lou Mat Lou Pro 435 CAA AGT TUG CTA Gin Ser Trp Lou

GCC

Ala

ACA

Thr

CGT

Arg

TTA

Lou

GUA

Gly 280

GGT

Gly

CTA

Lou

AGC

Ser

TTA

Lou

GGT

Gly 360

GAT

Asp

GCC

Ala

TTT

Ph.

ACT

Thr

CMA

Gin 440

ACC

Thr TTA MAC TTC Lou Asn Phe ATT TAT CAG Ile Tyr Gin ATT ACT CCA Ile Thr Pro 255 GTG TTA ACA Val Liou Thr 270 CAC CCT MAC His Pro Asn 285 GGG ACT CMA Gly Thr Gin AM& ACG CTT Lye Thr Lou GCA ACT GAT Ala Thr Asp 335 TAT GTC CTT Tyr Val Lou 350 ACT CAT OTT Thr His Val 365 GGC CGT CAT Gly Arg His GCT ATO ATG Ala Met Met GMA ATT GTT Glu le Val 415 ATG ATA A Met Ile Lyu 430 GCC TTA GCG Ala Lou Ala 445 MAA GUT GTT Lys Gly Val 3909 3957 4005 4053 4101 4149 4197 4245 4293 4341 4389 4437 4485 4533 45811 4629,

I

SUBSTITUTE, SHEET (RULE 26) WO 94/25567 PCTIUS9404495 132 450 GGC GGT Gly Gly 465 CCC GCT Pro Ala GCA TTA Ala Lou TTA AAA .Lou Lys ATT CCT I1 Pro 530 GGG ATC Gly 1i.

545 AAA CAA Lys Gln AAC AAA Ann Lys GCA TGG.

Ala Trp ACC CM Thr Gln 610 CGT TAT Arg Tyr 625

CGTTAT

Arg $'yr CAA TC Gln Ber ACA ACA Thr Thr CAA TTA Gin Lou 690 TTC AAA TCT Phe Lys Ser OCT AAA Ala Lys 485 GAT TCA Asp Ser 500 GTT TTG Val Leu GTA TTA Val Lou CCA TTT Pro Ph.

TTA GAT Lou Asp 565 CAT TTT His Ph.

580 ATG AAT Met Asn CCA CAA Pro Gln OTT GOT Val Gly CAA TAT Gin Tyr 645 TTT AGC Phe Ser 660 ATT CAT ie His GCT OCA Ala Ala 455 OAT GGT Asp Gly 470 OAT OCA Asp Ala CCT TTT Pro Ph.

TTA AAA Lou Lys AGT GGT Ser Gly 535 AAA TGG Lys Trp 550 ACC ACA Thr Thr OAA GOC Giu Gly TAT GCA Tyr Ala CAA AGC Gin Bar 615 AAT GAA Ann Glu 630 GGA CAA Giy Gln CAT GCT His Ala CTT CCC Lou Pro GOT ATT Gly 1i.

695 TCT ATT Ser Ile TTT GOGT Pho Gly CGC TTA Axg Lou 505 ATG CGG Met Arg 520 CGT CAT Arg His ATO GCA Met Ala TTA TCC Lou Bar ATT AAC Ile Ann 585 TCA ATG Bar Met 600 TGG CTC Trp Lou AGC TAT Bar Tyr TTG GAA Lou Glu GGA TGG Oly Trp 665 TAT AAC Tyr Asn 680

TTT

Phe

GGT

Gly 490

TCT

Ber

ATC

lie

CCA

Pro

TTA

Lou

GCC

Ala 570

OCT

Ala

GCA

Ala

GCC

Ala

GAA

Glu

ATT

Ile 650

GAT

Asp

GAK

Glu His 475

TTA

Lou

ACT

Thr

TAC

Tyr

ACT

Thr

OCA

Ala 555

OCT

Ala

OAA

Glu

ATA

110

ATA

Ii1

AAT

Ann 635

ATT

Ile

TGG

Trp

CTT

Lou 2is

OCA

Ala

TCA

Ser

ACC

Thr

GGG

Gly 540

OGA

Gly

TAT

Tyr

AGT

Ber

CAA

Gln

GCG

Ala 620

AAC

Ann

CCA

Pro

AAT

OAA

Oiu

CTT

Lau 700 460 CAC CAT TCA CA CAT TAC Ser Gin His Tyr 480 CCC AGT GTT TAT Pro Bar Val Tyr 495 GCA CAT GAG CGT Ala His Glu Arg 510 MAA GAG ACA CAA Lys Glu Thr Gln 525 TTG CAT AAA ATA Leu His Lys Ile ACC CCA GAT GGC Thr Pro Aar Gly 560 GCA AAA TTA GAC Ala Lys Lou Asp 575 GAG CCA GTC GOC Giu Pro Val Gly 590 CGA AGA GCA TCG Arg Arg Ala Ber 605 COC GOT TTT AGC Arg Gly Phe Ber AAC CGT TAT GOT Ken Arg Tyr Giy 640 OCT GAT TTA ACT Ala Asp Lou Thr 655 AGA TAT CCA GGT Arg Tyr Pro Gly 670 OCA AAA CTT AAT Ala Lys Lou Ann 685 TCA ACA GAA AGT Sor Thr Glu Ber 4677 4725 4773 4821 4869 4917 496J 5013 5061 5109 5157 5205 5253 5301 5349 r i ,r 1 GA GMA ATO TTO Glu Glu Mot Lou c -I T SUBSTITUTE SHEET (RULE 26)

S..

WO 94/25567 PCTfUS94/04495 133

TAC

Tyr 705

TTA

Leu

TCC

Ser

GAA

Glu

GCC

Ala

CAA

Gin 785

CC?

Pro

AG?

Bar

GAA

alu

AAT

Ann

GCT

Ala 865 Lys

AAG

Lys

ATG

met

A??

Ann

GAT

Asp TCT GGT GCA Bar Gly Ala CAC GGT CAC Hi~s Gly His TAT TTC TTA Tyr Phe Leu 740 AAT GAT GAT Ann Asp Asp 755 GTC CC? AAA Val Pro Lys 770 TTA GAT ACT Lou Asp Thr GCC GGC AAT Ala Gly Ann TAT CAA AAA Tyr Gin Lys 820 CAA TTA TTT Gin Lou Ph.

835 GAA AA? TAT Glu han Tyr 850 CCC GAA TAC Pro Glu Tyr GAT AAA ATA Asp Lys Ile TTA AAA TCA Lou Lys Bar 900 GTC ATG OCT Val Not Ala 915 CCT GAT TTA Pro Asp Lou 930 AAA GGT A&T Lys Gly Aen ACC CT? Thr Leu 710 AAA TAT Lys Tyr GAT AAk? Asp Ann CAA CAT Gin His CAG TCA Gin Ser 775 TTA AC? Lou Thr 790 TAT AAG Tyr Lye CAT ?CA His Bar ACA GCT ?hr Ala TA? GCA Tyr Ala 855 OTA TTA Val Lou 870 CAA GAk Gin Glu GAT GdA Asp Ala ATA CA 110 Gin TTA TAT Lou Tyr 935 ATC GAk Ile Glu AA? Ak? Ann Ann CAA CAA Gin Gin AGA GTT Arg Val 745 ACG ACC Thr Thr 760 GTG ATC Val Ile TTA Ak? Lou An CTC ACT Lou Thr CT? GA? Lou Asp 825 OTT AT? Val Ile 840 ATA OCT Ile Ala CAA CAT Gin His GAG GGA Glu Oly ACA TTA Thr Lou 905 AA? CAO Ann Gin 920 CAA GOT Gin Oly GTT AG? Val Ser AAC AG? Ann Ser 715 CAA AGC Gin Ser 730 ATT OCT Xie Ala OAk ACA Giu Thr AT? Ak? Ile Ann Ak? GCA Ann Ala 795 AAA OGA Lye Gly 810 GAT AGA Asp Arg TC T CAT Bar His ATC GAA 110 Oiu Ak? GA? Ann Asp 875 TAT GCT Tyr Ala 890 TTA TCC Lou Ser CAA TTA Gin Lou AGA OAk Arg Giu GTT TAT Val Tyr ATG TTT GCC Met Pho Ala TTA AGG GCA Leu Arg Ala TTA GGC TCA Lou Gly Ser 750 ACA CTA TTC Thr Lou Ph.

765 GGC AAA AAG Gly Lys Lys 760 GAT ACA TTA Asp Thr Lau CAA ACT GTA Gin Thr Val AK? TCA AA Asn Bar Lys 830 G? AAG GC1A Giy Lye Ala 845 GdA CAA AAT Ala Gin Ann 860 CAG CTC CAT Gin Lou His TTT TTT OAK Ph. Ph. 3lu AG? GA? GCG Bar Asp Ala 910 ACA TTA AG? Thr Lou Bar 925 AAA GAT CAA Lys Asp Gin 940 TC- ,-CGT CAT Soe";Arg HiAl 5397 5445 5493 5541 5589 5637 5685 5733 5781 5829 5877 5925 59*73 6021 6069 6117 I 7

U

SUBSTITI LTE SHEET (RULE 26) Int .ional applicatiqf No.I TNTFRNATIONAL SEARCH REPORT

'I

WO 94125567 PCTUS94/04495 134 945 950 955 960 ACA GCA GAA TCG CAA TCA ACA AAT AGT ACT ATT ACC GTA AAA GGA ATA Thr Ala Giu Ser Gin Ser Thr Asn Ser Thr Ile Thr Val Lys Gly Ile 965 970 975 TGG AAA TTA ACG ACA CCT CAA CCC GGT GTT ATT ATT AAG CAC CAC 3AAT Trp Lys Lou Thr Thr Pro Gin Pro Gly Val Ile Ile Lys His His Ann 980 985 990 AAC AAC ACT CTT ATT ACG ACA AC~A ACC ATA CAG GCA ACA CCT ACT GTT Ann Ann Thr Leu Ile Thr Thr Thr Thr Ile Gin Ala Thr Pro Thr Val 995 1000 1005 ATT AAT TTA GTT AAG TAAATTTCGT AACTTTTAAA CTAAAGAGTC TCGACATAAA Ile Asn Lou Vai Lys 1010 AATATCGAGA CT'CTTTTTAT T7AAAAAATTA AAAACAAGTT AACGAATGAA TTAATTATTT GAAAAATAAA AAATAAATCG ATAGCTTTAT TATTGATAAT AAATGTGTTG TGCTCAATGG TTATTTTGTT ATTCTCTGCG CGGATGCTTG GATCAATCTG GTTCAAGCAT ATCGCAAGCA CCAGAPACGAA AAAAGCCCCG GGT INFORMATION FOR SEQ ID Wi SEQUENCE CHMAZACTERISTICS: LENGTH: 1013 amino acids TYPE: amino acid TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID Met Leu Ile Lys Ann Pro Lou Ala His Ala Val Thr Leu Ser Leu Cys 10 Lou Ser Lau Pro A~la Gin Ala Lou Pro Thr Lou Ser His Giu Ala Pho 25 GIV Asp Ile Tyr Lou Phe Giu Gly Glu Lou Pro Ann Thr Lou Thr Thr 35 40 Ser Ann Ann Ann Gin Lou Ser Lou Sor Lys Gin His Ala Lys Asp Gly so 55 Giu Gin Ser Lou Lys Trp Gin Tyr Gin Pro Gin Ala Thr Lou Thr Lou 70 75 ~Ann Ile V~il Ann TIyr Gin Asp Asp Lys Ann Thr Ala Thr Pro Lou 8590 Phe Met Met Trp le Tyr Ann Giu Lys Pro Gin Ser Ser Pro Lou 100 105 110 Thr Lqu Ala Ph. Lys Gin Ann Ann Lys 119 Ala Lou Ser Ph. Ann Ala 1135 120 125 6165 6213 6261 6316 6376 6436 6496 6519

I'

I

iJE SUBSTITUTE SHEET (RULE 26) -n ,Ak vt~ne ol-m f~nA Irion re.4 hlo ((Cnfi-wiutk'n nf lpm I af frst shhki tiUbb I I I U I t bHtt I h= /D) WO 94/25567 Wo 9425567PCTIUS94/04495 135 Giu Met 145 Pro Pro Asn Met Thr 225 Ph.

Met His Gin Al a 305 Gin Lys Gin Tyr Lou 385 Tyr Ala Lou Lou Phe Sor Ala Asn 180 Val Gin Arg Tyr Lys 260 Gly Met Asp Ile Giu 340 Thr Arg Asn Thr Asp 420 Lou Trp Thr Gly Ala Thr 150 Gly Thr 165 Arg Trp Asn Thr Met Ph.

Asp Asp 230 Gin Gly 245 His Lou Sor Ile Lys Val Ala Asn 310 Tyr Lou 325 Giu Arg Arg Gly Giu Lou Asn Lou 390 Gly Arg 405 le Lou Pro Asp Lou Aun Gly Lou Phe Pro 185 Ser His Giu Sor Trp 265 Lys Val Arg Asp Lou 345 Tyr Ala Pro Glu Gin 425 Gin Ile Ile Ala Asp Gin 155 Asp Gin 170 Asp Tyr Lys Asn Tyr Pro Met Ala 235 Asp Lys 250 Giu Lys Ala Lou Ph. Sor Asp Val 315 Sor Lou 330 Gly Thr Gin Ile Trp Ph.

Thr Gin 395 Lys Asn 410 Lou Gin Arg Gin Lou Ser

PI

7, 7, SUBSTITUTE SHEET (RULE 26) WO 94/25567 CTU9I45 ICT[US94/04495 136 Gly 465 Pro Ala Leu Ile Gly 545 Lys Asn Ala Thr Arg 625 Arg Gin Thr Gin Tyr 705 Lou Ber Giu Ala Gin 785 Giy Phe Lys Ala Tyr Ala Lou, Ser Asp 500 Lys Asp Val 515 Pro Val Val 530 Ile Ala Pro Gin Lys Leu Lys Thr His 580 Trp Ala Met 595 Gin Ser Pro 610 Tyr Lau Vai Tyr Lou Gin Ser Gly Ph.

660 Thr Thr Ile 675 Lou Pro Ala 690 Ser Gly Ala His Gly His Tyr Ph. Lou 740 Asn Asp Asp 755 Val Pro Lys 770 Lou Asp Thr Asp 470 Asp Pro Lau Ser Lys 550 Thr Giu Tyr Gin An 630 Sly His Lou Giy Thr 710 Lys Asp Gin Gin Lou 790 Giy Ala Phe Lys Gly 535 Trp Thr Gly Ala Ser 615 Giu Gin Ala Pro Ile 695 Lou Tyr An His Ser 775 Thr Ser Ile Ph. Giy Arg Lou 505 Met Arg 520 Arg His Met Ala Lou Ser Ile Asn 585 Ser Met 600 Trp, Lou Ser Tyr Lou Giu Gly Trp 665 Tyr Asn 680 Giu. Giu Ann Asn Gin Gin Arg Vai 745 Thr Thr 760 Val Ile Lou An Ph.

Gly 490 Ser Ile Pro Lou Ala 570 Ala Ala Ala Giu Ile 650 His 475 Lou Thr Tyr Thr Ala 555 Ala Gili Ile 118 Asn 635 Ile Gin Ser His 510 Giu His Pro Lys Pro 590 Arg Giy Arg Asp Tyr 670 Lys Thr His Val 495 Giu Thr Asp Lou 575 Val Ala Ph.

Tyr Pro Lou Glu Tyr 480 Tyr Arg Gin Ile Giy 560 Asp Gly Ser Ser Giy 640 Thr Gly An Ser Lys 720 Lys le Ph.

An Asp 800 Aiup Trp Asn Glu Lou Giu Met Lou Lou 700 Ann Sor Met 715 Gin Sor Lou 730 119 Aia Lou Giu Thr Thr Ile Asn Gly 780 Ann Ala Ph. Ala Met Arg Aia Gly Sor 750 Lou Ph.

765 Lys Lys Thr Lou SUBSTITUTE SHEET (RULE 26) WO 94/25567 Wo 9425567PCTJUS94/04495 137 Pro Ser Giu Asin Ala 865 Lys Lys Met Asn Asp 945 Thr Trp Asn Ile Ala Gly Asn Tyr Gin Lys 820 Gin Leu Phe 835 Glu Aen Tyr 850 Pro Glu Tyr Asp Lys Ile Lou Lys Ser 900 Val Met Ala 915 Pro Asp Lou 930 Lys Gly Asn Ala Giu Ser Lys Leu Thr 980 ken Thr Leu 995 Asn Lou Val 1010 Leu 805 Gin Ala Giu Thr Thr 885 Ala Lys Aen Gin Gin 965 Thr le Lys Lou Thr Leu Asp 825 Val Ile 840 Ile Ala Gin His Giu Gly Thr Lou 905 ken Gin 920 Gin Gly Vai Ser ken Ser Pro Gly 985 Thr Thr 1000 Giy Arg His Giu Asp 875 Ala Ser Leu Giu Tyr 955 Ile le Ile Gin Ala Thr Pro Thr Val 1005 INFORMATION FOR SEQ ID 110:41: Wi SEQUENCE CHARACTERISTICS: LENGTH: 36 base pairs TYPE: nucleic acid STRANDEDNESS: single TOPOLOGY: linear (ii) MO0LECUJLE (iii) HYPOTHETIC (iv) ANTI- SENS] (xi) SEQUENCE I ATTTGCAGGA AATCTGC EYPE: DNA (genomic) :AL: NO I: NO )ESCRIPTION: SEQ ID 110:41: LTA TGCTAATAAA AAACCC SUBSTITUTE SHEET (RULE ko)

L

4

Claims

1. A purified isolated DNA fragment of Proteus vulgaris (P. vulgaris) which encodes chondroitinase I enzyme.

2. A purified isolated DNA fragment of P. vulgaris, which encodes amino acids numbered 1-1021 of SEQ ID NO:2 or a mutant thereof that retains chondroitinase I enzymatic activity.

3. The purified isolated DNA fragment of Claim 2, wherein the fragment has the sequence of the nucleotides numbered

119-3181 of SEQ ID NO: 1, or the nucleotides numbered 119-3181 of SEQ ID NO: 3. 4. A purified isolated DNA fragment of P. vulgaris, which encodes amino acids numbered 25-1021 of SEQ ID NO: 2 of a mutant thereof that retains chondroitinase I enzymatic activity. The purified isolated DNA fragment of Claim 4, wherein the fragment has the sequence of the nucleotides numbered

191-3181 of SEQ ID NO:1, or the nucleotides numbered 191-3181 of SEQ ID NO:3. 6. A purified isolated DNA fragment of P. vulgaris, which encodes the amino acids numbered 24-1021 of SEQ ID NO:5 or a mutant thereof that retains chondroitinase I enzymatic activity. 7. The purified isolated DNA fragment of Claim 6, wherein the fragment has the sequence of nucleotides numbered 188-3181 of SEQ ID NO:4. 34333 '3 3~ .5 S A 3, S *5 3 *3 33 3 3)333 3, 3 3' 3~ 33 3 3 0 3 3 3 3, 33 33 I- L -139- 8. A plasmid containing a purified isolated DNA fragment of P. vulgaris as defined in Claim 1. 9. The plasmid of Claim 8 wherein the plasmid is that designated pTM49-6 or that designated LP 2

1359. A host cell transformed with the plasmid of Claim 8. 11. The host cell of Claim 10 wherein the plasmid is that designated pTM49-6 (ATCC 69234) of that designated LP 2 1359 (ATCC 69598). 12. A method of producing chondroitinase I enzyme which comprises transforming a host cell with the plasmid of Claim 8 and culturing the host cell under conditions which permit expression of said enzyme by the host cell. 13. A method for the isolation and purification of the recombinant chondroitinase I enzyme of Proteus vulgaris from host cells, said method comprising the steps of: lysing by homogenization the host cells to release the enzyme into the supernatant; subjecting the supernatant to diafiltration to remove salts and other small molecules; passing the supernatant through an anion exchange resin-containing column; loading the eluate from step to a cation exchange resin-containing column so that the enzyme in the eluate binds to the cation exchange column; and 2~.2 I. 5# 3 .3 fl I I .5 I 2) .2 211.5 22 .2 I) .2-2 21 1 I .2 I 1.2 II 3) if -140- eluting the enzyme bound to the cation exchange column with a solvent capable of releasing the enzyme from the column. 14. The method of Claim 13, wherein prior to step the following two steps are performed: treating the supernatant with an acidic solution to precipitate out the enzyme; and (ii) recovering the pellet and then dissolving it in an alkali solution to again place the enzyme in a basic environment. "I A recombinant chondroitinase I enzyme isolated and purified by the method of Claim 13 or by the method of Claim 14. DATED this 19th day of August, 1998 AMERICAN CYANAMID COMPANY By Its Patent Attorneys DAVIES COLLISON CAVE 1 4 4 I'I I i 1- 1 r ,L 1 A