This invention relates to substances having tissue plasminogen activator-type (tPA) activity. More specifically, this invention relates to "recombinant" thrombolytiσ proteins, a process for obtaining the proteins from genetically engineered cells, and the therapeutic use of the substances as thrombolytic agents.

These proteins are active thrombolytic agents which, it is contemplated, possess improved fibrinolytic profiles relative to native human tPA. This may be manifested as increased affinity to fibrin, decreased reactivity with inhibitors of tPA, faster rate of thrombolysis, increased fibrinolytic activity, decreased or at least acceptable levels of fibrinogenolysis and/or prolonged biological half-life. It is also contemplated that proteins of this invention can be more conveniently prepared in more homo¬ geneous form than can native human tPA. An improved overall pharmaσo inetic profile is contemplated for these proteins permitting administration, when desired, by bolus injection of pharmaceutical compositions containing these proteins.

More particularly, in the course of research involving the preparation and study of a series of tPA variants, we have found that a particular class of such variants is characterized by advantageous properties such as are mentioned above. As described in greater detail hereinafter, this invention provides novel protein analogs of human tPA which are characterized structurally by an amino acid sequence substantially the same as that of native tPA, except within the 91-amino acid mature N-terminus of tPA where a series of amino acids found in the native peptide sequence are deleted and/or replaced with different amino acids resulting in a new synthetic peptide domain resembling the kringle- 1 domain of human plasminogen. The protein variants of this inventionmay optionally be additionallymodified, e.g. as described in published International Application No. WO 87/04722, for

instance at one or more of the N-linked glycosylation sites and/or at the plasmin cleavage site spanning tPA positions 275- 276. It is contemplated that the proteins of this invention possess improved fibrinolytic and pharmacokinetic profiles relative to both native human tPA and certain other modified forms of tPA. Illustrative variants are depicted in Table 1, below.

The polypeptide backbone of natural human tPA is known to contain four consensus Asn-linked glycosylation sites. It has been shown that two of these sites are typically glycosylated in tPA from melanoma-derived mammalian cells, i.e. at snn ₇ and sn ₄₄ g. sn*L ₃ is glycosylated sometimes and sn ₂ is is typically not glycosylated. tPA from melanoma-derived mammalian cells, e.g. Bowes cells, is also referred to herein as "native" or "natural" human tPA.

This invention, as mentioned above, involves novel protein analogs of human tPA which possess tPA-type thrombolytic activity. The proteins of this invention, as illustrated in Table 1, differ in structure from human tPA in that they contain modifications in peptide sequence (i) at up to three of the Asn-linked glycosylation sites present in native tPA; (ii) within the 91-amino acid mature N-terminus of tPA; and/or (iii) at the proteolytic cleavage site spanning Arg-275 and Ile-276. The numbering of amino acids is shown in the one-letter code sequence of Table 1. For convenience, a gap in numbering is included within the pentapeptide sequence "CE EID" between E-82 and E-85 so that conventional numbering of tPA amino acids is retained e.g. at glycosylation sites and at the plasmin cleavage site. It should be understood, however, that the polypeptide link between the various structural domains, including the link between C-81 and C-95, may be varied in amino acid composition and length, and that this invention encompasses compounds with such variations.

A. Modifications at the N-terminus

In one aspect of this invention the proteins are characterized by replacement of all or a part of the N-terminal domain of tPA (which spans amino acids 1-91 of the native sequence) with a peptide domain the same or substantially the same as the first kringle region (kringle-1) of plasminogen. In an illustrative example of the proteins of this invention, amino acids 1-84 of tPA are replaced with the 82 amino acid sequence substantially as depicted at the mature N-terminus in Table 1. These proteins thus comprise a class of chimeric proteins comprising, in N-terminal to C-terminal order, a fusion of a plasminogen kringle-1 protein domain with a tPA serine protease domain, linked in various embodiments by other tPA domains including the tPA kringle-1 and tPA kringle-2 domains.

B. Modifications at N-linked Glycosylation Sites

As illustrated in Table 1, the protein variants of this invention may further contain no N-linked carbohydrate moieties or may be only partially glycosylated relative to natural human tPA. A "partially glycosylated" protein, as the phrase is used herein, means a protein which contains fewer N-linked carbohydrate moieties than does fully-glycosylated native human tPA. This absence of glycosylation or only partial glycosylation results from amino acid substitution or deletion at one or more of the concensus N-linked glycosylation recognition sites present in the native tPA molecule. We have found that variant proteins of this invention embodying such modification at one or more N-linked glycosylation sites retain tPA-type thrombolytic activity with greater fibrinolytic activity in certain cases, may be more readily produced in more homogeneous form than native tPA, and in many cases have longer in vivo half-lives than native tPA.

Table 1: Exemplary Variants

-35 [MDAMK RGLCCVLLLC GAVFVSPSQE IHARFRRGAR]

1 SECKTGNGKN YRGTMSKTKN GATCQKWSST SPHRPRFSPA THPSEGLEEN

51 YCRNPDNDPQ GPWCYTTDPE KRYDYCDILE CE EID TRATCYEDQG

98 ISYRGTWSTA ESGAECTNW- R^IAQKPYS GRRPDAIRLG LGNHNYCRNP

148 DRDSKPWCYV FKAG YSSEF CSTPACSEGN SDCYFG-R ² A YRGTHSLTES

198 GASCLPWNSM ILIGKVYTAQ NPSAQALGLG.KHNYCRNPDG DAKPWCHVLK

248 NRRLTWEYCD VPSCSTCGLR QYSQPQFJIK GGLFADIASH PWQAAIFAKH

298 RRSPGERFLC GGILISSCWI LSAAHCFQER FPPHHLTVIL GRTYRWPGE

348 EEQKFEVEKY IVHKEFDDDT YDNDIALLQL KSDSSRCAQE SSWRTVCLP

398 PADLQLPDWT ECELSGYGKH EALSPFYSER LKEAHVRLYP SSRCTSQHLL

448 -R ³ VTDNMLC AGDTRSGGPQ ANLHDACQGD SGGPLVCLND GRMTLVGIIS

498 WGLGCGQKDV PGVYTKVTNY LDWIRDNMRP

compound R ¹ R ² R ³

1-0 NSS NGS NRT 1-1 QSS NGS NRT 1-2 NSS QGS NRT 1-3 NSS NGS QRT 1-4 QSS QGS NRT 1-5 QSS NGS -RT 1-6 NSS -GS QRT 1-7 QSS -GS -RT 1-8 1-9 N-Q N-S N-T 1-10 N— N— N— 1-21 TSS TGS QRT

J = an amino acid or peptide bond; R ¹ , R ² , and R ³ are independently selected from the group consisting of a peptide bond, amino acid, dipeptide or tripeptide' "-", "—" and " " = a peptide bond.

Additionally, variants of this invention may contain at position 245 either M or V (as shown above) , and typically begin their N- ter inus earlier in the depicted sequence than SEC... as shown, e.g. beginning with SQEIHARFRRGARSEC ... , RGARSECK... , and/or, most typically, GARSECK...

N-linked glycosylation recognition sites are presently believed to comprise tripeptide sequences which are specifically recog¬ nized by the appropriate cellular glycosylation enzymes. These tripeptide sequences are either asparagine-X-threonine or aspar- agine-X-serine, where X is usually any amino acid. Their location within the peptide sequence is shown in Table 1 as R ¹ , R ² and R ³ , respectively. A variety of amino acid substitutions or deletions at one or more of the three positions of a glycosylation recognition site results in non-glycosylation at the modified sequence. By way of example, Asn^y and sn*L ₈₄ may individually or both be replaced with Gin in one embodiment. In the case of the double Gin replacement, the resultant glycoprotein (Gln ₁₁₇ Gln ₁₈₄ ) should contain only one N-linked carbohydrate moiety (at Asn ₄₄₈ ) rather than two or three such moieties as in the case of native tPA. Those skilled in the art will appreciate that analogous glycoproteins having the same Asn4 ₄ g monoglycosy- lation may be prepared by deletion of amino acids or substitution of other amino acids at positions 117 and 184 and/or by deleting or substituting one or more amino acids at other positions within the respective glycosylation recognitions sites, e.g. at Ser ₁₁₉ and Ser ^* L ₈₆ , as mentioned above and/or by substitution, or more preferably by deletion, at one or more of the "X" positions of the tripeptide sites. In another embodiment Asn at positions 117, 184 and 448 are replaced with Gin. The resultant variants should contain no N-linked carbohydrate moieties, rather than two or three such moieties as in the case of native tPA. In other embodiments, potential glycosylation sites have been modified individually, for instance by replacing Asn, e.g. with Gin, at position 117 in one presently preferred embodiment, at position 184 in another embodiment and at position 448 in still another embodiment. This invention encompasses such non-glycosylated, monoglycoslyated, diglycosylated and triglycosylated variants.

Exemplary modifications at one or more of the three consensus N- linked glycosylation sequences, R ¹ , R ² and R ³ , as found in various embodiments of this invention are depicted in Table 2, below.

C. Modification at the Arcr-275/Ile-276 Cleavage Site In one aspect of this invention the variants are optionally mod¬ ified at the proteolytic cleavage site spanning Arg-275 and Ile- 276 by virtue of deletion of Arg-275 or substitution of another amino acid, preferably an amino acid other than ys, Cys or His, for the Arg. Thr is at present an especially preferred replacement amino acid for Arg-275 in the various embodiments of this invention. Proteolytic cleavage at Arg-275 of native tPA yields the so- called "two-chain" molecule, as is known in the art. Proteins of this invention which are characterized by modification at this cleavage site may be more readily produced in more homogeneous form than the corresponding protein without the cleavage site modification, and perhaps more importantly may possess an improved fibrinolytic profile and pharmacokinetic characteristic. As an alternative or supplement to modification at R-275, variants of this invention may be similarly modified at positions 276 and/or 277, or as described in published International application WO 86/01538.

This invention thus provides a family of novel thrombolytic proteins related to human tPA. This family comprises several genera of proteins.

For the sake of clarity and convenience, the following nomenclature is used: variants are identified by a multi-part designation indicating modifications of this invention at R ¹ , R ² , and R ³ ; and at position 275- (J) , in that order. "Δ" indicates an amino acid deletion, i.e. where the position in question is occupied by a peptide bond. Replacement amino acids are specifically indicated. Compounds containing "R" groups specifically identified in Table 1 may be identified by designations containing reference to the appropriate compound designation from that Table.

-,— and = a peptide bond

* = any amino acid

U = any amino acid except Asn, Thr or Ser

V = li II Asn, or a peptide bond

W = II II Ser, or a peptide bond

X - II II Gly, or a peptide bond

Y = II Arg, or a peptide bond z = II Thr or Ser, or a peptide bond wt = wild type, i.e., prior to mutagenesis

Thus, the variant containing all three consensus N-linked glycosylation sites is referred to as "compound 1-0"; the variant further modified by replacement of N's with Q's in R ¹ and R ² is referred to as "compound 1-4" or as the "Q-117,Q-184 variant"; and that variant further modified by the replacement of R-275 by T is designated "compound l-4,T-275" or the "Q-117,Q-184,T-275 variant".

In one genus the proteins are characterized by a peptide sequence substantially the same as that shown in Table 1. By "characterized by a peptide sequence substantially the same as the peptide sequence of Table 1," as the phrase is used herein, we mean the specific peptide sequence of Table 1, or a peptide sequence at least about 90%, and preferably at least about 95%, homologous thereto. Also encompassed are peptide sequences encoded by DNA sequences encoding thrombolytically active derivatives of variants of this invention where the DNA sequences are capable of hybridizing to DNA sequences of this invention under stringent hybridization conditions. By "stringent conditions" as the phrase is used herein we mean hybridization conditions as described on pages 387 - 389 of MOLECULAR CLONING (A LABORATORY MANUAL), T. Maniatis, E.F. Fritsch and J. Sambrook (Cold Spring Harbor Laboratory, 1982) , except that the salt concentration in step 11 (page 388) should be between ~0.1 and -3 X SSC, or in accordance with the Note provided in Step 11 therein. Thus the proteins of this invention include analogs of tPA characterized by the various modifications or combinations of modifications as disclosed herein, which may also contain other variations, e.g. allelic variations or additional deletion(s), substitution(s) or insertio (s) of amino acids which still retain thrombolytic activity, so long as the DNA encoding those proteins is still capable of hybridizing to a DNA sequence encoding a variant of this invention under stringent conditions, or would be so capable but for the use of synonymous codons reflecting the degeneracy of the genetic code. Thus this invention also encompasses variants

embodying modifications described herein, which may also be modified (a) within the region linking the novel N-terminus to the tPA kringle-1 domain, and/or (b) by the inclusion of a consensus N-linked glycosylation site prior to or within the novel N-terminus and/or linking region, and/or (c) by the inclusion of multiple copies of the novel N-terminal peptide domain.

In a second genus the proteins are characterized by a peptide sequence substantially the same as the peptide sequence of Table 1 wherein (a) one or more Asn-linked glycosylation sites are optionally deleted or otherwise modified to other than a consensus Asn-linked glycosylation site, and/or (b) Arg-275 is optionally deleted or replaced by a different amino acid, preferably other than lysine, cystein or histidine.

In one aspect of the invention the proteins contain at least one so-called "complex carbohydrate" sugar moiety characteristic of mammalian glycoproteins. As exemplified in greater detail below, such "complex carbohydrate" glycoproteins may be produced by expression of a DNA molecule encoding the desired polypeptide sequence in mammalian host cells. Suitable mammalian host cells and methods for transformation, culture, amplification, screening, and product production and purification are known in the art. See e.g. Gething and Sambrook, Nature 293:620-625 (1981), or alternatively, Kaufman et al.. Molecular and Cellular Biology 5_ (7) :1750-1759 (1985) or Howley et al., U.S. Patent No. 4,419,446.

A further aspect of this invention involves tPA variants as defined above in which each carbohydrate moiety is a processed form of the initial dolicol-linked oligosaccharide characteristic of insect cell-produced glycoproteins, as opposed to a "complex carbohydrate" substituent characteristic of mammalian glyco¬ proteins, including mammalian derived tPA. Such insect cell-type glycosylation is referred to herein as "high mannose" carbohydrate for the sake of simplicity. For the purpose of this disclosure.

complex and high mannose carbohydrates are as defined in Kornfeld et al., Ann. Rev. Biochem. 54: 631-64 (1985). "High mannose" variants in accordance with this invention are characterized by a variant polypeptide backbone as described above which contains at least one occupied N-linked glycosylation site. Such variants may be produced by expression of a DNA sequence encoding the variant in insect host cells. Suitable insect host cells as well as methods and materials for transformation/transfection, insect cell culture, screening and product production and purification useful in practicing this aspect of the invention are known in the art. Glycoproteins so produced also differ from natural tPA and from tPA produced heretofore by recombinant engineering techniques in mammalian cells in that the variants of this aspect of the invention do not contain terminal sialic acid or galactose substituents on the carbohydrate moieties or other protein modifications characteristic of mammalian derived glycoproteins.

The proteins of this invention which contain no N-linked carbohydrate moieties may also be produced by expressing a DNA molecule encoding the desired variant, e.g. compounds 1-7 through 1-11 of Table 1, in mammalian, insect, yeast or bacterial host cells, with eucaryotic host cells being presently preferred. As indicated above suitable mammalian and insect host cells, and in addition, suitable yeast and bacterial host cells, as well as methods and materials for transformation/transfection, cell culture, screening and product production and purification useful in practicing this aspect of the invention are also known in the art.

As should be evident from the preceding, all variants of this invention are prepared by recombinant techniques using DNA sequences encoding the analogs which may also contain fewer or no potential glycosylation sites relative to natural human tPA and/or deletion or replacement of Arg-275. Such DNA sequences may be produced by conventional site-directed mutagenesis of DNA

sequences encoding tPA in combination with ligation to synthetic DNA.

DNA sequences encoding tPA have been cloned and characterized. See e.g., D. Pennica et al.. Nature (London) 301:214(1983) and R. Kaufman et al., Mol. Cell. Biol..5(7) :1750 (1985). One clone, ATCC 39891, which encodes a thrombolytically active tPA analog is unique in that it contains a Met residue at position 245 rather than Val. The natural t-PA encoding DNA sequence encodes a leader sequence which is typically processed, i.e., recognized and removed by the host cell, followed by the amino acid residues of the full length protein, beginning with the sequence Gly.Al- a.Arg.Ser.Glu.Cys... . Depending on the media and host cell in which the DNA sequence is expresed, the protein so produced may begin with the Gly.Ala.Arg amino terminus or be further processed such that the first three amino acid residues are proteolytically removed. In the latter case, the mature protein has an amino terminus comprising SYQV... . tPA variants having either amino terminus are thrombolytically active. In this invention, the mature proteins have an amino-terminus beginning earlier, e.g. at SQEIHARFRRGARSECK... , RGARSECK... , and/or (most typically) GARSECK, etc. Variants in accord with this invention also include proteins having either Met ₂₄ 5 or Val 245' ^as well as other variants, e.g. alleliσ variations or other amino acid substitutions or deletions, which still retain thrombolytic activity.

This invention also encompasses compounds as described above which contain a further modification at position 219. Compounds of this embodiment are characterized by the presence of an amino acid other than Pro, and preferably other than Cys, at position 219. Such compounds may thus be susceptible to N-linked glyco¬ sylation at Asn-218 which is not glycosylated in melanoma-derived tPA or recombinant versions thereof.

As::mentioned above, DNA sequences encoding individual variants of this invention may be produced by conventional site-directed mutagenesis of a DNA sequence encoding human tPA or analogs or variants thereof, preferably in combination with ligation to a synthetic DNA sequence encoding the novel N-terminal domain. Such methods of mutagenesis include the M13 system of Zoller and Smith, Nucleic Acids Res. .10:6487-6500 (1982); Methods Enzymol. lΩD_>468-500 (1983); and DNA 2:479-488 (1984), using single stranded DNA1 and: the method of Morinaga et al., Bio/technology, 636-639 (J-ϋXy _/ 19^85 -) , using heteroduplexed DNA. Several exemplary oligpnucleotides used in accordance with such methods to effect deletions in the tPA N-terminus or to convert an asparagine residue to threonine or glutamine, for example, are shown in Table 4. It should be understood, of course, that DNA encoding each of the proteins of this invention may be analogously produced by one skilled in the art through site-directed mutagenesis using (an) appropriately chosen oligonucleotide(s) and/or by ligation to one or more synthetic DNA sequences. Expression of the DNA by conventional means in a mammalian, yeast, bacterial, or insect host cell system yields the desired variant. Mammalian expression systems and the variants obtained thereby are presently preferred.

The mammalian cell expression vectors described herein may be synthesized by techniques well known to those skilled in this art,. The components of the vectors such as the bacterial replicons, selection genes, enhancers, promoters, and the like may be obtained from natural sources or synthesized by known procedures. See Kaufman et al., J. Mol Biol. , 159:51-521 (1982); Kaufman, Proc Natl. Acad, Sci. 8_2:689-693 (1985). Such vectors containing a DNA sequence encoding a variant of this invention operably linked to a promoter capable of directing expression of the DNA sequence in transformed cells may thus be readily prepared. Preparation of such a vector, transformation of suitable host cells therewith and culturing of the transformed cells under conditions permitting expression of the vector-borne DNA thus

comprises a convenient method for producing the variants described herein.

Established cell lines, including transformed cell lines, are suitable as hosts. Normal diploid cells, cell strains derived from i vitro culture of primary tissue, as well as primary explants (including relatively undifferentiated cells such as hematopoetic stem cells) are also suitable. Candidate cells need not be genotypically deficient in the selection gene so long as the selection gene is dominantly acting.

The host cells preferably will be established mammalian cell lines. For stable integration of the vector DNA into chromosomal DNA, and for subsequent amplification of the integrated vector DNA, both by conventional methods, CHO (Chinese hamster Ovary) cells are presently preferred. Alternatively, the vector DNA may include all or part of the bovine papilloma virus genome (Lusky et al.. Cell. 3_6:391-401 (1984) and be carried in cell lines such as C127 mouse cells as a stable episomal element. Other usable mammalian cell lines include but are not limited to, HeLa, COS-1 monkey cells, mouse L-929 cells, 3T3 lines derived from Swiss, Balb-c or NIH mice, BHK or HaK hamster cells lines and the like.

Stable transformants then are screened for expression of the product by standard immunological or enzymatic assays. The presence of the DNA encoding the variant proteins may be detected by standard procedures such as Southern blotting. Transient expression of the DNA encoding the variants during the several days after introduction of the expression vector DNA into suitable host cells such as COS-1 monkey cells is measured without selection by activity or immunologic assay of the proteins in the culture medium.

In the case of bacterial expression, the DNA encoding the variant may be further modified to contain different codons for bacterial

expression as is known in the art and preferably is operatively linked in-frame to a nuσleotide sequence encoding a secretory leader polypeptide permitting bacterial expression, secretion and processing of the mature variant protein, also as is known in the art. The compounds expressed in mammalian, insect, yeast or bacterial host cells may then be recovered, purified, and/or characterized with respect to physicochemiσal, biochemical and/or clinical parameters, all by known methods.

-ess: compounds have been found to bind to Erythrina trypsin .Lnhibitor-linked resin, which is known in the art, and to monoclonal antibodies directed to human tPA, and may thus be recovered and/or purified by affinity chromatography using such reagents. Furthermore, these compounds possess tPA-type enzymatic activity, i.e., compounds of this invention effectively activate plasminogen in the presence of fibrin to evoke fibrinolysis, as measured in a conventional indirect assay using the plasmin chromogenic substrate S-2251.

The variants of this invention may also be derivatized to provide conjugates, e.g. as disclosed in Australian patent application AU-A-55514/86, EP-A-O 155 388, EP-A-0 152,736, EPA 85308533.0, EPA 85308534.8, and EPA 0 196 920, which may then be formulated with known carriers or exipients to provide other pharmaceutically useful compositions, as is described below.

This invention also encompasses compositions for thrombolytic ^• therapy which comprise a therapeutically effective amount of a variant described above in admixture with a pharmaceutically acceptable parenteral carrier. Formulations may be prepared, e.g., as described in GB 8612781, GB 8513358, GB 8521704, EPA 211592 and GB 2176703. Such compositions can be used in the same manner as that described for human tPA and should be useful in humans or lower animals such as dogs, cats and other mammals known to be subject to thrombotiσ cardiovascular problems. It is

contemplated that the compositions will be used both for treating and desirably for preventing thrombotic conditions such as myocardial infarction, deep vein thrombosis, and other indications for which thrombolytic therapy would be useful. The exact dosage and method of administration will be determined by the attending physician depending on the potency and pharmacokinetiσ profile of the particular compound as well as on various factors which modify the actions of drugs, for example, body weight, sex, diet, time of administration, drug combination, reaction sensitivities and severity of the particular case.

The following examples are given to illustrate embodiments of the invention. It will be understood that these examples are illustrative, and the invention is not to be considered as restricted thereto except as indicated in the appended claims.

In each of the examples involving insect cell expression, the nuclear polyhedrosis virus used was the L-l variant of the Autoαrapha californica, and the insect cell line used was the Spodoptera frucriperda IPLB-SF21 cell line (Vaughn, J.L. et al.. In Vitro (1977) 13, 213-217). The cell and viral manipulations were as detailed in the literature (Pennock G.D., et al., supra; Miller, D.W. , Safer, P., and Miller, L.K. , Genetic Engineering. Vol. 8, pages 277-298, J.K. Setlow and A. Hollaender, eds. Plenum Press, 1986) . The RF ml3 vectors, mpl8 and mp 11, are commercially available from New England Biolabs. However, those of ordinary skill in the art to which this invention pertains will appreciate that other viruses, strains, host cells, promoters and vectors containing the relevant cDNA, as discussed above, may also be used in the practice of each embodiment of this invention. The DNA manipulations employed are, unless specifically set forth herein, in accordance with Maniatis et al.. Molecular Cloning: A Laboratory Manual (Cold Spring Harbor, NY 1982) .

Table 4: Exemplary Oligonucleotides for Mutagenesis

No. Sequence Mutation

1. ACC AAC TGG ACC AGC AGC GCG ^Asn 117 ^>Tnr

2. CTAC TTT GGG ACT GGG TCA GC Asn ₁₈ >Thr

3. GTGCACCAACTGGCAGAGCAGCGCGTTGGC Asn ₁₁₇ >Gln

4. CAACTGGCAGAGCAGCG (#3)*

5. ACTGCTACTTTGGGCAGGGGTCAGCCTACC Asn ₁₈₄ >Gln

6. CTTTGGGCAGGGGTCAG (#5)*

7. CATTTACTTCAGAGAACAGTC Asn ₄₄₈ >Gln

14. GC CAG CCT CAG TTT _Δ ATC AAA GGA GGG C (R-275 del)

15. CT CAG TTT ACC ATC AAA G (—> T-275)

*Used for screening the mutation indicated in parenthesis (where a screening oligonucleotide is not indicated, the same oligonucleotide is used for mutagenesis and screening) . Codons for replacment amino acids are underlined, Δ indicates site of deletion. As those skilled in this art will appreciate, oligonucleotides can be readily constructed for use in deleting one or more amino acids or inserting a different (i.e. , replacement) amino acid at a desired site by deleting the codon(s) or substituting the codon for the desired replacement amino acid, respectively, in the oligonucleotide. Other mutagenesis oligonucleotides can be designed based on an approximately 20-50 nucleotide sequence spanning the desired site, with replacement or deletion of the original codon(s) one wishes to change.

TABLE 4A: Synthetic oligonucleotides for cassette construction

1. GAT CCG AGT GCA AGA CTG GGA ATG GAA AGA ACT ACA GAG GGA CGA TGT CCA AAA CAA AAA ATG GCG CCA CCT GTC AAA AA

2. ACT CCA TTT TTG ACA GGT GGC GCC ATT TTT TGT TTT GGA CAT CGT CCC TCT GTA GTT CTT TCC ATT CCC AGT CTT GCA CTC G

3. TGG AGT TCC ACT TCT CCC CAC AGA CCT AGA TTC TCA CCT GCT ACA CAC CCC TCA GAG GGA CTG GAG GAG AAC TAC TGC AGG AAT CCA G

4. CGT TGT CTG GAT TCC TGC AGT AGT TCT CCT CCA GTC CCT CTG AGG GGT GTG TAG CAG GTG AGA ATC TAG GTC TGT GGG GAG AAG TGG A

5. ACA ACG ATC CGC AGG GGC CCT GGT GCT ATA CTA CTG ATC CAG AAA AGA GAT ATG ACT ACT GCG ACA TTC TTG AGT GTG AAG AGA T

6. CGA TCT CTT CAC ACT CAA GAA TGT CGC AGT AGT CAT ATC TCT TTT CTG GAT CAG TAG TAT AGC ACC AGG GCC CCT GCG GAT

TABLE 4B: Synthetic DNA cassette encoding novel N-terminus

GA TCC GAG TGC AAG ACT GGG AAT GGA AAG AAC TAC AGA GGG ACG ATG G CTC ACG TTC TGA CCC TTA CCT TTC TTG ATG TCT CCC TGC TAC

TCC AAA ACA AAA AAT GGC GCC ACC TGT CAA AAA TGG AGT TCC ACT TCT AGG TTT TGT TTT TTA CCG CGG TGG ACA GTT TTT ACC TCA AGG TGA AGA

CCC CAC AGA CCT AGA TTC TCA CCT GCT ACA CAC CCC TCA GAG GGA CTG GGG GTG TCT GGA TCT AAG AGT GGA CGA TGT GTG GGG AGT CTC CCT GAC

GAG GAG AAC TAC TGC AGG AAT CCA GAC AAC GAT CCG CAG GGG CCC TGG CTC CTC TTG ATG ACG TCC TTA GGT CTG TTG CTA GGC GTC CCC GGG ACC

TGC TAT ACT ACT GAT CCA GAA AAG AGA TAT GAC TAC TGC GAC ATT CTT ACG ATA TGA TGA CTA GGT CTT TTC TCT ATA CTG ATG ACG CTG TAA GAA

GAG TGT GAA GAG AT CTC ACA CTT CTC TAG C

Plasmid Derivations

Mutagenesis of cDNAs at codons for the various amino acids was conducted using an appropriate restriction fragment of the cDNA in M13 vectors by the method of Zoller and Smith. Deletions within the cDNA were effected by loopout mutagenesis using an appropriate restriction fragment, e.g. the Sad fragment, of the cDNA either in M13 vectors or by heteroduplex loop-out in plasmid PSVPA4.

The plasmid pSVPA4 was constructed to allow the expression of tPA glycoprotein in mammalian cells. This plasmid was made by first removing the DNA encoding the SV40 large T polypeptide from the plasmid pspLT5 (Zhu, Z. et al. , 1984, J. Virology 5_1:170-180) . This was accomplished by performing a total Xho 1 digest followed by partial Bam-Hl restriction endonuclease digestion. The SV40 large T encoding region in pspLT5 was replaced with human tPA-encoding sequence by ligating a cohesive Sail/ BamHl tPA encoding restriction fragment, isolated by digesting plasmid J205 (ATCC No. 39568) with Sal I and BamHl, to the parent Xhol/BamHl cut vector pspLT5 prepared as described above. Consequently, tPA will be transcribed in this vector under the control of the SV40 late promoter when introduced into mammalian cells. This final contruct is designated pSVPA4.

Plasmid pLDSG is an amplifiable vector for the expression of tPA in mammalian cells such as CHO cells. pLDSG contains a mouse DHFR cDNA transcription unit which utilizes the adenovirus type 2 major late promoter (MLP), the simian virus 40 (SV40) enhancer and origin of replication, the SV40 late promoter (in the same orientation as the adenovirus MLP) , a gene encoding tetracycline resistance and a cDNA encoding human tPA (Met-245) in the proper orientation with respect to the adenovirus type 2 MLP. The preparation of pLDSG from pCVSVL2 (ATCC No. 39813) and a tPA encoding cDNA has been described in detail as has cotransformation

with, and amplification of, pLDSG in CHO cells. Kaufman et al., Mol. and Cell. Bio. 5(7): 1750-1759 (1985).

Plasmid pWGSM is identical to pLDSG except that the cDNA insert encodes Val-245 human tPA. pWGSM may be constructed using cDNA from plasmid J205 (ATCC No. 39568) or pIVPA/1 (ATCC No. 39891) with the desired mutagenesis at position 245. Throughout this disclosure pWGSM and pLDSG may be used interchangeably, although as indicated previously, the former vector will produce Val-245 proteins and the latter Met-245 proteins.

pIVPA/1 (ATCC No. 39891) is a baculoviral transplacement vector containing a tPA-encoding cDNA. pIVPA/1 and mutagenized derivatives thereof are used to insert a desired cDNA into a baculoviral genome such that the cDNA will be under the transcriptional control of the baculoviral polyhedrin promoter.

pMT2pc is a mammalian expression vector which contains the adenovirus-VA genes, SV40 replication origin including enhancer, adenovirus major late promoter including tripartite leader and 5" donor splice site, 3' splice acceptor site, DHFR cDNA insert, SV40 early polyadenylation signal and pBR322 sequences. pMT2pc contains a unique Pst I cloning site. The vector has been deposited with the American Type Culture Collection as ATCC No. 40348.

pMT2pc-tPA and derivatives thereof are themselves derivatives of pMT2pσ produced by destroying an existing Bglll site on pMT2pc, digesting pMT2pc with PstI, ligating the linearized DNA to blunt ended linkers, and then ligating the linkered DNA to the Ball restriction fragment of pWGSM containing the tPA-encoding cDNA or mutagenized derivatives thereof. Vectors so produced may be analyzed by conventional restriction analysis for insertion of the cDNA in the correct orientation, given the known restriction map of tPA, e.g. as presented hereafter.

Preparation of expression vector

The mammalian expression vector pMT2pc-FE787 ^• is prepared by first digesting 10 μg of plasmid pMT2pσ DNA with the restriction endonuσlease Bgl II. This is followed by filling in the cut ends using DNA polymerase I (Klenow fragment) in the presence of nucleotides. The cut and filled plasmid is then diluted and ligated. The plasmid is used to transform E. coli HB101, and a colony containing Bglll-resistant plasmid is selected. The purified plasmid is then cut with Pst I. This is followed by extraction with phenol/chloroform, then ethanol precipitation.

The following blunt-ended adaptor is then ligated to the Pst I linearized vector at a 50:1 (adaptor:vector) molar ratio:

(5') HO-CTAGAGGCCTCTGCA-OH (3 ^« ) (3') HO GATCTCCGGAG-P (5 ¹ )

The vector is then cut with Ball and the mixture is run on a

0.7% agarose gel and the -4.8 kbp expression vector DNA is excised and purified. This vector can now accept the tPA cDNA insert.

The cDNA insert encoding a variant tPA sequence is prepared for insertion into the vector prepared as above by agarose gel purification of the -2.1 kbp restriction fragment produced by digestion of the plasmid pWGSM-FE787" with the restriction endonuclease Ball. pWGSM-FE787' is a derivative of pWGSM which encodes the tPA variant lacking amino acids 6-86 and containing Q instead of N at position 117 and M instead of V at position 245. That variant (compound 2-1/N-22/R-275) is described in published International patent application No. WO 87/04722, as is preparation of a cDNA and vector for its expression. The Ball fragment of pWGSM-FE787• contains the DNA sequence required for full length translation of a Q-117 tPA variant lacking the "finger" and "epidermal growth factor" domains found in the 91-amino aci N-terminus of the native protein as described in publishe International application WO 87/04722. Insertion of the Bal

fragment into pMT2pc produces the vector pMT2pc-FE787' . pMT2pc- FE787' may be converted into pMT2pσ-tPA or derivatives thereof by (1) digestion of pMT2pc-FE787• with Bgl II and Apa I to remove the DNA sequence encoding a large portion of the tPA variant beginning prior to the mature N-terminus and extending through the third consensus N-linked glycosylation site, and (2) inserting in its place the corresponding Bglll/Apal restriction fragment from pWGSM or other tPA-encoding vectors. Alternatively, heteroduplex mutagenesis may be conducted on pMT2pc-FE787' to effect any desired mutagenesis, e.g. at one or more glycosylation sites and/or at the plasmin cleavage site.

Preparation of derivatives of tPA cDNAs: M13 method The following schematic restriction map illustrates a cDNA encoding human tPA (above) with cleavage sites indicated for specific endonucleases (indicated below) :

The initiation codon, ATG, and the cDNA regions encoding R ¹ , R ² and R ³ are indicated. Thus, mutagenesis at the N- terminus may be effected using the Sad fragment or the Bglll/Narl fragment, for example. Mutagenesis at Arg-275 and/or at R ¹ and/or R ² may be effected using, e.g., the SacI fragment or Bglll/SacI fragment. Mutagenesis at R ³ may be effected using, e.g. a Sacl/Xmal or Sacl/Apal frag¬ ment. The choice of restriction fragment may be determined based on the convenience of using particular vectors for mutagenesis and/or for expression vector construction.

Generally, the cDNA restriction fragment to be mutagenized may be excised from the full-length cDNA present, e.g., in

pWGSM, pMT2pc-tPA, pIVPA/1 or pSVPA4, using the indicated endonuclease enzyme(s), subcloned into appropriate vectors, and then mutagenized, e.g. with the oligonucleotides shown in Table 4 or other oligonucleotides designed for the desired mutagenesis.

Exemplary mutagenized cDNA fragments which may thus be prepared are shown in Table 5, below. Such fragments may then be used to replace nucleotide sequences in pMT2pc-tPA or mutagenized derivatives thereof as mentioned above.

Table 5: Exemplary Mutagenized cDNA Fragments

* indicates site(s) of mutagenesis; cDNA fragments I through IV are prepared by digesting pWGSM or pSVPA4 with Sad, inserting SacI fragment into M13 vector, mutagenizing with desired oligonucleotide(s) , and digesting mutagenized M13/tPA DNA with SacI; alternatively, I-IV may be excised from mutagenized M13/tPA with Bglll and SacI and the Bglll/SacI fragment encoding the peptide domain spanning the N-terminus, R ¹ , R ² & Arg-275 may be inserted into Bglll/SacI-digested pIVPA; cDNA fragments V and VI are prepared as described in Examples 2 and 1, respectively, below. New N-terminal sequence precedes Cla I site.

Following mutagenesis the fragment, with or without further mutagenesis, may then be excised from the M13 vector and ligated back into an expression vector containing the full-length or partial cDNA previously cleaved with the same enzyme(s) as were used for excising the mutagenized fragment from the M13 vector. By this method the full-length cDNA, mutagenized as desired, may be re-assembled using one or more mutagenized fragments as restriction fragment cassettes.

cDNAs encoding the illustrative compounds (see Table 1) may be prepared as follows.

For production of DNA encoding compound 1-1, the set of overlapping oligonucleotides depicted in Table 4A was first synthesizedby conventionalmeans using a commercial automated DNA synthesizer following the supplier's instructions. Odd numbered oligonucleotides are "sense" strands, even numbered oligonucleotides are "anti-sense" strands. Oligonucleotides 2, 3, 4 and 5 were separately kinased using conventional pro¬ cedures. Oligonucleotide pairs 1 and 2, 3 and 4, 5 and 6, were then each annealed to one another under conventional con¬ ditions, e.g. 85 mM tris, pH 7.5, 50 mM NaCl, 8.5 mM MgCl ₂ , and 4.2 picomoles ( of each oligonucleotide)/λ of solution, with heating to 80 C followed by slow cooling over -2 h to 37 C to form a set of overlapping synthetic duplexes: 1 3 5

The individual mixtures of duplexes were then combined and concentrated, and the duplexes were ligated to one another under standard conditions, e.g. 50 mM tris, pH 7.4, 10 mM MgCl ₂ , 10 mM DTT, and 1 mM ATP and 5 Weiss units of T4 ligase (New England BioLabs) at 4 C overnight (~16 h) . The mixture was electrophoresed through a 2% low gelling

temperature agarose gel and a band of -250 bp was excised from the gel. That DNA molecule so produced encodes the synthetic kringle region on a cassette with BamHl and Cla I overhangs at the 5' and 3' termini, respectively.

pMT2pc-FE787' plasmid DNA prepared in a dam" bacterial host such as E. coli GM161 was digested with Bgl II and with Cla I (partial) to excise the coding region beginning within the Arg-(-l) codon and continuing into the Ile-5 codon of the Ile(5)-Asp(87) fusion site present in pMT2pc-FE787• . pMT2pc-FE787' , so digested, was then ligated to the BamHl- Clal synthetic cassette by conventional means.

The above-described construction creates a new pMT2pc vector, pMT2pc-PKl-FE787' , in which the codons for Arg-(-l) and Ile-5 (at the Ile-5-Asp-87 fusion site) are recreated, between which the cassette sequence is inserted. Thus, the resultant vector contains a modified cDNA insert which now encodes compound 1-1.

The following represents a restriction map of the coding sequence of pMT2pc-FE787• :

where "*" indicates the position of the sequence encoding the lie(5)-Asp(87) fusion and R ¹ is modified as described.

A restriction map of the synthetic cassette is as follows:

The following represents a restriction map of the coding sequence of pMT2pc-PKl-FE787 ' , which incorporates the BamHI/Clal cassette:

where "*"s indicate the location of the synthetic cassette. It should be noted that the BamHI/Bglll fusion at the site of ligation of the 5' terminus of the cassette into the vector results in the destruction of both the BamHl and Bglll restriction sites. Additionally, insertion of the cassette into the vector recreates the Cla I site at the 3• terminus of the cassette, and introduces a new Narl site spanning the Gly-Ala encoding nucleotides witin the cassette sequence.

Any further mutagenesis desired, e.g. at one or more of the consensus N-linked glycosylation sites and/or at the plasmin cleavage site, may be conveniently effected by conventional heteroduplex mutagenesis using pMT2pc-PKl-FE787' and approp¬ riate oligonucleotides such as are depicted in Table 4.

Plas ids pIVPA, pSVPA4 or pMT2pc-tPA, in addition to utility as expression vectors, may also be used as a "depot" in the construction of cDNAs having any desired permutation of mutagenized sites. Thus, mutagenized (via M13 or hetero- duplexing) plasmids containing a desired modification in the cDNA, e.g. at any combination of Arg-275, R ¹ , R ² and/or R ³ -encoding regions, may then be digested with one or more appropriate restriction enzymes, the desired fragment may then be identified, isolated and ligated into a correspondingly digested expression vector such as pMT2pc- PK1-FE787'.

EXAMPLES Example 1 : Preparation of pMT2pc-PKl-FE787' A. Preparation of cDNA

A cDNA molecule encoding the polypeptide sequence of Δ6-86, Q-117 tPA was prepared using the oligonuσleotide-directed mutagenesis method of Zoller and Smith, essentially as described in WO 87/04722. Specifically, the mutagenesis vector RF M13/tPA containing a Met-245 tPA gene was constructed from the mammalian tPA expression plasmid pSVPA4. RF M13/tPA was constructed by first digesting pSVPA4 to completion with the restriction endonucleases Hindlll and Xmal. The approximately 1,860 base pair (bp) Hindlll/Xmal fragment encodes a large portion of the polypeptide sequence of M-245 tPA and includes the nucleotide sequences encoding the consensus N-linked glycosylation sites encompassing asparagines 117,184, and 218. This -1,860 bp (hereinafter 1.9 kbp) fragment was purified by preparative agarose gel electrophoresis.

The Hindlll/Xma I fragment of the tPA cDNA, obtained as above, was ligated to a linearized double-stranded RF M13mpl8 DNA vector which had been previously digested with Hindlll and Xmal. The ligation mixture was used to transform transformation competent bacterial JM101 cells. M13 plaques containing tPA-derived DNA produced from transformed cells were identified and isolated by analytical DNA restriction analysis and/or plaque hybridization. Radiolabeled oligo¬ nucleotides (~17mers, of positive polarity) derived from within the Hindlll/Xmal restriction sites of the tPA-encoding nucleotide sequence were used as probes when filter hybridization was employed to detect viral plaques containing tPA DNA. All oligonucleotides were prepared by automated synthesis with anApplied Biosyste s DNA synthesizer according to the manufacturer's instructions.

Several of the positive plaques detected by restriction or

hybridization analysis were then further cloned by conventional plaque purification. Purified M13/tPA bacteriophage obtained from the plaque purification procedure was used to infect JM101 cells. These infected cells produce cytoplasmic double-stranded "RF" M13/tPA plasmid DNA. The infected cells also produce bacteriophage in the culturemediumwhichcontainsingle-strandedDNAcomplimentary to the 1.9 kbp Hindlll/Xmal fragment of tPA cDNA and to M13 DNA. Single-stranded DNA was purified from the M13/tPA-con- taining phage isolated from the culture medium. This single- stranded M13/tPA DNA was used as a template in two rounds of mutagenesis according to the method of Zoller and Smith, using oligonucleotides #1 and #10 in Table 7 of International application WO 87/04722 as described therein. This mutagene¬ sis changes the Asn codon to a Gin codon at position 117 and deletes nucleotides encoding amino acids 6 through 86 of the subsequently obtained coding strand of DNA. Following the mutagenesis reactions, the DNAwas transformed intothebacte¬ rial strain JM 101. To identify mutagenized cDNA's, the ss form of M13 from the transformant plaques was seguenced.

RF M13/tPA plasmid DNA was purified from JM 101 cells infec¬ ted with purified M13 phage containing mutagenized tPA cDNA. The RF M13/tPA plasmid thus obtained contains the Hindlll/ Xmal restriction fragment of tPA DNA mutagenized as describ¬ ed. This mutagenized restriction fragment can thenbe further mutagenized, e.g. at one ormoreN-linkedglycosylation sites, at the plasmin cleavage site at Arg-275, or otherwise, as described in published International Patent Application WO 87/04722 if desired, again by the method of Zoller and Smith, using appropriate oligonucleotides such as disclosed in Table 4. Of particular interest are the variants of this invention in which the first (R ¹ ) , the first and second (R ¹ and R ² ) or all three (R ¹ , R ² and R ³ ) of the consensus N- linked glycosylation sites are modified to be other than consensus N-linked glycosylation sites.

B. Preparation of Expression vector pMT2pc-FE787'

In this case, the mutagenized cDNA encoding Δl-86, Q-117 tPA is excised from the M13 vector as the Hindlll/Xmal fragment. That fragment is then ligated to Hindll /Xmal- digested pWGSM by conventional means to produce pWGSM- FE787'. pMT2pc-FE787' was then prepared from pMT2pc and pWGSM-FE787' as previously described.

C. Preparation of Synthetic Cassette

For production of vector encoding compound 1-1, the set of overlapping oligonucleotides depicted in Table 4A was first synthesizedby conventional means using a commercial automated DNA synthesizer following the supplier's instructions. Odd numbered oligonucleotides are "sense" strands, even numbered oligonucleotides are "anti-sense" strands. Oligonucleotide pairs 1 and 2; 3 and 4; and 5 and 6, were each annealed to one another, following appropriate kinasing, under conventional conditions as previously described to form a set of synthetic duplexes. The individual mixtures of duplexes were then combined and the duplexes were ligated to one another as previously described. The mixture was electrophoresed through a 2% low gelling temperature agarose gel and a band of -250 bp was excised from the gel. That DNA molecule so produced encodes the synthetic kringle region on a cassette with BamHl and Cla I overhangs at the 5' and 3' termini, respectively.

D. Preparation of pMT2pc-PKl-FE787' pMT2pσ-FE787' grown up in E. coli GM161 was digested with Bgl II and Cla I (partial) to excise the coding region beginning within the Arg-(-l) codon and continuing into the Ile-5 codon of the Ile(5)-Asp(87) fusion site present in pMT2pc- FE787'. pMT2pc-FE787', so digested, was then ligated to the BamHI-Clal synthetic cassette by conventional means.

The above-described construction creates a new pMT2pc vector, pMT2pc-PKl-FE787' , in which the codons for Arg-(-l) and Ile-5 (at the Ile-5-Asp-87 fusion site) are recreated, between which the cassette sequence is inserted. Thus, the resultant vector contains a modified cDNA insert which now encodes compound 1-1.

Example 2 : cDNAs encoding other compounds of this invention

£SJ mentioned above, pMT2pσ-PKl-FE787* can be conveniently mutagenized using conventional heteroduplex mutagenesis methods and other conventionally prepared synthetic oligon¬ ucleotides which modify the DNA sequence as desired, e.g. to restore a consensus N-linked glycosylation site at R ¹ , and/or to convert one or both of R ² and R ³ to other than N- linked glycosylation sites, and/or to modify the plasmin cleavage site spanning position 275.

Example 3 : preparation of variants in mammalian cells

Expression vectors such as pMT2pc derivatives containing a cDNA molecule encoding a desired polypeptide sequence are prepared as described above.

Transformation of mammalian host cells, selection of transfor ants, amplification of gene copy number, and cell culture leading to the production of the desired product may be effected by the method of Kaufman et al., supra, (CHO host cells) or by the method of Howley et al., U.S. Patent No. 4,419,446 (1983) (using BPV expression systems) to yield the correspondingmammalian-derived variant proteins. In the case of CHO cell expression, the vector DNA was introduced into CHO cells by conventional protoplast fusion, and amplified by the method of Kaufman, supra. The transformed and amplified CHO cells produce the desired variant in good yield which may be detected in the culture

medium by human tPA specific antibodies (presumably directed to epitopes other than at the N-terminus) . The variant may then be recovered and purified by immunoaffinity chroma¬ tography or other conventional methods such as those involving ETI resin. Similarly, other variants of this invention may also be expressed in mammalian cells, recovered and purified by such methods.

The variant expressed by CHO cells transformed with pMT2pc- PK1-FE787', as described above, was characterized by an in vivo half life, as measured in rat and rabbit models, at least about 10 to 20 times greater than that of wild type tPA and a specific activity, as measured by the indirect chromogenic substrate assay using S-2251, significantly higher than that of the WHO tPA standard, e.g. up to about 50-100% higher, if not more.