CA1340971C - Metalloproteinase inhibitor sequence recombinant vector system for using the same and recombinant-dna method for the manufacture of same - Google Patents

Abstract

A portable DNA sequence is disclosed which is capable of directing intracellular production of metalloproteinase inhibitors. Vectors containing this portable DNA sequence are also set forth, including the vector pUC9-FS/237P10 (ATCC
Accession No. 53003). A recombinant-DNA method for the microbial production of a metalloproteinase inhibitor, which method incorporates at least one of the portable DNA sequences and the vectors disclosed herein.

Description

1340 97' METAL~L:OSPRCYrEINASE INHIBITOR SEQUENCE RECONBINANT VEC~R
SYSTEM FOR USING SAME AND REC1~I~IBINANT-DNA METHOD FOR THE
MANUFACTURE OF SAME
BACK;GROUND OF THE INVE,'NTION
Endogenc>us proteolytic enzymes serve to degrade invading organisms, antigen-antibody complexes and certain tissue proteins which are no longer necessary or useful to the on~anism. In a normally functioning Organ7.STil, proteolytic enzymes are produced in a limited quantity and are recfiulated in part through specific inhibitors.
Metalloproteinases are enzymes present in the body which are often involved in the degradation of connective tissue. While some , connective tissue degradation is necessary for normal functioning of an organism, an excess of connective tissue degradation occurs in several disease states and _Ls believed to be attributable, at least in part, to excess metalloprote:~_nase. It is believed that metalloproteinases are at least implicated in periodontal disease, corneal and skin ulcers, rheumatoid arthriti:~ and the spread of cancerous solid tumors.
These diseases generally occur in areas of the body which contain a high proportion of collagen, a particular form of connective tissue. An examination of ;patients with these diseases of connective tissue has revealed an excessive breakdown of the various components of connective tissues, including collagen proteoglycans and elastin.
Therefore, it has been deduced that an excessive concentration of a particular metallopz~oteinas~=_, for example collagenase, proteoglyconase, gelatinase, and certain elastases, may cause or exacerbate the connective tissue destruction associated with the aforerr~ntioned diseases.
In the normal state, the body possesses metalloproteinase inhibitors which bir..d to met=alloproteinases to effectively prevent these enzymes from acting on thei~_ connective tissue substrates. Specifically, in a healthy organism, metalloproteinase inhibitors are present in concentrations sufficient to interact with metalloproteinases to an extent which allows sufficient quantities of metalloproteinase to remain ' active while binding the excess metalloproteinase so that the ~i 134pg71 connective tissue damages seen in the various diseases does not occur.
It is postulated that one immediate cause of the con-nective tissue destruction present in the foregoing disease states is an imbalance i.n the relative metalloproteinase/metallo-proteinase inhibitor concentrations. In these situations, either due to an excessive amount of active metalloproteinase or a defi-ciency in the amount of active metalloproteinase inhibitor, the excess metalloproteinase~ is believed to cause the connective tis-sue degradation responsible for causing or exacerbating the dis-ease. It is postulated that, by treating persons with connective tissue diseases with metalloproteinase inhibitors, the degradative action of the excess metalloproteinase may be cur-tailed or halted. Therefore, particular metalloproteinase inhib-itors of specific interest to the present inventors are collage nase inhibitors because it is believed that these inhibitors would be pharmaceutically useful in the treatment or prevention of connective tissue diseases.
The exi:>tence of metalloproteinase and metallopro-teinase inhibitors has been discussed in the scientific litera-ture. For exampls~, Sellers et al., Biochemical And Biophysical Research Communications 87:581-587 (1979), discusses isolation of rabbit bone collactenase inhibitor. Collagenase inhibitor iso-lated from human :;kin fi'broblasts is discussed in Stricklin and Welgus, J.B.C. 25Et:12252-12258 (1983) and Welgus and Stricklin, J.B.C. 258:12259-1.2264 (1983). The presence of collagenase in-hibitors in naturally-occurring body fluids is further discussed in Murphy et al., Biochem. J. 195:167-170 (1981) and Cawston _et al., Arthritis and Rheumatism, _27:285 (1984). In addition, met-alloproteinase inhibitors are discussed by Reynolds et al. in Cellular Interactions, D.ingle and Gordon, eds., (1981). Although these articles characterize particular, isolated metallopro-teinase inhibitors and discuss, to some extent, the role or potential role of metalloproteinases in connective tissue disease treatment and speculate on the ability of metalloproteinase inhibitors to counteract this destruction, none of these re-searchers had previously been able to isolate a portable DNA

~ 340 97 ~

sequence capable of directing intracellular production of metal-loproteinase inhibitors or to create a recombinant-DNA method for the production of these :inhibitors.
Surprisingly, 'the present.inventors have discovered a portable DNA sequence capable of directing the recombinant-DNA
synthesis of metalloproteinase inhibitors. These metallopro-teinase inhibitors are biologically equivalent to those isolated from human skin fibroblaat cultures. The metalloproteinase in-hibitors of the present :invention, prepared by the recombinant-DNA methods set forth herein, will enable increased research into prevention and treatment of metalloproteinase-induced connective tissue diseases. Zn add:ition, the metalloproteinase inhibitors of the present invention are useful in neutralizing metallopro-teinases, including the excess metalloproteinase associated with disease states. Therefor e, it is believed that a cure for these,, diseases will be developed which will embody, as an active ingre-dient, the metalloproteinase inhibitors of the present invention.
Furthermore, the metalloproteinase inhibitors of the present invention are capable of interacting with their metalloproteinase targets in a manner which allows the development of diagnostic tests for degradative connective tissue diseases using the newly discovered inhibitors.
The recombinant metalloproteinase inhibitors discussed herein interact stoichiornetrically (i.e., in a 1:1 ratio) with their metalloproteinase targets. In addition. these metallopro-teinase inhibitors are hE~at resistant, acid stable, glycosylated, and exhibit a high isoelEactric point.
SUMMARY OF THE INVENTION
The present invention relates to metalloproteinase in-hibitors and a recombinant-DNA method of producing the same and to portable DNA sequences capable of directing intracellular pro-duction of the metalloproteinase inhibitors. Particularly, the present invention relates to a collagenase inhibitor, a recombi-nant-DNA method for producing the same and to portable DNA se-quences for use in the rEacombinant method. The present invention also relates to a series of vectors containing these portable DNA
sequences.

1 340 97 ~

One object of tree present invention is to provide a metalloproteinase inhibitor, which can be produced in sufficient quantities and purities to provide economical pharmaceutical compositions which posses metalloproteinase inhibitor activity.
An additional object of the present invention is to provide a recombinant-DNA method for the production of these metalloproteinase inhibitors. The recombinant metalloproteinase inhibitors produced by this method are biologically equivalent to the metalloproteinase inhibitor isolable from human skin fibroblast cultures.
To facilitate the recombinant-DNA synthesis of these metalloproteinase inhibitors, it is a further object of the present invention to provide portable DNA sequences capable of directing intracellular production of metalloproteinase inhibitors. It is also an object of the present invention to provide cloning vectors containing these portable sequences. These vectors are capable of.~
being used in recombinant systems to produce pharmaceutically useful quantities of metalloproteinase inhibitors.
Additional objects and advantages of the invention will be set forth in part i:n the description which follows, and in part will be obvious from the description or may be learned from practice of the invention. The objects and advantages may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.
To achieve the objects and in accordance with the purposes of the present invention, metalloproteinase inhibitors are set forth, which are capable o:E stoichiometric reaction with metalloproteinases.
These metalloproteinase inhibitors are remarkably heat resistant, acid stable, glycosylated, and exhibit a high isoelectric point.
Furthermore, these tnetalloproteinase inhibitors are biologically equivalent to those inhibitors isolated from human skin fibroblast cultures.
To furthe:= achieve the objects and in accordance with the purposes of the present invention, as embodied and broadly described herein, portable DNA sequences coding for metalloproteinase inhibitors are provided. Theses sequences comprise s y nucleotide sequences capable of directing intracellular produc-tion of metalloproteinase inhibitors. The portable sequences may be either synthetic sequences or restriction fragments ("natural"
DNA sequences). In a preferred embodiment, a portable DNA se-quence is isolated from a human fibroblast cDNA library and is capable of directing intracellular production of a collagenase inhibitor which is biologically equivalent to that inhibitor which is isolable from a human skin fibroblast culture.
The coding strand of a first preferred DNA sequence which has been discoverEad has the following nucleotide sequence:

GTTGTTGCTG TGGCTGATAG CC:CCAGCAGG GCCTGCACCT GTGTCCCACC CCACCCACAG

ACGGCCTTCT GCAATTCCGA CC:TCGTCATC AGGGCCAAGT TCGTGGGGAC ACCAGAAG~C

AACCAGACCA CCTTATACCA GC:GTTATGAG ATCAAGATGA CCAAGATGTA TAAAGGGTTC

CAAGCCTTAG GGGATGCCGC TGACATCCGG TTCGTCTACA CCCCCGCCAT GGAGAGTGTC

TGCGGATACT TCCACAGGTC CC:ACAACCGC AGCGAGGAGT TTCTCATTGC TGGAAAACTG

CAGGATGGAC TCTTGCACAT CACTACCTGC AGTTTCGTGG CTCCCTGGAA CAGCCTGAGC

TTAGCTCAGC GCCGGGGCTT CACCAAGACC TACACTGTTG GCTGTGAGGA ATGCACAGTG

TTTCCCTGTT TATCCATCCC CTGCAAACTG CAGAGTGGCA CTCATTGCTT GTGGACGGAC
490 500 510 520 ~ 530 540 CAGCTCCTCC AAGGCTCTGA AAAGGGCTTC CAGTCCCGTC ACCTTGCCTG CCTGCCTCGG
i GAGCCAGGGC TGTGCACCTG GC:AGTCCCTG CGGTCCCAGA TAGCCTGAAT CCTGCCCGGA

GTGGAAGCTG AAGCCTGCAC AGTGTCCACC CTGTTCCCAC TCCCATCTTT CTTCCGGACA

ATGAAATAAA GAGTTACCAC CC:AGCAAAAA AAAAAAGGAA TTC
The nucleotides represented by the foregoing abbreviations are set forth in the Detailed Description of the Preferred Embodi-ments.
A second prefe:rrred DNA sequence has been discovered which has an additional nucleotide sequence 5' to the initiator sequence. This sequence, which contains as the eighty-second through four-hundred-thirty-second nucleotides nucleotoides 1 through 351 of the first. preferred sequence set forth above, has the following nucleotides sequence:

GGCCATCGCC GCAGAT~~CAG CGCCCAGAGA GACACCAGAG AACCCACCAT GGCCCCCTTT

GACCCCTGGC TTCTGC,ATCC TGITTGTTGCT GTGGCTGATA GCCCCAGCAG GGCCTGCACC

TGTGTCCCAC CCCACC~~ACA .Gp,CC;GCCTTC TGCAATTCCG ACCTCGTCAT CAGGGCCAAG

TTCGTGGGGA CACCAG,AAGT CF,ACCAGACC ACCTTATACC AGCGTTATGA GATCAAGATG

ACCAAGATGT ATAAAGGGTT CC'.AAGCCTTA GGGGATGCCG CTGACATCCG GTTCGTCTAC

ACCCCCGCCA TGGAGA'uTGT CTGCGGATAC TTCCACAGGT CCCACAACCG CAGCGAGGAG

TTTCTCATTG CTGGAAAACT GC'.AGGATGGA CTCTTGCACA TCACTACCTG CAGTTTCGTG
_.' : ~ a 134097 ~
_7_ GCTCCCTGGA AC
A third preferred DNA sequence which incorporates the 5' region of the second preferred sequence and the 3' sequence of the first preferred sequence, has the following nucleotide se-quence:

GGCCATCGCC GCAGATCCAG CGiCCCAGAGA GACACCAGAG AACCCACCAT GGCCCCCTTT

GACCCCTGGC TTCTGC.4TCC TGTTGTTGCT GTGGCTGATA GCCCCAGCAG GGCCTGCACC
1.30 140 150 160 ~ 170 180 TGTGTCCCAC CCCACCCACA GA.CGGCCTTC TGCAATTCCG ACCTCGTCAT CAGGGCCAAG

TTCGTGGGGA CACCAGi~AGT CAACCAGACC ACCTTATACC AGCGTTATGA GATCAAGATG

ACCAAGATGT ATAAAGGGTT CCAAGCCTTA GGGGATGCCG CTGACATCCG GTTCGTCTAC

ACCCCCGCCA TGGAGAGTGT CTGCGGATAC TTCCACAGGT CCCACAACCG CAGCGAGGAG

TTTCTCATTG CTGGAAAACT GCAGGATGGA CTCTTGCACA TCACTACCTG CAGTTTCGTG

GCTCCCTGGA ACAGCCiCGAG CTTAGCTCAG CGCCGGGGCT TCACCAAGAC CTACACTGTT

GGCTGTGAGG AATGCAC:AGT GT'TTCCCTGT TTATCCATCC CCTGCAAACT GCAGAGTGGC

ACTCATTGCT TGTGGAC:GGA CC.AGCTCCTC CAAGGCTCTG AAAAGGGCTT CCAGTCCCGT
I

_8_ CACCTTGCCT GCCTGCc;TCG GGAGCCAGGG CTGTGCACCT GGCAGTCCCT GCGGTCCCAG

ATAGCCTGAA TCCTGC(:CGG AGTGGAAGCT GAAGCCTGCA CAGTGTCCAC CCTGTTCCCA

CTCCCATCTT TCTTCCGGAC AATGAAATAA AGAGTTACCA CCCAGCAA.AA AAAAAAAGGA
Currently, for expression of the instant met-alloproteinase inhibitors in animal cells, the inventors most prefer a method which utilizes a fourth preferred DNA sequence.
The coding strand of this sequence reads as follows:

GGCCATCGCC GCAGATC'.CAG CGCCCAGAGA GACACCAGAG AACCCACCAT GGCCCCCTTT

GAGCCCCTGG CTTCTGGCAT CC'CGTTGTTG CTGTGGCTGA TAGCCCCCAG CAGGGCCTGC

ACCTGTGTCC CACCCCA.CCC ACAGACGGCC TTCTGCAATT CCGACCTCGT CATCAGGGCC

AAGTTCGTGG GGACACCAGA AG9t'CAACCAG ACCACCTTAT ACCAGCGTTA TGAGATCAAG

ATGACCAAGA TGTATAAAGG GTTCCAAGCC TTAGGGGATG CCGCTGACAT CCGGTTCGTC

TACACCCCCG CCATGGAGAG TGTCTGCGGA TACTTCCACA GGTCCCACAA CCGCAGCGAG

GAGTTTCTCA TTGCTGGAAA ACTGCAGGAT GGACTCTTGC ACATCACTAC CTGCAGTTTC

GTGGCTCCCT GGAACAGCCT GAGCTTAGCT CAGCGCCGGG GCTTCACCAA GACCTACACT

GTTGGCTGTG AGGAATGCAC AGTGTTTCCC TGTTTATCCA TCCCCTGCAA ACTGCAGAGT' J.

_9_ 1 3 4 0 9 7 1 GGCACTCATT GCTTGTGGAC GGACCAGCTC CTCCAAGGCT CTGAAAAGGG CTTCCAGTCC

CGTCACCTTG CCTGCCTGCC TCGGGAGCCA GGGCTGTGCA CCTGGCAGTC CCTGCGGTCC

CAGATAGCCT GAATCCTGCC CGGAGTGGAA GCTGAAGCCT GCACAGTGTC CACCCTGTTC

CCACTCCCAT CTTT~~TTCCG GACAATGAAA TAAAGAGTTA CCACCCAGCA
GGAATTC
To facilitate identification and isolation of natural DNA sequences for use in the present invention, the inventors have developed ;3 human skin fibroblast cDNA library. This li-brary contains the genetic information capable of directing a cell to synthesize the metalloproteinase inhibitors of the pres-ent invention. Other natural DNA sequences which may be used in the recombinant DNA methods set forth herein may be isolated from human genomic libraries.
Additionally, portable DNA sequences useful in the pro-cesses of the present invention may be synthetically created.
These synthetic DNA sequences may be prepared by polynucleotide synthesis and sequencing techniques known to those of ordinary skill in the art..
Additionally, to achieve the objects and in accordance with the purposes of the present,invention, a recombinant-DNA
method is disclosed which results in microbial manufacture of the instant metalloproteinase inhibitors using the portable DNA se-quences referred to above. This recombinant DNA method com-prises:
(a) preparation of a portable DNA sequence capable of directing a host microorga-nism to produce a protein having metallo:proteinase inhibitor activity, preferably collagenase inhibitor activity;
s a 134097 ~
-lo-(b) cloning i:he portable DNA sequence into a vector capable of being transferred into and replicating in a host microorganism, such vector containing operational ele-ments for the portable DNA sequence;
(c) transfers-ing the vector containing. the portable DNA sequence and operational elements into a host microorganism capa-ble of expressing the metalloproteinase inhibitor protein;
(d) culturing the host microorganism under conditions appropriate for amplification of the vector and expression of the in-hibitor; and (e) in either order: , (i) harvesting the inhibitor; and (ii) cau~~ing the inhibitor to assume an active, tertiary structure whereby it possesses metalloproteinase in-hibitor activity.
To further accomplish the objects and in further accord with the purposes of they present invention, a series of cloning vectors are provided comprising at least one of the portable DNA
sequences discussed above. In particular, plasmid pUC9-F5/237P10 is disclosed.
It is understood that both the foregoing general de-scription and the following detailed description are exemplary and explanatory only and. are not restrictive of the invention, as claimed.
The accompanying drawing, which is incorporated in and constitutes a part of this specification, illustrates one embodi-ment of the invention anl, together with the description, serves to explain the principles of the invention.
B;EtIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a partial restriction mad of the plasmid pUC9-F5/237P10.

_11_ 1 3 4 0 9 7 1 DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Reference will now be made in detail to the presently preferred embodiments of the invention, which, together with the drawing and the following examples, serve to explain the princi-ples of the invention.
As noted above, the present invention relates in part to portable DNA sequences capable of directing intracellular pro-duction of metalloprote:inase inhibitors in a variety of host microorganisms. "Portable DNA sequence" in this context is in-tended to refer either to a synthetically-produced nucleotide se-quence or to a restriction fragment of a naturally occuring DNA
sequence. For purposes of this specification, "metalloproteinase inhibitor" is intended t=o mean the primary structure of the pro-tein as defined by the codons present in the deoxyribonucleic , acid sequence which dirEacts intracellular production of the amino acid sequence, and which may or may not include post-transla-tional modifications. 7:t is contemplated that such post-transla-tional modifications include, for example, glycosylation. It is further intended that the term "metalloproteinase inhibitor"
refers to either the form of the protein as would be excreted from a microorganism or the methionyl-metalloproteinase inhibitor as it may be present in microorganisms from which it was not ex-creted.
In a preferred embodiment, the portable DNA sequences are capable of directing intracellular production of collagenase inhibitors. In a particularly preferred embodiment, the portable DNA sequences are capable of directing intracellular production of a collagenase inhibitor biologically equivalent to that previ-ously isolated from human skin fibroblast cultures. By "biologi-cally equivalent," as used herein in the specification and claims, it is meant that an inhibitor, produced using a portable DNA sequence of tile present invention, is capable of preventing collagenase-induced tissue damage of the same type, but not nec-essarily to the same degree, as a native human collagenase inhib-itor, specifically that native human collagenase inhibitor isolable from human skin fibroblast cell cultures.
,~,;~.i . 1340971 A first. preferred portable DNA sequence of the present invention has a nucleotide sequence as follows:

GTTGTTGCTG TGGCTGATAG CCCCAGCAGG GCCTGCACCT GTGTCCCACC CCACCCACAG

ACGGCCTTCT GCAATTCCGA CCTCGTCATC AGGGCCAAGT TCGTGGGGAC ACCAGAAGTC

AACCAGACCA CCTTATACCA GCGTTATGAG ATCAAGATGA CCAAGATGTA TA.AAGGGTTC

CAAGCCTTAG GGGATGCCGC TGACATCCGG TTCGTCTACA CCCCCGCCAT GGAGAGTGTC

TGCGGATACT TCCACA.GGTC Cc:ACAACCGC AGCGAGGAGT TTCTCATTGC TGGAA.AAC:TG

CAGGATGGAC TCTTGCACAT CACTACCTGC AGTTTCGTGG CTCCCTGGAA CAGCCTGAGC

TTAGCTCAGC GCCGGGGCTT CACCAAGACC TACACTGTTG GCTGTGAGGA ATGCACAGTG

CAGCTCCTCC AAGGCTCTGA AAAGGGCTTC CAGTCCCGTC ACCTTGCCTG CCTGCCTCGG

GAGCCAGGGC TGTGCACCTG GC:AGTCCCTG CGGTCCCAGA TAGCCTGAAT CCTGCCCGGA

GTGGAAGCTG AAGCCTGCAC AGTGTCCACC CTGTTCCCAC TCCCATCTTT CTTCCGGACA

ATGAAATAAA GAGTTACCAC CC:AGCAAAAA AAAAAAGGAA TTC
S

wherein the following nucleotides are represented by the abbrevi-ations indicated below.
Nucleotides Abbreviation Deoxyadenylic acid A
Deoxyguany~.ic acid G
Deoxyc~rtidylic acia C
Thymid;rlic acid T
A second preferred portable DNA sequence of the present invention has thEa following nucleotide sequence:

GGCCATCGCC GCAGA7:'CCAG CGCCCAGAGA GACACCAGAG AACCCACCAT GGCCCCCTTT

GACCCCTGGC TTCTGC:ATCC TGTTGTTGCT GTGGCTGATA GCCCCAGCAG GGCCTGCACC

TGTGTCCCAC CCCACC:CACA GACGGCCTTC TGCAATTCCG ACCTCGTCAT CAGGGCCAAG

TTCGTGGGGA CACCAGAAGT CAACCAGACC ACCTTATACC AGCGTTATGA GATCAAGATG

ACCAAGATGT ATAAAGGGTT CCAAGCCTTA GGGGATGCCG CTGACATCCG GTTCGTCTAC

ACCCCCGCCA TGGAGAGTGT CTGCGGATAC TTCCACAGGT CCCACAACCG CAGCGAGGAG

TTTCTCATTG CTGGAAAACT GCAGGATGGA CTCTTGCACA TCACTACCTG CAGTTTCGTG

GCTCCCTGGA AC
In this second preferred sequence, an open reading frame exists from nucleotides 1 through 432. The first methionine of this reading frame is encoded by nucleotides by 49 .through 51 and is the site of translation initiation. It should be noted that the amino acid sequence prescribed by nucleotides 49 through 114 is ' s r~ , not found in the mature metalloproteinase. It is believed that this sequence is the :Leader peptide of the human protein..
A third preferred portable DNA sequence has the nucleotide sequence:

GGCCATCGCC GCAGATCCAG CGCCCAGAGA GACACCACAC AACCCACCAT GGCCCCCTTT

GACCCCTGGC TTCTGCATCC T~~TTGTTGCT GTGGCTGATA GCCCCAGCAG GGCCTGCACC

TGTGTCCCAC CCCACC.'CACA G:~.CGGCCTTC TGCAATTCCG ACCTCGTCAT CAGGGCCAAG

TTCGTGGGGA CACCAC~AAGT CAACCAGACC ACCTTATACC AGCGTTATGA GATCAAGA'~'G

ACCAAGATGT ATAAAGGGTT CCAAGCCTTA GGGGATGCCG CTGACATCCG GTTCGTCTAC

ACCCCCGCCA TGGAGA.GTGT CTGCGGATAC TTCCACAGGT CCCACAACCG CAGCGAGGAG

TTTCTCATTG CTGGAAAACT GCAGGATGGA CTCTTGCACA TCACTACCTG CAGTTTCGTG

GCTCCCTGGA ACAGCCTGAG CTTAGCTCAG CGCCGGGGCT TCACCAAGAC CTACACTGTT

GGCTGTGAGG AATGCACAGT G7.'TTCCCTGT TTATCCATCC CCTGCAAACT GCAGAGTGGC

ACTCATTGCT TGTGGACGGA CC'AGCTCCTC CAAGGCTCTG AAAAGGGCTT CCAGTCCCGT
610 620 630 640 ' 650 660 CACCTTGCCT GCCTGCCTCG GGAGCCAGGG CTGTGCACCT GGCAGTCCCT GCGGTCCCAG
X. r.

_15_ 1 3 4 0 9 7 1 ATAGCCTGAA TCCTGC:CCGG AGTGGAAGCT GAAGCCTGCA CAGTGTCCAC CCTGTTCCCA

CTCCCATCTT TCTTCfGGAC A~ATGAAATAA AGAGTTACCA CCCAGCAAAA P,AAAAAAGGA
This third sequence contains the 5' nontranslated region of the second preferred sequence and the 3' region of the first pre-ferred sequence. It is envisioned that this third preferred se-quence is capable of directing intracellular production of a met-alloproteinase analogous; to a mature human collagenase inhibitor in a microbial or mammalian expression system.
Currently, for' expression of the instant met-alloproteinase in~hibitor~s in animal cells, the inventors most prefer a method which utilizes a fourth preferred DNA sequence.
The coding strand of this sequence reads as follows:

GGCCATCGCC GCAGAT(:CAG CGCCCAGAGA GACACCAGAG AACCCACCAT GGCCCCCTTT

GAGCCCCTGG CTTCTGGCAT CC'I'GTTGTTG CTGTGGCTGA TAGCCCCCAG CAGGGCCTGC

ACCTGTGTCC CACCCCACCC ACAGACGGCC TTCTGCAATT CCGACCTCGT CATCAGGGCC

AAGTTCGTGG GGACACC.AGA AGTCAACCAG ACCACCTTAT ACCAGCGTTA TGAGATCAAG

ATGACCAAGA TGTATAAi~GG GTTCCAAGCC TTAGGGGATG CCGCTGACAT CCGGTTCGTC

TACACCCCCG CCATGGAGAG TGTCTGCGGA TACTTCCACA GGTCCCACAA CCGCAGCGAG

GAGTTTCTCA TTGCTGGp,AA ACTGCAGGAT GGACTCTTGC ACATCACTAC CTGCAGTTTC

GTGGCTCCCT GGAACAGCCT GAG<:TTAGCT CAGCGCCGGG GCTTCACCAA GACCTACACT
~r GTTGGCTGTG AGGA,ATGCAC AGTGTTTCCC TGTTTATCCA TCCCCTGCAA ACTGCAGAGT

GGCACTCATT GCTT(3TGGAC GGACCAGCTC CTCCAAGGCT CTGAAAAGGG CTTCCAGTCC

CGTCACCTTG CCTGCCTGCC TCGGGAGCCA GGGCTGTGCA CCTGGCAGTC CCTGCGGTCC
670 680 690 700 710 72p CAGATAGCCT GAATC:CTGCC CGGAGTGGAA GCTGAAGCCT GCACAGTGTC CACCCTGTTC

CCACTCCCAT CTTTC:TTCCG GACAATGAAA TAAAGAGTTA CCACCCAGCA A,~~A.AAAAAAA
It mu~;t be borne in mind in the practice of the present invention that the alteration of some amino acids in a protein sequence may not affect the fundamental properties of the pro-tein.' Therefore, it is also contemplated that other portable DNA
sequences, both those capable of directing intracellular produc-tion of identical amino acid sequences and those capable of directing intracellular production of analogous amino acid se-quences which also possess metalloproteinase inhibitor activity, are included within then ambit of the present invention.
It is contemplated that some of these analogous amino acid sequences will be substantially homologous to native human metalloproteinase inhiY~itors while other amino acid sequences, capable of functioning as metalloproteinase inhibitors, will not exhibit substantial homology to native inhibitors. By "substan-tial homology," as used herein, is meant a degree of homology to a native metalloproteinase inhibitor in excess of 50%, preferably in excess of 60%, preferably in excess of 80%. The percentage homology as discussed Herein is calculated as the percentage of amino acid residues found in the smaller of the two sequences that align with identical amino acid residues in the sequence being compared when four gaps in a length of 100 amino acids may be introduced to assist. in that alignment as set forth by Dayhoff, M.O. in Atlas of Protein Sequence an_d Structure Vol. 5, p. 124 (1972), National. Biochemical Research Foundation, Washington, D.C.

X34097 ~

As noted above, the portable DNA sequences of the pres-ent invention may be synthetically created. It is believed that the means for synthetic creation of these polynucleotide se-quences are generally known to one of ordinary skill in the art, particularly in light of the teachings contained herein. As an example of the current state of the art relating to. poly-nucleotide synthesis, one is directed to Matteucci, M.D. and Caruthers, M.H., in J. Am. Chem. Soc. 103: 3185 (1981) and Beaucage, S.L. and Caruthers, M.H. in Tetrahedron Lett. _22: 1859 (1981).
Additionally, the portable DNA sequence may be a frag-ment of a natural_ sequence, i.e., a fragment of a polynucleotide which occurred in nature and which has been isolated and purified for the first time by the present inventors. In one embodiment, the portable DNA sequence is a restriction fragment isolated from a cDNA library. In this preferred embodiment, the cDNA library is created from ruuman skin fibroblasts.
In an alternative embodiment, the portable DNA sequence is isolated from a human genomic library. An example of such a library useful in this embodiment is set forth in Lawn et al.
Cell 15: 1157-1174 (1978), As also noted above, the present invention relates to a series of vectors, each containing at least one of the portable DNA sequences described herein. It is contemplated that addi-tional copies of the portable DNA sequence may be included in a single vector to increase a host microorganism's ability to pro-duce large quantities of the desired metalloproteinase inhibitor.
In add ition, the cloning vectors within the scope of the present invent=ion may contain supplemental nucleotide se-quences preceding or subsequent to the portable DNA sequence.
These supplementa~_ sequences are. those that will not interfere with transcription of the portable DNA sequence and will, in some instances as set 1'orth more fully hereinbelow, enhance transcrip-tion, translation, or the ability of the primary amino acid structure of the resultant metalloproteinase inhibitor to assume an active, tertiary form. ' -18- ~ 3 4 0 9 7 ~
A preferred vector of the present invention is set forth in Figure 1. This vector, pUC9-F5/237P10, contains the preferred nucleotide sequence set forth above. Vector pUC9-F5/237P10 is present in the C600/pUC9-FS/237P10 cells on de-posit in the American Type Culture Collection in Rockville, Maryland under Accession No. 53003.
A preferred nucleotide sequence encoding the metallo-proteinase inhibitor is identified in Figure 1 as region A.
Plasmid pUC9-FS/?37P10 also contains supplemental nucleotide se-quences preceding and subsequent to the preferred portable DNA
sequence in region A. These supplemental sequences are identi-fied as regions F3 and C, respectively.
In alternate :preferred embodiments, either one or both of the preceding or subsequent supplemental sequences may be re-moved from the vector o:E' Fig. 1 by treatment of the vector with restriction endonucleasEas appropriate for removal of the supple-mental sequences. The supplemental sequence subsequent to the portable DNA sequence,' identified in Fig. 1 as region C, may be removed by treatment of the vector with a suitable restriction endonuclease, preferabl~~ H~iAI followed by reconstruction of the 3' end of region A using synthetic oligonucleotides and ligation of the vector with T-4 DNA ligase. Deletion of the supplemental sequence preceding the portable DNA sequence, identified as re-gion B in Fig. 1, would be specifically accomplished by the meth-od set forth in E:Xample 2.
In preferred embodiments, cloning vectors containing and capable of expressing the portable DNA sequence of the pres-ent invention cons=ain various operational elements. These "oper-ational elements," as discussed herein, include at least one pro-moter, at least one Shine-Dalgarno sequence, at least one terminator codon. Preferably, these "operational elements" also include at least one operator, at least one leader sequence, and for proteins to be exported from intracellular space, at least one regulator and any ot)ner DNA sequences necessary or preferred for appropriate transcription and subsequent translation of the vector DNA.

X34097' Additional embodiments of the present invention are en-visioned as emplo~ring other known or currently undiscovered vectors which would contain one or more of the portable DNA se-quences described herein. In particular, it is preferred that these vectors have some or all of the following characteristics:
(1) possess a minimal number of host-organism sequences; (2) be stable in the desired host; (3) be capable of being present in a high copy number i.n the desired host; (4) possess a regulatable promoter; (5) have at least one DNA sequence coding for a se-lectable trait present on a portion of the plasmid separate from that where the portable 17NA sequence will be inserted; and (6) be integrated into the vecto r.
The following, noninclusive, list of cloning vectors is believed to set forth vectors which can easily be altered to meet the above-criteria and are therefore preferred for use in the present invention. Such alterations are easily performed by those of ordinary skill in the art in light of the available lit-erature and the te,achings~ herein.
TABLE I
HOST Vectors Comments E. coli pUCB Many selectable replicons pUC9 have been characterized.
pBR322 Maniatis, T. et al. (1982), pGW7 Molecular Clonin ~ A
placIq Laboratory Manual, Cold pDPB Spring Harbor Laboratory.
BACILLUS pUB110 Genetics and Biotechnology B. subtilis p~5A0501 of Bacilli, Ganesan and B. amyloliquefaciens p;5A2100 Hoch, eds., 1984, Academic B. stearothermophilus p13D6 Press.
p13D8 p'.~ 12 7 1 340 97 ~

PSEUDOMONAS RSF1010 Some vectors useful in P. aeruginosa Rms149 broad host range of gram-P. putida pKT209 negative bacteria including RK2 Xanthomonas and Agrobacterium.
pSa727 CLOSTRIDIUM pJUl2 Shuttle plasmids for E.
C. perfringens pJU7 coli and C.
perfringens pJUlO construction ref. Squires, pJUl6 C. et al. (1984) Journal pJUl3 Bacteriol. 159:465-471.
SACCHAROMYCES YEp24 Botstein and Davis in S. cerevisiae YIpS Molecular Biology of the .
YRpl7 Yeast Saccharomyces, Strathern,. Jones, and Broach, eds., 1982, Cold Spring Harbor Laboratory.
It is to be understood that additional cloning vectors may now exist or will be <9iscovered which have the above-identified prop-erties and are the refore suitable for use in the present inven-tion. These vectors are also contemplated as being within the scope of the disclosed series of cloning vectors into which the portable DNA sequences may be introduced, along with any neces-sary operational elements, and which altered vector is then in-cluded within the scope of the present invention and would be capable of being used in the recombinant-DNA method set forth more fully below.
In addition to the above list, an E. coli vector sys-tem, as set forth in Example 2, is preferred in one embodiment as a cloning vector. Moreover, several vector plasmids which auton-omously replicate in a broad range of Gram Negative bacteria are preferred for use as cloning vehicles in hosts of the genera Pseudomonas. The:;e are described by Tait, R.C:, Close, T.J., Lundquist, R.C., fiagiya, M., Rodriguez, R.L., and Kado, C.I. in Biotechnology, Ma~~, 1983, pp. 269-275; Panopoulos, N.J. in 134097 ~

Genetic Engineering_in the Plant Sciences, Praeger Publishers, New York, New York, pp. 163-185, (1981); and Sakaguchi, K. in Gl~rrent Topic in Microbiology and Imnn~nology 96:31-45, (1982).
One particularly preferred construction employs the plasmid RSF1010 and derivatives thereof as described by Bagdasarian, M., Bagdasarian, M.M., C:oleman, S., and Timmis, K.N. in Plasmids of Medical Environmental and Ccm~nercia:l Importance, Timmis, K.N. and Puhler, A.
eds., Elsevier/North Holland Biomedical Press, (1979). The advantages of RSF1010 are that it is relatively small, high copy number plasmid which is readily transformed into and stably maintained in both E. coli and Pseudomonas species. In this system, it is preferred to use the Tac expression system as described for Escherichia, since it appears that the, E. coli trp promoter is readily recognized by Pseudomonas RNA polymerase as set forth by Sakaguchi, K. in Clzrrent Topics in Microbiology and Immunology 96:31-45 (1982) and Gray, G.L., McKeown, K.A., Jones, A.J.S., Seeburg, P.H., and Heyneker, H.L. in Biotechnology Feb. 1984, pp. 161-165. Transcriptional activity may be further maximized by requiring the exchange of the prompter with, e.g., an E. coli or P. aeruginosa trp promoter.
In a preferred e~~nbodiment, P. aeruginosa is transformed with vectors directing the synthesis of the metalloproteinase inhibitor as either an intracellu:Lar product or as a product coupled to leader sequences that will ssffect its processing and export from the cell. In this embodiment, there leader sequences are preferably selected from the group consisting of )seta-lactainase, OmpA protein, the naturally occurring human signal peptide, and that of carboxypeptidase G2 from Pseudomonas. Translation may be coupled to translation initiation for any of the E. coli proteins as described in Example 2, as well as to initiation sites for any of the highly expressed proteins of the host to cause intracellular expression of the metalloproteinase inhibitor.
.~~" . ~

~y'~.

In those cases vvhere restriction minus strains of a host Pseudomonas species are not available, transformation efficiency with plasmid constructs isolated from E. coli are poor. Therefore, passage of the Pseudomonas cloning vector through an r- m+ strain of another species prior to transformation of 'the desired host, as set forth in Bagdasarian, M., et al., Plasmids of Medical, Environmental and Commercial Importance, pp. 411-422, Timmis and Puh:Ler eds., Elsevier/North Holland Biomedical Press (1979), is de:;ired.
Furthern~~re, a preferred expression system in hosts of the genera Bacillus involves us:Lng plasmid pUB110 as the cloning vehicle. As in other host vector systema, it is possible in Bacillus to express the metalloproteinase inhibitor: of the present invention as either an , intracellular or a secreted protein. The present embodiments include both systems. Shuttle vectors that replicate in both Bacillus and E. coli are available for constructing and testing various genes as described by Dubnau, D., Gryczan, T., Corrtente, S., and Shivakumar, A.G. in Genetic Encrineering, Vol. 2, Setlow and Hollander eds., Plenum Press, New York, New York, pp. 115-131, (198C)). For the expression and secretion of metalloproteinase inhibitors from B. subtilis, the signal sequence of alpha-amylase is preferably coupled to the coding region for the metalloproteinase inhibitor. For synthesis of intracellular metalloproteinase irihibitor, the portable DNA sequence will be translationally coupled to the ribosome binding site of the alpha-amylase leader sequence.
Transcription of either of these constructs is preferably directed by the alpha-amylase promoter or a derivative thereof. This derivative contains ~=he RNA polymerase recognition sequence of the native alpha-amylase promoter but incorporates the lac operator region as well.
Similar hybrid promot=ers constructed from the penicillinase gene promoter and the lac operator have been shown to function in Bacillus hosts in a regulatable fashion as set forth by Yansura, D.G. and Henner in Genetics and Biotechnology of Bacilli, Ganesan, A.T. and Hoch, 1340 g7 ~

J.A., eds., Academic: Press, pp. 249-263, (1984). The lacI gene of lacIq would also be included to effect regulation.
One pre=ferred construction for expression in Clostridium is in plasmid pJZTl2 described loy Squires, C. H. et al in J. Bacteriol.
159:465-471 (1984), transfo:cmed into C. perfrinc~ens by the method of Heefner, D. L. et al.. as described in J. Bacteriol. 159:460-464 (1984).
Transcription is directed b~~ the promoter of the tetracycline resistance gene. Translation is coupled to the Shine-Dalgarno sequences of this same tetr gene in a inner ~;trictly analogous to the procedures outlined above for vectors suitable for use in other hosts.
Maintenance of foreign DNA introduced into yeast can be effected in several ways (Botstein, D., and Davis, R. W., in The Molecular Bioloctv of the Yeast Saccharomyces, Cold Spring Harbor , Laboratory, Strathern, Jones and Broach, eds., pp. 607-636 (1982). One preferred expression system for use with host organisms of the genus Sacchammyces harbors the anticollagenase gene on the 2 micron plasmid.
The advantages of th~= 2 micron circle include relatively high copy number and stability when i:ztroduced into cir° strains. These vectors preferably incorporave the replication origin and at least one antibiotic resistance marker from pBR322 to allow replication and selection in E.
coli. In addition, t=he plasmid will preferably have 2 micron sequences and the yeast LEU2 ge=ne to serve the same purposes in LEU2 mutants of yeast.
The regi:~latable promoter from the yeast GAL1 gene will preferably be adapted to direct transcription of the portable DNA
sequence gene. Tram>lation of the portable DNA sequence in yeast will be coupled to the leader sequen~~e that directs the secretion of yeast alpha-factor. This will cause fozm~tion of a fusion protein which will be processed in yeast arid result in secretion of a metalloproteinase inhibitor. Alternatively, a methionyl-metalloproteinase inhibitor will be translated for inclusion within the cell.

134pg7 ~

It is anticipated that translation of mRNA coding for the metalloproteinase inhibitor in yeast will be~ more efficient with the preferred codon usage of yeast than with the sequence present in pUCB-Fic, as identified in Example 2, which has been tailored to the prokaryotic bias. For this reason, the portion of the 5' end of the portable DNA sequence beginning at the Tth111I site is preferably resynthesized. The new sequence fa-vors the codons most frequently used in yeast. This new sequence preferably has the following nucleotide sequence:
HgiA:I
S' GAT CCG TGC ACT TGT GTT CCA CCA CAC
GC ACG 'FGA ACA CAA GGT GGT GTG
CCA CAA ACT GCT TTC TGT AAC TCT GAC C
GGT GTT TGA (:GA AAG ACA TTG AGA CTG GA 3' As will be seen from an examination of the individual cloning vectors and systems contained on the above list and de-scription, various. operational elements may be present in each of the preferred vectors of the present invention. It is contem-plated any additional ope rational elements which may be required may be added to these vec tots using methods known to those of or-dinary skill in the art, particularly in light of the teachings herein.
In practice, ii. is possible to construct each of these vectors in a way that al7Lows them to be easily isolated, assem-bled, and interchanged. This facilitates assembly of numerous functional genes from connbinations of these elements and the coding region of the metalloproteinase inhibitor. Further, many of these elements will beg applicable in more than one host.
At least one origin of replication recognized by the contemplated host microorganism, along with at least one select-able marker and at least one promoter sequence capable of initiating transcription of the portable DNA sequence are contem-plated as being included in these vectors. It is additionally contemplated that the vectors, in certain preferred embodiments, will contain DNA sequences capable of functioning as regulators ("operators"), and other DNA sequences capable of coding for _ .

regulator proteins. In preferred vectors of this series, the vectors additionally contain ribosome binding sites, transcrip-tion terminators and leader sequences.
These regulators, in one embodiment, will serve to pre-vent expression of the portable DNA sequence in the presence of certain environmental conditions and, in the presence of other environmental conditions, allow transcription and subsequent ex-pression of the protein coded for by the portable DNA sequence.
In particular, it. is preferred that regulatory segments be in-serted into the vector such that expression of the portable DNA
sequence will not. occur in the absence of, for example, isopro-pylthio- ~ -d-gal<ictoside. In this situation, the transformed mi-croorganisms containing the portable DNA may be grown to a de-sired density pr:Lor to initiation of the expression of the metalloproteinasEa inhibitor. In this embodiment, expression of the desired protease inhibitor is induced by addition of a sub-stance to the microbial environment capable of causing expression of the DNA sequence after the desired density has been achieved.
Additionally. it is preferred that an appropriate se-cretory leader sequence be present, either in the vector or at the 5' end of they portable DNA sequence, the leader sequence being in a position which allows the leader sequence to be imme-diately adjacent.to the initial portion of the nucleotide se-quence capable oi= directing expression of the protease inhibitor without any intervening transcription or translation termination signals. The presence of the leader sequence is desired in part for one or more of the following reasons: 1) the presence of the leader sequence nnay facilitate host processing of the initial product to the mature recombinant metalloproteinase inhibitor; 2) the presence of t:he lea~aer sequence may facilitate purification of the recombinant metalloproteinase inhibitors, through directing the met:allopr~oteinase inhibitor out of the cell cytoplasm; 3) the presence of the leader sequence may affect the ability of the recombinant metalloproteinase inhibitor to fold to its active structure through directing the metalloproteinase in-hibitor out of the cell cytoplasm. , In particular, the leader sequence may direct cleavage of the initial translation product by a leader peptidase to re-move the leader sequence and leave a polypeptide with the amino acid sequence which has the potential of metalloproteinase inhib-itory activity. In some species of host microorganisms, the presence of the appropriate leader sequence will allow transport of the completed protean into the periplasmic space, as in the case of E. coli. In the case of certain yeasts and strains of Bacillus and Pseudomonas, the appropriate leader sequence will allow transport of the protein through the cell membrane and into the extracellula.r medium. In this situation, the protein may be purified from ex.tracel:Lular protein.
Thirdly, in the case of some of the metalloproteinase inhibitors prepared by the present invention, the presence of the leader sequence may be necessary to locate the completed protein in an environment where it may fold to assume its active struc-ture, which structure possesses the appropriate metalloproteinase activity.
Additional operational elements include, but are not limited to, ribosome-b;Lnding sites and other DNA sequences neces-sary for microbial expression of foreign proteins. The opera-tional elements as disc ussed herein can be routinely selected by those of ordinary skil:L in the art in light of prior literature and the teachings contained herein. General examples of these operational elements are~set forth in B. Lewin, Genes, Wiley &
Sons, New York (1983).
Various examples of suitable operational elements may be found on the vectors discussed above and may be elucidated through review of the publications discussing the basic charac-teristics of the aforementioned vectors.
In one preferred embodiment of the present invention, an additional DNA sequence is located immediately preceding the portable DNA sequence which codes for the metalloproteinase in-hibitor. The additional DNA sequence is capable of functioning as a translational coupler, i.e., it is a DNA sequence that encodes an RNA which serves to position ribosomes immediately adjacent to the ribosome binding site of the metalloproteinase inhibitor RNA with which it is contiguous. , 1 340 97 ~

Upon synthesis and/or isolation of all necessary and desired component parts of the above-discussed cloning vectors, the vectors are assembled by methods generally known to those of ordinary skill in the ari:. Assembly of such vectors is believed to be within the duties and tasks performed by those with ordi-nary skill m the art an<i, as such, is capable of being performed without undue experiments tion. For example, similar DNA se-quences have been ligate<i into appropriate cloning vectors, as set forth in Schoner et al., Proceedings of the National Academy of Sciences U.S.A., 81:5403-5407 (1984).
In construction of the cloning vectors of the present invention, it should additionally be noted that multiple copies of the portable DNA sequcance and its attendant operational ele-ments may be inserted ini_o each vector. In such an embodiment;
the host organism would produce greater amounts per vector of the desired metalloproteinase inhibitor. The number of multiple copies of the DNA sequence which may be inserted into the vector is limited only by the ability of the resultant vector, due to its size, to be transferred into and replicated and transcribed in an appropriate host microorganism.
Additionally, .it is preferred that the cloning vector contain a selectable marker, such as a drug resistance marker or other marker which causes expression of a selectable trait by the host microorganism. In a particularly preferred embodiment of the present invention, the gene for ampicillin resistance is in-cluded in vector p~UC9-F5,~237P10.
Such a drug resistance or other selectable marker is intended in part to facilitate in the selection of transformants.
Additionally, the presence of such a selectable marker on the cloning vector may be of use in keeping contaminating microorga-nisms from multiplying in the culture medium. In this embodi-ment, such a pure culture of the transformed host microorganisms would be obtained by culturing the microorganisms under condi-tions which require the :induced phenotype for survival.
It is noted that, in preferred embodiment, it is also desirable to reconstruct the 3' end of the coding region to allow .,., 1,340 97 1 assembly with 3' non-translated sequences. Included among these non-translated s~equence~s are those which stabilize the mRNA or enhance its transcription and those that provide strong tran-scriptional terminatior,~ signals which may stabilize the vector as they are identified by Gentz, R., Langner, A., Chang, A.C.Y., Cohen, S.H., and Bujard,,. H. in Proc. Natl. Acad. Sci. USA
78:4936-4940 (19;91).
This invention also relates to a recombinant-DNA method for the producti~~n of metallproteinase inhibitors. Generally, this method includes:
(a) p:ceparation of a portable DNA sequence c.3pable of directing a host microorga-nism to produce a protein having , m~stallop~roteinase inhibitor activity;
(b) c:Loning the portable DNA sequence into a vector capable of being transferred into and replicating in a host microorganism, such vector containing operational ele-ments for the portable DNA sequence;
(c) transferring the vector containing the portable DNA sequence and operational elements into a host microorganism capa-b:Le of expressing the metalloproteinase inhibitor protein;
(d) culturing the host microorganism under conditions appropriate for amplification o:E the vector and expression of the in-h:Lbitor; and (e) in either order:
(:i) harvesting the inhibitor; and (:ii) causing the inhibitor to assume an active, tertiary structure whereby it possesses metalloproteinase in-hibitor activity.
In thia method, the portable DNA sequences are those , synthetic or naturally-occurring polynucleotides described above.

In a preferred embodiment of the present method, the portable DNA
sequence has the nucleotide sequence as follows:

GTTGTTGCTG TGGCTGA'rAG CCf.CAGCAGG GCCTGCACCT GTGTCCCACC CCACCCACAG

ACGGCCTTCT GCAATTC~~GA CC'rCGTCATC AGGGCCAAGT TCGTGGGGAC ACCAGAAGTC

AACCAGACCA CCTTATA~CA GCCiTTATGAG ATCAAGATGA CCAAGATGTA TAAAGGGTTC

CAAGCCTTAG GGGATGCCGC TGP~CATCCGG TTCGTCTACA CCCCCGCCAT GGAGAGTGTC
250 260 270 280 ' 290 300 TGCGGATACT TCCACAGGTC CCFvCAACCGC AGCGAGGAGT TTCTCATTGC TGGAAAACTG

CAGGATGGAC TCTTGCACAT CAC'.TACCTGC AGTTTCGTGG CTCCCTGGAA CAGCCTGAGC

TTAGCTCAGC GCCGGGGCTT CAC'.CAAGACC TACACTGTTG GCTGTGAGGA ATGCACAGTG

CAGCTCCTCC AAGGCTCTGA AAAGGGCTTC CAGTCCCGTC ACCTTGCCTG CCTGCCTCGG

GAGCCAGGGC TGTGCACCTG GCAGTCCCTG CGGTCCCAGA TAGCCTGAAT CCTGCCCGGA

GTGGAAGCTG AAGCCTGCAC AGTGTCCACC CTGTTCCCAC TCCCATCTTT CTTCCGGACA

ATGAAATAAA GAGTTACCAC CCAGCAAAAA AAAAAAGGAA TTC

The vectors contemplated as being useful in the present method are those described above. In a preferred embodiment, the cloning vector pUC9-F5/237P10 is used in the disclosed method.
The vect~~r thus obtained is then transferred into the appropriate host microorganism. It is believed that any microorganism having the ability t.o take up exogenous DNA and express those genes and attendant operational elements may be chosen. It is preferred that the host microorganism k~e an an,~erobe, facultative anaerobe or aerobe.
Particular hosts which may 'oe preferable for use in this method include yeasts and bacteria. Specific yeasts include those of the genus Saccharomyces, and especial=Ly Saccharomvces cerevisiae.
Specific bacterua include those of the genera Bacillus and Escherichia and Pseizd~nas. Various other preferred hosts are set forth in Table I, s-upra. In other, alternatively preferred embodiments of the present invention, E,acillus subtilis. Escherichia coli or Pseudomonas aerucrinosa are selecaed as the host microorganisms.
After a host organism has been chosen, the vector is transferred into the=_ host organism using methods generally known by those of ordinary skill in the art. Examples of such methods may be found in Advanced Bacterial Genetics by R. W. Davis et al., Cold Spring Harbor Press, Cold Spring Harbor, Pdew York, (1980). It is preferred, in one embodiment, that the transformation occur at low temperatures, as temperature regulation is contemplated as a means of regulating gene expression through the use of operational elements as set forth above.
In another embodiment, if osmolar regulators have been inserted into the vector, regulation of the :salt concentrations during the transformation would be required to insure' appropriate control of the synthetic genes.
If it is contemplated that the recombinant metalloproteinase inhibitors will ultimately be expressed in yeast, it is preferred that the cloning vector first beg transferred into Escherichia coli, where the vector would be allowed to replicate and from which the vector would be obtained and purified after amplification. The vector would then be transferred into tr.e yeast for ultimate expression of the metalloproteinase inhibitor.
,.

The host microorganisms are cultured under conditions appropriate for the expression of the metalloproteinase inhib-itor. These conditions are generally specific for the host or-ganism, and are readily determined by one of ordinary skill in the art, in light of the published literature regarding the growth conditions for such organisms, for example Bergey's Manual of Determinative Bacteriology, 8th Ed., Williams & Wilkins Compa-ny, Baltimore, Maryland..
Any conditions. necessary for the regulation of the ex-pression of the D.NA sequence, dependent upon any operational ele-ments inserted into or F>re$ent in the vector, would be in effect at the transformation and culturing stages. In one embodiment, the cells are grown to a, high density in the presence of appro-priate regulatory conditions which inhibit the expression of the, DNA sequence. When optimal cell density is approached, the envi-ronmental conditions are altered to those appropriate for expres-sion of the portable DNA sequence. It is thus contemplated that the production of the metalloproteinase inhibitor will occur in a time span subsequent to the growth of the host cells to near optimal density, and that the resultant metalloproteinase inhib-itor will be harvested at some time after the regulatory condi-tions necessary for its expression were induced.
In a preferred embodiment of the present invention, the recombinant metal7.oproteinase inhibitor is purified subsequent to harvesting and pr3.or to assumption of its active structure. This embodiment is preferred as the inventors believe that recovery of a high yield of re-folded protein is facilitated if the protein is first purified. However, in one preferred, alternate embodi-ment, the metalloproteinase inhibitor may be allowed re-fold to assume its active structure prior to purification. In yet another preferred, alternate embodiment, the metalloproteinase inhibitor is caused to assume its re-folded, active state upo n recovery from the cultur:Lng medium.
In certain circumstances, the metalloproteinase inhib-itor will assume its proper, active structure upon expression in the host microorganism and transport of the protein through the , cell wall or memt~rane or into the periplasmic space. This will generally occur if DNA coding for an appropriate leader sequence has been linked to the I)NA coding for the recombinant protein.
The preferred metalloprotienase inhibitors of the present inven-tion will assume their mature, active form upon translocation out of the inner cell membrane. The structures of numerous signal peptides have been published, for example by Marion E.E. Watson in Nuc. Acid Res. 12:515-5164, 1984.
It is intendec9 that these leader sequences, together with portable DNA, will direct intracellular production of a fusion protean which will be transported through the cell membrane and will have the leader sequence portion cleaved upon release from the cell.
In a preferred embodiment, the signal peptide of the E.coli OmpA protein is used as a leader sequence and is~located.~
in a position contiguous with the portable DNA sequence coding for the metalloproteines~e inhibitor structure.
Additionally preferred leader sequences include those of beta-lactamase, carbo:Kypeptidase G2 and the human signal pro-tein. These and other leader sequences are described.
If the metalloproteinase inhibitor does not assume its proper, active structure,, any disulfide bonds which have formed and/or any noncovalent interactions which have occurred will first be disrupted by denaturing and reducing agents, for exam-ple, guanidinium chlorides and ~ -mercaptoethanol, before the metalloproteinase inhibitor is allowed to assume its active structure following dilution and oxidation of these agents under controlled conditions.
The transcription terminators contemplated herein serve to stabilize the vector. In particular, those sequences as de-scribed by Gentz _et. al., in Proc. Natl. Acad. Sci. USA _78:
4936-4940 (1981), are contemplated for use in the present invention.
It is to be understood that application of the teach-ings of the present: invention to a specific problem or environ-ment will be within the capabilities of one having ordinary skill in the art in light. of the teachings contained herein. Examples of the products of the present invention and representative processes for their isolation and manufacture appear in the following examples.
EXAMPLES

Preparation of Poly A+ RNP, from HEF-SA Fibroblasts HEF-SA cells were grown to near confluence in 75 cm2 T-flasks. Cells were washed twice in Dulbecco's phosphate buffered saline solution and harve:~ted by the addition of 2 ml of lOmM Tris, pH

7.5 containing 1% w/v SDS (obtained from BDH Chemicals, Ltd., Poole, England), 5mM EDTA and 20 ug/ml protease K (obtained from Boehringer Mannheim Biochemica.ls, Indianapolis, Indiana). Each flask was subsequently washed with an additional milliliter of this same solution.
The pooled aliquots from the cell harvest were made to 70 ug/ml in protease K and incubated at 40°C for 45 minutes. The , proteolyzed solution was brought to a NaCl concentration of 150 mM by the addition of 5 M stock and subsequently extracted with an equal volume of phenol: chloroform 1:1. The aqueous phase was reextracted with an equal volume of chloroform. Two volumes of ethanol were added to the aqueous phase and incubated overnight at -20°C. The precipitated nucleic acids were recovered by centrifugation at 17,500 TM
xg for 10 minutes i:n a Beckman J2-21 centrifuge, Beckman Instruments, Palo Alto, California, and were redissolved in 25 ml of 0.1% w/v SDS.
This solution was a~3ain extracted with an equal volume of chloroform.
The aqueous phase w~~s added to two volumes of cold ethanol and kept at -20°C for 2 hours. The precipitate was collected by centrifugation at 10,000 xg for 15 minutes a:nd redissolved in 10 ml of 1 mM Tris, 0.5 mM
EDTA, 0.1% SDS, pH '7.5. R1VA was precipitated from this solution by the addition of 10 rnl of 4 M LiCl, 20 mM NaoAc, pH 5.0 and incubated at -20°C for 18 hours. The precipitate was again recovered by centrifugation and washed twice with 2 M LiCl before redissolving in 1 mM Tris, 0.5 mM EDTA, 0.1% SDS, pH 7.5. This solution was stored at -70°C.

Chromatography on Oligo dT Cellulose Total cellular RNA prepared as above was ethanol pre-cipitated and redissolved in 0.5 M NaCl. Five ml of RNA at 0.45 mg/ml were applied to a 1 ml column of washed type VII oligo dT
cellulose (obtained from PL Biochemicals, Milwaukee, Wisconsin).
The column was then washed with 10 ml of 0.5 M NaCl and eluted with 2.0 ml of sterile H20. The eluted poly(A+) fraction of RNA
was ethanol precipitated and dissolved to give a 1 mg/ml solution in 1 mM Tris, 0.1 mM EDTA, pH 8Ø This was stored at -70°C.
cDNA Synthesis Poly(A+) RNA Hras primed with oligo dT (obtained from PL
Biochemicals, Milwaukee, Wisconsin) to serve as a template for cDNA synthesis by AMV reverse transcriptase (obtained from Life Sciences, Inc., St. Petersburg, Florida). Following the synthe-sis reaction, the RNA was hydrolyzed by the addition of 0.1 voi~
ume of 3 N NaoH and incubated at 67°C for 10 minutes. The solu-tion was then neutralized and the cDNA purified by gel filtration rM
chromatography on Biogel A 1.5 (obtained from BioRad Laboratories, Rictunond, California) in a 0.7x25 cm column in a 10 mM Tris, 5 mM EDTA, and 1% SDS, pH 7.5 solution. Fractions con-taining cDNA were pooled and concentrated by ethanol precipita-tion. The cDNA was dG trailed and purified by gel filtration using the procedure set forth above. Second strand synthesis was primed with oligo dC and polymerized in an initial reaction with the large (Klenow) fragment of DNA polymerase (obtained from Boehringer Mannheim). Following second strand synthesis, _E. coli DNA polymerase I (obtain<~d from Boehringer Mannheim) was added and incubation continued to form blunt ends. The double stranded cDNA was again purified by chromatography. EcoRI restriction sites within the cDNAs were modified by the action of EcoRI
methylase, obtained from New England Biolabs, Beverly, Mas-sachusetts. The cDNA wa~~ again purified and ligated to synthetic EcoRI linkers. Finally, the ends were then trimmed with the endonuclease and t'he cDNA, purified by gel filtration. This DNA
was ligated into a unique EcoRI site in lambda'gtl0 DNA packaged in vitro and used to infect _E. coli strain hflA according to the method set forth b:~~ Huynh, T.V., Young, R.A., and Davis, R.W., in 1~4097~

DNA Cloning Technirn;~es, A Practical Approach (ed. Glover, D.M.) (IRL
Press Oxford), in pz~ess. Approximately 25,000 recombinants were amplified in this manner.
Screenincr Recombinant-phage-containing sequences of interest were selected by their preferential hybridization to synthetic oligo-nucleotides encoding portions of the primary structure of the desired metalloproteinase inhibitor, hereinafter referred to as FIBAC. These portions of the protein seq~zence correspond in part to those set forth in the published literature by Stricklin, G.P. and Welgus, H.G., J. Biol.
Chem. 258: 12252-12258 (198:3). Recombinant phage were used to infect E.
coli strain hflA and plated at a density of approximately 2x103 pfu/150 mm petri dish. Phao~e were blotted onto nitrocellulose filters (BA85, Schleicher & Schuell Inc., Keene, New Hampshire), and DNA was denatured and fixed essentially as described by Benton and Davis in Science 196:180-182 (1979).
Using that procedure, the filters were treated sequentially for 10-15 minutes each in 0,.5 M NaCl, then 1.0 M Tris, 1.5 M NaCl pH 8.0, and finally submerged in 2x SSPE. (2x SSPE is 0.36 M NaCl, 20 mM NaHzP04, 2 mM EDTA pH 7.4). Filters were blotted dry and baked 75°-80°
for 3-4 hours. Duplicate filters were made of each plate. Filters were prehybridized for 1-3 hours at 37° in 5x SSPE containing O.lx SET, 0.150 NaPPi, and lx Denhardts solutions. Filters were then hybridized for 72 hours at 37° in this same solution containing 5x105cpm/ml of 5' end-labeled 51-mer oligonucleotide specific activity approximately 106 cpm/pmole). Following hybridization, filters were washed six times in 5x SSPE containing O.lx SET and 0.05% sodium pyro-phosphate at 37°, then three times in 2x SSPE at 21.°. These were then blotted dry and autorad_iographed on .Kodak'"' YP.R-5 film at -70° with a Kodak lightening-plus intensifying screen. ~~ignals clearly visible from duplicate filters were used to pick ph~ge for plaque purification. Filter preparations and hybridization procedires for plaque purification steps were the same as above.
.;

The washing procedure was simplified to 6 changes of 2x SSPE at 37°. Six isolates purified by repetit=ive plating were then arranged on a single lawn of _E. coli strain C600 for testing with subsequent probes.
Preferential hybridization of the 17-mer to each of the isolates (as opposed to control plaques) was observed under a condition identical to that u:~ed for ;plaque purification. Probe C was used in a similar test, except: that the SSPE concentration during hybridization was reduced to 4x. Aga__n, each of the isolates demonstrated stronger hybridization to the probe than did control plaques.
Phag_e Purification <rnd cDNA Characterization Quantities of each of the six isolated phage were made by the plate stock technique and purified by serial CsCl block gradient , centrifugation. DNi~ was extracted from these by dialysis against 50%
forn~amide as described by Davis, R.W., Botstein, D., Roth, J.R., in A
Manual for Genetic l~gineering~ Advanced Bacterial Genetics, 1980, Cold Spring Harbor L~abor,~tory. DNA from each of the isolates was digested with EcoRI and the produ~~ts were: analyzed by agarose gel electrophoresis. The insert from one of the larder clones, lambda FIBACT"' 5, was found to lack internal sites for SalI, Hi.ndIII, BamHI, and EcoRI. The cDNA insert was released from lambda FIBAC 5 DNA and the lambda arms digested by co-digesting with these four enzymes. The fragments were then ethanol-precipitated and ligated into the EcoRI site of plasmid pUC9 without further purification. There plasmids _were then used to transform E. coli strain JM83. Transforn~ants were selected on ampicillin containing plates. Plasmids from several trans-formants were purified and characterized on tr.e basis of the EcoRI digestion products. One was selected which had an insert co-migrating with the insert from lambda FIBAC 5 on agarose gel electrophoresis. This plasmid has been named pUC9-F5/237P10.
Mapping and Subclonin The insert in pUC9-F5/237P10 was mapped with respect to internal PstI site:>. Double digests with EcoRI and Pst demonstrated three internal Pstl recognition sites. The entire insert :"'~ , and the component. pieces were subcloned into M13 bacteriophage mpl9 and mpl8, respectively. Sequencing of the pieces was per-formed by the dic9eoxynucleotide method described by Sanger et al.
in Sanger, F., Nicklen, S., and Coulson, A.R., Proc. Natl. Acad.
Sci. USA 74:5463--5467 (1977), , The sequence of the DNA insert from pUC9-F5/237P10 showed an open rEaading frame which encodes the primary structure of a mature fibroblast collagenase inhibitor biologically equiva-lent to that isol.able from human skin fibroblasts. The salient features of the :sequence are:
(1) The insert is flanked by EcoRI restric-tion sites and by G/C and A/T homo-polymeric tracts consistent with the cloning methodology;
(2) The coding strand is presented in the 5' to 3' convention with poly C at the 5' en.d and poly A at the 3' end, again con-sistent with the techniques employed;
(3) If' the first G in the sequence GTTGTTG
i~runediately adjacent to the 3' end of the poly C tract is considered as nucleotide 1, then an open reading frame is presented which encodes the primary structures of the mature human fibroblast collagenase inhibitor beginning at nucleotide 34 and continuing through nucleotide 585;
(4) The termination codon TGA at nucleotides 586 through 588 defines the carboxy ter-minus of the translation product which is the same as that of the mature pro-tein;
(5) Nucleotides 1 through 33 define an amino acid sequence which is not found in the primary structure of the processed pro-tein, but: which is probably a portion of :.

_38_ 1 3 4 0 9 7 1 a leader. peptide characteristic of se-creted proteins;
(6) The three internal PstI sites have as their first base nucleotides 298, 327, and 448;
(7) There is a single recognition sequence for the restriction enzyme Tth111I
beginning at nucleotide 78; and (8) There is. a single recognition sequence for the restriction endonuclease NcoI
b,eginnin.g at nucleotide 227.
The sequence of nucleotides 1 through 703 and restriction site analysis are shovan.
# SITES FRAGMENTS FRAGMENTS ENDS
ACC 1 (GTVWAC) 1 214 495 (69.8) 214 709 214 (30.2) 1 214 ALU 1 (AGCT) 4 35B 358 (50.5) 1 358 363 124 (17.5) 482 606 48.2 119 (16.8) 363 482 6015 103 ( 14 . 5 ) 606 709 ( 0.7) 358 363 AVA 1 (CQCGPG) 1 53Ei 536 (75.6) 1 536 173 (24.4) 536 709 AVA 2 ( GGRCC ) 3 25i' 257 (36.2) 1 257 47f 220 (31.0) 257 477 572 137 (19.3) 572 709 95 (13.4) 477 572 BBV 1 (GCTGC) :l 269 440 (62.1) 269 709 269 (37.9) 1 269 .,~.:r<.
.~

# SI'T'ESFRAGM ENTS FRAGMENTS ENDS

BST N1(CCRGG) 3 344 344 (48.5) 1 344 544 200 ( 28. 344 544 2 ) 55;~ 152 (21.4) 557 709 13 ( 1.8) 544 557 DDE 1 (CTNAG) 4 lBEi 344 (48 365 709 . 5 ) 355 186 (26.2) 1 186 360 169 (23.8) 186 355 365 5 ( 0.7) 360 365 5 ( 0.7) 355 360 ECO R1(GAATTC) 1 69E3 698 ( 98 1 698 . 4 ) 11 ( 1.6) 698 709 FNU4H 1 (GCNGC)2 19E. 440 ( 62 269 709 . 1 ) 269 196 (27.6) 1 196 73 (10.3) 196 269 FOK 1 (GGATG) 4 192 274 (38.6) 435 709 204 192 (27.1) 1 192 303 132 (18.6) 303 435 435 99 (14.0) 204 303 12 ( 1.7) 192 204 HAE 2 (PGCGCQ) 1 368. 368 (51.9) 1 368 341 (48.1) 368 709 134pg71 # SI'.CESFRAGMENTS FRAGMENTS ENDS

HAE 3 (GGCC) 3 30 616 (86.9) 93 709 63 30 ( 4.7) 30 63 93 30 ( 4.2) 63 93 30 ( 4.2) 1 30 HGI A1 (GRGCRC) 55:! 552 (77.9) 1 552 157 (22.1) 552 709 HHA 1 (GCGC) 1 369 369 (52.0) 1 369 ~

340 (48.0) 369 709 HINC 2 (GTQPAC) 1 118 591 (83.4) 118 709 118 (16.6) 1 118 HINF 1 (GANTC):2 308 308 (43.4) 1 308 587 279 (39.4) 308 587 122 (17.2) 587 709 HPA 2 (CCGG) 4 207 224 (31.6) 372 596 372 207 (29.2) 1 207 596 165 (23.3) 207 372 654 58 ( 8.2) 596 654 55 ( 7.8) 654 709 HPH 1 (GGTGA) 2.

380 380 (53.6) 1 380 519 190 (26.8) 519 709 139 (19.6) 380 519 MBO 2 (GAAGA) 1 650 650 (91.7) 1 650 ' 59 ( 8.3) 650 709 ~~j ;
~u w. 134097 1 # SITES FRAGM ENTS FRAGMENTS ENDS

MNL (CCTC) 5 S:l 193 (27.2) 81 274 274 174 (24.5) 535 709 406 132 (18.6) 274 406 486 81 (11.4) 1 bl 5 3!i 80 ( 11 406 486 . 3 ) 49 ( 6.9) 486 535 MST (CCTNAGG)1 18!i 524 ( 73 185 709 ' . 9 ) 185 (26.1) 1 185 NCI (CCSGG) 2 37a 372 (52.5) 1 372 59!i 223 (31.5) 372 595 114 (16.1) 595 709 NCO (CCATGG) 1 ' 22',7 482 (68.0) 227 709 227 (32.0) 1 227 NSP B2 (CVGCWG) 1 197 512 (72.2) 197 709 197 (27.8) 1 197 PST 1 (CTGCAG) 3 298 298 (42.0) 1 298 32T 261 (36.8) 448 709 448 121 (17.1) 327 448 29 ( 4.1) 298 327 SAU 1 (CCTNAGG) 1 185 524 (73.9) 185 709 185 (26.1) 1 185 SAU 3A (GATC) :l , 150 559 (78.8) 150 709 ,r.' # SITES FRAGMENTS FRAGMENTS ENDS
150 (21.2) 1 150 SAU96 1 (GGNCC) 5 29 220 (31.0) 257 477 '92 165 (23.3) 92 257 257 137 (19.3) 572 709 477 95 (13.4) 477 572 572 63 ( 8.9) 29 92 29 ( 4.1) 1 29 SCR F1 (CCNGG)5 344 344 (48. 1 344 5 ) 3'72 172 (24. 372 544 3 ) 544 114 (16.1 595 709 ) 5li7 38 ( 5.4) 557 595 5!~5 28 ( 3.9) 344 372 13 ( 1.8) 544 557 SFA N1 (GATGC)1 1<.~3 516 (72.8) 193 709 193 (27.2) 1 193 (GACNNNGTC) 79 630 (88.9) 79 709 79 (11.1) 1 79 The following appear:
do not B.AL1 BAMH1 BCL 1 BGL 1 CfR 1 CLA1 ECO R5 FNUD

G;DI2 HAE1 HGA 1 HGI C1 Ni~R1 NDE1 NRU 1 NSP C1 GTTGTTGCTG TGGCTGATAG C:CCCAGCAGG GCCTGCACCT GTGTCCCACC CCACCCACAG
SH
AA
UE

ACGGCCTTCT GCAAT'TCCGA C:CTCGTCATC AGGGCCAAGT TCGTGGGGAC ACCAGAAGTC
H T M SH H
A T N AA I
E H L UE N' AACCAGACCA CCTTA'rACCA G'~CGTTATGAG ATCAAGATGA CCAAGATGTA TAAAGGGTTC
S
A
U
A

CAAGCCTTAG GGGATGCCGC T'GACATCCGG TTCGTCTACA CCCCCGCCAT GGAGAGTGTC
SD FS FN F H A N
AD OF ITS O P C C
UE KA 1:JP K A C O
11 11 :l2 1 2 1 1 TGCGGATACT TCCACAGGTC CCACAACCGC AGCGAGGAGT TTCTCATTGC TGGAAAACTG
A B M P
V B N S
A V L T

~~T~ .1 CAGGATGGAC TCTTGCACAT CACTACCTGC AGTTTCGTGG CTCCCTGGAA CAGCCTGAGC
F H P B D A D
O I S S D L D
K N T T E (1 E

TTAGCTCAGC GCCGGGGCTT CACCAAGACC TACACTGTTG GCTGTGAGGA ATGCACAGTG
A D HH N H M
L D AH C P N
U E EA I H L

TTTCCCTGTT TATCCATCCC CTGCAAACTG CAGAGTGGCA CTCATTGCTT GTGGACGGAC
F P A
O S V
K T A

CAGCTCCTCC AAGGCTCTGA AP~AGGGCTTC CAGTCCCGTC ACCTTGCCTG CCTGCCTCGG
A M H MA
L N P NV
U L H LA

GAGCCAGGGC TGTGCA~CCTG GC'.AGTCCCTG CGGTCCCAGA TAGCCTGAAT CCTGCCCGGA
B H B A H NH
S G S V I CP
T I 'T A N IA

1340 97 ~

GTGGAAGCTG AAGCCTGCAC AGTGTCCACC CTGTTCCCAC TCCCATCTTT CTTCCGGACA
A M H

U O A

ATGAAATAAA GAGTTA.CCAC CCAGCAAAAA AAAAAAGGAA TTC
E
C
O
i EXAMPLE 2-EXPRESSION OF COLLAGENESE INHIBITOR IN E. COLI
In this Example, a preferred method of coupling a pre-ferred portable DNA sequence to the 5' end of the cloned cDNA is set forth. This involv<as making a nucleolytic cleavage at a specified point within the coding sequence and reconstructing the desired portion of the (:oding sequence by means of synthetic oligonucleotides in a manner that allows its excision and recom-bination (i.e., by incorporating useful restriction sites).
Trimming the 5' end of the coding region will be accomplished by synthesizing both strands of the DNA extending from the Tth111I site in the 5' direction and ending in a BamHI
overhang. This synthet~~c oligonucleotide, referred to as FIBAC
A, has the following features:
(1) Codon selection has been biased toward those most frequently found in the genes of highly expressed bacterial. proteins;
(2) A methionine codon from which to initiate transla-tion has been provided immediately upstream from the cyste~ine which begins the coding region of, human processed FIBAC;
(3) The spacing of the BamHI site to the methionine codon is such that when cloned into pUCB, the coding region of FIBAC will be in-frame with the 5' end o~: the beta-galactosidase gene;
L.

(4) An in-frame stop codon and Shine Dalgarno sequence are also presented. Translation of this frame for t:he amino terminal portion of the beta-<~alactosidase is terminated at the TAA codon, and translation of FIBAC should be initiated at the following ATG;
(5) <:odons have been selected to create a HgiAI site beginning with the G in the FIBAC initiation c:odon; and (6) There is a PvuI site separated by one base from t:he 3' end of the BamHI sequence.
The structure of FIHAC A is GA TCC: GCG A'TC GGA GTG TAA GAA ATG TGC ACT
G. CGC T,AG CCT CAC ATT CTT TAC ACG TGA
TGC GTT CCG ~~CG CAT CCG CAG ACT GCT TTC
ACG CP,A GGC GGC GTA GGC GTC TGA CGA AAG
TGC AP,C TCT ~3AC C
ACG TT'G AGA CTG GA
FIHAC A is synthesized using the ABI DNA synthesizer (Foster City, California) as a series of four component oligonucleotides.
Component ol:igonucleotide FA1 is:
GATCC GCGAT CGGAG TGTAA GAAAT GTGCA CTTGC
Component ol:igonucleotide~FA2 is:
GGAACG CAAGT GCACA TTTCT TACAC TCCGA TCGCG
Component oligonucleotide FA3 is:
GTTC CGCCG CATCC GCAGA CTGCT TTCTG CAACT CTGAC C
Component ol3.gonucleotide FA4 is:
AGGTC AGAGT TGCAG AAAGC AGTCT GCGGA TGCGG C
The remainder of the coding portion of the FIBAC gene is isolated as the 3' Tth111I to EcoRI fragment generated by a double digest of pUC9-P'S/237P10 with these enzymes.
a .w A synthetic linker is made to couple the 3' end of the Tth111I to EcoRI fragment to a Sall site. These oligonucleotides will be designed to recreate the SalI site and destroy the EcoRI
site. The linker is comprised of the oligonucleotides linker A1 and linker A2.
Linker A1 is: AATTGGCAG
Linker A2 is: TCGACTGCC
These oligonuc:Ieotides and oligonucleotides FA1-FA4 are kinased separately and annealed in equal molar ratios with the Tth111I to EcoRI 3' end of the cDNA and BamHI/SalI cut mpl9RF
DNA. The ligated DNA is used to transfect JM105. Plaques are picked by their color in the presence of IPTG and X-gal and by hybridization to oligonucleotide FA2. Several positive plaques are to be sequenced. Tt~:ose containing the designed sequence are subcloned into BamHI/Sal.I digested pUCB. Translation of the FIBAC gene in this construct is coupled to translation initiated for beta-galactosidase. This expression vector is referred to as pUCB-Fic.
Coupling tran~olation of FIHAC to translation initiated for other highly expressed proteins is similarly arranged. For example, a portion of the OmpA gene which contains the Shine-Dalgarno and initiator methionine sequences has been synthesized.
This sequence encodes the entire signal peptide of OmpA protein and had convenient restriction sites, including those for EcoRI, EcoRV, Pvul, and StuI.
The sequence of the sene~e strand is:

GAATTCGATA TCTCGT'rGGA GF~TATTCATG ACGTATTTTG GATGATAACG AGGCGCAAAA
E T E F M H
C A C O N H
O Q O K L A

,1~ ~~ ,i 134097 ~

AATGAAAAAG ACAGCTATCG CGATCGCAGT GGCACTGGCT GGTTTCGCTA CCGTA
A NF PS
L RN V.A
U UU UU
1 12 lA

GCGCA GGCCTCTGGT AAAAGC'rT
H S H M HA
H T A N IL
A U E L NU

This sequence is hereinafter referred to as OmpA lead-er. Coupling the trans:Lation of FIBAC to OmpA is accomplished,t~y cutting the pUCB-Fic wii_h PvuI and SalI and isolating the coding region. This, together with the EcoRI to PvuI fragment isolated from OmpA leader, will tie cloned into EcoRI/SalI-cut pUCB. As in the prior example, transcription is driven by the lac promoter and regulated by the lac I gene product at the lac operator.
This FIBAC expression vector is referred to as pUCB-F/OmpAic.
To effect the translocation of FIBAC across the inner cell membrane, an appropriate leader sequence is added to the amino terminus of FIBAC. The protein thus produced will be translocated and processed to yield the mature form.
To effect such a translocation, a FIBAC gene encoding the signal peptide of the E. coli OmpA protein continuous with the structural region of FIBAC is created. This particular FIBAC
gene necessitates having' in frame stop codons at the 5' end of the FIBAC coding region changed. To accomplish this, the portion of the 5' coding region from pUCB-Fic that extends from the HgiAI
site to the NcoI cite is isolated. Upstream sequences are resynthesized as ,3 linker having cohesive ends from BamHI and HgiAI and containing an internal StuI site. This is synthesized as two oligonucleotides, linker B1 and linker 82.
Linker 131 is: GATCCCAGGCCTGCA.
Linker 132 is: GGCCTGG
r 134097 ~
Linker:: B1 and B2 are kinased separately and annealed in equal molar ratios with the HgiAI to NcoI fragment described above and BamHI/rfcoI cut pUCB-Fic. The resulting construct has the coding sequence of FIBAC in frame with the translation of the amino terminus of beta-galactosidase. Translation of this se-quence forms a fusion protein with FIBAC. This plasmid is referred to as pUC8-Ff.
Attaching the OmpA leader sequences to the coding re-gion of FIBAC is accomp:Lished by ligating EcoRI/StuI cut pUCB-Ff with an excess of the purified EcoRI to StuI fragment of OmpA
leader. Following transformation, plasmids from several colonies will be characterized by hybridization. Those that have incorpo-rated the OmpA leader fragment are characterized further to veri-fy the structure. This plasmid, pUCB-F OmpAl, will direct the synthesis of a fusion protein beginning in the signal peptide o'f the E. coli OmpA protein and ending in human FIBAC. The signals present in the OmpA portion of the protein effect the protein's export from the cytopla~~m and appropriate cleavage from the pri-mary structure of FIBAC.
If the efficiency of expression were to be compromised by the sequence of the leader peptide or its combination with FIBAC either at t',he protein or at the nucleic acid level, the gene could be altered to encode any of several known E. coli leader sequences.
Transcription of all of the genes discussed is effected by the lac promoter. As in the case of initiation sites for translation, the promoter and operator region of the gene may be interchanged. FIBAC may also be expressed from vectors incor-porating the lambda PL promoter and operator (OL), and the hybrid promoter operator,, Tac as described in Amann, E., Brosius, J., and Ptashne, M. Gene _25:167-178 (1983).
Excision of those portions of the gene including ribosome binding site structural region and 3' non-translated sequences and insertion in alternate vectors contain-ing the PL or Tac promoter makes use of the unique restriction sites that flank these structures in pUCB-F/OmpAic and pUCB-F/OmpAl. In:;ertion of the EcoRI to SalI fragment from I
. ..., -5~- ~ 3 4 ~ 9 7 either into similarly digested plasmid pDP8 effects transcription of these genes directed by i:he lambda PI, promoter. Transcriptional regulation would k>e temp<~rature sensitive by merit of the cI857 mutation harbored on thia same plasmid.
Putting simila~_ gene fragments into the transcription unit of the Tac promoter will be accomplished by first isolation the EcoRv to SalI fragment. This, together with the synthetic Tac promoter sequence which is flanked by BamHI and PvuII sites and which contains the lac operator will be inserted into the BamHI to SalI sites of pBR322 or preferably der=ivatives. The derivatives in this case refer to constructs containing either the lacI gene or the Iq gene.
Expression of FIBAC in host microorganisms other than Escherichia is considered. Yeast and bacteria of the genera Bacillus, Pseudomonas, and Clostridium may each offer particular advantages.
The processes outlined above could easily be adapted to others.
In general, expression vectors for any microorganism will embody features analogous to those which we have incorporated in the above mentioned vectors of E. coli. In some cases, it will be possible to simple move t:he specific gene constructs discussed above directly into a vector compatible with the new host. In others, it may be necessary or desirable to alter certain operational or structural elements of the gene.

The human collagenase inhibitor may be readily purified after expression in a variety of microbes. In each case, the spectrum of contaminant proteins will differ. Thus, appropriate purification steps will be selected from a variety of steps already known to give a good separation of the human collagenase inhibitor from other proteins and from other procedures which are likely to work.
If the inhibitor is not secreted from the microbes, it may form inclusion bodies inside the recombinant microbes. These bodies are separated from other proteins by differential centrifugation after disruption of the cells with a French Press. The insoluble inclusion bodies are solubil.ized in 6 M guanidine W.

hydrochloride or 8 M urea, and the inhibitor protein is more com-pletely solubilized by reaction of its cysteines with sodium sulfite. At any time subsequent to this step, the cysteines are converted back to their reduced form with dithiothreitol. Once the inhibitor protein is solubilized from inclusion bodies, immunoaffinity chromatography using antibodies raised against the unfolded inhibitor are used for purification before refolding.
The inhibitor can be refolded according to the protocol mentioned in Example 6, infra. After refolding of the inhibitor, or if the inhibitor is :secreted from the microbes, purification from other proteins is accomplished by a variety of methods.
Initial steps include ul.trafiltration through a SO K dalton cut-off membrane or ammonium sulfate fractionation. Other useful methods include, but are not limited to, ion-exchange chromatography, gel filtration, heparin-sepharose chromatography, reversed-phase chromatography, or zinc-chelate chromatography.
All of these step;s have been successfully used in purification protocols. Addit:ional.high resolution steps include hydrophobic interaction chrom~3tography or immunoaffinity chromatography.
After purification, the metalloproteinase inhibitor is preferably at least 90-95$ pure.
FY~MD1-F d Purification of Human Collagenase Inhibitor from Human Amniotic Fluid Human annniotic-fluid obtained from discarded amniocentesis samples was pooled and 6 liters were subjected to ultrafiltration through a 100 kD MW cutoff filter, obtained from Millipore Corporation, in a Millipore Pellicon Cassette System.
The eluate was concentrated through a 10 kD cutoff filter, TM
obtained from Millipore Corporation, then through an Amicon PM-10 membrane. Aliquots (10 ml) of concentrated amniotic fluid were eluted through a 2.5 x 100 cm column of Ultrogel AcA54, obtained from LKB Corporation, which was equilibrated with pH 7.6, 0.05 M
hepes, 1 M sodium chloride, 0.01 M calcium chloride, and 0.02$
sodium azide (all chemicals were obtained from~Sigma Chemical Company). Fractions containing the inhibitor were collected and pooled, dialyzed against pH 7.5, 0.025 M Hepes buffer containing 0.01 M calcium chloride and 0.02% sodium azide, and loaded onto a TM
1.5 x 28 cm heparin-Sepharose CL-6B (obtained from Pharmacia, Inc.) column equilibrated with the same buffer. This column was rinsed with 1 liter of ~:he above buffer and eluted with a linear gradient of 0-0.3 M sodium chloride. The fractions from the largest peak of inhibitor activity, eluting at about 0.1-0.15 M
sodium chloride, were pooled, concentrated to 1 ml, and loaded rM
onto a Synchropak rp-8 reverse phase HPLC column equilibrated with 0.05% trifluoroacet_ic acid (Aldrich Chemical Company). The column was eluted with a linear gradient of 0-40% acetonitrile (J. T. Baker Chemical Company) at 1/2% per minute. All fractions were immediately dried in a Savant speed-vac concentrator to re-move acetonitrile, and redissolved in pH 7.5, 0.1 M Hepes before assay. The inhibitor eluted between 32-38% acetonitrile. Frac-tions containing the inhibMtor were pooled, and 100 ul aliquots~
were eluted over a Bio-rad biosil-TSK 250 HPLC gel filtration column. The pooled peaka of inhibitor activity contained 0.1 mg of inhibitor, which was over 95% pure as judged by SDS- poly-acrylamide gel electrophoresis.
FY~MDT.F S
Purification of Human Fibroblast Collagenase Inhibitor from Human Embryonic Skin Fibroblast Serum-Free Medium Human embryonic skin fibroblasts were grown in serum-free tissue culture medium. Ten liters of this medium were col-lected, dialyzed <igainst pH 7.5, 0.02 M hepes buffer containing 0.02% sodium azide and 0.01 M calcium chloride, and applied to a 2.8 x 48 cm column of heparin-sepharose CL-6B (Pharmacia, Ine.) equilibrated with the same buffer. The column was rinsed with 2 liters of this bu1'fer and was then eluted with linear gradient of 0-0.3 M sodium chJ.oride contained in this buffer. The fractions obtained were tested for the presence of inhibitor by their abil-ity to inhibit human fibroblast collagenase. The fractions cor-responding to the peak of activity were those obtained near 0.15 M sodium chloride. These fractions were concentrated to about 5 ml by ultrafiltrat:ion through an Amicon YM10 filter and the con-centrate was applied in four separate runs to a 250 x 4.1 mm Synchropak rp-8 reverse phase HPLC column, equilibrated with 1% ' s _~ 'S:..

134097 ~

trifluoroacetic acid. The column was eluted with a 0-60% linear gradient of acetonitrile in 0.1% trif:luoroacetic acid. The gradient was run at 1/2o acetontrile per minute. The inhibitor eluted in two sharp peaks between 26-29% acetonitrile. All fractions were immediately dried in a Savant speed-vac concentrator, redissolved in pH 7.5, 0.1 M Hepes, and assayed. At least 1.2 mg of: collagenase inhibitor was recovered, which was 90-95% pure. This material gives single band when run on a 17.50 reducing SDS gel. After cax-boxymethylation of the cysteines and elution through the same RP-8 column under identical conditions, the inhibitor is suitably homogenous for protein sequencing.

It is contemplated that the human collagenase inhibitor can be readily refolded into its native structure from its denatured state after expression of its gene in a microbe and separation of the collagenase inhibitor from most of the other proteins produced by the microbe. By analogy to the conditions necessary for the refolding of other disulfide-containing proteins as set forth by Freedman, R. B. and Hillson, D. A., in "Formation of Disulfide Bonds," in: The Enzymology of Post-Translational N.odificat-ion of Proteins, Vol. 1, R. B. Freedman and H. C. Hawkins, eds., pp. 158-207 (1980), refolding of the human collagenase inhibitor should occur in solutions with a pH of 8.0 or greater. At this pH, the c;~steines of the protein are partially ionized, and this condition is neces:~ary for the attainment of native disulfide bond pairings. 'I'he inhibitor concentration should be relatively low, less than 0.1 mg/ml, to min_~mize the formation of intermolecular disulfide-linked aggregates which will interfere with the refolding process.
Since th~~ stabil.ity of the refolded (native) disulfide bonded structure relative to the enfolded (reduced) structure depends on both the solution oxidation-reduction potential and the concentrations of other redox-active molecule,, it is conter~lated that the redox potential should be buffered with a rc=_dox buffer giving a potential equivalent to a reduced: oxidized c~lutathione ratio of 10. The preferred concentration range of reduced glutathione would be 0.1-1.0 mM. At higher concentrations, mixed disulfides will form with protein, reducing t:he yield of the refolded (native) structure. The relative stabilities of the unfolded protein and the native structure, and thus the rate and yie:Ld of refolding, will also depend on other solution variables, such as the pH, temperature, type of hydrogen-ion buffer, ionic strength, and the presence or absence of particular anions or cations as discussed in Privalov, P. L., "Stability of Proteins, Small Globular 1?roteins," in Advances in Protein Chemistry, Vol. 33, pp. 167-2 .6, (19'79) .
These conditions vary for every protein and can be determined experimentally.. It is contemplated that addition of any molecule that strongly prefers to bind the native (as opposed to the unfolded) structure, and which can be readily separated afterwards from the native (refolded) protein, will increase not only the yield but the rate of re-folding. These molecules include monoclonal antibodies raised against the native structure, and other proteins which tightly bind the native collagenase inhibitor, such as the mamalian enzymes collagenase or gelatinase.
Example 7 The second preferred sequence as set forth herein, i.e., GGCCATCGCC GCAGATCC'AG CGCC:CAGAGA GACACCAGAG AACCCACCAT GGCCCCCTTT
H F XB HH N S
A N HI AH C A
E U ON EA O U

GACCCCTGGC TTCTGCAT'CC TGT7.'GTTGCT GTGGCTGATA GCCCCAGCAG GGCCTGCACC
B SF S H
S FO A A
T AK U E

i...
A

1 34n g7 ~

TGTGTCCCAC CCCACCCACA GA.CGGCCTTC TGCAATTCCG ACCTCGTCAT CAGGGCCAAG
H T M SH
A T N AA
E H L UE

TTCGTGGGGA CACCAGAAGT CAACCAGACC ACCTTATACC AGCGTTATGA GATCAAGATG
H S
I A
N U

ACCAAGATGT ATAAAGCiGTT CC.AAGCCTTA GGGGATGCCG CTGACATCCG GTTCGTCTAC
SD FS FN F H A
AD OF NS O P C
UE KA UP K A C

ACCCCCGCCA TGGAGAGTGT CT~~CGGATAC TTCCACAGGT CCCACAACCG CAGCGAGGAG
N A B M
C V B N

TTTCTCATTG CTGGAAAACT GC,AGGATGGA CTCTTGCACA TCACTACCTG CAGTTTCGTG
P F H P
S O I S
T K N T
1 1 1 ~ 1 ~.~~'7 -56- ~ 3 4 0 9 7 ~

GCTCCCTGGA AC
B
S
T

has the following restriction sites:
SITES FRAGMENTS FRAGMENTS ENDS
ACC 1 (GTVWAC) :l 295 295 (68.3) 1 295 137 (31.7) 295 432 AVA 2 (GGRCC) 1.
338 338 (78.2) 1 338 94 (21.8) 338 432 BBV 1 (GCTGC) 350 350 (81.0) 1 350 82 (19.0) 350 432 BIN 1 (GGATC) 14 418 (96.8) 14 432 14 ( 3.2) 1 14 BST N1 (CCRGG) 65 360 (83.3) 65 425 425 65 (15.0) 1 65 7 ( 1.6) 425 432 DDE 1 (CTNAG) SITES FRAGMENTS FRAGMENTS ENDS
267 267 (61.8) 1 267 165 (38.2) 267 432 FNU4H 1 (GCNGC) 8 269 (62.3) 8 277 277 82 (19.0) 350 432 350 73 (16.9) 277 350 8 ( 1.9) 1 8 FOK 1 (GGATG) c4 76 197 (45.6) 76 273 273 99 (22.9) 285 384 285 76 (17.6) 1 76 384 48 (11.1) 384 432 12 ( 2.8) 273 285 HAE 2 (PGCGCQ) 19 413 (95.6) 19 432 19 ( 4.4) 1 19 HAE 3 (GGCC) 1 258 (59.7) 174 432 51 60 (13.9) 51 111 111 50 (11.6) 1 51 144 33 ( 7.6) 111 144 174 30 ( 6.9) 144 174 1 ( 0.2) 1 1 HHA 1 (GCGC) 20 412 (95.4) 20 432 20 ( 4.6) 1 20 , _:.s~~.

~ 34~ 97 ~

SITES FRAGMENTS FRAGMENTS ENDS
HINC 2 (GTQPAC) :L
199 233 (53.9) 199 432 199 (46.1) 1 199 HINF 1 (GANTC) 1.
389 389 (90.0) 1 389 43 (10.0) 389 432 HPA 2 (CCGG) 288 288 (66.7) 1 288 144 (33.3) 288 432 MNL 1 (CCTC) 162 193 (44.7) 162 355 355 162 (37.5) 1 162 77 (17.8) 355 432 MST 2 (CCTNAGG) 266 266 (61.6) 1 266 166 (38.4) 266 432 NCO 1 (CCATGG) 47 261 (60.4) 47 308 308 124 (28.7) 308 432 47 (10.9) 1 47 NSP B2 (CVGCWG) 278 278 (64.4) 1 278 154 (35.6) 278 432 1340 97 ~

# SITES FRAGMENTS FRAGMENTS ENDS
PST 1 (CTGCAG) 3751 379 (87.7) 1 379 40E1 29 ( 6.7) 379 408 ~4 ( 5.6) 408 432 SAU 1 (CCTNAGG) 266~ 266 (61.6) 1 266 166 (38.4) 266 432 SAU 3A (GATC) :2 14 217 (50.2) 14 231 231 201 (46.5) 231 432 14 ( 3.2) 1 14 SAU96 1 (GGNCC) S1 165 (38.2) 173 338 110 94 (21.8) 338 432 173 63 (14.6) 110 173 338 59 (13.7) 51 110 51 (11.8) 1 51 SCR F1 (CCNGG) 65 360 (83.3) 65 425 425 65 (15.0) 1 65 7 ( 1.6) 425 432 SFA N1 (GATGC) 75 199 (46.1) 75 274 274 158 (36.6) 274 432 75 (17.4) 1 75 -6a- ~ 3 4 0 9 7 ~
# S:LTES FRAGMENTS FRAGMENTS ENDS
STY 1 (CCRRGG) X47 261 (60.4) 47 308 308 124 (28.7) 308 432 47 (10.9) 1 47 TTH111 1 (GACNNNGTC) 1E~0 272 (63.0) 160 432 160 (37.0) 1 160 XHO 2 (PGATCQ) 13 419 (97.0) 13 432 ., 13 ( 3.0) 1 13 The following not do appear:

nSSH BST E2 CFR 1 CLA 1 _....y r P.

The salient features of this cDNA are:
1. The coding strand is presented in the 5' to 3' convention with the polyC tract at the 5' sand.
2. If the :First G in the sequence GGC CAT
CGC CGC is considered as nucleotide 1, then an open reading frame exists from nucleotide 1 through nucleotide 432, which is the 3' end of this partial cDNA.
3. The fir~~t methionine in this reading frame is encoded by nucleotides 49 through 51 and represents the initiation site of translation.
4. The amino acid sequence prescribed by nucleotides 49 through 114 is not found i;z the primary structure of the mature protein, but it is the sequence of the leader peptide of human protein.
5. Tile sequence of nucleotides 82 through 432 is identical to the sequence of nucleotides numbered 1 through 351 in the insert from the first preferred se-quence of Example 1.
6. The amino .acid sequence of the mature protein displays two consensus sequences for sugar attachment. These sequences, -N-Q-T- prescribed by nucleotides 202 through 210 and -N-R-S- prescribed by nucleotides 346 through 354, are amino acid residues 30 through 32 and 78 through 80, respectively, in the mature protein. Both sites are glycosylated in the human inhibitor protein.
FYLMD1.F Q
A series of expression vectors have been constructed which direct the transcription and translation of the FIBAC gene in E.~coli.

134pg71 A. Expression Vector pFib51 The first of these constructions is a derivative of plasmid pUCB which contains the coding region of the human fibroblast collagenase inhibitor ("FIBAC") gene arranged in such a way as to allow its transcription to be directed and regulated by the lac promoter and opc>rator. This, construct, pFib5l, was made with minor modification of the procedure outlined in Example Two for the assembly of expression vector pUCB-Fic.
Trimming of t:he S' end of the coding region was effected by isolating that piece of DNA extending from the HaeIII
site at nucleotid a 93 through the 3' EcoRI site beginning at nucleotide 698. Reconstruction of the 5' end was accomplished as described except that oligonucleotides FA3 and FA4 were each lengthened by 12 bases to extend the reconstruction to the HaeIII
recognition site. Hence, the structure of FIBAC A' was create:
GA Tt:C GCG ATC GGA GTG TAA GAA
G CGC TAG CCT CAC ATT CTT

TAC AC:G TGA ACG CAA GGC GGC GTA
CCG CP~G ACT GCT TTC TGC AAC TCT
GGC GTC TGA CGA AAG ACG TTG AGA
GAC CTG GTG ATC AGG G
CTG GA.C CAC TAG TCC C
The salient features of FIBAC A' have remained as described in Example 2.
A synthetic linker was made to couple the 3' end of the HaeIII to EcoRI fragment to a SalI site. These oligonucleotides were designed to recreai:e the SalI site and to destroy their EcoRI site. In addition, the linkers were modified from the original description to include an internal K~nI site. The new linker is comprised of t:he oligonucleotides "modified-linker A1"
and "modified-linker A2,."
Modified-linker A1 is: AATTGGTACCAG
Modified-linker A2 is: TCGACTGGTACC
r Ligation into M13 mpl9, cloning, and selection were es-sentially as described in previous Examples. The coding region of FIBAC was removect from a clone with the designed sequence by di-gestion with BamFiI and HindIII and subcloned into these restric-tion sites in pUC'8. The resulting plasmid is pFib5l. In this ~lasmid, the transcription of the FIBAC gene is directed by the lac promoter. Translation of methionyl FIBAC is coupled to translation initiated for beta-galactosidase.
B. Exyressioia Vector pFib55 Creation of the Hc~iAI restriction site at the 5' end of the mature FIBAC coding sequence would allow the entire FIBAC
coding sequence to be portable via a H~CiAI/SalI, KpnI, or HindIII
double digestion in plasmid pFib5l, except for another H~iAI site at position 552 within the coding sequence. This restriction sitetwas removed using in vitro oligonucleotide-directed, site-' specific mutagenesis essentially as described by Zoller and Smith in Methods in Enzymology 100:468.
Single-stranded DNA isolated from a derivative of bacteriophage mpl8 containing the met-FIBAC gene translationally coupled to lacZ as described above was annealed to the synthetic oligonucleotide GGGCTTTGCACCTGGCAG. This oligonucleotide anneals to the FIBAC coding region across the H~CiAI site with a single mismatch. The resultant. DNA was then incubated with the Klenow fragment of E. coli DNA polymerase, T4 DNA ligase and a mixture of all four deoxy-nucleotidetriphosphates.
The cov,alently-closed, double-stranded phage DNA thus obtained was used to tra.nsfect E. coli strain JM107. Plaques were assayed for the presence of the mutant sequence by their hy-bridization to the mutagenic oligonucleotide shown at~ove at 58°C
in 6x SSC.
The selected clone had the leu codon CTT replacing CTG
at amino acid position 173. The coding region from this clone was excised as a l3amHI to HindIII fragment and ligated into a similarly-digested pUC8. The resulting plasmid, pFib55, has all of the features o:E pFib51 as well as a more easily mobilized FIBAC coding region.
...

C. Expression Vector pFib56 Alternate methods of translational coupling to beta-galactosidase or any E., coli protein can be similarly con-structed. For one embodiment, a clone has been designed which is similar from a regulatp.on of expression point of view to plasmid pFib51 (i.e., FIBAC translationally coupled to lacZ in pUCB), but which uses an alternate translational coupler. To create pFib56, the following pieces of: DNA were synthesized:
EcoRI end CAI end ' A A T T C C .)1 A G G. A G A A A T A A A T G T G C A 3 ' H~iAI end EcoRI end 5' C A T T T A 't T T C' T C C T T G G 3' This double-stranded EcoRI/Hc~iAI fragment was then combined with the ~AI/HindII:L FIBAC coding sequence and plasmid pUCB that had been digested wit=h EcoRI and HindIII and ligated. The resulting plasmid has been called pFib56. When this plasmid is transformed into E. coli strain JM107, the strain, JM107/pFib56, can be in-duced to express even more methionyl FIBAC than JM107/pFib51 or JM107/pFib55. From this, it has been concluded that the transla-tional coupler in pFib5~6 is more efficient than that in pFib5l.
D. ~>ressio:n Vectors pFiblO and pFibll To direct the expressed FIBAC to the periplasmic space of E. coli, a leader peptide is added to the amino terminus of FIBAC. The leader peptide will effect the transport of the fu-sion protein out of the inner cell membrane. Cellular processing removes the signal peptide and yields the mature form of FIBAC.
Two signal sequences have been separately fused to FIBAC for this purpose. They are the leader peptides of _E. coli ompA and phoS gene products. Both om~AL-FIBAC and phoSL-FIBAC
fusion proteins contain signals to direct them to be located in the periplasm of E. colil and to allow proteolytic processing of the fusion to ompA or phoS leader fragments and native FIBAC.
Plasmid pFiblO is a derivative of pUCB into which the coding sequence for the om~AL-FIBAC fusion protein has been in-serted. In addition, some 5' nontranslated sequences from the ompA gene are included. Transcription of this plasmid is e:~.

directed by the lac promoter/operator of pUCB. Translation is initiated at the methionine codon beginning the ompA leader se-quence and uses. the Slzine-Dalgarno sequence found in the ompA
gene.
The plasmid was constructed by liga,ting the FIBAC
coding sequence contained on a HgiAI/HindI~I fragment of pFit~55 together with synthetic oligonucleotides encoding the ompA leader peptide and with EcoR:L/HindIII-digested pUCB. The coding strand of the synthesized oligonucleotide is:
o-~-'~i.
EcoRI end 5' A A T T C G A T A T C T C G T T G G A G A T A T T C A T G A C
G T A T T T T G G Fv T G A T A A C G A G G C G C A A A A A A T
PvuI
G A A A A A G A C A, G C T A T C G C G A T C G C A G T G G C A
C T G G C T G G T T' T C G C T A C C G T A G C G C A G G C C T
G C A 3' ~iAI end The s;rnthesis was accomplished as four oligonucleo-tides, two for each of the strands. Together, the double-stranded DNA features:
(a) An EcoRI cohesive end at the 5' end and a HgiAI
cohesive end at the 3' end of the coding strands;
(b) An open reading frame encoding the leader peptide of the ompA gene product which is in frame with t:he FIB.~C gene when ligated at the HgiAI site;
(c) Dfon-coding sequences 5' to the translated portion which contain the ribosome binding site normally found in the ompA gene; and (d) ~l,n inte:rnal, unique PvuI site.
Plasmi.d pFib:ll is a derivative of plasmid pKK223-3 and is identical to pFiblO except that its pUC8 portion has been re-placed by the 4550 by EcoRI/HindIII fragment of plasmid pKK223-3.
In this construct, transcription is directed and regulated by the hybrid tac promo~ter/ope rator. Additionally, this plasmid con-tains a transcriptiona:l terminator which may stabilize the , plasmid in a high expression system.
> _-.

~~~097' E. Expression Vector pFibl3 In plasmid pfibl3, the 5' non-coding region of the o_mpAL sequence has been eliminated and the om~AL-FIBAC fusion gene is translationallh coupled directly to the N-terminal por-tion of the lacZ gene as it appears in plasmid pUCB. This has been accomplished by li.gating the.EcoRI/PvuI (3200 bp) fragment of plasmid pFiblO to a synthetic oligonucleotide shown here.
EcoRI end 5' A A T T C C A A G G A G A A A T A A A T G A A A A A G A C A G C
T A T C G C G A T 3' PvuI end PvuI end 5' C G C G A T A G C T G T C T T T T T C A T T T A T T T C T C C T
T G G 3' EcoRI end F. ~~ression Vector pFib31 The ~hoS gene of E. coli codes for the phosphate bind-ing protein and has been described and sequenced by Surin, B.D.
et al., in J. Bacteriol. 157:772-778 (1984), This protein i.s periplasmic, with a 25-amino-acid leader sequence that directs it to that compart-ment. The leader sequence is proteolytically removed during the translation process to leave only mature phoS protein in the periplasm. We have synthesized the phoS leader sequence as two double-stranded DNA fragments with EcoRI and CAI ends as shown:
EcoRI/ClaI
EcoRI end 5' A A T T C A T G A A A G T T A T G C G T A C C A C C G T C G C
A A C T G T T G T C G C C G C G A C C T T A T 3' ClaI end ClaI end 5' C G A T A A G G T C G C G G C G A C A A C A G T T G C G A C G
G T G G T A C G C A ~,~ A A C T T T C A T G 3' EcoRI end -67- ~ 34~ 97 1 ClaI/HgiAI
ClaI end HgiAI end 5' C G A T G A G T G C 'T T T C T C T G T G T T T G C G T G C A 3' HgiAI end ClaAI.e.nd 5' C G C A A A C A C A G A G A A A G C A C T C A T 3' These fragments were ligated together at a ClaI site internal to the phoS leader. These fragments were simultaneously combined with the HgiAI/HindIII 1~IBAC coding fragment described above and plasmid pKK223-3 that had been digested with EcoRI and HindIII.
The resulting plasmid has been called pFib3l.
RXAMDf.R 4 EXPRESSION OF THE FIBAC GENE IN E. COLI
Three methods have been employed to qualitatively de-.
termine the amount and !'orm of FIBAC produced by _E. coli cells harboring the plasmids described above. They are (1) specific reaction of FIBAC antibody to _E. coli proteins produced following induction of the FIBAC
gene resolved by polyacrylamide-SDS-el~~ctrophoresis and subsequently bound to ni~~rocellulose paper (western blotting);
(2) labeling of E. coli proteins with 35S-cysteine, 35;x-methionine, or 35S04 following induction of the FIBAC gene; and (3) inspection of polyacrylamide SDS gels containing E. coli proteins and FIBAC without antibody cou-pling or radioactive-labeling.
These mEathods have allowed not only comparison of the amounts of FIHAC produced by each strain, but also purification of the FIHAC folly>wing expression without the need for functional metalloproteinase inhibition. All of the plasmids discussed in Example 8 have been expressed in the background of _E. coli strain JM107. The expre~;sion of FIBAC in E. coli has so far been greater in those systems designed to transport the protein out-side of the inner cell membrane. This is possibly due to degra-dation of the expressed ;protein in the cytoplasm.
~,: '( ..

Processing om.~aL-FIBAC fusion protein to yield the ma-ture form of FIBAC has been found to be partially dependent on the phase of growth of the cells. Cells induced with IPTG in early log phase of growth accumulate a mixture of processed and unprocessed FIBA~" while cell cultures induced in late log phase accunn~late only processed PIBAC. Strains expressing the phoSL-FIBAC fusion appear to process the protein completely independent of growth phase.
All of the expression vectors described above in Exam-ple 8 have ampic:illin resistance as the selectable marker. For the purposes of production, it might be preferable to have a tetracycline resistance marker. One plasmid generally useful as an expression vector has been constructed with a tetracycline re-sistance marker. This plasmid is a derivative of pKK223-3 in which the truncated tetracycline resistance gene has been re- '"
placed with fully functional tetr gene adapted from pBR322.

PURIFJ:CATION OF FIBAC EXPRESSED IN E. COLI
The recombinant human co.lleagenase inhibitor (FIBAC) has been purified! from E. coli strain JM107 transformed with plasmid pFibll. In this strain, JM107/pFibll, FIBAC can be made to accumulate as an insoluble aggregate. In this example, the conditions for growth o:E the cells, induction of FIBAC gene ex-pression, cell harvesting, and purification of FIBAC from the in-soluble fraction of a total cell lysate are described. The same protocol may also be used to substantially enrich Fibac from the soluble fraction of total cell lysate.
A. Insoluble Fraction Luria broth containing ampicillin at a concentration of 100 ug/ml was innoculatc~d with an overnight culture of JM107/pFibll to an initp.al OD600 of 0.15-0.20. The shake flask cell culture was allowed to grow at 37°C to an OD600 of 1.5, at which time the culture medium was supplemented with IPTG to a final concentration of 0.5 mM. Incubation at 37°C was then con-tinued for 2.5 to 3 hours. The cell culture was rapidly cooled to 4°C in an ice bath and the cells harvested by centrifugation.
A one-liter culture growrn as described above yielded on the . 9 _.

average 3 gr of cells (wet weight). The cells were washed once in cold lysis buffer (50 mM MES, pH 6.0, 4 mM EDTA) and then resuspended in th.e lysi;s buffer to a final concentration of 0.26 g cells (wet weight)/ml. The cell suspension was frozen at -70°C
until further processinc3.
Total ceell lysate was prepared by passing the cell sus-TM
pension two times throuc3h a French pressure cell (SLM-Aminco model #FA-079 fitted wii:.h piston #FA-073 and operated at 20,000 psti, SLM Instruments, 'Inc., Urbana, Illinois). The resultant cell lysate was incubatE~d with 10 ug of DNase I/gr of cells (wet weight) for 2-3 hours on ice. The cell lysate was then divided into small aliquots and stored frozen at -70°C until further pro-cessing.
A cell lysate supernatant and cell lysate pellet frac-tion were obtained by centrifugation of a 5 ml portion of cell~~
lysate for 30 minutes ai: 4°C in an Eppendorf micro centrifuge.
The resultant pellet was washed twice with 3 ml of 50 mM
Tris-HC1, pH 8.0, 4 mM EDTA, 50 mM DTT. The supernatants from these washes were poolec! and saved for further analysis. The washed pellet was then solubilized by resuspension in 3 ml of 50 mM MES, pH 6.0, 4 mM EDTA, 50 mM DTT, 10 M urea, and incubated at room temperature for 15 minutes. Should protein carbamylation occur due to urea solubi.lization the side reaction could be quenched by the addition of a one hundred fold excess of a suit-able nucleophile over total protein amino groups. The resultant solution was clarified by centrifugation for 15 minutes at 4°C in an Eppendorf micro centrifuge. The supernatant contained essen-tially all of the protein from the solubilization procedure and was saved for further analysis. The remaining small pellet con-sisted mostly of cell wall debris and few proteins (none of which were FIBAC). This was discarded.
The identification of FIBAC in the various fractions prepared as described above was accomplished by SDS-PAGE and probing of western blots with anti-FIBAC antibodies. SDS-gel analysis of total cell l.ysate protein obtained from IPTG-induced JM107/pFibll shows the presence of a protein band of approxi-mately 20,000 dalton apparent molecular mass that is absent in ' 134087 ~
_70_ the gel pattern From a cell Lysate of non-included JM107/pFibll.
The 20,000 Da protein and a faster migrating band, presumably a degradation product of FIBAC, react with anti-FIBAC antibodies in the western blot analysis. The IPTG induction-dependent presence of this protein, the molecular weight, and the reactivity with anti-FIBAC antibodies suggest that the protein represents the ex-pressed recombinant FIB,AC. Analysis of the cell lysate supernatant fraction obtained from IPTG-induced JM107/pFibll and the cell lysate pellet wash revealed little FIBAC. The bulk of the FIBAC was found, however, in the urea-DTT solubilized cell lysate pellet fraction. This was interpreted to mean that, upon IPTG induction, the FIBAC accumulates an insoluble fraction and can be isolated in a substantially purified form .from the washed cell lysate pellet.
The urea-DTT solubilized cell lysate pellet was used~as the starting material for the CM-chromatography. A 1.2 ml ali-quot of solubilized cel.'L lysate pellet (16 mg protein) was di-luted to 25 ml with cold CM-buffer (50 mM MES, pH 6.0, 6 M urea, 14 mM 2-ME). The samplsa was then applied to a carboxymethyl cel-lulose column (25x130 mnn) previously equilibrated with CM-buffer at 4°C. After sample application, the column was washed with CM-buffer until the A28C1 returned to baseline. Adsorbed protein was eluted with a linear sodium chloride gradient (0-200 mM) in CM-buffer. Total gradient volume was 400 ml, flow rate was 26 ml/hr, and 5 ml fractions were collected. The "flow-through"
fraction (CM-FT) and ths: peak fractions 58-61 were pooled and analyzed by SDS-PAGE. 9fie electrophoretic and immunological analysis of these fractions revealed that the recombinant FIBAC, including some degraded FIBAC, eluted at approximately 120 mM
NaCl without any other detectable proteins. Under the chromatographic conditions employed, most non-FIBAC proteins did not adsorb to the CM-column and were found in the "flow-through"
fraction.
The amount of CM-purified FIBAC obtained in this proce-dure was estimates to represent 1.3% of the total cell protein.
Bradford protein assays were used for quantitation throughout the isolation and purification procedure.

Two ml (100 ug) of purified FIBAC in 50 mM MES, 6 M
urea, 14 mM 2-mercaptoethanol were concentrated to 200 ul by Centricon centrifugation and adjusted to pH 8.5 by addition of 2 M Tris-HC1, pH 8.5 to a final Tris concentration of 0.5 M. The cysteine residuee~ of FIBAC were then carboxymethylated using 3H-iodoacetic acid. The alkylation reaction mixture was desalted by reverse phase HPLC. The modified FIBAC eluted at 30$
acetonitrile and was collected for further analysis. An aliquot of the modified F'IBAC isolated from the HPLC was subjected to SDS-PAGE analysis.. Comparison of the CM-purified FIBAC that was used for the cart~oxymetlzyla'tion (starting material) and the FIBAC
after alkylation and HPLC desalting showed that the modified FIBAC migrated slightly slower in the SDS gel than the non-modified FIBAC. This is not an unusual observation, particularly in view of the substantial number of modified cysteines present, in FIBAC.
The carboxymethylated FIBAC was then applied to an Applied Biosystems (model 470A) gas-phase protein sequencer (Fos-ter City, CA) for automatic Edman degradation, and the amino acid sequence for the first :>.4 residues was identified. The sequencing data for the first 6 cycles of the Edman degradation are shown in the table below. The data clearly establish that the N-terminal amino acid sequence of the purified FIBAC
(C-T-V-P-P...) is identical to that previously determined for na-tive FIBAC. It is thene:fore concluded that pFibll properly pro-cesses the recombinant f'IBAC by cleaving the ompA-FIBAC fusion protein at its ala-cys junction to produce the mature form of the FIBAC protein.
N-TERMINAL AMINO ACID SEQUENCE ANALYSIS OF PURIFIED
RECOMBINANT FIBAC
(cysteine residues were labeled with 3H-iodoacetic acid prior to sequencing of the protein) ~34097~

3 12 fi 90 CYS

4 :3 3 9 VAL

1.45 PRO

6 :!55 PRO

B. Soluble fraction Because of the additional contaminants in the starting material, the procedures discussed herein does not initially result in a homogeneous. preparation of FIBAC. However, the pro-cedure does provide sufficient purification to allow refolding of FIBAC to its native cor,~formation. Subsequent purification steps may then be used to complete the isolation procedure.
In this example, the cell growth, induction, harvest ing, and preparation of a total cell lysate were as in the previ-ous example. In order to demonstrate the ability of the present procedure to purify FIBAC from any fraction of the cell lysate, the original cent=rifugal fractionation of the homogenate was omitted. Instead, the total cell lysate was made to 10 M urea, 4 mM EDTA, 50 mM D'.CT, 50 mM MES, pH 6.0 and incubated at 22°C for minutes. The solution was then centrifuged for 15 minutes in an Eppendorf microcentrifuge at 4°C. The pellet was discarded and the supernatant diluted with CM-buffer (50 mM MES, pH 6.0, 6 M urea, 14 mM :>.-mercaptoethanol) and chromatographically frac-tionated on carboxymethyl cellulose as in the previous example.
FIBAC eluted at approximately 120 mM salt along with some de-graded FIBAC and several immunologically non-related contaminants. Estimation of the purity of FIBAC by SDS-PAGE
analysis showed !.t to be greater than 50 percent of the total protein in this fraction. At this level of purity, it is possi-ble to refold they FIBAC to its native conformation using the refolding procedure below. The refolded FIBAC is fully function-al as a metalloprotease inhibitor and may be further purified by anion exchange chromatography.
Anion e~xchang~e chromatography was effected on a column TM
10x100 mm of What.man DE~-52 (Whatman Inc., Clifton, New Jersey) , , equilibrated in 600 mM urea, SO mM Tris, pH 9.6. The solution containing the impure, refolded FIBAC was titrated to a pH of 9.6 by drop-wise addition of 5 N NaOH and applied to the DEAE cellu-lose. Analysis of the flow-through fraction demonstrated that the FIBAC was not retained. The immunologically non-related contaminants bound to t:he matrix and were thereby removed from the solution. The flo4r-through fractions can be concentrated on a CM-cellulose column as shown below.
The flow-through fraction from the anion exchange col-umn was titrated to pH 7.5 by the addition of 5 N HC1. This so-lution was applied to a CM-cellulose column (25x130 mm) previous-ly equilibrated with 600 mM urea and 50 mM Tris, pH 7.5. The column was washed with this same buffer until no protein could be spectrally observed to elute. The FIBAC was then eluted with the above buffer made to 250 mM in NaCL. The protein peak was poo~,ed and dialyzed to equilibrium against 50 mM Tris, pH 7.5.
Electrophoret.ic, immunological, and functional assays of the resulting FIBAC demonstrate an active, refolded col-lagenase inhibitor of greater than 90% purity. The only de-tectable contaminant is a FIBAC degradation product as in the pu-rification from the cell lysate pellet. Because of the identical amino terminal sequence of this contaminant and its apparent mo-lecular mass, it has been concluded that a proteolytic clip has been made close to the carboxy terminus. This material can be removed in further purification steps (e. g., higher resolution ion exchange chromatography or affinity chromatography of the refolded FIBAC).

CONSTRUCTION OF A YEAST FIBAC EXPRESSION CLONE
Another organism in which gene expression and protein export vectors have been constructed is the yeast Saccharomyces cerevisiae. The yeast alpha-factor is a mating hormone which. is produced intrace:Llularly and exported to the growth medium. A
single peptide sequence directs this transport. The FIBAC coding sequence has been cloned into a yeast expression vector to create a fusion of the breast alpha-factor leader sequence to FIBAC. A
construct was first made as a derivative of p~S385. Plasmid pGS385 has the following features:

(a) I:t contains portions of the yeast alpha-factor gene including the promoter, leader peptide, polyadenylation signal, and transcriptional termi-nation aignals;
(b) T'he portion of the alpha-factor gene between the two most distant HindIII sites has been deleted, creating a unique HindIII site;
(c) It contains a unique SalI site 3' to the HindIII
site; and (d) It has t:he pBR322 origin of replication and ampicill.in resistance gene to allow replication and selection in E. coli.
The plasmid F>GS385 was digested with HindIII and SalI.
The HindIII site defines the carboxy terminus of the alpha-factor leader sequence. The ~;ynthetic octanucleotide 5'-AGCTTGCA-3' was used to bridge the alpha-factor C-terminus to the N-terminus of FIBAC. The alpha-factor transcription-translation sequences drive gene expression in this vector. The entire alpha-factor-FIBAC fusion was then removed by digestion with EcoRI and in-serted into plasmid YIPS, a derivative of plasmid pBR322, con-tains the yeast ura3 gene. This plasmid is suitable for use as an expression vector following digestion at a unique StuI site in ura3 and transformation into S. cerevisiae to direct integration of the entire plasmid into the chromosomal ura3 locus. Such integrants of an alpha-FIBAC fusion derivative of YiP5 have been obtained and herE~ produced and secreted immunoreactive material as determined by colony screening techniques.
FYnMDT.F 1 7 EXPRESSION OF RECOMBINANT FIBAC IN ANIMAL CELLS
Two expression systems for the production of FIBAC in animal cells are proposed. The first incorporates the SV40 late promoter to direct transcription in COS-1 cells. This expression system is primarily useful for studying the expression, protein synthesis, post-t:ranslational modification, and transport of FIBAC in COS-1 cells. .Although the system is~rapid and conve-nient, it is limited to the COS-1 monkey cell line. The SV40 ex-pression plasmid will be a derivative of pJC119, the construction ''~:Y~ a 134097 ~

of which is described in derail by Sprague, J., Condra, J.H., Arnheiter, H. and Lazzarini, R.A., in J. Virol. 45:773-781 (1983). The complete FIBAC coding region, including the naturally-occurring signal sequence, has been assembled from the partial cDNA clones. The NcoI site coinciding with the initiator methionine for the leader peptide will be linked via a short synthetic olic~onucleot~ide to the unique XhoI site in PJC119. The entire FIBAC coding region <~nd 3' nontranslated sequences are inserted at this site where transcription is directed by the SV40 late promoter. The plasmid would thus contain 3V40 origin of replication, the pBR322 origin of replication and the ampicillin resistance selectable marker.
The second system would be preferred for production of FIBAC
in animal cells because it will result in the stable and continuous , expression of FIBAC from a human cell line. This vector will be a derivative of pBPV5~:-1, described by Florkiewicz, R.Z., Smith, A., Bergmann, J.E., and Rose, J.K. in Cell Biol. 97:1381-1388 (1983). This plasmid features the bovine papilloma virus origin of replication, pBR322 origin of replicatic>n, beta-lactamase gene, and the 69o transforming fragment of BPV DNA. The S'J40 origin of replication and the SV40 early promoter will be cloned into this plasmid. Sequences from pBR322 that interfere with the replication of the vector in human cells will be excluded. As with the previous plasmid, the entire coding portion of the FIBAC cDNA will be inserted so as to direct its transcription by the SV40 early promoter.
It is expected that purification of FIBAC from the medium following expression and secretion from these cells will be possible essentially as described previously in Examples 4 and 5.

REFOLDING FIBAC
Two assays have been used to monitor the refolding of FIBAC.
Both assays measure the apps=_arance of the functional capacity of FIBAC as its native structure:. The :First assay is an inhibition assay which measures the inhibitory efff=_ct of the W ~

134097 ~

sample on the ability o:E human fibroblast collagenase to degrade 14C-Labeled collagen. 'the second assay is a modified ELISA which measures the binding of the refolded FIBAC to human collagenase.
The collagenase binding ELISA is the primary assay by which FIBAC activity was detected. Here, collagenase is coated overnight at 4°C in 96-well Immulon II plates (1.0 ug./ml in 50 mM
Tris, pH 8.2, 5 mM CaCl~~; 100 ul per well). After the wells are blocked for 45 minutes with 150 ul/well of 3% BSA in washing buffer (50 mM Tris, pH 7.5, 5 mM CaCl2. 0.02% Tween-20), varying dilutions of FIBAC standards or unknown samples diluted in blocking buffer are pipe~tted into the wells (100 ul/well). Fol-lowing a 45-minute incubation period (37°C), the wells are washed three times with washing buffer. Affinity-purified rabbit and anti-FIBAC is added to the wells (diluted 1/100, 100 ul/well) and incubated at 37°C for 45 minutes. The wells are again washed and alkaline phosphat.ase-conjugated goat anti-rabbit IgG (Sigma, di-luted 1/1000 in washing buffer) is then added to the wells (100 ul/well). Following a one-hour incubation period at 37°C, the wells are waslZed a last time and alkaline phosphatase sub-strate (Sigma #104-105, 1 mg/ml in 10% diethanolamine, 100 mM
MgCl2, pH 9.8) is added to the wells. Color development is moni-TM
tored at 495 nm uaing a Titertek Multiskan MC ELISA reader (Flow Laboratories). N<itive FIBAC serves as a standard curve against which unknown samples may be quantitated.
In the collagenase inhibition assay, 14C-labeled guinea pig skin collagen pellets (25 ul/pellet = 2100 cpm) is digested with 50 microliters of trypsin-activated collagenase (approxi-mately 75 ug/ml in 50 mM Tris, pH 7.5, 10 mM CaCl2), which re-leases 14C into solution. After incubating the pellets for 1-3 hours (depending on the rate of digestion), the reaction is stopped by adding 100 ul of Tris buffer and centrifuging for minutes at 10,000 rpm. The supernatant is then pipetted into scintillation vials containing 3 mls of scintillation fluid and counted. Preincubation of the collagenase with 50 ul of varying dilutions of standard or purified FIBAC prior to adding the solu-tion to the collagen pellet should inhibit digestions of the col-lagen and the subsequent release of 14C into solution. The '34087 ~
_7,_ quantitation of inhibitory activity of an unknown sample depends on the amount of active collagenase used in the assay. From this, the activity of the unknown sample may be calculated by assuming a 1:1 molar ratio between inhibitor and enzyme.
Using these assays, the efficiency of the refolding process with respect to protein concentration, oxidised gluthathione concentration, pH, and temperature has been exam-ined. While all combinations of these parameters have not been exhaustively examined, a procedure has been developed which allows efficient renaturation.
The refolding of FIBAC depends greatly on the concen-tration of FIBAC at which refolding is performed. Under oxidizing conditions, dp~lute solutions prevent the formation of interchain disulfides, which eventually lead to the precipitation of aggregates. ~ "
Purified recombinant FIBAC can be refolded and remains soluble with 100% recovery of protein by following the protocol described below:
(1) Dilute, purified FIBAC (less than 300 ug/ml) in 6 Id urea, 50 MES, pH 6.0, 14 mM 2-mercaptoethanol, is incubated in the presence of 70 mM oxidized glutathione.
(2) After a 10-minute incubation period at room tem-perature, the sample is diluted ten-fold with 50 mM Tris, pH 9Ø
(3) The sample is then incubated overnight at 4°C.
SDS-PAG1: analysis of this reactivated material indi-cates the presence of both intact and degraded FIBAC. The de-graded FIBAC is carried through from the purification procedure and does not appear to degrate further during the reactivation procedure.
This sol.ubilized FIBAC has been demonstrated to have both collagenase binding activity (ELISA) and collagenase inhib-itory activity. The amount of activity relative to the amount of protein appears to be greater than 90% as determined by the col-lagenase inhibition assay. This number was derived from calcula-tions based on the estimated amount of collagenase in the assay.

_,8_ Although this is an estimate, it is not believed it to be off by more than 50%. l3inding activity as measured against the FIBAC
standard has enabled us to monitor the relative reactivation of different sample:.
The re:Eolding process has also been shown to work on 50% pure FIHAC preparations. The nature and amount of contaminating protein that can be tolerated is still uncertain, however, some act=ive FIBAC has been detected in total cell lysates without purification demonstrating that at least some yields are obtainable at less than 5% purity.
It will. be apparent to those skilled in the art that various modifications and variations can be made in the processes and products of t:he present invention. Thus, it is intended that the present invention cover the modifications and variations of this invention provided they come within the scope of the appended claims and their equivalence.
._s ,.

Claims (28)

1. A DNA sequence capable of directing intracellular production of collagenase inhibitor, wherein the first 6 amino acids at the N-terminus of said collagenase inhibitor are Cys-Thr-Cys-Val-Pro-Pro.

2. The DNA sequence of claim 1 wherein said sequence capable of directing intracellular production of collagenase inhibitors is:

GTTGTTGCTG TGGCTGATAG CCCCAGCAGG GCCTGCACCT GTGTCCCACC CCACCCACAG

ACGGCCTTCT GCAATTCCGA CCTCGTCATC AGGGCCAAGT TCGTGGGGAC ACCAGAAGTC

AACCAGACCA CCTTATACCA GCGTTATGAG ATCAAGATGA CCAAGATGTA TAAAGGGTTC

CAAGCCTTAG GGGATGCCGC TGACATCCGG TTCGTCTACA CCCCCGCCAT GGAGAGTGTC

TGCGGATACT TCCACAGGTC CCACAACCGC AGCGAGGAGT TTCTCATTGC TGGAAAACTG

CAGGATGGAC TCTTGCACAT CACTACCTGC AGTTTCGTGG CTCCCTGGAA CAGCCTGAGC

TTAGCTCAGC GCCGGGGCTT CACCAAGACC TACACTGTTG GCTGTGAGGA ATGCACAGTG

TTTCCCTGTT TATCCATCCCC CTGCAAACTG CAGAGTGGCA CTCATTGCTT GTGGACGGAC

CAGCTCCTCC AAGGCTCTGA AAAGGGCTTC CAGTCCCGTC ACCTTGCCTG CCTGCCTCGG

GAGCCAGGGC TGTGCACCTG GCAGTCCCTG CGGTCCCAGA TAGCCTGAAT CCTGCCCGGA

GTGGAAGCTG AAGCCTGCAC AGTGTCCACC CTGTTCCCAC TCCCATCTTT CTTCCGGACA

ATGAAATAAA GAGTTACCAC CCAGCAAAAA AAAAAAGGAA TTC

3. The DNA sequence of claim 2 wherein said sequence is capable of directing intracellular production of a collagenase inhibitor biologically equivalent to that isolable from human skin fibroblasts.

4. A recombinant-DNA cloning vector comprising a nucleotide sequence capable of directing intracellular production of metalloproteinase inhibitors, wherein said vector comprises a nucleotide sequence containing the following nucleotides:

GTTGTTGCTG TGGCTGATAG CCCCAGCAGG GCCTGCACCT GTGTCCCACC CCACCCACAG

ACGGCCTTCT GCAATTCCGA CCTCGTCATC AGGGCCAAGT TCGTGGGGAC ACCAGAAGTC

AACCAGACCA CCTTATACCA CTCGTTATGAG ATCAAGATGA CCAAGATGTA TAAAGGGTTC

CAAGCCTTAG GGGATGCCGC TGACATCCGG TTCGTCTACA CCCCCGCCAT GGAGAGTGTC

TGCGGATACT TCCACAGGTC CCACAACCGC AGCGAGGAGT TTCTCATTGC TGGAAAACTG

CAGGATGGAC TCTTGCACAT CACTACCTGC AGTTTCGTGG CTCCCTGGAA CAGCCTGAGC

TTAGCTCAGC GCCGGGGCTT CACCAAGACC TACACTGTTG GCTGTGAGGA ATGCACAGTG

TTTCCCTGTT TATCCATCCC CTGCAAACTG CAGAGTGGCA CTCATTGCTT GTGGACGGAC

CAGCTCCTCC AAGGCTCTGA AAAGGGCTTC CAGTCCCGTC ACCTTGCCTG CCTGCCTCGG

GAGCCAGGGC TGTGCACCTG GCAGTCCCTG CGGTCCCAGA TAGCCTGAAT CCTGCCCGGA

GTGGAAGCTG AAGCCTGCAC AGTGTCCACC CTGTTCCCAC TCCCATCTTT CTTCCGGACA

ATGAAATAAA GAGTTACCAC CCAGCAAAAA AAAAAAGGAA TTC

5. The vector pUC9-F5/237P10.

6. A recombinant-DNA method for microbia production of a collagenase inhibitor comprising:
(a) preparation of a portable DNA sequence capable of directing a host microorganism to produce a protein having collagenase inhibitor activity;
(b) cloning the portable DNA sequence into a vector capable of being transferred into and replicating in a host microorganism, such vector containing operational elements for the portable DNA sequence;
(c) transferring the vector containing the portable DNA
sequence and operational elements into a host microorganism capable of expressing the collagenase inhibitor protein;
(d) culturing the host microorganism under conditions appropriate for amplification of the vector and expression of the inhibitor;
(e) in either order;
(i) harvesting the inhibitor; and (ii) causing the inhibitor to assume an active, tertiary structure whereby it possesses collagenase inhibitor activity, wherein the first 6 amino acids at the N-terminus of said inhibitor are: Cys-Thr-Cys-Val-Pro-Pro.

7. The method of claim 6 wherein said metalloproteinase inhibitor is collagenase inhibitor.

8. The method of claim 6 wherein said portable DNA
sequence contains the following sequence:

GTTGTTGCTG TGGCTGATAG CCCCAGCAGG GCCTGCACCT GTGTCCCACC CCACCCACAG

ACGGCCTTCT GCAATTCCGA CCTCGTCATC AGGGCCAAGT TCGTGGGGAC ACCAGAAGTC

AACCAGACCA CCTTATACCA GCGTTATGAG ATCAAGATGA CCAAGATGTA TAAAGGGTTC

CAAGCCTTAG GGGATGCCCC TGACATCCGG TTCGTCTACA CCCCCGCCAT GGAGAGTGTC

TGCGGATACT TCCACAGGTC CCACAACCGC AGCGAGGAGT TTCTCATTGC TGGAAAACTG

CAGGATGGAC TCTTGCACAT CACTACCTGC AGTTTCGTGG CTCCCTGGAA CAGCCTGAGC

TTAGCTCAGC GCCGGGGCTT CACCAAGACC TACACTGTTG GCTGTGAGGA ATGCACAGTG

TTTCCCTGTT TATCCATCCC CTGCAAACTG CAGAGTGGCA CTCATTGCTT GTGGACGGAC

CAGCTCCTCC AAGGCTCTGA AAAGGGCTTC CAGTCCCGTC ACCTTGCCTG CCTGCCTCGG

GAGCCAGGGC TGTGCACCTG GCAGTCCCTG CGGTCCCAGA TAGCCTGAAT CCTGCCCGGA

GTGGAAGCTG AAGCCTGCAC AGTGTCCACC CTGTTCCCAC TCCCATCTTT CTTCCGGACA

ATGAAATAAA GAGTTACCAC CCAGCAAAAA AAAAAAGGAA TTC

9. The method of claim 6 wherein said vector containing said portable DNA sequence is pUC9-F5/237P10.

10. The method of claim 6 wherein said host microorganism is a bacterium.

11. The method of claim 10 wherein said bacterium is a member of the genus Bacillus.

12. The method of claim 11 wherein said bacterium is Bacillus subtilis.

13. The method of claim 10 wherein said bacterium is Escherichia coli.

14. The method of claim 10 wherein said bacterium is a member of the genus Pseudomonas.

15. The method of claim 14 wherein said bacterium is Pseudomonas aeruginosa.

16. The method of claim 6 wherein said host microorganism is a yeast.

17. The method of claim 6 wherein said yeast is Saccharomyces cerevisiae.

18. The method of claim 6 where said inhibitor is harvested prior to being caused to assume said active, tertiary structure.

19. The method of claim 6 wherein said inhibitor is caused to assume said active, tertiary structure prior to being harvested.

20. A collagenase inhibitor, which is biologically equivalent to the collagenase inhibitor isolable from human skin fibroblasts, has a molecular weight of about 20 kDa, the first 6 amino acids at the N-terminus of said collagenase inhibitor are Cys-Thr-Cys-Val-Pro-Pro, and is produced by the method of claim 6.

21. The microorganism C600/pUC9-F5/237P10 having ATCC
Accession No. 53003.

22. The DNA sequence of claim 1 wherein said nucleotide sequence is:

GGCCATCGCC GCAGATCCAG CGCCCAGAGA GACACCAGAG AACCCACCAT GGCCCCCTTT

GACCCCTGGC TTCTGCATCC TGTTGTTGCT GTGGCTGATA GCCCCAGCAG GGCCTGCACC

TGTGTCCCAC CCCACCCACA GACGGCCTTC TGCAATTCCG ACCTCGTCAT CAGGGCCAAG

TTCGTGGGGA CACCAGAAGT CAACCAGACC ACCTTATACC AGCGTTATGA GATCAAGATG

ACCAAGATGT ATAAAGGGTT CCAAGCCTTA GGGGATGCCG CTGACATCCG GTTCGTCTAC

ACCCCCGCCA TGGAGAGTGT CTGCGGATAC TTCCACAGGT CCCACAACCG CAGCGAGGAG

TTTCTCATTG CTGGAAAACT GCAGGATGGA CTCTTGCACA TCACTACCTG CAGTTTCGTG

GCTCCCTGGA ACAGCCTGAG CTTAGCTCAG CGCCGGGGCT TCACCAAGAC CTACACTGTT

GGCTGTGAGG AATGCACAGT GTTTCCCTGT TTATCCATCC CCTGCAAACT GCAGAGTGGC

ACTCATTGCT TGTGGACGGA CCAGCTCCTC CAAGGCTCTG AAAAGGGCTT CCAGTCCCGT

CACCTTGCCT GCCTGCCTCG GGAGCCAGGG CTGTGCACCT GGCAGTCCCT GCGGTCCCAG

ATAGCCTGAA TCCTGCCCGG AGTGGAAGCT GAAGCCTGCA CAGTGTCCAC CCTGTTCCCA

CTCCCATCTT TCTTCCGGAC AATGAAATAA AGAGTTACCA CCCAGCAAAA APAAAAAGGA

23. A recombinant-DNA method for the production of metalloproteinase inhibitor from animal cells, comprising:
(a) preparation of a portable DNA sequence capable of directing a host animal cell to produce a protein having metalloproteinase inhibitor activity;
(b) cloning the portable DNA sequence into a vector capable of being transferred into and replicating in a Host animal cell, said vector containing operational elements for the portable DNA
sequence;
(c) transferring the vector containing the portable DNA sequence and operational elements into a host animal cell capable of expressing the metalloproteinase inhibitor protein;
(d) culturing the host animal cells under conditions appropriate for expression of the inhibitor; and (e) in either order;
(i) harvesting the inhibitor; and (ii) causing the inhibitor to assume an active, tertiary structure whereby it possesses metalloproteinase inhibitor activity.

24. The method of claim 23 wherein said vector is amplified in a microbial host prior to transfer into the animal cell host.

25. The method of claim 24 wherein said vector contains a portion of the following DNA sequence:

GGCCATCGCC GCAGATCCAG CGCCCAGAGA GACACCAGAG AACCCACCAT GGCCCCCTTT

GAGCCCCTGG CTTCTGGCAT CCTGTTGTTG CTGTGGCTGA TAGCCCCCAG CAGGGCCTGC

ACCTGTGTCC CACCCCACCC ACAGACGGCC TTCTGCAATT CCGACCTCGT CATCAGGGCC

AAGTTCGTGG GGACACCAGA AGTCAACCAG ACCACCTTAT ACCAGCGTTA TGAGATCAAG

ATGACCAAGA TGTATAAAGG GTTCCAAGCC TTAGGGGATG CCGCTGACAT CCGGTTCGTC

TACACCCCCG CCATGGAGAG TGTCTGCGGA TACTTCCACA GGTCCCACAA CCGCAGCGAG

GAGTTTCTCA TTGCTGGAAA ACTGCAGGAT GGACTCTTGC ACATCACTAC CTGCAGTTTC

GTGGCTCCCT GGAACAGCCT GAGCTTAGCT CAGCGCCGGG GCTTCACCAA GACCTACACT

GTTGGCTGTG AGGAATGCAC AGTGTTTCCC TGTTTATCCA TCCCCTGCAA ACTGCAGAGT

GGCACTCATT GCTTGTCGAC GGACCAGCTC CTCCAAGGCT CTGAAAAGGG CTTCCAGTCC

CGTCACCTTG CCTGCCTGCC TCGGGAGCCA GGGCTGTGCA CCTGGCAGTC CCTGCGGTCC

CAGATAGCCT GAATCCTGCC CGGAGTGGAA GCTGAAGCCT GCACAGTGTC CACCCTGTTC

CCACTCCCAT CTTTCTTCCG GACAATGAAA TAAAGAGTTA CCACCCAGCA AAAAAAAAAA
GGAATTC

26. A yeast cell transformed with a recombinant DNA
molecule comprising a DNA sequence as claimed in claim 1.

27. A COS-1 monkey cell or a human cell transformed with a recombinant DNA molecule comprising a DNA sequence as claimed in claim 1.

28. A bacterial cell transformed with a recombinant DNA
molecule comprising a DNA sequence as claimed in claim 1.