CA2070393A1 - Biosynthetic constructs of tgf-beta - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE

Disclosed are constructs of truncated transforming growth factor-beta (TGF-.beta.) produced by expression of recombinant DNA in a prokaryotic host cell. These constructs include at least one polypeptide chain of fewer than about 112 amino acides and fewer than 9 cysteine residues. The sequence of amino acids in these constructs is sufficiently duplicative of the sequence of native transforming growth factor-beta such that it is capable of inducing an anti-proliferative effect on mammalian epithelial cells in vitro. Also disclosed are methods of producing analogs of TGF-.beta. using recombinant DNA technology, and methods of using such analogs.

Description

!
~`7~3~3 BIOSYNTHETIC CONSTR~E OF T~F-~

B~ckuround of the Invention This invention relates to biosynthetic peptide constructs o~ transforming growth factor-beta, to synthetic genes encodi~g polypeptides having transforming growth factor-beta-like biological activity, to methods of producing such synthetic genes using recombinant DNA technology, and to the use of such biosynthetic peptide constructs as regulators of cell proliferation and growth.

Transforming growth factor-beta (TGF-B) is a multifunctional peptide regulator of activity involving cellular or tissue response to injury or stress. This factor has the ability to stimulate cell proliferation in cells of mesenchymal origin, while also being able to inhibit the growth of epithelial cells, embryonic fibroblasts, endothelial cells, and T and B lymphocytes. In addition, TGF-B
has a number of other regulatory activitites which appear to be related uniquely to the specialized function of a particular cell type. For example, TGF-B stimulates the production of matri~ components, e.g., inhibiting the synthesis and secretion of proteolytic enzymes which act on tnese components, thereby regulating the synthesis and degradation of extracellular matriY ~for a reviow, see, e.g., Sporn et al. (1987) ~. Cell Biol. lQ~:1039-104S).
TGF-B-type activities have been identified in many normal fibronectin- and collagen-producing , -2-- 2Q70393 fibroblasts, as well as tissues such as kidney (Roberts et al. (1983) Biochem. 22:5692-5698), placenta (Frolik et al. (1983) Proc. Natl. Acad. Sci.
(USA) 80:3676-3680), and platelets (Childs et al.
(1982) Proc. Natl. Acad. Sci. (USA) 79:5312-5316; and Assoian et al. (1983) J. ~iol. Chem. 258:7155-7160), as well as in tumor cells (see, e.g., Roberts et al.
(1980) Proc. Natl. Acad. Sci. (USA) 77:3494-3498).

There are four known molecular configurations of TGF-B, each having an apparent molecular weiqht of about 25,000 daltons. Three of these species result from homodimeric (TGF-~l, TGF-B2) and heterodimeric (TGF-Bl.2) combinations of the monomeric subunits, Bl and B2. The fourth species is a homodimer of a B3 subunit. Each subunit is processed from a precursor of about 390 amino acids, and the mature subunit protein includes approximately 112 amino acids of its carbosy terminus. The Bl and B2 subunits have about a 70%
amino acid sequence homology in their N-terminal portions, and are highly conserved between species.
The deduced amino acid sequence of TGF-B3 shares about 80~ homology with types Bl and B2, with many of the differences being conservative substitutions.
TGF-Bl was originally isolated from hu~an platelets and placenta ~EP 0128849), and bovine kidney (Roberts et al. iki~-). TGF-B2 was originall~ identified as cartilage-inducing factors tCIF) isolated from bovine bone tUS 4,774,228). TGF-Bl.2 has been found in porcine platelets and other cells which coexpress the Bl and B2 chains (Cheifetz et al. (1988) J. ~iol.
Chem. ~~:10783-10789). TGF-B3 has been identifie~

2~7~393 in both human and chicken S~uke et al. (1985) Proc.
Natl. Acad. Sci~ (USA) 85:4715-4719).

The TGF-B's belong to a larger gene family, the members of~which encode struct~rally similar proteins that have similar regulatory activities (reviwed in Massague (1987) Cell 4~:437-438).
Included in this family are: (1) Ygl, a protein involved in mesoderm formation during XenoDus development; (2) decapentaplegic complex (DPP), a polypeptide encoded by a Drosophila gene responsible for development of the dorsoventral pattern in the embryo: (3) OPl, a region of a native osteogenic protein sequence encoded by exons of a genomic DNA
sequence retrieved by applicants; (4) cartilage inducing factors ~CIFs) isolated from bovine bone (US
4,774,228); (5) mammalian osteogenic bone matri~
proteins CBMP-2a, C~MP-2b, and CBM~-3, discovered by applicants (see W089/01453); and (6) B-inhibin-a and b, gonadal proteins that suppress pituitary secretion of follicle stimulating hormone. All of these proteins are believed to dimerize during refolding, and are inactive when reduced to the monomeric form.
In addition, many include portions of a common precursor peptide.

Identification of the regulatory activities of the proteins in the TGF-B family, and the elucidation of their amino acid sequences, have resulted in research efforts directed to the production o these proteins by recombinant means.
For e~ample, EP 0200341 discloses nucleic acid sequences encoding native TGF-B and precursors thereof which can ~e expressed in a host eukaryotic 2~3~3 cell transformed therewith. EP 0150572 discloses the manufacture of structural genes coding for TGF-Bl and analogs thereof and the expression of these genes in microorganisms. ~owever, the desi~n and expression of consensus protein constructs having considerable sequence homology with a number of the proteins in the TGF-B family, and displaying TGF-B-like activity, has heretofore not been contemplated.

Accordingly, it is an object of this invention to provide novel analogs of TGF-B having TGF-B biological activities. Another object is to provide an efficient method of producing novel, active TGF~B analogs. Yet another object is to provide genes encoding novel, non-native, TGF-B
species and methods for their production using recombinant DNA techniques. Another object is to provide novel truncated forms of T5F-B and structural designs for proteins with TGF-B biological activity.
A further object is to provide methods of regulating cell proliferation using TGF-B analogs.

These and other objects and features of the invention will be apparent from the description, drawings, and claims which follow.

s~m~aFy o~_thç_Inventinn 2 0 7 0 3 9 3 It has been discovered that forms of native TGF-B which have been truncated at the N-terminus, and which have fewer than the native number of cysteine residues demonstrate TGF-~-like biological activity, including the ability to induce an anti-proliferative effect on mammalian epithelial cells n vit~o. It has also been une~pectedly discovered that truncated analogs ~f other structurally similar proteins in the TGF-B family known to have unrelated biological activities also possess this TGF-B-like activity. These discoveries enabled the design and construction of DNAs encoding novel, non-native protein constructs which individually and combined are capable of inhibiting the proliferation of mammalian epi~helial cells in culture.

Thus, in one aspect, the invention comprises a truncated TGF-B analog produced by e~pression of recombinant DNA in a host cell and capable of inducing an antiproliferative effect in mammalian epithelial cells Ln Vi~LQ- This protein construct includes two polypeptide chains, each including a biologically active domain, and each having ewer than 9, and preferably 6 or ~, cysteine (Cys) residues. It may further be characterized as being unglycosylated.

In another aspect, the invention comprises a protein produced by expression of recombinant DNA in a prokaryotic host cell and including a pair of polypeptide chains of fewer than about 112 amino 2~7~3~3 acids each. The sequence of amino acids in each chain is sufficiently duplicative of the sequence of TGF-B such that the protein is capable of inducing an anti-proliferative effect on mammalian epithelial cells in vi~o. Preferably, the polypeptide chain contains fewer than 9, and more preferably 6 or 8, cysteine residues, and further, may be unglycosylated.

The cysteine residues are involved in the formation of intra- and inter-chain disulfide bonds (folding), the correct formation of which results in an active construct having TGF-B-l~ke activity. In eucaryotes, the synthesis and proper folding of the protein can occur at least within those cells known to express TGF-~; in prokaryotic calls, folding must be performed Ln Vit~Q, a difficult feat in that any number of combinations of disulfide linkages e~ist between two polypeptide chains, each having less than 9, and prefera~ly 6 or 8 cysteine residues. An important aspect of this inventior. is the discovery that truncated constructs of the type described herein may be post-translationally modified and folded (by o~idation) in vit~Q to produce TGF-B-like activity.

Several forms of TGF~ monomers are known in nature, ~1, B~, and B3. Investigation of the properties and structure of these native forms enabled the development of a rational design for non-native ( i . Q ., not known to be e~pressed in nature), truncated protein constructs which also are capable of differentially reyulating cell proliferation in various cell types. Further, upon examination of a number of unrelated proteins wi ~ ~ 7 ~ 3 9 3 some amino acid sequence homology, it was une~pectedly discovered that they, too, possess TGF-B-like activity.

Based on this knowledqe, a series of consensus DNA sequences were designed with the goal of producing active TGF-B analogs. The sequences were based on partial amino acid sequence data obtained from native TGF-B species and from observed homologies with genes reported in the literature encoding proteins of the TGF-B family (including Vql and DPP), or on the amino acid seque~ces they encode, having a presumed or demonstrated developmental function. Several of the biosynthetic consensus sequences have been expressed as fusion proteins in prokaryotes, purified, cleaved, refolded, applied to a mammalian L~ vitrQ assay system, and shown to have TGF-B-like anti-proliferative activity.

In preferred aspects of tne invention, the proteins encoded by these consensus sequences include the generic amino acid sequences:

CXXXXLYXXFXXDXGWX~WXXXPXGYXAXXCXGXCPXXXX~X~3XXX~X~3XXX

XXLXXXXXPXXX~XXXCCVXK~K~JO~W UCL~XXXX~XXMXVXXCXCX

2~393 and 10 20 30 40 so LYXXFXXnXGWX~3WX~PXGYiCAXXCXGXCPXXXX6~3X~lX~3XXX

~LXxxxxpx~axxxccw~ 3~vxxcxcx wherein the letters indicate the amino acid residues of standard single letter code, each ~X~
independently represents one of the naturally occurring amino acid or a derivative thereof, and each "~" independently represents an amino acid or a peptide bond.

The currently preferred active peptide constructs comprise amino acid sequences derived from the three ~onomeric subunits of TGF-B (1, 2, and 3), DPP, and OPl. The amino acid seguences of these proteins are set forth in FIGURE 1 relative to the sequence of TGF-Bl. Pre~erred amino acid sequences within the foregoinq qeneric sequences are:

CCVRQLYIDFRKDLGWK-WIHEPKGYHANFCLGPCPYIWS--L-DTQ--Y-SKV
L P KR N A A S H R
Q V Y S R A T ~
KK H VE - V QN IA Q M Y Y E PLTEI NGSN AIL
RRHS S DD V L D Y H K F ADHF S V

LALYNQHNPGAS-AAPCCVPQALEPLPIVYYVGRKPKVEQL-SNMIVRSCKCS
S TI E S S D T L I KT I K
G L V
QT VHS E D-IPL TKMS ISM F DNNDNV LRHYE A DE G R
NN K V KA Q DSVA LNDQST KN QE T VG

2Q~03~3 and LYIDFRKDLGWX~WIHEPKGYHANFCLGPCPYIWS--L-DTQ--Y-SKV
P KR N A A S H R
Q V Y S R A T T
VE - V QN IA Q M Y Y E PLT~I NGSN AIL
S DD V L D Y H K F ADHF S V

LALYNQHNPGAS-A~PCCVPQALEPLPIvYYVGRKPKVEQL-SNMIVRSCKCS
S TI E S S D T L I KT I K
G L V
QT VHS E D-IPL TKMS ISM F DNNDNV LRHYE A DE G R
NN K V KA Q DSVA LNDQST KN QE T VG

wherein, at each position where more than one amino acid is shown vertically, any one of the amino acids shown may be used alternatively in various combinations, and "-" and n__n represent a peptide bond. Note that the numbering of amino acids is selected solely for purposes of facilitating comparisons among alternative sequences. These generic sequences have fewer than 9, and preferably 6-8 cysteine residues where inter- and/or intramolecular disulfide bonds can form, and contain other critical amino acids which influence the tertiary structure of the proteins. Similar structural features are found in the above named known proteins of the TGF-~ family whose amino acid sequences previously have been published. However, of these only the TGF-B species (1, 2, and 3) have been describad as capable of inducing an anti-prolierative effect in mammalian epithelial cells Ln vi~~Q.

Particularly useful sequences include analogs having the following amino acid sequences:

TGF-~l CCVRQLYIDFRKDLGWKWIHEPKGYHANFCLGPCPYIWSLPTQYSKVLAL

YNQHNPGASAAPCCVPQALEPLPIVYYVGRKPKVEQLSNMIVRSCKCS;

TGF-~2 CCLRPLYIDFKRDLGWKWIHEPKGYNANFCAGACPYLWSSDTQHSRVLSLY

NTINPEASASPCCVSQDI.EPLTILYYIGKTPKIEQLSNMIVKSCKCS;

TGF-~3 CCVRPLYIDFRQDLGWKWVllEPKGYYANFCSGPCPYLRSAD'rTHSTVLGL

YNTLNPEASASPCCVPQDL,EPt.TIL.YYVGK'I`PKVF:QLSNMVVKSCKCS;

Vgl CKKRHLYVEFKDVe,WQNWVIAPQGYMANYCYGEe'PYPL'rEILNGSNtlAIL
gO 100 QTI.VHSIEiPEDIPL,PCCVPTKMSPISMl.ti`Y[)NNDNVVLRHYENMAVDECGCR;

DPP

CRRHSLYVDFSDVGWDDWIVAPLGYDAYYCHGKCPFPLADHFNSTNHAVV

QTLVNNNNPGKVPKACCVPTQLDSVAMLYLNDQSTVVLKNYQEMT W GCGCR;

~, . . .

--ll--and CCVRQLYIDFKRDLGWKWVHEPKGYAANFCAGACPYLWSADTQHSRVLA

LYNTANPEASAAPCCVPQDLEPLTILYYVGRTPKVEQLSNM W KSCKCS.

or more truncated analogs such as:
TGF-~1 LYIDFRKDLGWKWIHEPKGYHANFCL.GPCPYIWSLDTQYSKVLAL

YNQHNPGASAAPCCVPQALEPLPIVYYVGRKPKVEQLSNMIVRSCKCS;

TGF-~2 LYIDFKRDLGWKWIllEPKGYNANFCAGACPYLWSSD'rQHSRVLSLY

NTINPEASASPCCVSQl)l,EPLTILYYIGKTPKIEQl.SNMIVKSCKCS;

TGF`-~3 LYIDF'RQl)l.GWKWVlll.PK(;YYANFC'SGPt~PYl.l~SAl)T'r~lS'rVt,GL

YNTI,NPE`ASASlCCVPQl)l,l`lLTII,YYVGRl`PKVEQl.SN~lVVKSCKCS;

V~31 ~0 50 LYVEFKDVGWQNWVIAEQGYMANYCYGECPYPLlEIl,NGSNHAIL

QTLVHSIEPEDIPLPCCVYTKMSPISMLFYDNNDNVVLRtlYENMAVDECGCR;

;,~

DPP

LYVDFSDVGWDDWIVAPLGYDAYYCHGKCPFPLADHFNSTNHA W

QTLVNNNNPGKVPKACCVPTQLDSVAMLYLNDQSTVVLKNYQEMTVVGCGCR;

and LYIDFKRDLGWKWVHEPKGYAANFCAGACPYLWSADTQHSRVLALYN

TANPEASAAPCCVYQDLEPI,TILYYVGRTPKVEQ~,SNM WKSCKCS.

The name given to each of these sequences designates the natural source DNA sequence encoding the amino acid sequence which, as far as applicants are aware, exhibits the most homology with the recited TGF-~analog.

The invention further includes DNA sequellces encoding these constructs and a prokaryotic host cell en~ineered to express these DN~ seq11ences. In a preferred aspect of the invention, the p1okaryotic host cell (~.g. E. coli) is tral1seGteci with a V~CtOI
., ~ ., ~ .~ ,,=
inC1Udin~ tl1C! TGF ~-enCOd illCJ I~N~ seq~1enc~. The tral1sor1ned cell i* cl1ltr11tcd to express thc protc~
which .is then puliried ,~l1d ac~.ivated by oxidation ln v.itro. T~1~? pl~o~eil1 so tl`~r~t~ )lcls t~1~ ahility to induce an allt:i-ploli~etclt.ive effect on cultuled mammalian epithelial cel1s.

The biosynthetic constructs disclosed herein may be used to regulate cellular activities such as proliferation and growth. In this regard, these .

2~703~3 constructs have wide potential clinical applications, for example, by controlling the proliferation of various tumor cell lines, or by enhancing the growth rate of T and B lymphocytes in immunosuppres~ed (e.g.
Acquired Immunodefficiency Syndrome (AIDS) patients.
The constructs also may be used in cell cultures to modulate growth of various types of eucaryotic cells.

Brief ~es~iption of th Q rawin~ 2 Q 7 0 3 9 3 The foregoing and other objects of this invention, the various features thereof, as well as the invention itself, may be more ully understood from the following description, when read together with the accompanying drawings, in which:

FIGURE 1 is a comparison of the amino acid sequence of various proteins in the TGF-B family to TGF-Bl; and FIGUREs 2A and 2B are rep--esentations of a DNA sequence and corresponding amino acid sequence of a modified trp-LE leader sequence, two FB domains of protein A, an Asp-Pro cleavage sit~, and (A) the 6 Cys TGF-Bl sequence, and (B) the 8 Cys TGF-Bl sequence.

DescriP~ion 2070~93 Nucleic acid sequences encoding truncated TGF-B analogs were designed based on sequence data reported in the literature, codons inferred from known amino acid sequences, and observations of partial homology with known genes of the TGF-B
family. These sequences have been refined by comparison with the sequences present in certain regulatory genes from the TGF-B family.

The naturally occurring proteins of the TGF-B family are made as precursors~ with a large and poorly conserved N-terminal domain and a characteristic C-terminal domain in which a pattern of 7 cysteines (Cys) residues is highly conserved.
In addition to these Cys residues, certain other amino acids are found in members of the family very nearly in the same relative positions in sequence as set forth below:

CXXXXLYXXFXXDXGWX6WXXXPXGYXAXXCXGXCPXXXX6~X~XXX6~X~XXX
~0 90 100 XXLXXXXXP~XX0XXXCC ~ ~XXMXVXXCXCX

wherein each X independently repre3ents a naturally occurring amino acid, and ~ or ~ represents an amino acid or peptide bond.

The N-~erminal sequence o~ mature ~GF-B and other related proteins contains a variable number of Cys residues which appear to be crosslinked among each other or with a residue of another amino acid chain, but not to Cys residues in :he C-terminal 2~!7~393 dom~in. Maturation of the precursor to the mature form of these proteins occurs by trypsin-like cleavages between the precursor and the mature protein, and possibly also within the precursor form as other similar cleavage sites ar~ presen~ therein.

All members of the Vgl-related subyroup of the TGF-B ~amily (includinq Vgl, DPP, OPl, CBMP-2a, CBMP-2b, and CBMP-3) share the feature of two basic residues (i.e., Lys-Lys, Arg-Arg, ~rg-Lys) following the first Cys in the conserved C-terminal domain.
The conserved double basic residues may represent another secondary maturation site. Cleavage by trypsin or related protease releases a C-terminal domain containing only 6 Cys residues. Since the precursor region of TGF-B contains up to S Cys residues which are not crosslinked to the C-terminal domain, the first of the 7 Cys residues may not be crosslinked to the C-terminal domain either.
Therefore 6 Cys residues appear to be sufficient for a properly folded C-terminal domain.

In view of this disclosure, skilled genetic engineers can design and synthesize genes which encode a number of appropriate amino acid sequences.
These genes can be e~pressed in various types of eucaryotic cells but, or reasons of efficiency are preferably produced in prokaryotic host cells, thereby providing large ~uantities of active synthetic proteins such as truncated analogs, muteins, fusion proteins, and other constructs, all mimicking the biological activity af native TGF-B, including the ability to induce an anti-proliferative e~fect on cultured mammalian epithelial cells.

2~393 More specifically, the DNA sequences designed according to the above criteria and logic were constructed using known techniques involving assembly of oligonucleotides manufactured in a DNA
synthesizer. The sequences may be expressed using well established recombinant DNA tachnologies in various prokaryotic host cells, and the e~pressed proteins may be cleaved from precursors, osidized, and refolded in vitro for biological activity. This approach has been successful in producing a number of novel protein constructs not found in nature (as far as applicants are aware) which hava TGF-B-like activity, i.e., the ability to induce an anti-proliferative effect on cultu~ed mammalian epidermal cells.

The design and production of such biosynthetic proteins, and other m3terial aspects concerning the nature, utility, how to make, and how to use the subject matter claimed herein will be further understood from the following non-limiting e~amples, which constitute the best method currently known for practicing the various aspects of the invention.

~,~

1. Consensus Sequence Design Published amino acid sequences for TGF-B2, TGF-B3, Vgl, and DPP-C were used to determine which amino acids showed strong homology with the TGF-Bl sequence. FIGURE 1 compares the amino acid sequences ~ 8-~7~3~3 of these proteins with the sequence of TGF-~denotes a match), and TABLE 1 summaries the e~tent of homology.

comDarisonno. of matches % homology TGF-B2/TGF-B1 78/115 67.8 TGF-B3/TGF-B1 85/115 73.9 DPP/TGF-Bl 34/115 29.6 Vgl/TGF-Bl 35/115 29.6 -In determining an appropriate consensus amino acid sequence for TGF-B analogs, from which encoding nucleic acid sequences can be determined, the following points were considered: ~1) the known amino acid sequence of natural source TGF-~l, 2, and 3 are ranked highest; (2) where an amino acid in the sequence matches for all three proteins, it is used in the synthetic gene sequence; (3) matching amino acids in DPP and Vgl are used; (4) if Vgl or DPP
diverge, but either one were matched by TGF-Bl, 2, or 3, this matched amino acid was chosen; and (5) where all sequences diverge, the amino acid residue alanine was chosen, provided that the secondary structure is maintained.

2Q~3~3 Using these criteria, the preferred sequence i s :

3~ 40 50 CCVRQLYIDFKRDLGWKWVHEPKGYAANFCAGACPYLWSADTQHSRVLA

6~ 70 80 90 LYNTANPEASAAPCCVPQDLEPLTILYYVGRTPK~EQLSNMVVXSCKCS.

In addition, the first consensus sequence was designed to preserve 8 of the disulfide crosslinks and the apparent structural homology among the related proteins, while the secoffd more highly truncated consensus sequence was d~signed to preserve 6 disulfide bonds. That sequence is:

LYIDFKRDLGWKW~HEPKGYAANFCAGACPYLW.CADTQHSRVLALYN

TANPEASM PCCVPQDLEPLTILYYVGRTPKVE~SNMVVKSCKCS.

2. Gene Preparation and Expression The synthetic genes designed using the criteria set forth above are produced by assembly of chemically synthesized oligonucleotides. 15-lOOmer oligonucleotides are synthesized on a Biosearch DNA
~odel 8600 Synthesizer, and are p~rified by pol~acrylamide gel electrophoresis ~PAGE) in Tris-Borate-EDTA buffer (TBE). T~.e DNA is then electroeluted from the gel. Overlapping oligomers are phosphorylated by T4 polynucl~otide kinase, and then ligated into larger blocks which may also be -2~-2Q7~393 purifed by PAGE. Alternatively, n~tural gene sequences and cDNAs may be used for e~pression. The two resulting genes are shown as tne latter portion of the fusion sequences in FIGUREs 2A and 2B. The sequence shown in 2A is a truncated form of 2B; five amino acids at the N-terminus have been eliminated.

To enable the expression ~f the synthetic gene shown in FIGURE 2A in an E. Ç:~li host, the qene is modified by cassette mutagenesis, The N-terminus is replaced up to the ClaI site wi.h a hinge region that provides for release of the TGF-B protein from the leader, preceded by a BamHI sïte for attachment to leader peptides. The modified ~ene is then attached to an FB-dimer leader at :he BamHI site.
The complete fusion gene is shown in FIGURE 2A.

The fusion gene is then inserted as an EcoRI
to PstI f raqment into an e~pression vector based on pBR~22 and containing a synthetic tryptophan (trp) promoter/operator and a modified tIp-LE (MLE) leader (which is similar to the one described by Huston et al. in Proc. Natl. Acad. Sci. (USA) (1988) 85:5~79-5883, but having only a single Eco~I site at the hinge of the MLE leader). The vector is opened at the EcoRI and PSTI restriction iites, and a F~-F~TGF-B gene fragment is then inserted therebetween, where FB is f ragment B of Staphylococcal Protein A. The resulting e~pression vector includes the TGF-B gene to a ragment encoding FB.

~21-~Q~93 3. Production of Active Analogs The protein constructs are expressed in ~
ÇQli host strain JM101 (e.g.) grown in minimal medium (M9) after starvation for trp and induction by indoacrylic acid (I~A). The cells are lysed and the inclusion bodies collected by differential centrifugation. The fusion proteins are purified from the inclusion bodies by urea or quanidine solubilization. The FB sequence is then chemically cleaved from the TGF-B protein construct at the hinge region of the fusion protein. The hinge region has the sequence Asp-Pro-Asn-Gly which can be cleaved at the Asp-Pro site with dilute acid, or at the Asn-Gly site with hydrosylamine. The resulting cleavage products are passed through a Sephacryl-200HR column which separates most of the uncleaved fusion products from the TGF-~ analogs.

Protein refolding is performed under the conditions of 50 mM Tris-HCl, pH 8.0, 3 M guanidine hydrochloride (GuHCl), 10 mM dithiothreitol ~DTT), and 1-10 mM o~idized glutathione.

4. TGF-B Activity Assay This assay is based on the ability of TGF-~to inhibit DNA synthesis in the mink lung epithelial cell line, ATCC no. CCL 64. A confluent culture o CCL-64 maintained in Eagle's minimum essential medium (EM~M) supplemented with 10% fetal bovine serum (F~S), 200 units/ml penicillin, and 200 ~g/ml streptomycin, is used to seed a 48-well cell culture plate at a cell density of 200,000 cells per well.

2~393 When the culture becomes confluent, the media is replaced with 0.5 ml of EMEM containing 1% FBS and penicillin/streptomycin. The culture is incubated for 2~ hours at 37C. The TGF-B test samples in EMEM
containing 0.5% FBS are then zdded to the wells, and the cells are then incubated for another 18 hours.
After incubation, 1.0 ~Ci of 3H-thymidine in 10 ~1 is added to each well. The cells are incubated for 4 hours at 37C. The media is then removed and the cells washed once with ice-cold phosphate-buffer saline. The DNA is precipitated by adding 0.5 ml of 10% TCA to each well and incubating at room temperature for 15 min. The cells are then washed three times with ice-cold distilled water and lysed with 0.5 ml 0.4 M NaOH. The lysate from each well is then transferred to a scintillation vial and the radioactivity recorded using a scintillation counter (Smith-Kline Beckman).

Each test sample is assay~d in triplicate.
A TGF-~ control is included in each assay. The inhibition activity of each sample is espressed as the 50% effective dose ~ED50), which is defined as the amount of material in ng/ml required to induce 50% reduction in ma2imal incorporation o 3H-thymidine.

The invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the oregoing 207~93 description, and all changes which come within the meaning and range of eguivalency of the claims are therefore intended to be embraced therein.

What is claimed is:

Claims (32)

1. A truncated transforming growth factor-beta construct produced by expression of recombinant DNA
in a host cell, capable of inducing an anti-proliferative effect in mammalian epithelial cells in vitro, and comprising two polypeptide chains, each of said chains comprising an active domain having fewer than 9 cysteine residues.

2. The construct of claim 1 wherein each of said active domains comprises 8 cysteine residues.

3. A protein produced by expression of recombinant DNA in a prokaryotic host cell, said protein comprising a pair of polypeptide chains, each of which has fewer than about 112 amino acids in a sequence sufficiently duplicative of the sequence of transforming growth factor-beta such that said protein is capable of inducing an anti-proliferative effect on mammalian epithelial cells in vitro.

4. The protein of claim 3 wherein each of said polypeptide chains comprises less than 7 cysteine residues.

5. The protein of claim 4 wherein each of said polypeptide chains comprises 6 cysteine residues.

6. The protein of claim 1 or 3 further characterized by being unglycosylated.

7. The protein of claim 1 or 3 comprising the amino acid sequence:

wherein each "X" independently represents one of the naturally occurring amino acid or a derivative thereof, and each ".THETA." independently represents an amino acid or a peptide bond.

8. The protein of claim 1 or 3 comprising the amino acid sequence:

wherein each "X" independently represents one of the naturally occurring amino acid or a derivative thereof, and each ".THETA." independently represents an amino acid or a peptide bond.

9. The protein of claim 7 comprising the amino acid sequences:

wherein, at each position where more than one amino acid is shown, any one of said amino acids shown may be in that position, and "-" and "--" represent a peptide bond.

10. The protein of claim 9 comprising the amino acid sequence:

.

11. The protein of claim 9 comprising the amino acid sequence:

.

12. The protein of claim 9 comprising the amino acid sequence:
.

13. The protein of claim 9 comprising the amino acid sequence:
.

14. The protein of claim 9 comprising the amino acid sequence:
.

15. The protein of claim 9 comprising the amino acid sequence:
.

16. The protein of claim 8 comprising the amino acid sequences:

wherein, at each position where more than one amino acid is shown, any one of said amino acids shown may be in that position, and "-" and "--" represent a peptide bond.

17. The protein of claim 16 comprising the amino acid sequence:

.

18. The protein of claim 16 comprising the amino acid sequence:
.

19. The protein of claim 16 comprising the amino acid sequence:
.

20. The protein of claim 16 comprising the amino acid sequence:
.

21. The protein of claim 16 comprising the amino acid sequence:
.

22. The protein of claim 16 comprising the amino acid sequence:

.

23. A DNA sequence encoding the amino acid sequence of one of the chains of the construct of claim 1 or 3.

24. A prokaryotic cell engineered to express the DNA sequence encoding the amino acid sequence of one of the chains of the construct of claim 1 or 3.

25. The prokaryotic cell of claim 24 wherein said cell is E. coli.

26. A method of producing a protein construct having the activity of transforming growth factor-beta, said method comprising the steps of:

(a) transforming a prokaryotic host cell with a vector comprising the DNA sequence of claim 23;

(b) culturing said transformed host cell to express said protein construct;

(c) purifying said protein construct; and (d) activating said protein construct by oxidation in vitro to produce a two chain disulfide-linked transforming growth factor-beta analog, said activated protein construct having an anti-proliferative effect on mammalian epithelial cells in vitro.

27. The method of claim 26 wherein said activated protein comprises a dimer.

28. The method of claim 26 wherein said activated protein comprises fewer than 9 cysteine residues.

29. The method of claim 28 wherein said activated protein comprises 6-8 cysteine residues.

30. The method of claim 28 wherein said activated protein comprises 8 cysteine residues.

31. The method of claim 28 wherein said activated protein comprises 6 cysteine residues.

32. The method of claim 28 wherein said activated protein is unglycosylated.