* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Hybrid tryptophan aporepressor containing ligand binding sites
Ancestral sequence reconstruction wikipedia , lookup
Promoter (genetics) wikipedia , lookup
Monoclonal antibody wikipedia , lookup
Molecular cloning wikipedia , lookup
Protein–protein interaction wikipedia , lookup
Genetic code wikipedia , lookup
Western blot wikipedia , lookup
Ribosomally synthesized and post-translationally modified peptides wikipedia , lookup
Deoxyribozyme wikipedia , lookup
Endogenous retrovirus wikipedia , lookup
Ligand binding assay wikipedia , lookup
Expression vector wikipedia , lookup
G protein–coupled receptor wikipedia , lookup
Transcriptional regulation wikipedia , lookup
Paracrine signalling wikipedia , lookup
Gene expression wikipedia , lookup
Clinical neurochemistry wikipedia , lookup
Nucleic acid analogue wikipedia , lookup
Biochemistry wikipedia , lookup
Metalloprotein wikipedia , lookup
Vectors in gene therapy wikipedia , lookup
Silencer (genetics) wikipedia , lookup
Biosynthesis wikipedia , lookup
Point mutation wikipedia , lookup
Signal transduction wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
US005190873A United States Patent [191 [11] Patent Number: Lernhardt et a1. [45] [54] HYBRID TRYPTOPHAN APOREPRESSOR CONTAINING LIGAND BINDING SITES [75] Inventors: Waldemar Lernhardt, Solana Beach; [73] Assignee: Mario Bourdon, San Diego; Phil Youderian, Ramona, all of Calif. California Institute of Biological Research, La Jolla, Calif. [21] Appl. No: 720,222 [22] Filed: Jun. 21, 1991 [51] Int. Cl.5 .................... .. C07K 13/00; C07K 17/00; [52] US. Cl. ................................. .. 435/177; 435/697; C07K 17/02; C12P 21/00 435/691; 530/350; 530/812; 930/250 [58] Field of Search ..................... .. 435/91, 69.7, 69.1, [56] 435/177; 530/350, 812; 930/250 References Cited U.S. PATENT DOCUMENTS 4,703,039 10/1987 l-lawiger et a1. .................... .. 514/21 FOREIGN PATENT DOCUMENTS 9100912 1/1991 PCT lnt’l Appl. . OTHER PUBLICATIONS Botstein, et al. Science 229:1193-1201 1985. Bass, et a]. Science 242:240-245 1988. Gunsalus, et al. PNAS 77(12) 7117-7121 1980. Date of Patent: 5,190,873 Mar. 2, 1993 Joachimiak et al., J. Biol. Chem, 258:12641-12643 (1983). Lawson et al., Nature, 333:869-871 (1988). Luisi et al., Biochim. Biophys. Acta, 1048:113-126 (1990). Marrnorstein et al., J. Biol. Chem, 262:4922-4927 (1987). Pfau et al., Nucl. Acids Res, 18:6165 (1990). Schevitz et al., Nature, 317:782-786 (1985). Arrowsmith et al., Biochem, 29:6332-6341 (1990). Arvidson et al., J. Biol. Chem, 261:238-243 (1986). Arvidson et al., Genetics, 128:29-35 (1991). Bourdon et al., J. Cell Biol., 108:1149-1155 (1989). Joachimiak et al., Proc. Natl. Acad. Sci. USA, 80:668-672 (1983). Zhang et al., Nature, 327:591-597 (1987). Engel et al., Biochem. 30: 3161—3169 (1991). Primary Examiner-David L. Lacey Assistant Examiner—-Donald E. Adams Attorney, Agent, or Firm-Thomas Fitting [57] ABSTRACT Hybrid proteins containing repressor proteins and sub‘ stituted receptor binding sites, amino acid and DNA sequences encoding the hybrid proteins are provided. Methods for preparing the hybrid proteins are also described. 19 Claims, 6 Drawing Sheets US. Patent Mar. 2, 1993 m m2wzeo.0Sm;:>w“Eéf. (/ Q 09.6 235 ‘\ E.250 Sheet 1 of 6 US. Patent Mar. 2, 1993 Sheet 2 of 6 5,190,873 @:m2z53k02 .6 FO w I0 2‘ lo wm 10 m .LNBWHOVllV ‘n30 m8; US. Patent Mar. 2, 1993 Sheet 3 of 6 \\ \ x x\ \\ _\ x x\ \\o x Ix x x IO QN \ \\ \com BONHOSHHOH‘H BALLV'I 3H .LNBINHOVllV 1130 5,190,873 mQI U.S. Patent mIOzuwiom Mar. 2, 1993 5,190,873 Sheet 4 of 6 NaIw|1o 2mw ¢O_, 000-09O-_ -_b—:byp.- - IN. I: IO._ lm.O IQO .LHO iwd Ind ind’ .LNHWHOViiV 1130 211M0Q:9E2. .0 I w US. Patent w0E2O5mQ6J m|O_oaElxmrum 864 5. 3 7I.2 Mar. 2, 1993 Sheet 5 of 6 5,190,873 _ _ _ O 0. VN TrpR-TN5 TrpR wt FIG. 5 US. Patent Mar. 2, 1993 Sheet 6 of 6 5,190,873 4000 NO |__| TREATMENT ‘_ a200 Lzu E I Q LM609 2400 U 4 t< I600 _| ....| 8 aoo~ o ' 7// FIG. 6 1 5,190,873 2 of the B cell cycle of human and murine preactivated blasts (Melchers, supra (1985); Bohnsack and Cooper J. HYBRID TRYPTOPHAN APOREPRESSOR CONTAINING LIGAND BINDING SITES Immunol, 141:2569 (1988)). CR2 is also phosphorylated during B cell stimulation, a common property of This invention was made with government support 5 growth factor receptors (Changelian and Fearon J. under government contract 5 29 CA45506 awarded by Exp. Med, 161:101 (1986)). the National Institutes of Health. The government has CR2 occurs on normal and malignant B lymphocytes certain rights in the invention. (Cooper et al., supra (1988); Hatzfeld et al , J. Immunol, 140:170 (1988)), on epithelial cells (Young et al., The FIELD OF THE INVENTION 10 Lancet, 240 (1986)), and, to a lesser extent, on immature This invention relates to hybrid proteins constructed from prokaryotic repressor proteins and receptor bind ing sites, and to the methods for producing the hybrid proteins. ' I BACKGROUND OF THE INVENTION Receptors are molecules, typically proteins or glyco proteins, found on the surface of cells, including mam malian cells, that possess speci?c af?nity for other mole thymocytes and follicular dendritic cells (Tsoukas and Lambris, Eru. J. Immunol. 18:1299 (1988); Reynes et al., J. Immunol, 135:2687 (1985)). The primary struc ture of CR2 has been deduced from the DNA sequence 15 of its clone. Human CR2 is a membrane glycoprotein of 145 kd, and has sequence similarity to other members of the family of complement binding proteins. The recognition sites on CR2 for C3d and EBV have been located on the N-terminal part of this longitudinal cules known as ligands. Ligands may be small or large molecule. The sequence motifs on C3 and EBV coat (macro) molecules such as proteins. Binding of ligands protein, gp350, that mediate binding to CR2 receptor to receptors on the surface of mammalian cells elicits dramatic responses or “signals” in the cells such as have also been de?ned (Lambris et al., Proc. Natl. Acad. Sci. USA, 8224235 (1985); Nemerow et al., Cell 56:369 proliferation and adhesion. These cellular responses (1989)) (Table 1). Synthetic hexapeptides with the se involve protein-protein interactions and intercellular 25 quence of the CR2 binding site on C3 inhibit human and interactions that regulate important physiological pro murine B cell proliferation (Lernhardt et al., Immunol. cesses, such as the humoral immune response. Defects in receptor structure and function may interfere with Rev., 99:239 (1987)). Thus CR2 receptor can bind both monomeric C3d and aggregated C3d as ligands, as well recognition of ligands by the cell-bound receptors re sulting in disease or dysfunction and death. as the major epitope of EBV capsid protein. Many cell surface receptors and their corresponding ligands have been identi?ed and characterized structur CR2 ligands act in concert with other B cell growth modulators, including growth factors, lymphokines, and cytokines. Thus, the growth-inducing effect of anti-CR2 monoclonal antibody OKB7 is T cellsdepend ent, and requires T cell-derived B cell growth factors only short stretches of amino acids, e.g. at least three 35 (Cooper et al., supra (1988)). It has been shown that amino acids, on the surfaces of mammalian proteins are optimal cell cycle progression and cell division occurs both necessary and suf?cient to bind speci?c receptors only in the presence of both anti-lg antibodies and IL-2 and thereby elicit dramatic cellular responses. or IL-5. Examples of characterized receptors include comple The CR2 receptor is of clinical interest, because it is ment receptor type 2 (CD21 or “CR2") (Weigle et al., 40 the receptor for Epstein-Barr virus (EBV) (Frade supra In Complement, Muller-Eberhard and Miescher, Eds, (1985)). EBV is the causative agent of infectious mono Springer-Verlag, Berlin, p. 323 (1985)). The clonal ex nucleosis (Huang et al., Int. J. Cancer 14:580 (1974)), pansion of mature, antigen-reactive B lymphocytes in and possibly is a human cancer virus, because its pres the humoral immune response is regulated both by di ence is correlated with nasopharyngeal carcinoma and rect intercellular interactions (with T helper lympho 45 Burkitt’s lymphoma (Henle et al., Science, 157:1064 cytes and accessory cells), and by interactions with (1967)). In addition, EBV may play a role in the onset of soluble growth factors (Unanue, Adv. Immunol, 15:95 B cell neoplasia observed in a substantial fraction of (1972)). These soluble factors include B cell growth AIDS patients (Y archoan et al., J. Clin. Invest, 78:439 factors, interleukins, interferons, and components of the (1986)). At the least, a substantial fraction of AIDS complement system (O'Garra et al., Immunol. Today, 50 patients have chronic EBV infections. Exposure of 8:45 (1988); Weigle et al., supra). Members of the latter pregnant women to individuals infected with and shed two classes of proteins bind speci?c cell surface recep ding EBV poses a signi?cant risk to fetal development. tors, including CR2. CR2 is the B lymphocyte receptor It would be useful to better understand the mechanism for the proteolytic activation products C3bi, C3dg and of CR2 ligand action and to design and engineer prote C3d of complement component C3 (Cooper et al., Ann. 55 ins that function as recombinant inhibitors of EBV in fection and lymphoma proliferation. Rev. Immunol, 6:85 (1988); Aggregated C3b and C3d induce B cell proliferation (Erdei et al., Eur. J. Prokaryotic repressors are small, multimeric proteins that are easy to manipulate genetically. Prokaryotic Immunol, 15:184 (1985); Melchers et al., Nature, 317:264 (1985)). CR2 is also the receptor for Epstein repressors bind short stretches of DNA called oper'a Barr virus (EBV), a potent polyclonal B cell activator 60 tors. Aporepressor proteins bind operators poorly and (Fingeroth et al., Proc. Natl. Acad. Sci. USA, 81:4510 must complex with other small molecules called core pressors, such as tryptophan or S-adenosylmethionine (1985); Frade et al., Proc. Natl. Acad. Sci. USA, 82:1490 (1985)). CR2 plays a central role in signalling B cell to form active repressor complexes, or simply, repres proliferation (Cooper et al., supra (1988)). Several sors. Corepressors act as “keystones” that ?t into and monoclonal antibodies and polyclonal anti-CR2 antisera 65 stabilize the hydrophobic cores of their aporepressors. stimulate T cell-dependent B cell proliferation (Cooper The E. Coli Tryptophan (Trp) aporepressor mono et al., supra (1988)). Furthermore, ligand binding to mer is a peptide 108 amino acids long (M,= 12,356 CR2 is necessary for the transition from G1 to S phase daltons) (Gunsalus and Yanofsky, Proc. Nat. Acad. Sci. ally and biochemically. The study of protein-mediated intercellular signalling yields the surprising result that 3 5,190,873 USA, 77:7117-7121 (1980)) that assembles as a dimer 4 (Joachimiak et al., Proc. Natl. Acad. Sci. USA, 80:668-672 (1983); Arvidson et al., J. Biol. Chem, tially disordered, and the residues organized as D in the crystal do not form an a-helix, but rather comprise some sort of surface loop (the “D Loop”) (Arrowsmith 261:238-243 (1986)). Trp aporepressor binds DNA poorly in the absence of the corepressor ligand, L-tryp et al., supra (1990)). The binding of the corepressor, L-tryptophan, restricts the motion of the D loop; amide tophan (or the analog 5-methyltryptophan, S-MT). Ap protons of residues in D become less solvent-accessible in the presence of corepressor. However, corepressor binding does not elicit a coil-to-helix transition, because these protons remain uninvolved in H-bond formation orepressor assembles with tryptophan or S-MT to form active Trp repressor complex, a global repressor that binds operator sites to regulate the initiation of tran scription from at least three different E. coli promoters. 10 in the repressor complex and in the speci?c repressor In addition, aporepressor can form inactive Trp /operator complex [Arrowsrnith and Jardetsky, supra. pseudorepressor complexes with indole-3-propionic Trp aporepressor is usually stable, and may be puri ?ed in large quantities (Arvidson eta1., In Protein Puri acid (IPA) or indole-rr-acrylic acid (IAA); these pseu do-repressor complexes bind operator DNA more poorly than aporepressor (Doolittle and Yanofsky, J. Bacteriol, 95:1283-1294 (1968); Baker and Yanofsky, Proc. Natl. Acad. Sci USA, 60:313-320 (1968)). ?cation: Micro to Macro, UCLA Symp. Mol. Cell Biol (Ed. Burgess), Alan R. Liss, NY; (1986); Smith et al., Proc. Natl. Acad. Sci, USA. 82:6104-6108 (1985)). In addition, to understand how particular amino acids Trp aporepressor controls three operons that com contribute to the structure and function of Trp repres prise a system to maintain the concentration of L-tryp sor, methods have been developed for both mismatch tophan in E. coli homeostatically, within levels neces 20 primer (Arvidson et al., Genetics, 128 (1991) and cas sary for efficient protein synthesis. When concentra sette-style (Pfau and Youderian, Nuc. Acids Res, tions of intracellular tryptophan are low, TrpR exists 18:6165 (1990) mutagenesis of either single or multiple predominantly as an aporepressor that cannot bind trp adjacent codons of TroR. Mutagenesis may be coupled operator DNA, and 'the trpEDCBA biosynthetic genes with a rapid screen for Trp repressor function; this are expressed maximally. When L-tryptophan levels are 25 screen depends on the color of colonies made by a strain high, a substantial fraction of TrpR is active repressor, of bacteria, CGl03, which overproduces Trp repressor. and tryptophan biosynthesis slows (Cohen and Jacob, Comparisons of the NMR structures of Trp apore CR. Acad. Sci. Paris, 248:3490-3492 (1959); Yanofsky, pressor and repressor suggests that the binding of indole J. Amer. Med. Assoc, 218:1026-1035 (1971); Bennet et analogs results in subtle changes in the orientation of D al., Proc. Natl. Acad. Sci. USA, 73:2351-2355 (1976); 30 and E relative to the stationary hydrophobic core of the Zurawski et al., J. Mol. Biol, 145:47-73 (1981); Ya protein. The TrpR dimer has two identical, independent nofsky et al., J. Bacteriol, 158: 1018-1024 (1984)). Trp binding sites for corepressor (Arvidson et al., supra aporepressor regulates a biosynthetic pathway in re (1986); Marmorstein et al., J. Biol. Chem, sponse to the amount of an end product; thus, it func 262:4922-4927 (1987)); surprisingly, these are formed tions as a rheostat, rather than an on/off switch. In 35 by the side chains of residues from both monomers in a contrast, y. and other phage repressors control binary dimer (Schevitz et al., supra (1985)). The interactions that each corepressor is predicted to make with apore developmental decisions, and are not known to respond to small ligands. Other ligand-activated DNA-binding proteins have been studied to lesser extents. The X-ray crystal structures of two forms of Trp repressor (Schevitz et al., Nature, 317:782-786 (1985); Lawson et al., Proteins, 3:18-31 (1988)), aporepressor (Zhang et al., Nature, 327:591-597 (1987)), and pseudorepressor (Lawson et al., Nature, 333:869-871 (1988)) have been determined, and show that, when 45 pressor are primarily hydrophobic. Presumably, the binding of corepressor restricts the ensemble of pre ferred conformations of the DNA-binding domains of an aporepressor to a subset of conformations that bind DNA with lower free energies (pseudorepressor bind ing restricts aporepressor conformations to a subset that binds DNA with higher free energies). crystallized, the peptide monomer is a bundle of six Attempts have been made to construct hybrids be tween structural proteins and receptor binding sites. a-helices with a disordered, ll-residue N-terminal arm. For example, hybrids between proteins having highly The TrpR dimer has a remarkable subunit interface, in repetitive sequences such as silk-like protein (SLP) and which four of each subunit’s six a-interface, helices (A, the ten-residue RGD motif of ?bronectin have been B, C, and F) are interlocked. The amino acid sequence 50 described (Cappello and Crissman, Chemical and Engi of the two ?exible a-helices, D and E, resembles the neering News, pp. 26-32 (July 16, 1990)). Although the conserved “helix-turn-helix” DNA-binding motif char acteristic of many prokaryotic repressors, and pairs of the 2° substructures formed by the D loop and E are hybrid protein is active in vitro, its highly repetitive gene is unstable. It would be advantageous to provide a method for positioned on the surface of Trp repressor to contact 55 producing hybrid proteins containing receptor binding successive major grooves of trp operator DNA. Ge netic analyses of mutant TrpR genes show that residues from both D and E are critical for DNA-binding (Bass et al., Science, 242:240-245 (1988); Kelly et al., Proc. Natl. Acad. Sci USA, 79:3120-3124 (1982)). sites, that are active as ligands for mammalian cell re ceptors to design reagents for a variety of applications including treatment of diseases resulting from recep tor/ligand dysfunction. SUMMARY OF THE INVENTION The present invention provides such method and Recently, Arrowsmith, Jardetsky and coworkers have determined the structure of Trp repressor in solu tion, using lH-NMR spectroscopic methods (Arrow hybrid proteins. The hybrid proteins are produced from smith et al., Biochemistry, 2926332 (1990); Arrowsmith and Jardetsky, submitted for publication (1991). Their a prokaryotic repressor protein and a peptide segment results show that the structure of Trp repressor in solu tion resembles the crystal structures closely, with two major differences. In solution, the ?rst half of A is par 65 that is heterologous to (not naturally present in) the repressor. Preferably, the peptide segment de?nes a binding site from a ligand reactive with a mammalian cell surfacereceptor. Examples of prokaryotic repres 5 5,190,873 sor protein that may be used include TrpR aporepressor 6 DETAILED DESCRIPTION OF THE INVENTION protein, MET aporepressor protein, bacteriophage lambda, Lac repressor, bacteriophage P22 Arc repres sor and the like. The invention also provides DNA DEFINITIONS sequences encoding the amino acid sequences of the invention, vectors containing the DNA sequence and host cells transfected with the vectors. The receptor binding site can be obtained from inter chemical digestion (hydrolysis) of a polypeptide at its peptide linkages. The amino acid residues described Amino Acid Residue: An amino acid formed upon herein are preferably in the “L” isomeric form. How feron alpha, ?brinogen gamma, tenascin, ?bronectin, and the like. Viral immunogenic pathogen derived pro teins, such as proteins from HIV, EBV, hepatitis B and ever, residues in the “D” isomeric form can be substi tuted for any L-amino acid residue, as long as the de the like can also be used. NH; refers to the free amino group present at the amino terminus of a polypeptide. COOI-I refers to the free carboxy group present at the carboxy terminus of a sired functional property is retained by the polypeptide. The invention also provides a method for preparing a hybrid protein prokaryotic repressor protein containing polypeptide. In keeping with standard polypeptide no menclature (described in J. Biol. Chem, 243:3552-59 (1969) and adopted at 37 CPR. l.822(b)(2)), abbrevia at least one substituted binding site from a ligand reac tive with an eukaryotic cell surface receptor by l) mu tating the DNA codons for at least one selected amino acid in the amino acid sequence encoding the prokary tions for amino acid residues are shown in the following otic repressor to encode a contiguous sequence of 20 Table of Correspondence: amino acids that encodes a selected receptor binding TABLE OF CORRESPONDENCE site from a ligand reactive with an eukaryotic cell re SYMBOL ceptor to produce a mutated amino acid sequence; 2 a l-Letter B-Letter AMINO ACID hybrid protein encoded by the mutated DNA sequence that has the activity of the selected receptor binding 25 site; and isolating the hybrid protein produced. The hybrid protein is then puri?ed and tested for biological activity as a binding site for eukaryotic cell receptors. The hybrid protein is produced by recombinant means or by chemical synthesis. A preferred prokaryotic re pressor protein is Trp aporepressor. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram illustrating the Construction of pTrpR-TN Plasmids. Single stranded DNA template 35 was prepared by M13 infection of C1236 transformed with pPYlSO. Site direct mutagenesis was done on sin gle stranded DNA with oligonucleotides (38-45 mer) and transformation into CGl03F.Kan. Tac P/O tac promotor/operator fragment rpoc+: rpoC transcrip Y G F Tyr Gly Phe tyrosine glycine phenylalanine M A S I L T V Met Ala Ser Ile Leu Thr Val methionine alanine serine isoleucine leucine threonine valine proline P Pro K Lys lysine 1-! His histidine Q Gln glutamine E 2 Glu Glx glutamic acid Glu and/or Gln W R D Trp Arg Asp tryptophan arginine aspartic acid N B Asn Asx asparagine Asn and/or Asp tion terminator. C Cys cysteine J Xaa Unknown or other FIG. 2 illustrates cell Attachment to TrpR-TNS. Shown is cell attachment (relative fluorescence) of Di-I stained U251 human glioma cells to wells elated with It should be noted that all amino acid residue sequen increasing amounts of TrpR-TNS as described in Exam 45 ces represented herein by formulae have a left-to-right ple 7. orientation in the conventional direction of amino ter FIG. 3 illustrates cell Attachment to TrpR-TNS and minus to carboxy terminus. In addition, the phrase TrpR-TNI. Shown is cell attachment of U251 cells to “amino acid residue” is broadly de?ned to include the TrpR-TNS, TrpRTNl and TrpR in the presence (+) or 50 amino acids listed in the Table of Correspondence and absence 6(—-) of lmg/ml SRRGDMS peptide as de modi?ed an unusual amino acids, such as those listed in scribed in Example 7. 37 CF. R. l.822(b)(4), and incorporated herein by refer FIG. 4 illustrates peptide Inhibition of Cell Attach ence. Furthermore, it should be noted that a dash at the ment to Trp-TNS. Increasing concentrations of peptide beginning or end of an amino acid residue sequence SRRGDMS-NI-I; and GRGDSP-NHZ were present in 55 indicates a peptide bond to a further sequence of one or cell attachment assays on TrpR-TN5 as described in more amino acid residues or a covalent bond to an Example 7. The control peptide MGSRSRD did not amino-terminal group such as NH; or acetyl or to a inhibit cell attachment. carboxy-terminal group such as COOH. Nucleotide: A monomeric unit of DNA or RNA FIG. 5 illustrates a,- Bg-liposome binding to TrpR TNS. Receptor liposomes were prepared by dialyses of receptor and phosphatidylcholine (PC) plus 3H-PC as described in Example 7. Shown is liposome binding to ?bronectin (FN) vitronectin (VN), TrpR-TNS, and TrpR coated wells. 60 consisting of a sugar moiety (pentose), a phosphate, and a nitrogenous heterocyclic base. The base is. linked to the sugar moiety via the glycosidic carbon (1' carbon of the pentose) and that combination of base and sugar is a nucleoside. When the nucleoside contains a phosphate FIG. 6 illustrates inhibition of Cell Attachment by 65 group bonded to the 3’ or 5' position of the pentose it is referred to as a nucleotide. A sequence of operatively LM609 Anti av [33. Shown is cell attachment in media linked nucleotides is typically referred to herein as a (or monoclonal antibody 81C6 1/200) and in presence “base sequence” or “nucleotide sequence”, and their of LM 609 1/200 as described in Example 7. 7 5,190,873 grammatical equivalents, and is represented herein by a 8 not having a common biological origin, i.e., evolution arily different, are said to be “heterologous”. formula whose left to right orientation is in the conven tional direction of 5’-terminus to 3'-terminus. Vector: a rDNA molecule capable of autonomous replication in a cell and to which a DNA segment, e.g., gene or polynucleotide, can be operatively linked so as Base Pair (bp): A partnership of adenine (A) with thymine (T), or of cytosine (C) with guanine (G) in a double stranded DNA molecule. In RNA, uracil (U) is substituted for thymine. Nucleic Acid: A polymer of nucleotides, either single to bring about replication of the attached segment. Vec tors capable of directing the expression of genes encod ing for one or more proteins are referred to herein as or double stranded. “expression vectors”. Particularly important vectors allow cloning of cDNA (complementary DNA) from mRNAs produced using reverse transcriptase. Polynucleotide: a polymer of single or double stranded nucleotides. As used herein “polynucleotide” and its grammatical equivalents will include the full range of nucleic acids. A polynucleotide will typically Receptor: A receptor is a molecule, such as a protein, glycoprotein and the like, that can speci?cally (non-ran refer to a nucleic acid molecule comprised of a linear domly) bind to another molecule. strand of two or more deoxyribonucleotides and/or 15 Antibody: The term antibody in its various grammati ribonucleotides. The exact size will depend on many factors, which in turn depends on the ultimate condi tions of use, as is well known in the art. The polynucleo tides of the present invention include primers, probes, RNA/DNA segments, oligonucleotides or “oligos” (relatively short polynucleotides), genes, vectors, plas mids, and the like. cal forms is used herein to refer to immunoglobulin molecules and immunologically active portions of im munoglobulin molecules, i.e., molecules that contain an 20 antibody combining site or paratope. Exemplary anti body molecules are intact immunoglobulin molecules, Gene: A nucleic acid whose nucleotide sequence codes for an RNA or polypeptide. A gene can be either RNA or DNA. 25 substantially intact immunoglobulin molecules and por tions of an immunoglobulin molecule, including those portions known in the art as Fab, Fab’, F(ab')z and F(v). Antibody Combining Site: An antibody combining cule comprising two strands of substantially comple site is that structural portion of an antibody molecule comprised of a heavy and light chain variable and mentary polynucleotides held together by one or more hypervariable regions that speci?cally binds (im Duplex DNA: a double-stranded nucleic acid mole munoreacts with) an antigen. The term immunoreact in hydrogen bonds between each of the complementary bases present in a base pair of the duplex. Because the 30 its various forms means specific binding between an antigenic determinant-containing molecule and a mole nucleotides that form a base pair can be either a ribonu cleotide base or a deoxyribonucleotide base, the phrase cule containing an antibody combining site such as a “duplex DNA” refers to either a DNA-DNA duplex whole antibody molecule or a portion thereof. comprising two DNA strands (ds DNA), or an RNA Monoclonal Antibody: The phrase monoclonal anti DNA duplex comprising one DNA and one RNA 35 body in its various grammatical forms refers to a popu strand. lation of antibody molecules that contains only one Complementary Bases: Nucleotides that normally pair up when DNA or RNA adopts a double stranded con?guration. species of antibody combining site capable of im munoreacting with a particular antigen. A monoclonal antibody thus typically displays a single binding affinity Complementary Nucleotide Sequence: A sequence of 40 for any antigen with which it immunoreacts. A mono nucleotides in a single-stranded molecule of DNA or clonal antibody may therefore contain an antibody mol RNA that is sufficiently complementary to that on an other single strand to speci?cally hybridize to it with consequent hydrogen bonding. ecule having a plurality of antibody combining sites, each immunospeci?c for a different antigen, e.g., a bis peci?c monoclonal antibody. Conserved: A nucleotide sequence is conserved with 45 Fusion Protein: A protein comprised of at least two respect to a preselected (reference) sequence if it non polypeptides and a linking sequence to operatively link randomly hybridizes to an exact complement of the the two polypeptides into one continuous polypeptide. preselected sequence. The two polypeptides linked in a fusion protein are Hybridization: The pairing of substantially comple mentary nucleotide sequences (strands of nucleic acid) 50 typically derived from two independent sources, and therefore a fusion protein comprises two linked poly to form a duplex or heteroduplex by the establishment peptides not normally found linked in nature. of hydrogen bonds between complementary base pairs. Upstream: In the direction opposite to the direction It is a speci?c, i.e. non~random, interaction between two of DNA transcription, and therefore going from 5' to 3’ complementary polynucleotides that ca be competi tively inhibited. Nucleotide Analog: A purine or pyrimidine nucleo 55 on the non-coding strand, or 3' to 5' on the mRNA. Downstream: Further along a DNA sequence in the direction of sequence transcription or read out, that is tide that differs structurally from A, T, G, C, or U, but is sufficiently similar to substitute for the normal nucle traveling in a 3’- to 5'-direction along the non’coding otide in a nucleic acid molecule. DNA Homolog: Is a nucleic acid having a prese lected conserved nucleotide sequence and a sequence transcript. strand of the DNA or 5'- to 3’-direction along the RNA coding for a receptor capable of binding a preselected Cistron: Sequence of nucleotides in a DNA molecule coding for an amino acid residue sequence and includ ligand. ing upstream and downstream DNA expression control elements. cule produced by operatively linking two DNA seg 65 Stop Codon: Any of three codons that do not code Recombinant DNA (rDNA) molecule: a DNA mole ments. Thus, a recombinant DNA molecule is a hybrid DNA molecule comprising at least two nucleotide se for an amino acid, but instead cause termination of pro tein synthesis. They are UAG, UAA and UGA and are quences not normally found together in nature. rDNA’s also referred to as a nonsense or termination codon. 5,190,873 The present invention concerns a method for prepar ing hybrid proteins consisting of prokaryotic repressors 10 the formula, read from left to right and in the direction of amino-terminus to carboxyterminus: and receptor binding sites. In the method of the invention, surface features of a stable, prokaryotic repressor, such as Trp aporepressor, are replaced with short peptide segments (oligopep titles) that are active as ligand binding sites for receptors produced by eukaryotic cells, such as antibodies, T-cell receptor, hormone receptors, and the like. The prokary In the formula, Z is an amino-terminal portion of the tryptophan aporepressor, and U is a carboxyterminal portion of the tryptophan aporepressor. The lengths of Z and U are selected (adapted) such that B, the peptide segment, is expressed on the surface of the hybrid pro otic repressor acts as a scaffold, or “context” protein, 10 tein. Factors considered when adapting the length of Z for presentation of the peptide segment. and U are the length in residues of B, whether B is The peptide segment is typically at least 3, and prefer substituted or inserted without substitution into the ably at least 5, and usually no more than about 20, and aporepressor’s native sequence, and in view of forming preferably no more than about 10, amino acid residues hybrid protein whose overall amino acid residue se in length. Usually, the peptide segment will be in the quence length is about 100 to about 120 residues, prefer range of 6 to 9 (inclusive) residues in length. The length ably about l05 to about 115 residues, and more prefera must be suf?cient for the segment to form a binding site bly about 108 to 112 residues. Typically, when the E. such that the hybrid protein mimics at least one biologi coli tryptophan aporepressor is used as the context pro cal activity, as evidenced by binding, of the native pro 20 tein, the hybrid protein length will be in the range of tein ligand from which it is derived. 108 residues because that is the length of the native An exemplary hybrid protein is a tryptophan apore aporepressor. For example, when a peptide segment pressor containing within its amino acid residue se that is 15 residues in length is to be inserted between quence a peptide segment that is heterologous to (not naturally occurring within) the aporepressor. The pep ?anking portions of the E.c0li tryptophan aporepressor tide segment is located within the amino acid residue 25 protein, about 15 residues should be deleted from the native aporepressor sequence to a hybrid protein length sequence of the hybrid protein such that it is expressed equivalent to that of the native aporepressor, i.e. about on the protein’s solvent-accessible surface. Typically, 108 residues. The residues to be deleted are those adja the peptide sequence is ?anked by amino-terminal and cent to the insertion site, i.e. interior residues, while the carboxy-terminal sequences corresponding to respec terminal portions, i.e. the amino- and carboxy- most tive amino-and carboxy-terrninal sequences of the apo (terminal) residues, are left intact. The dashes in the repressor protein. However, the peptide segment can formula represent peptide bonds that link the ?anking also be located at the amino- or carboxyterminus of the sequences to the peptide segment. Preferred hybrid proteins are represented by the amino acid residue se When the peptide segment is inserted or substituted quences SEQ ID NO 1 through SEQ ID NO 19 listed 35 between two ?anking sequences, the amino acid residue below. sequence of the hybrid protein can be represented by hybrid protein. SEQ ID NO (1) MAQQ-B AAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPDEREALGT RVRIVEELLRGEMSQRELKNELGAGIATITRGSNSLKAAPVELRQW LEEVLLKSD (2) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD (3) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRIVEE-B SQRELKNELGAGlATITRGSNSLKAAPVELRQWLEEVLLK SD EREALGTRVRlVEELLRGEM-B ELGAGIATITRGSNSLKAAPVELRQWLEEVLLKSD (4) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD (5) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD (6) EREALGTRVRIVEELLRGEMSQRELK-B ATITRGSNSLKAAPVELRQWLEEVLLKSD EREALGTRVRIVEELLRGEMSQRELKN-B ITRGSNSLKAAPVELRQWLEEVLLKSD MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRIVEELLRGEMSQRELKNELGAGIATITRGSNSLKAA PVELRQWLEEVLLKSD-B (7) LHLPLLNLMLTPDEREA LGTRVRIVEELLRGEMSQRELKNELGAGIATITRGSNSLKAAPVEL RQWLEEVLLKSD (s) MAQQ-B SPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPDERE ALGTRVRIVEELLRGEMSQRELKNELGAGIATITRGSNSLKAAPVE LRQWLEEVLLKSD (9) LHLPLLNLMLTPDEREAL GTRVRIVEELLRGEMSQRELKNELGAGIATITRGSNSLKAAPVELR QWLEEVLLKSD (10) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRIVEELL-B QRELKNELGAGlATlTRGSNSLKAAPVELRQWLEEVLLK SD (11) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRlVEELLRGEMSQ-B 5,190,873 11 -continued SEQ ID NO GIATITRGSNSLKAAPVELRQWLEEVLLKSD (12) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRIVEELLRGEMSQRE-B ATITRGSNSLKAAPVELRQWLEEVLLKSD (13) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRIVEELLRGEMSQRELKNELGAGIATITRGSNSLKAA PVELRQWLEEVLL-B (14) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLH LPLLNLML-B TRVRl VEELLRGEMSQRELKNELGAGIATITRGSNSLKAAPVELRQWLEEV LLKSD (15) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRIVEELLRGEM SQR-B IATITRGSNSLKAAPVELRQWLEEVLLKSD (l6) MAQQS-B AAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPDEREALGTR VRIVEELLRGEMSQRELKNELGAGIATITRGSNSLKAAPVELRQWL EEVLLKSD (17) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRIVEEL-B QRELKNELGAGIATITRGSNSLKAAPVELRQWLEEVLLK SD (18) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRlVEELLRGEMS-B ELGAGIATITRGSNSLKAAPVELRQWLEEVLLK SD ( 19) MAQQSPYSAAMAEQRHQEWLRFVDLLKNAYQNDLHLPLLNLMLTPD EREALGTRVRIVEELLRGEMSQRELKN-B ATITRGSNSLKAAPVELRQWLEEVLLK SD (1985)) and, typically, an additional restriction selection step as described Wells, Philos. Trans. R. Soc. Lond. Ser. A 317:415-423 (1986)). In this procedure, the gene en When the peptide segment is located at the amino- or carboxy-terminus of the hybrid protein, the hybrid pro tein is represented by the formula B-U or Z-B, respec tively, where Z and U are carboxy- or amino-terminal 30 coding the prokaryotic repressor, for example Trp apo repressor, is prepared, for example by introducing aporepressor sequences as previously described, and B is the peptide segment. unique restriction endonuclease sites, and is inserted The phrases “surface accessible” and “surface expres into an appropriate plasmid, such as pPYlSO, a rop't' sion” indicate that the peptide segment is positioned derivative of pBR322, which expresses trpR from the within the primary structure of the hybrid protein so 35 lacUVS promoter (Bass, Science 242:240-245 (1988)) or that it is available for speci?c binding by a receptor pTACTERM, a top pBR322 derivative and site when the hybrid is in a physiologically tolerable aque directed mutagenesis to alter the sequence of the repres ous solution. Windows or zones which are available for sor to encode a desired peptide segment receptor bind surface expression or which are surface accessible in the ing site). This method can be used to make single changes, or to “randomize" a codon by replacing it with the mixed sequence NNS, in which N can be any E. coli tryptophan aporepressor are de?ned by residues 5-, 44-52, 61-67, 67-77, 71-81 and 108 of the native protein sequence as shown in SEQ ID NO 20. of the four DNA nucleotides A, C, G or T and S can be -C or G. This sequence represents all 20 amino acids with 32 codons. the proteins interferon alpha (IFN) and gp350 (EBV coat protein) that interact with one major eukaryotic B 45 Substitutions are made in the prokaryotic repressor lymphocyte surface receptor, CR2, are substituted or using site-directed mutagenesis as described above to inserted as peptide segments into the E. coli Trp apore alter the DNA sequence to replace a stretch of contigu pressor protein to form novel hybrid proteins. ous DNA in the sequence of the repressor with the In another embodiment, the hybrid proteins, at the sequence of at least one selected binding site. Prefera DNA level, are substrates for subsequent site-directed bly, the selected binding site substituted into the prokar In one embodiment of the invention, binding sites on mutagenesis to generate mutant binding sites with al tered biological properties, e.g. greater affinity for re yotic repressor comprises at least three, and preferably ?ve amino acids. The gene encoding the repressor and substituted receptor binding site is then cloned and ceptor, to provide additional hybrid proteins with novel activities. expressed using standard molecular biology procedures. _ Peptide segments forming binding sites can be substi tuted for or inserted into regions of the aporepressor 55 The hybrid protein products are then puri?ed to pro vide a hybrid repressor protein containing the substi regions whose conformation is restricted by corepres sor binding, without disturbing the corepressor binding pocket. Corepressor analogs may be used to modify the conformations of the peptide segments. Thus, several different conformations of each peptide segment in the hybrid Trp aporepressor may be made with different whether the protein binds t cells containing the receptor reactive with the binding site. Hybrid Trp repressor resulting activities. protein containing the tenascin receptor binding site The method of the invention for preparing hybrid tuted receptor binding site or sites. Activity of the resulting hybrid protein is con?rmed using a binding assay appropriate for the ligand being mimicked. Typically, this involves determining Bourdon et al., J. Cell. Biol, 10811149-1155 (1989) as proteins is carried out as follows. Mutagenesis is per— 65 described in Example 7, infra, an assay to determine formed on single or multiple codons of the prokaryotic whether the hybrid protein binds to human tumor cells repressor gene using a procedure such as that described is carried out using the all attachment assay described by Kunkel (Proc. Natl. Acaa’. Sci. USA 82:488-492 by Bourdon et al., supra. 5,190,873 13 - 14 The present invention also contemplates performing further mutagenesis in the receptor binding site sequen gens and immunogenic viral proteins. This favors the induction of cellular immunity which plays a key role in tumor rejection. A suitable virus, for example vaccinia virus derived from a plaque-puri?ed virus of the Wyeth smallpox, vaccine (New York City Board of Health - ces after prior mutagenesis of the repressor protein to provide the binding site sequence in the repressor pro tein to identify hybrid proteins with desired activities. Changes are made in the flanking codons for amino acid residues of the binding site in the hybrid protein to optimize the presentation of the site to its corresponding receptor. In addition, single amino acid changes are made in the binding site to optimize the primary se quence of the site. These will include mutations that increase the speci?c activity of the site as determined by standard binding assays. strain), is used to construct a recombinant virus contain ing the coding sequence of the hybrid protein under control of the vaccinia virus “7.5K” promoter (Hue et al., J. Virol. 62:176-180 (1988)). The recombinant virus may then be administered intravenously as a vaccine to protect against infection. In addition to Trp repressor protein, any prokaryotic repressor proteins may be used as the context protein for expressing, in a conformationally restricted manner, invention may be used as reagents for treatment of or 15 a receptor binding site. For example, the MET repres sor, bacteriophage lambda repressor phage P22Arc introduction into humans to combat infection or disease repressor, and the like, can be used in place of the Trp caused by defects in the interaction of the receptor and repressor to carry out the invention. its ligands. Thus, binding sites for receptors from prote Hybrid proteins produced by the methods of the‘ Receptor binding sites for use in the present invention ins involved in infection, for example from EBV _virus are likewise not limited to those shown in the examples herein, i.e. ?bronectin, tenascin, or IFN . Any binding site that recognizes a cell surface receptor for which the amino acid or DNA sequence is known may be used to substitute or insert into the repressor protein to form a proteins, may be engineered in hybrid proteins using the methods of the invention to combat EBV infection. Such proteins may mimic a compound such as inter feron in vivo to block the binding or subsequent interac tions of ligands to or with the receptors for EBV, pre venting the effects of infection by this virus. 25 Therapeutic applications of the hybrid proteins of the invention are carried out using pharmaceutical compo sitions containing a pharmaceutically effective amount hybrid. The invention described herein further comprises DNA sequences encoding the hybrid monomers previ ously described. As used herein, the term “DNA se quences” encompasses both double-stranded DNA and of the hybrid protein and a pharmaceutically acceptable carrier. The compositions may additionally include 30 single-stranded DNA containing information equiva other reagents for treatment. Such compositions are administered using conventional modes of administra lent to that of the amino acid sequences as determined by the genetic code; such single-stranded sequences can tion including, but not limited to, topical, intravenous, be either in the sense strand orientation or the antisense intraperitoneal, oral and intralymphatic introduction. strand orientation. The hybrid proteins are used in a variety of dosage forms which include, but are not limited to, liquid solu monomer according to the present invention can be tions or suspensions, tablets, pills, powders, supposito ries, polymeric microcapsules or microvesicles, lipo somes, and injectable or infusible solutions. The pre ferred form depends upon the mode of administration 40 and the therapeutic application. The compositions containing the hybrid proteins of A double-stranded DNA sequence encoding a hybrid operatively linked to a transcription-effecting DNA sequence capable of effecting transcription of the DNA sequence encoding the monomer. This transcription effecting sequence is preferably one of Escherichia coli Lac promoter, E. coli trp promoter, bacteriophage lambda P1; promoter, and tac promoter, a hybrid trp-lac the invention also preferably include conventional promoter, but other promoters are known in the art and pharmaceutically acceptable carriers and adjuvants can be used. known in the art such as human serum albumin, ion 45 exchangers, alumina, lecithin, buffer substances such as phosphates, glycine, sorbic acid, potassium sorbate, and - This DNA sequence, comprising a DNA sequence encoding a hybrid monomer operatively linked to a transcription-effecting DNA sequence, can be incorpo salts or electrolytes such as protamine sulfate. The most effective mode of administration and dos rated into a plasmid capable of stably transforming matography, ion exchange chromatography electro phoresis, and the like. Puri?ed hybrid proteins may be ular Cloning Techniques” (Berger & Kimmel, eds., prokaryotic host cells to form a vector. The plasmid age regimen for the compositions of the invention de 50 preferably has both a drug resistance marker and a replication origin. Many suitable plasmids are known in pends on the severity and course of the disease or dys the art, including, but not limited to, the following: functions, the patient’s health and response to treat pBR322 and its derivatives; pUCl8, pUCl9, and their ment, and the judgment of the treating physician. Ac derivatives; bacteriophage u-derived plasmids, and bac cordingly, the dosages of the compositions are titrated 55 teriophage M13-derived plasmids. Suitable host-vector to the individual patient. systems are described, for example, in B. Perbal, “A Quantities of the hybrid proteins of the invention may Practical Guide to Molecular Cloning” (2d ed., John be readily made and puri?ed using additional standard Wiley & Sons, New York, 1988), and “Guide to Molec procedures such as immunoaf?nity, gel exclusion chro used as vaccines to immunize against certain disease. Procedures for preparing such vaccines are known in Academic Press, New York, 1987; Volume 152 of Meth ads in Enzymologv). Preferably, the prokaryotic host cells are E. coli. Prokaryotic, eukaryotic and archaebac terial host cells stably transformed with such a vector the art (see, eg Estin et al., Proc. Natl. Acad. Sci. (USA) are also within the present invention. 85:1052 (1988)). Brie?y, recombinant viruses are con structed for expression of the cloned gene encoding the 65 EXAMPLES hybrid protein. Cells infected with the recombinant The following description provides details of the viruses will express the hybrid protein at the surface of manner in which particular embodiments of the present the cells together with the host’s incompatibility anti 15 5,190,873 invention may be made and used. This description, while exemplary of the present invention, is not to be construed as speci?cally limiting the invention. Varia tions and equivalents, now known or later developed, which would be within the understanding and technical 16 complex. The aporepressor has been crystallized in more than eleven different forms, most of which dif fract clearly to a resolution of less than 2 A. The crystal structure of aporepressor reveals that the protein is comprised of six alpha helices joined by short surface competence of one skilled in this art are to be consid turns. Four of these alpha helices, A,B,C, and F, are involved in the formation of an extraordinary stable hydrophobic core; the aporepressor dimer denatures at>90° C. The ?rst three helices are intertwined such ered as falling within the scope of this invention. 1. Oligonucleotide Design for Producing Trp Aporepressor Hybrid Recombinant DNA that many of the van der Waals interactions that stabi lize the core of the aporepressor dimer are intermolecu lar. In addition the alpha-helix F from one monomer In this invention, surface features of a stable procary otic aporepressor, in this instance E. coli tryptophan aporepressor (TrpR), are replaced with short oligopep makes contacts with both the A and C helices from its tides that are active as ligands or binding sites for recep partner monomer. The extraordinary stability of the tors on eucaryotic cells. The procaryotic aporepressor - 5 core is emphasized by the fact that the coordinates of then acts as a scaffold or context protein to allow pre atoms in these four-helices vary little from crystal struc sentation of the oligopeptide binding site. The nature of ture to crystal structure. the conformation of a receptor binding site can then be In contrast, the coordinates of helices D and E vary evaluated with respect to the speci?city of its interac greatly from structure to structure. When aporepressor tions with its receptor. 20 and repressor are crystallized in the same space group, To achieve this invention, different regions of the the major difference between the two structures is gene encoding TrpR were separately subjected to Oligonucleotide-mediated site-directed mutagenesis as described below to produce receptor binding sites in found in the coordinates of helices D and E, leading to a postulate that these two secondary substructures com prise a “flexible reading head” that can interact with those locations. The nucleotide sequence for a potential 25 DNA when its structure is altered by tryptophan (core integrin binding site derived from human tenascin was pressor) binding. This idea is also a supported by results engineered into eight different regions on the surface of showing that changes at several critical residues in D TrpR. The regions were mutagenized such that a fusion (Lys72) and E (Thr8l) either abolish DNA-binding, or protein having the amino acid residue sequence SRRGDMS (SEQ ID NO 21) flanked by TrpR amino alter the speci?city of the Trp repressor/DNA binding interaction as described by Bass et al., Science, acid residues was produced. Multiple sites on the TrpR gene which is 324 base pairs in length were chosen to provide a reasonable 242:240-245 (1988). Two-dimensional NMR analyses of both aporepressor and repressor indicate that "alpha opportunity to place the engineered recombinant bind X-ray diffraction analysis indicates that D and E form helix D” is not ordered as a helix in solution although ing site in a variety of structural contexts, one or more 35 helix-turn-helix conformation. Both structural models of which may be appropriate for receptor binding in the sites for site-directed mutagenesis was based on the predict that substitutions of a SRRGDMS peptide in between D and E helices will be solvent accessible and unlikely to disrupt the structure of the rest of the TrpR extensive nuclear magnetic resonance (NMR) and protein. fusion protein. The rationale for selecting particular X-ray crystallographic characterization of Trp apore All synthetic oligonucleotides for use in accomplish pressor protein. See Joachimiak et al., J. Biol. Chem, 258:1264l-l2643 (1983); Luisi et al., Biochim. Biophys. Acza., 1048:113-126 (1990); and Arrowsmith et al., Bio ing this invention are listed in Table 1 with the corre sponding SEQ ID NO. The oligonucleotides were syn thesized on an Applied Biosystems DNA synthesizer, model 381A, following the manufacturer’s instructions. chem, 29:6332 (1990). Brie?y, E.c0li Trp aporepressor is a small dimeric 45 The use of these oligonucleotides in the construction of regulatory protein that binds two molecules of the core TrpR binding site recombinants by site-directed muta pressor, L-tryptophan, to form the active repressor genesis is described in Example 2. 55 65 5,190,873 17 18 on : 0 . on ac I 00 mo POP 00 0<0 (CL. E. 00 mm F0 on 00 em 5. P0 F0 no 00 00 560mi0.3uAN:no.25w3cm-0u:.»av5 mZP an <0 00 2. 0P0 3. <0 Na g 21 5,190,873 2. Preparation of a Recombinant TrpR Plasmid TERM was performed following the site-directed mu Expression Vector tagenesis procedure for the construction of hybrid TrpR containing receptor binding sites as described a. Puri?cation of TrpR Gene A 1.3 kilobase insert from E. coli containing the TrpR gene was engineered into a plasmid RPGS, a derivative of pBR322, following procedures described by Gun salus et al., J. BacterioL, 140:106-113 (1979) and Gun salus et al., Proc., Natl. Acad. Sci, 77:7ll7-712l (1980). Brie?y, MC4lOO pheA905 thrzMu c(Ts) [lambdapl(209) 22 In the instant invention, subcloning into pTAC below. For the tenascin construct, a 409 bp Barn HI fragment containing the entire TrpR gene and the re ceptor binding site prepared in Example 3a was sub cloned into a Barn HI linearized pTACTERM vector to form a TrpR expression vector as described in Example O 6. was constructed and a lysate of this lysogen was pre 3. Preparation of TrpR-Heterologous Receptor Binding pared by UV light induction to infect MC4lO0 aroF922 fal thr-90O trpR (lambdacI857S7). Lambda thr+trans ducing phage were recovered and the thr+transduc Site Recombinant pPYlSO Plasmid Vectors a. TrpR Recombinant Hybrids Having an Integrin Receptor Binding Site 1) Tenascin The tenascin-derived peptide sequence SRRGDMS has been shown to be important in mediating cell bind ing to the extracellular matrix protein tenascin Bourdon tants were assayed for anthranilate synthetase activity for measurement of those carrying the trpR gene. Eco RI restriction digestion fragments from the trpR con taining lysogens were ligated into the single Eco RI site of the CM Tc plasmid vector pACYCl 84. The recipient strain for the transformation was W31 l0SRT4 and was 0 et al., J. Cell Biol, 109:317-330 (1989). The peptide is a mutant in trpR. Transformants containing the trpR gene were selected and plasmid DNA was purified following competitive inhibitor of speci?c RGD-dependent re ceptor binding to tenascin and ?bronectin. In addition, antisera speci?c for SRRGDMS block cell binding to tenascin. Oligonucleotide mediated site-directed muta techniques well known to those skilled in the art. Maniatis, Molecular Cloning: A Laboratory Manual, 2 ed, Cold Spring Harbor Laboratory Press, NY, NY. genesis was used to introduce nucleotide sequences (1989). This plasmid was designated pRPGS and con tained the entire TrpR gene. For construction of pPYl5O as described below in which the TrpR gene was mutagenized, pRPGS was first digested with HpaII to form a smaller TrpR gene fragment consisting of 424 coding for the SRRGDMS binding site into the TrpR gene in the pPYlSO expression vector prepared in Ex ample 2b. The SRRGDMS sequence was introduced at 8 sites within the TrpR gene. These sites included the amino terminus, a site between alpha-helix A and alpha-helix B, two sites between alpha-C and alpha-D, three sites in alpha-D, or the alpha-D/alpha-E transition, and at the bp. Bam I-II linkers were constructed on this fragment and then was ligated into a Bam HI linearized pBR322 plasmid. The ligated plasmid containing the smaller TrpR gene was designated pRPG47 and was subse quently digested with Eco RI and Pvu II to form ‘a carboxyl-terminus of TrpR. The oligonucleotides used to encode the binding peptide sequence at each site are shown in Table 1. Three substitutions of the TrpR gene containing fragment. This fragment was ligated to a similarly digested fragment from plasmid pZ152 described Zagursky et al., Gene, 27:183-191 SRRGDMS sequence in the D/E transition designated TN4, 5 and 6 were made. Substitution of residues for the region between Lys72 and Thr8l was performed to (1984) which contains an M13 origin to form a ligated plasmid which contains the TrpR gene and an M13 replace this ?exible portion of aporepressor without origin necessary for subsequent site-directed mutagene disrupting the core structure of TrpR. The amino termi nal and carboxy terminal sites were chosen because they were solvent exposed, disordered structures as revealed sis and helper phage rescue. b. Insertion of the TrpR Gene-Containing Fragment into an Expression Vector l) pPYlSO Expression Vector The plasmid expression vector pPYl50, a rop-l" derivative of pBR322, was used as a recipient cloning 45 by both NMR and X-ray diffraction data which would likely mimic the active linear synthetic peptides. The alpha-B helix is between the highly stable alpha A and alpha-C helices of the hydrophobic core. Substi tution of the SRRGDMS sequence at the beginning of vector for insertion of the 424 bp TrpR gene fragment 50 alpha-B was determined to not be likely to disrupt core structure but its location between the highly stable as described in Example 2a. The resulting ligated plas structures of the alpha-A and alpha-C would define a mid carried the M13 origin from pZ152 and a fusion of ' the PlacUVS promoter to the TrpR structural gene particular set of constrained conformations as a result. The site in alpha-C/alpha-D transition is solvent ex from pRPG47. The expression of the TrpR was driven 55 posed, but due to its proximity to the hydrophobic core, from the lacUV5 promoter. would provide a unique conformational environment 2) pTACTERM Expression Vector for the SRRGDMS peptide. This SRRGDMS substitu tion site would be unlikely to disrupt the core structure In an alternative to expressing TrpR protein in and limit the ability to obtain NMR data. pPYl50, the aporepressor alone or containing a hybrid binding site can be expressed from a rop pBR322 deriv 60 Oligonucleotide-mediated site-directed mutagenesis of TrpR pPYlSO expression vector was performed by ative designated pTACTERM 62 which has a stronger the method described by Kunkel, Proc. Natl Acad. Sci, tac promoter. The advantage of using pTACTERM USA,_ 82:488-492 (1985) using a Muta-Gcne In Vitro over pPYlSO is that plasmids replicate to a higher plas mid copy number due to the absence of the rop gene. When expressed from pTACTERM, wild-type TrpR and mutant aporepressors comprise l to 5% of the total cell protein. Paluh et al., Nuc. Acids Res, 14:7851-7860 (1986). Mutagenesis Kit (Catalog Number 170-3576, Bio-Rad Laboratories, Richmond, Calif). The pPYlSO plasmid containing the TrpR gene prepared in Example 2bl) was transformed using standard procedures well known to one skilled in the art in the bacterial strain C1236 (Bio 23 5,190,873 this bacterial host strain was de?cient for dUTPase and uracil~N-glycosylase which results in an occasional substitution of uracil for thymine in newly synthesized DNA. The resulting DNA, when transformed into a wild-type host with an active uracil-N-glycosylase is inactivated. The uracil-containing DNA was then used as a template for oligonucleotide-mediated site-directed mutagenesis wherein the complementary strand was synthesized using the mutagenic oligonucleotide to 24 ammonium acetate at pH 7.5 and 20% polyethylene glycol (PEG). The resulting admixture was maintained at 25° C. for 15 minutes. The admixture was centrifuged at 11,000X g for 20 minutes to pellet the single-stranded phage present. The resulting supernatant was removed Rad Laboratories). In an overview of the procedure, 10 from the phage pellet. The phage pellet was resuspended in 300 pl of TE buffer consisting of 10 mM Tris-HCl (Tris [hydroxyll aminomethane hydrochloride) pH 8.0 and 1 mM EDTA (ethylene-diamine tetraacetic acid). The result prime DNA synthesis. The resultant double-stranded wild-type/mutant heteroduplex molecule was then ' ing solution was admixed with an equal volume of phe transformed into a host with an active uracil-N-glycosy ferred to a fresh tube. The aqueous phase was reex lase resulting in the inactivation of the uracil-containing wild-type parental strand and the enrichment of the tracted with phenol/chloroform until no signi?cant interface between the organic and aqueous phase was observed. The aqueous phase was then admixed with an nol/chloroform and the resulting aqueous phase trans mutant strand. Brie?y, the uracil-containing single-stranded tem plate required for the site-directed mutagenesis proce dure was prepared ?rst by maintaining a 5 milliliter (ml) culture of XLl—blue cells (Stratagene, La Jolla, Calif.) containing the pPYlSO plasmid at 37° C. for 3 hours. One pl of a solution containing 1X 1011 plaque forming units (pfu) per ml of M13K07 helper phage provided 20 with the kit was then added to the 5 ml culture and the culture was further maintained at 37° C. for 3 hours. 25 equal volume of chloroform and the single-stranded DNA isolated from the aqueous phase by ethanol pre cipitation using ammonium acetate. The amount of single-stranded pPYlSO DNA isolated was determined by gel electrophoresis. This single stranded pPYlSO DNA was then used as a template in the site-directed mutagenesis procedure below. The nucleic acid segment coding for the tenascin receptor binding site, SRRGDMS, (SEQ ID NO 45) During this time, single-stranded M13 phage containing was then inserted into the plasmide pPY150 using the the single-stranded pPY15O DNA were formed and released from the XLl-blue bacteria. The bacteria in this culture were killed by heating the pPY150 single-stranded DNA template prepared above culture to 68° C. for 15 minutes. The killed bacteria 30 were removed from culture, by centrifuging the culture at 10,000Xg to produce a supernatant containing the pPYlSO single-stranded phage. One ml of this superna tant was added to a previously established 5 ml culture and the oligonucleotides, TNl through TN8, listed in Table I in the site directed mutagenesis protocol de scribed in the manufacturer’s instructions provided with the Mutagenesis Kit (Bio-Rad). The nucleic acid seg ment coding for the SRRGDMS was placed at eight different locations within the pPYlSO template DNA. Five ug of the pPYlSO single-stranded template pre of C1236 bacteria that was in the late log phase of 35 pared above were admixed with 50 ng of each of the growth. The culture was maintained at 37° C. for 10 oligonucleotides, TNl through TN8, to form a 10 ul minutes and then 10 rnicroliters (pl) of the culture was admixture. The admixture was maintained for 10 min plated onto LB bacterial plates (LB media contains 10 utes at 68° C. followed by 5 minutes at 25° C. to allow grams/Liter (g/L) of bactotryptone, 5 g/L of yeast the oligonucleotide to anneal to the single-stranded extract, 5 g/L of NaCL containing both 50 pg/ml of 40 template. The complementary strand was then synthe ampicillin and 30 pg/ml of chloramphenicol. The re sized by admixing 4 ul of a solution containing 10 mM of sulting bacterial plates were maintained at 37° C. for adenosine triphosphate (ATP), 4 ul of a buffer contain 12-18 hours to allow individual colonies of CJ236 bac ‘ing 0.66 M Tris-HCl at pH 7.6, 50 mM MgClz, 50 mM teria containing the pPYlSO single-stranded M13 phage dithiothreitol (DTI‘), 2 ul of a solution containing 2.5 to form. 45 mM each of dCTP, dATP, dGTP, and dTTP, 1 ul of T4 One of the resulting colonies was selected and used to DNA ligase (Stratagene) and 1 ul of the Klenow frag initiate a 5 ml bacterial culture in superbroth media ment of DNA Polymerase I (Stratagene) to form a consisting of 35 g/L bactotryptone, 20 g/L yeast ex mutagenesis reaction admixture. The mutagenesis reac tract, 5 g/L sodium chloride at pH 7.5. This 5 ml cul tion was terminated by admixing 1 ul of 50 inM EDTA. ture was maintained at 37° C. for 6 hours and then trans 50 The resultant mutagenized closed circular cDNA was ferred to 150 mls of superbroth in a 250 ml ?ask and then transformed into competent CG103F’ Fan bacteria further maintained at 37° C. for one hour with constant (Bio-Rad) and kanamycin resistant clones containing shaking. Then 100 pl of a solution containing 1X10ll the mutagenized DNA were selected. DNA was then PFU/ml of M13K07 helper phage (Stratagene) was isolated from the resulting transformants using the added to the culture and the culture was maintained at 55 DNA isolation procedures of Sambrook et al., Molecu 37° C. for 12-18 hours with constant shaking. During this time, a culture of C1236 bacteria containing the Iar Cloning: A Laboratory Manual, Second Edition, Cold Spring Harbor Laboratory, Cold Spring Harbor, New pPYlSO containing TrpK gene single-stranded DNA York, N.Y. (1989). The presence of mutations were template was produced. con?rmed by dideoxynucleotide sequencing of the iso This culture was then used to isolate the pPY150 60 lated double stranded DNA using the manufacturer’s single-stranded template using the single-stranded phage DNA isolation procedures described in the p Bluescript II manual (Stratagene). Brie?y, 50 pl of the culture containing pPY150 single-stranded phage pro instructions in the AMV Reverse Transcriptase 35S dATP Sequencing Kit (Stratagene). Following site-directed mutagenesis and veri?cation of the construction of the hybrid TrpR-SRRGDMS duced above was centrifuged at 17,000X g to produce a 65 containing plasmid, the region containing the entire clari?ed supernatant. Approximately 1.2 ml of this su mutagenized TrpR gene was subcloned into a pTAC pernatant was transferred to an eppendorf centrifuge TERM expression vector for expression of the hybrid tube containing 300 pl of a solution consisting of 3.5 M protein as described in Example 6. 5,190,873 25 26 for binding to ?bronectin receptors as described in Ex 2) Fibronectin ample 8. A peptide, GRGDSP (SEQ ID NO 46), derived from 4. Preparation of TrpR-Heterologous Receptor Binding the extracellular matrix protein ?bronectin containing Site Recombinants to Produce Cysteine Disulfide the integrin receptor binding site RGD has been de 5 scribed. Pierschbacher et al., J. Cell Biol, 28:115-126 Loops (1985). The receptors alpha 5 betal, alpha,beta1, alphav betas, for example recognize an RGD-binding site in a. TrpR Recombinant Hybrids Having an lntegrin Receptor Binding Site Containing Cysteine Disul?de Loops matrix ligands. The receptors have overlapping but distinct ligand speci?cities. Peptides from ?bronectin The conformational context of the SRRGDMS (SEQ ID NO 21) or GRGDSP (SEQ ID NO 46) sequence (GRGDSP) or tenascin (SRRGDMS) act as competi tive inhibitors of receptors binding ?bronectin, tenascin and vitronectin and GRGDS (SEQ ID NO 46) peptides within TrpR is further altered by creating cysteine di sul?de bridges to create a loop containing the binding can inhibit tumor metastasis, tumor invasiveness and in site sequence. The rationale for this is several fold. First, vitro angiogenesis. The peptides do not have, however, 15 modeling predictions (Cachau et al., J. Mol. Res, the receptor speci?city of the natural matrix ligand. 22179-186 (1989)) and results with cyclic peptides Conformation must play an integral role in receptor (Pierschbacher et al., J. Biol. Chem, 262:17294-17298 recognition of the RGD binding sites. This has impor (1987)) and a helical dimers (Engel et al., Bic/rem, tant implications since this conformational recognition 30:3161-3169 (1991)) indicate that the most active con can result in very distinct cellular responses. Fibronec tin is a strong cell adhesion and spreading factor, while tenascin prompts reduces cell adhesion and increased motility but not spreading. The two proteins contrast and complement each other in modulating adhesion and cell motility. The GRGDSP sequence is placed at the same sites as formations are likely to be contained in a turn structures or loops. Second, the snake venom disintegrin proteins have the RGD site ?anked by cysteines (Gould et al., P.S.E.B.M., 195:168-171 (1990)). Third, the active TrpR-TNS hybrid protein produced in Example 6 is 25 likely to reside within an extended or loop structure. Finally, cysteine disul?de bridges constrain the binding with the tenascin peptide in TrpR as described in Exam sequence in ways not possible through the TrpR con text alone. By selecting the ?anking distance from the ple 3al). Oligonucleotide-mediated site-directed muta genesis is performed as described using the ?bronectin binding-site encoding oligonucleotides listed in Table l. The resulting TrpR-GRGDSP hybrid constructs are binding sequence, loops of various sizes and varying then subcloned as described in Example 6 into a pTAC ial cyclic peptides. However, unlike cyclic peptides the conformation contexts are generated. The formation of disul?de bridges is analogous to the formation of artific TERM expression vector for expression of the hybrid NH; and COOH protein context is not lost. This strategy is most applicable to TrpR-hybrids at for binding to ?bronectin receptors as described for 35 the NH; and COOH-termini (TrpR-TNl, TrpR-TN7) protein. The expressed hybrid protein is then assayed tenascin receptors in Example 7. and D/E transition (TrpR-TN4, TNS, and TN6), be b. TrpR Recombinant Hybrids Having a Lymphocyte cause these sites are ?exible regions of Trp aporepressor and changes in these regions are least likely to disrupt the hydrophobic core. Core disruption would make Receptor Binding Site, CR2 1) Alpha Interferon (IFNa) NMR spectroscopy and X-ray crystallographic analysis exceedingly difficult. Because both TrpR-Tn] (NH; The peptide QLNDLEACV (SEQ ID NO 47) is the receptor binding site on IFNa which is recognized by the B lymphocyte receptor called CR2. Oligonucleo terminal) and TrpR-TNS (D/E) are both active recep tor binding proteins, cysteine bridges in these mutants tide-mediated site-directed mutagenesis was performed are generated ?rst. Cysteine pairs ?anking the as described using the interferon binding-site encoding 45 SRRGDMS sequence are introduced into the TrpR oligonucleotides listed in Table 1. The resulting TrpR TNl and TNS by a single round of site-directed muta genesis as described in Example 3 using paired oligonu cleotides coding for a cysteines ?anking the GRGDSP hybrid constructs were then subcloned as described into a pTACTERM expression vector as described in Example 6 for expression of the hybrid SRRGDMS substitution site. The oligonucleotides used protein. The expressed hybrid protein was then assayed 50 for generating hybrid TrpR-SRRGDMS fusion prote ins having disul?de loops are listed in Table 2 below. TABLE 2 CYS-CYS LOOPS: SSL 1A(TN1) (SEQ ID NO 48) l 2 3 M A C ATT ATG GCC CAA TGC _ S TCA R CGC R CGC G GGA C TGC l0 A GCA D GGC M ATG S TCA D GAC M ATG 81 S C T TCC TGC ACG l1 M ATG 12 A GCA SSL IBaCI'Nl) (SEQ ID NO 49) 1 CGA CAT A'I'T 3 4 M C Q Q S R ATG TGC CAA CAA TCA CGC C TGC l2 A GCA l3 E GAA l4 Q CAG 15 R CGT C TGC S TCA R CGC R CGC G GGA SSL] BbCI‘Nl) (SEQ ID NO 50) 8 M S ATG TCA 9 A GCA l0 A GCG SSL 5A(TN5) (SEQ ID NO 51) 68 69 Q R CAG CGT 70 E GAG 71 L TTA SSLS Bb(TN5) (SEQ ID NO 52) 82 83 l T ATI' ACG 8 CG