Background of the Invention The three-dimensional structure of a protein is determined by its amino acid sequence. However, previous attempts to predict and design protein structures have not been successful [D. Jones and J. Thornton, J. Computer-Aided Mol. Design, 7: 439 (1993) ; J. W. Bryson et al., Science, 270: 935 (1995)]. Comparisons among mutant proteins suggest that not all residues contribute equally to three-dimensional structure [M. H. Hecht et al., Proc. Natl. Acad. Sci. USA, 80: 2676 (1983); B. W.

Matthews, Biochemistry, 26: 6885 (1987); J. F. Reidhaar-Olson and R. T. Sauer, Science, 241: 53 (1988); D. D. Loeb et al., Nature, 340: 397 (1989); D. W. Heinz et al., Proc. Natl. Acad. Sci. USA, 89: 3751 (1992); M. E. Milla et al., Nature Structural Biol., 1: 518 (1994).]. Protein native structures are stabilized primarily by the aversion of hydrophobic side chains to polar solvent [W. Kauzmann, Adv. Protein Chem., 14 : 1 (1959); C. Chothia, Nature, 248: 338 (1974); F. M. Richards, Annu.

Rev. Biophys. Bioeng, 6 : 151 (1977); K. A. Dill, Biochemistry, 24: 1501 (1985); K.

A. Dill, Biochemistry. 29: 7133 (1990); J. R. Livingstone et al., Biochemistrv, 30 : 4237 (1991) ; K. A. Sharp et al., Biochemistry, 30 : 9686 (1991) ; C. N. Pace, J. Mol.

Biol., 226: 29 (1992)]. This aversion, which is referred to as the hydrophobic effect, distributes the hydrophobic and polar residues in the protein interior and on the surface, respectively, such that the pattern of hydrophobic and polar residues in the linear sequence of a protein is a significant determinant of three-dimensional structure [Jones and Thornton, cited above; Bryson, cited above]. A second important determinant of structure is amino acid helical propensity, which reflects the entropic cost of incorporating a residue into an ordered secondary structure element [P. C. Lyu

et al., Science, 250: 669 (1990); K. T. O'Neil and W. F. DeGrado, Science, 250: 646 (1990); S. Padmanabhan et al., Nature, 344: 268 (1990) ; T. P. Creamer and G. D.

Rose, Proc. Natl. Acad. Sci. USA, 89: 5937 (1992)]. Protein design efforts, sequence analysis of structurally related and unrelated proteins and theoretical models of proteins consisting of only hydrophobic and polar residues suggest that a number of protein conformations are compatible with a specific hydrophobic/polar and helical propensity profile [Jones and Thornton, cited above; Bryson, cited above; K. A. Dill et al., Protein Science, 4 : 561 (1995)]. A unique structure is therefore attained by additional determinants, for example, in a few native proteins and in simple designed proteins by incorporating binding sites for metal ions or other prosthetic groups [J.

Miller et al., EMBO J., 4: 1609 (1985); H. Paulsen et al., Eur. J. Biochem., 215: 809 (1993) ; T. M. Handel et al., Science, 261: 879 (1993); D. E. Robertson et al., Nature, 368: 425 (1994)]. In the majority of proteins, hydrogen bonds and electrostatic interactions have been proposed to stabilize unique native structures [Jones and Thornton, cited above; Bryson et al., cited above; D. F. Stickle et al., J. Mol. Biol., 226: 1143 (1992); Z. S. Hendsch and B. Tidor, Protein Science, 3: 211 (1994)]. Yet, polar and charged residues are not highly conserved in evolution [D. J. Barlow and J.

M. Thornton, J. Mol. Biol., 168: 867 (1983)] and in at least one example all of the buried salt bridges can be eliminated with no effect on structure [C. D. Waldburger et al., Nature Structural Biol., 2 : 122 (1995)]. Since the determinants that allow proteins to adopt unique well-packed structures are not understood, it has been virtually impossible to modify the three-dimensional structure of a protein such that it exhibits novel properties. Thus, even though it is well-appreciated that the amino acid sequence of a protein determines its three-dimensional structure, no amino acid substitutions have been described that change the three-dimensional structure of a protein.

What is needed are methods for altering protein structures to provide designed proteins. Desirably, these designed proteins may have novel functions, or a change in their functional properties relative to the native protein from which they were derived.

Preferably, these designed proteins retain the function of the native protein from

which they are derived, but have some other advantage, such as enhanced stability, improved binding, lower molecular mass, or the like.

Summary of the Invention In one aspect, the present invention provides a method for altering the three-dimensional structure of proteins. The method involves the steps of : identifying a native protein with three-dimensional structure to be altered; identifying the hydrophobic residues within this protein; distinguishing the hydrophobic residues on the basis of side chain size into large and small hydrophobic residues; providing mutants of the protein having substitutions in the hydrophobic residues distinguished above; and assaying the mutant proteins for a switch in three-dimensional structure.

In another aspect, the method of the invention further involves the steps of generating mutants of the selected protein having at least one of the large hydrophobic residues within its hydrophobic core substituted by hydrophobic residues with small side chains and assaying the mutant proteins for ability to at least maintain the biological function of the native protein. Optionally, the method of the invention involves substituting at least one of the small hydrophobic residues within the hydrophobic core of the protein with large hydrophobic residues.

In yet another aspect, the present invention provides an altered protein comprising p53 oligomerization domain altered according to the above method, wherein the side chain size of the hydrophobic amino acid with the largest side chain in the p53 protein hydrophobic core (Phe341) has been decreased and the side chain of another hydrophobic amino acid in the hydrophobic core (Leu344) has been increased.

In another aspect, the invention provides an altered p53 protein oligomerization domain designed as described above, which has further been modified to contain a Lys at amino acid position 340.

In still another aspect, the invention provides a p53 protein containing the altered p53 oligomerization domains as described above in place of the native p53 oligomerization domain.

In a further aspect, the invention provides p53 fusion proteins comprising an altered p53 oligomerization domain fused to a heterologous protein.

In still a further aspect, the present invention provides nucleic acid sequences encoding the altered proteins according to the present invention.

In yet a further aspect, the invention provides vectors comprising nucleic acid sequences of the invention under the control of suitable regulatory sequences.

In another aspect, the invention provides host cells transformed with the vectors of the invention. Also provided are pharmaceutical compositions containing the nucleic acid sequence of the invention and method of administering same.

Other aspects and advantages of the present invention are described further in the following detailed description of the preferred embodiments thereof.

Brief Description of the Drawings Fig. 1 illustrates the three-dimensional structure of the p53wt oligomerization domain corresponding to residues 325-355 of SEQ ID NO: 2. Residue type is indicated by the single letter code: F, Phe; L, Leu.

Fig. 2 illustrates the three-dimensional structure of the p53KIY oligomerization domain corresponding to residues 327-353 of human p53 [SEQ ID NO: 2]. Residue type is indicated by the single letter code: Y, Tyr; I, Ile.

Fig. 3A illustrates the packing of residues with large hydrophobic side chains, specifically, Phe341 in p53wt [SEQ ID NO: 2]. The a-helix of one of the subunits is not shown for clarity, also only two of the four subunits are shown. Residue type is abbreviated: L, Leu; F, Phe; N, Asn.

Fig. 3B illustrates the packing of Tyr344 in p53KIY [SEQ ID NO: 4].

Residue type is abbreviated: L, Leu; Y, Tyr; N, Asn.

Detailed Description of the Invention This invention provides a general method for changing the three-dimensional structure of proteins. The invention further provides modified proteins generated by this method and nucleic acid sequences encoding them. These modified proteins and

nucleic acid sequences are particularly useful in pharmaceutical compositions and therapeutic regimens, and also in biotechnology and other industrial applications.

Suitably, the altered proteins of the invention, have an altered three-dimensional structure and are further characterized by maintaining substantially the same level of a desired biological function of the native protein. However, the altered proteins of the invention advantageously can be designed or selected such that they are also characterized by improved functional properties relative to the native (wild-type) domain, e. g., improved binding ability, improved stability, or the like.

The methods, altered proteins, nucleic acids, and pharmaceutical compositions of the invention are discussed in more detail below.

I. Method For Altering Three-Dimensional Structure of Proteins In one aspect, the present invention provides a method for altering the three-dimensional structure of proteins without denaturing the proteins. The examples provided herein demonstrate alteration of the three-dimensional structure of the oligomerization domain of wild-type p53. However, other proteins may be readily altered. Desirably, the method of the invention is performed upon a protein that has a hydrophobic core (i. e., upon essentially all proteins). Most preferably, however, the protein is useful for industrial, therapeutic or diagnostic purposes. Included among desirable proteins for application of the method of the invention are non-linear, proteins including, without limitation, industrial enzymes, such as proteases, lipases, chymases, etc., and therapeutically useful proteins, such as the members of the globin family, in which the side chain sizes of specific hydrophobic residues are highly conserved [D. Bashford et al., J. Mol. Biol., 196: 199 (1987)].

Desirably, the three-dimensional (or crystalline) structure of the protein to be altered is known. Alternatively, the three-dimensional structure may be determined using known techniques, e. g., NMR spectroscopy, x-ray crystallography and the like.

Once selected, the hydrophobic residues within this protein are identified and distinguished into large and small hydrophobic residues on the basis of

side chain size. For the purposes of this specification, amino acid residues Phe, Tyr and Trp are considered large and amino acid residues Ala, Val and Ile are considered small. Leu is intermediate in size and may be considered either large or small.

More particularly, for the performance of the method of the invention, not only is the total surface area, but also the surface area buried upon folding with is relevant, as the latter determines the strength of the hydrophobic effect [J. R.

Livingstone et al., Biochemistry, 30 : 4237-4244 (1991)]. By way of example, the total surface areas of residues [G. D. Rose el al., Science, 229: 834-838 (1985)] and the surface areas buried upon folding are provided in the table below. All values are in square Angstroms Amino acid Total Surface Area Surface Area Buried Upon Folding Alanine 118 80-86 Valine 164 127-141 Isoleucine 181 154-158 Leucine 193 156-164 Tyrosine 236 137-177 Phenylalanine 222 168-194 Tryptophan 266 177-224 Once so distinguished, mutant proteins are generated by increasing the size of one or more of the small residues and/or decreasing the size of one or more of the large residues. Since Leu cannot be unambiguously classified as large or small, mutants are generated which replace Leu with both larger and smaller amino acids.

As a first step, it is preferable to decrease the size of the largest residue in the hydrophobic core (e. g., Phe is substituted with Ile, Val or Ala) and increase the size of one of the smaller residues in the hydrophobic core (e. g., Val is substituted with Phe or Trp). All substitutions are made in such a manner as to preserve the hydrophobic character of the native residue. In general, one or two substitutions are made for each secondary structure element whose packing in the three-dimensional structure of the

protein is to be altered. When two substitutions are performed, one decreases the side chain size of a large hydrophobic amino acid and the other increases the side chain size of a small hydrophobic amino acid.

The mutant proteins may be generated using conventional techniques.

For example, the peptides may be synthesized using a commercially available automatic synthesizer according to standard procedures. Alternatively, other standard techniques may be utilized. See, e. g., Merrifield, J. Amer. Chem. Soc., 85: 2149-2154 (1963). Preferably, however, the mutant proteins of the invention are generated recombinantly, making use of a variety of well-known techniques (such as site- directed mutagenesis [see, Gillman & Smith, Gene, 8: 81-97 (1979) and S. Roberts et al, Nature, 328: 731-734 (1987)]) and, desirably, the nucleic acid sequences of the invention. See, e. g., Sambrook et al., Molecular Cloning. A Laboratory Manual., 2d Edit., Cold Spring Harbor Laboratory, New York (1989).

The mutant proteins generated by the method of the invention may be assayed for biological function as a preliminary screening step. In this manner, mutant proteins having the desired biological function (e. g., binding ability) may be selected.

The assays for examining the function of the protein will depend on the particular function that needs to be assayed. Such assays are well known to those of skilled in the art and are not a limitation on this invention. For example, if the protein altered according to the invention is a DNA binding protein, then one could use a DNA binding assay to examine the function of the mutant proteins. If engineering involves a protein that forms oligomers, e. g., the oligomerization domain of p53, then one could use a glutaraldehyde crosslinking assay [W. Lee et al, Nature Structural Biology, 1: 877 (1994)] to examine the function of the mutant proteins. Other suitable functional assays will be readily apparent to one of skill in the art, based on the function of the native protein which has been altered by the method of the invention.

The mutant proteins generated according to the method of the invention may then be screened for a change in the three-dimensional structure.

Biophysical methods to probe protein structure include NMR spectroscopy, X-ray

crystallography, [G. M. Clore et al., Science, 265: 386 (1994); W. Lee et al., Nature Structural Biol., 1 : 877 (1994); G. M. Clore et al., Nature Structural Biol., 2: 321 (1995); P. D. Jeffrey et al., Science, 267: 1498 (1995)], among other techniques. The applications of a DNA binding assay, a glutaraldehyde crosslinking assay and NMR spectroscopy are illustrated in the examples of engineered p53 oligomerization domains appropriate for determining three-dimensional conformation.

In a currently preferred embodiment, the p53 oligomerization domain has been altered according to the method of the invention. Thus, the method of the invention may involve generating mutant of fragments of a useful protein which is responsible for biological activity. As described above, similar mutants may be generated using the hydrophobic core of other selected proteins, or full-length proteins, as desired.

II. Altered Profeins of the Invelltion Using the method of the invention, the inventor has altered the oligomerization domain of p53 to provide a p53 protein with an altered three-dimensional structure and oligomerization stoichiometry relative to the native (wild-type) domain. All references to p53 residue numbers herein refer to the numbering scheme provided by Zakut-Houri et al., EMBO J, 4 : 1251-1255 (1985) [GenBank Code Hsp53] for human p53. The nucleotide and amino acid sequences of human p53 are reproduced as SEQ ID NOS : 1 and 2, respectively. Although reference is made by way of example to human p53, one of skill in the art could readily substitute other non-human p53 sequences. Alignment of the highly conserved p53 sequences is provided in Soussi eí al., Oncogene, 5: 945-952 (1990).

The p53 tumor suppressor protein is a sequence-specific transcription factor with fundamental significance to the pathogenesis and therapy of human cancer [C. C. Harris, Science, 262: 1980 (1993); L. J. Ko and C. Prives, Genes Dev., 10 : 1054 (1996)]. The tumor suppressor activity of p53 requires homo-oligomerization [M. J. F. Waterman et crl., Cancer Res., 56: 158 (1996)], which is mediated by a thirty residue domain at the C-terminus of the protein [H. Sakamoto 6/cl/., Proc. Natl.

Acad. Sci. USA, 91: 8974 (1994); P. Wang et al., Mol. Cell. Biol., 14: 5182 (1994); J. L. F. Waterman etal., EMBO J., 14 : 512-519 (1995)]. The native thirty residue p53 oligomerization domain has a 13-strand, a tight turn and an a-helix in each subunit. [Clore et al., cited above (1994); Lee et al., cited above; Clore et al., cited above, (1995); Jeffrey et al., cited above]. This oligomerization domain folds independently with antiparallel packing of its a-helices.

In a preferred embodiment, the altered p53 generated according to the invention contains substitutions of residues Phe341 and Leu344 [SEQ ID NO: 2] in the a-helix by other hydrophobic amino acids, that decreased the side chain size at position 341 and increased the side chain size at position 344 [corresponding to SEQ ID NO: 2], resulting in an altered p53 domain that assembles as a dimer instead of a tetramer. The three-dimensional structure of a mutant dimeric domain determined in solution by NMR spectroscopy differs substantially from the wild-type structure, since the a-helices are packed parallel, rather than antiparallel, and are rotated significantly relative to each other and to the ß-strands. The engineered p53 oligomerization domain was found to drive the sequence-specific DNA binding function of the modified p53 protein.

Thus, in another aspect, the present invention provides altered p53 oligomerization domains that assemble as dimers, rather than tetramers, and which have a different three-dimensional structure relative to wild-type p53. These altered p53 oligomerization domains of the invention [SEQ ID NO: 3] desirably contain the following residues, which differ from the residues in native human p53 [SEQ ID NO: 2].

Position 341 Position 344 Val Phe Val Trp Val Tyr Ile Phe Ile Trp Ile Tyr Leu Phe Leu Trp Leu Tyr ---Trp Optionally, the altered p53 oligomerization domains described above may be further modified to contain Lys at position 340 [SEQ ID NO: 4]. The inventor has found that the latter modification increases the solubility of the mutant p53 oligomerization domains with altered three-dimensional structure. For example, at a 1-2 mM concentration a mutant p53 domain with Ile at position 341 and Tyr at position 344 [SEQ ID NO: 3] precipitates within one hour when heated to 40°C. In contrast, a mutant p53 domain [SEQ ID NO: 4] with Lys at position 340, Ile at position 341 and Tyr at position 344 remains soluble under the same conditions.

Based on the functional activities of other dimeric p53 proteins [constructed by replacing the native p53 oligomerization domain with a leucine zipper, which is a dimerization domain found in certain nuclear DNA binding proteins; Pietenpol et al., Proc. Natl. Acad. Sci. USA, 91: 1998 (1996)], it is anticipated that the altered p53 proteins described herein will retain their tumor suppressor function.

Because these p53 proteins are dimers, rather than tetramers, their molecular mass is half of that of wild-type p53 and therefore have certain advantages over wild-type p53. For example, they can be introduced more easily into cells. In addition, because the altered p53 oligomerization domain has a different three-dimensional structure than the wild-type p53 domain, the two types of domains will not hetero-oligomerize.

Accordingly, a p53 tumor suppressor protein will not be sequestered into inactive hetero-oligomers with tumor-derived p53 mutant proteins. Thus, a p53 protein of the invention can be delivered by gene therapy vectors and not be suppressed by the mutant p53 protein present in tumor cells.

As desired, the altered p53 oligomerization domains of the invention may be engineered on an otherwise unmodified p53wt protein. The altered p53 oligomerization domain may also be fused to a selected heterologous protein.

Alternatively, the altered p53 oligomerization domains may be engineered on p53 proteins bearing additional modifications. One suitable modification is substitution of residue threonine 284 [SEQ ID NO: 2] with Arginine. This substitution enhances the tumor suppressor function of wild-type p53 5-to 7-fold [Wieczorek et al, Nature Medicine, 2 : 1143 (1996)].

Suitable heterologous proteins include those which in the past have been fused to a leucine zipper. Leucine zippers have the disadvantage that they may interact with host leucine-zipper bearing proteins, which interaction may compromise the biological activity of the chimeric protein bearing the leucine zipper. The altered p53 oligomerization domains described in this application do not exist in nature, and do not form oligomers with any host proteins. However, because both leucine zippers and the altered p53 oligomerization domain of the invention have similar topologies, i. e., both types of domains have parallel a-helices, they may be used in many similar applications.

An example of suitable heterologous proteins include single chain antibody variable chains (scFv antibodies). The prior art has described scFv antibodies fused to Jun and Fos leucine zippers to produce dimeric antibodies that have higher affinity for their ligands, because they are bivalent [Kruif and Logtenberg, J. Biol. Chem., 271: 7630 (1996)]. According to the present invention, scFv antibodies can be fused to the altered p53 oligomerization domain of the invention by using the sequences encoding the altered p53 oligomerization domain in place of the leucine zipper sequences of the prior art. This will lead to homodimeric (hence monospecific), bivalent (high affinity) antibodies.

Another example of a suitable heterologous protein includes a soluble interleukin-2 (IL-2) receptor. Wu et al, J. Biol. Chem., 270: 16039 (1995) has described a soluble IL-2 receptor complex formed by attaching leucine zippers to the C-terminus of the extracellular domain of the receptor. According to the present invention, the leucine zipper domain of the prior art is replaced by the altered p53 oligomerization domain of the invention. Such soluble domains are useful for screening ligands (drugs) that bind to the native receptors. Such soluble domains are also useful therapeutically as decoys competing for ligand binding with the endogenous receptors of patients. The methods for attaching the altered p53 oligomerization domain are readily apparent to those of skill in the art. For example, the extracellular domains of the T-cell receptor can be isolated in a soluble form, then fused to the modified p53 domains, using a modified version of the technique described in Chang et al., Proc. Natl. Acad. Sci. USA, 91: 11408 (1994). Where desired or needed, heterodimers can be isolated from homodimers by conventional protein chromatography or other suitable techniques.

As another example, the altered p53 oligomerization domains of the invention may be fused to a transmembrane receptor. Many such receptors become physiologically activated by ligand-induced dimerization. Fusion of a dimerization domain to such receptors can therefore constitutively activate them. Such an example is provided by the Trp-Met fusion receptor, in which the Trp protein provides a leucine zipper which induces dimerization and activation of the Met receptor [Rodriques and Park, Mol. Cell Biol., 13: 6711 (1993)]. Similarly, the altered p53 oligomerization domains of the invention may be used to activate a receptor of choice.

Depending upon the type of receptor to which the modified p53 oligomerization domain of the invention is fused, the outcome could be cell proliferation or cell death.

The altered p53 oligomerization domain of the invention can also be used to induce dimerization of DNA binding proteins. Many DNA binding proteins, for example c-Myc, bind DNA as dimers. c-Myc will not homodimerize, but will bind DNA as a hetero-dimer with a protein called Max. However, c-Myc will homo- dimerize if its native leucine zipper is replaced by the leucine zipper of GCN4, since

the latter zipper has a high tendency to homo-oligomerize. A c-Myc fusion bearing a GCN4 leucine zipper binds DNA with a very high affinity [Halazonetis and Kandil, Science, 255: 464 (1992)]. Thus, a c-Myc protein whose native C-terminus is fused to the modified p53 oligomerization domain of this invention, would bind to DNA with high affinity and could compete for DNA binding of the native Myc/Max heterodimer without interfering with native proteins which contain leucine zippers. Furthermore, if the chimeric Myc protein lacked the N-terminal domain of p53, which is required for carcinogenic transformation of cells overexpressing c-Myc, then it can be used to revert the tumorigenic phenotype of cells overexpressing c-Myc, such as many leukemia and lymphom cells. Other proteins that bind DNA and are implicated in cancer development, such as EWS-ATF-1 [Fujimura et al., Oncogene, 12 : 159 (1996)] and the E2A-HLF [Yoshihara et al., Mol. Cell. Biol., 15: 3247 (1995)], can be similarly modified.

As described above, altered p53 produced according to the method of the invention is used by way of example only. Other proteins (e. g., enzymes, antibodies and members of the globin family) may be similarly altered and used for a variety of purposes, as described herein.

III. Nucleic Acid Sequences The present invention further provides nucleic acid sequences encoding the altered proteins of this invention. In addition to the coding strand, the nucleic acid sequences of the invention include the complementary DNA sequence representing the non-coding strand, the messenger RNA sequence, the corresponding cDNA sequence and the RNA sequence complementary to the messenger RNA sequence. Variants of these nucleic acids of the invention include variations due to the degeneracy of the genetic code and are encompassed by this invention. Such variants may be readily identified and/or constructed by one of skill in the art. In certain cases specific codon usage may be employed to optimize expression. The above nucleotide sequences can be included within larger DNA or RNA fragments, or may be interrupted by introns.

A. Expression Vectors In another embodiment, the nucleic acids encoding the proteins of the invention are present in the context of vectors suitable for amplification in prokaryotic or eukaryotic cells. Many such vectors are known and many of these are commercially available. For example plasmids with bacterial or yeast replication origins allow amplification in bacteria or yeast, respectively. Such vectors allow the production of large quantities of nucleic acids encoding the proteins of the invention, which nucleic acids can be used for gene therapy or for expression of the proteins of the invention, e. g., p53.

In yet another embodiment the nucleic acids encoding the proteins of the invention are present in the context of vectors suitable for expression in cell-free extracts or lysates or in prokaryotic or eukaryotic cells. Many such vectors are known [Ausubel et al, Current Protocols in Molecular Biology, Greene Publishing Associates and John Wiley & Sons, NY (1997)] and many of these are commercially available. For example, the vector pGEM4 (Promega, Madison, WI) is suitable for expression of the proteins in cell-free lysats, while the vector pSV2 [ATCC] is suitable for expression in mammalian cells. Such vectors allow the production of the proteins of the invention in vitro for analysis of their functional properties or for delivery to patients.

B. Gene Therapy Vectors The nucleic acid sequences of the invention may be inserted into a vector capable of targeting and infecting a desired cell, either in vivo or ex vivo for gene therapy, and causing the encoded modified protein of this invention to be expressed by that cell. Many such viral vectors are useful for this purpose, e. g., adenoviruses, retroviruses and adeno-associated viruses (AAV) [Schreiber et al., Biotechniques, 14: 818-823 (1993); Davidson et al., Nature Genetics, 3: 219-223 (1993); Roessler et al., J. Clin. Invest., 92: 1085-1092 (1993); Smythe et al., Ann.

Thorac. Surs.. 57 : 1395-1401 (1994); Kaplitt et al., Nature Genetics, 8: 148-154 (1994)]. There has already been success using viral vectors driving expression of

wild-type p53 [Fujiwara et al., Cancer Res., 53: 4129-4133 (1993); Fujiwara et al., Cancer Res., 54: 2287-2291 (1994) ; Friedmann, Cancer, 70 (6 Suppl) : 1810-1817 (1992); Fujiwara et al., Curr. Opin. Oncol., 6: 96-105 (1994b); Roth et al., Nature Medicine, 2: 985-991 (1996)].

For use in gene therapy, these viral vectors containing nucleic acid sequences encoding a protein of the invention, e. g., an altered p53 protein, are prepared by one of skill in the art with resort to conventional techniques (see references mentioned above). For example, a recombinant viral vector, e. g. an adenovirus, of the present invention comprises DNA of at least that portion of the viral genome which is capable of infecting the target cells operatively linked to the nucleic acid sequences of the invention. By"infection"is generally meant the process by which a virus transfers genetic material to its host or target cell. Preferably, the virus used in the construction of a vector of the invention is rendered replication-defective to remove the effects of viral replication on the target cells. In such cases, the replication-defective viral genome can be packaged by a helper virus in association with conventional techniques.

Briefly, the vector (s) containing the nucleic acids encoding an altered protein of the invention is suspended in a pharmaceutically acceptable carrier, such as saline, and administered parenterally (or by other suitable means) in sufficient amounts to infect the desired cells and provide sufficient levels of modified protein to achieve the desired therapeutic or prophylactic effect, e. g., sufficient p53 activity to arrest abnormal cellular proliferation. Other pharmaceutically acceptable carriers are well known to those of skill in the art. A suitable amount of the vector containing the chimeric nucleic acid sequences is between about 106 to 109 infectious particles per mL carrier. The delivery of the vector may be repeated as needed to sustain satisfactory levels of biological activity. For example, where modified p53 is administered, activity may be determined by monitoring clinical symptoms.

As desired, this therapy may be combined with other therapies for the disease or condition being treated. For example, therapy involving the administration of a vector capable of expressing an altered p53 protein of the

invention is well suited for use in conjunction with conventional cancer therapies, including surgery, radiation and chemotherapy.

Nucleic acid sequences driving expression of a protein of the invention may also be introduced by"carriers"other than viral vectors, such as liposomes, nucleic acid-coated gold beads or can simply be injected in situ [Fujiwara et al (1994b), cited above; Fynan et al., Proc. Natl. Acad. Sci. USA, 90: 11478-11482 (1993); Cohen, Science, 259: 1691-1692 (1993); Wolffet al., Biotechniques, 11 : 474-485 (1991)].

IV. Pharmaceutical Compositions The altered proteins and nucleic acid sequences of this invention may also be formulated into pharmaceutical compositions and administered using a therapeutic regimen compatible with the particular formulation. When administered in the form of nucleic acid sequences, the composition may contain"naked"DNA, or a vector containing the nucleic acid sequences. As used herein, the term"naked DNA" means substantially pure DNA which is not associated with a protein, lipid, carbohydrate or contained within a cell or an artificial delivery system such as a liposome.

Pharmaceutical compositions within the scope of the present invention include compositions containing an altered protein of the invention (or a nucleic acid sequence encoding a modified protein) in an effective amount to have the desired physiological effect, e. g. to arrest the growth of cancer cells without causing unacceptable toxicity for the patient.

Suitable formulations for parenteral administration include aqueous solutions of the active compounds in water-soluble or water-dispersible form, e. g. saline. Alternatively, suspensions of the active compounds may be administered in suitable conventional lipophilic carriers or in liposomes.

The compositions may be supplemented by active pharmaceutical ingredients, where desired. Optional antibacterial, antiseptic, and antioxidant agents in the compositions can perform their ordinary functions. The pharmaceutical

compositions of the invention may further contain any of a number of suitable viscosity enhancers, stabilizers, excipients and auxiliaries which facilitate processing of the active compounds into preparations that can be used pharmaceutically. Preferably, these preparations, as well as those preparations discussed below, are designed for parenteral administration. However, compositions designed for oral or rectal administration are also considered to fall within the scope of the present invention.

Those of skill in the pharmaceutical art should be able to derive suitable dosages and schedules of administration. As used herein, the terms"suitable amount"or"effective amount"means an amount which is effective to treat the conditions referred to below. A preferred dose of a pharmaceutical composition containing a protein of this invention is generally effective above about 0.1 mg modified protein per kg of body weight (mg/kg), and preferably from about 1 mg/kg to about 100 mg/kg. These doses may be administered with a frequency necessary to achieve and maintain satisfactory activity levels. Although a preferred range has been described above, determination of the effective amounts for treatment of each type of tumor or other condition may be determined by those of skill in the art.

Dosage units of such pharmaceutical compositions containing the proteins of this invention preferably contain about 1 mg to 5 g of the protein.

V. Therapeutic Indications The nucleic acids encoding altered p53 proteins and the altered p53 proteins themselves can be introduced into human patients for therapeutic benefits in conditions characterized by insufficient wild-type p53 activity. Such conditions have been described in the art. See, e. g., PCT/US95/15353 (June 6,1996). For example, the pharmaceutical compositions of the invention, including the gene therapy vectors, may be employed to induce the cellular defense to DNA damaging agents such as sunlight UV irradiation, as well as radiation and chemotherapeutics used for cancer treatment. The therapeutic indications include inducing apoptosis of specific cells,

such as proliferating lymphocytes, the prevention of transplant rejection, and the treatment of autoimmune diseases, e. g., systemic lupus erythrematosis, rheumatoid arthritis and the like.

The pharmaceutical compositions of this invention may also be employed to restore p53 function in tumor cells and to suppress cell proliferation in diseases other than cancers, which are characterized by aberrant cell proliferation.

Among such diseases are included psoriasis, atherosclerosis and arterial restenosis.

Pharmaceutical compositions containing other altered proteins of the invention (or nucleic acids encoding them) may also be readily prepared and used for a variety of indications which will be readily apparent to one of skill in the art.

VI. Antibodies The altered proteins of the invention are useful for generating antibodies, which may be used as diagnostic reagents, for example, to monitor the presence of modified protein or protein domain.

Specific antisera may be generated using known techniques. See, Sambrook, cited above, Chapter 18, generally, incorporated by reference. Similarly, antibodies of the invention, both polyclonal and monoclonal, may be produced by conventional methods, including the Kohler and Milstein hybridoma technique, recombinant techniques, such as described by Huse et al, Science, 246: 1275-1281 (1988), or any other techniques known to the art.

The invention further encompasses functional fragments of the antibodies of the invention, including, Fab, Fv, and F (ab') 2 fragments, the binding site of the antibodies, and the complementarity determining regions (CDRs). Optionally, the binding site and/or CDRs may be contained in a synthetic molecule which provides antibody framework regions. Further, these functional fragments may be used in the production of recombinant antibodies, including bifunctional antibodies, chimeric antibodies, and humanized antibodies, which preferably retain the antigen binding specificity of the antibodies of the invention. Such recombinant antibodies may be constructed and produced according to known techniques [see, e. g., S. D. Gillies et

al, J. Immunol. Meth., 125: 191-202 (1989); and G. E. Mark and E. A. Pladlan, "Humanization of Monoclonal Antibodies", The Handbook of Experimental Pharmacology, Vol. 113, Chapter 4, pp. 105-133, Springer-Verlag (June, 1994)].

These functional fragments and recombinant antibodies may be used for a variety of purposes, including any of those described herein for the antibodies of the invention.

In general, polyclonal antisera, monoclonal antibodies and other antibodies which bind to an altered protein as antigen are useful as research tools, as diagnostic reagents, as therapeutic agents, and for producing other antibodies (as described above) which are similarly useful.

VII. Diagnostic Reagents The altered proteins of the invention may be used therapeutically or as diagnostic reagents. These reagents may optionally be labeled using diagnostic labels, such as radioactive labels, colorimetric enzyme label systems and the like conventionally used in diagnostic or therapeutic methods. Alternatively, the N-or C- terminus of an altered protein of the invention may be tagged with a detectable label which can be recognized by a specific antisera. For example, the reagents derived from p53 may be used in diagnosis of a variety of conditions associated with p53 and/or aberrant cell proliferation, including autoimmune diseases, e. g., systemic lupus erythrematosis, rheumatoid arthritis and the like, cancers, psoriasis, atherosclerosis and arterial restenosis. For example, in tissue biopsies, the presence of p53 could be directly verified by RT-PCR or immunostaining. Reagents produced from other altered proteins of the invention, e. g., antibodies and the like, may similarly be utilized as diagnostic reagents. The selection of the appropriate assay format and label system is within the skill of the art and may readily be chosen without requiring additional explanation by resort to the wealth of art in the diagnostic area.

These examples illustrate the method of the invention as performed in connection with p53 and the preparation of modified p53 proteins of the invention.

These examples are illustrative only and do not limit the scope of the invention.

Example 1-Method for altering the three-dimensional structure of wild-type p53 Wild-type p53 was identified as the native protein with three-dimensional structure to be altered, according to the method of the invention.

The three-dimensional structure of this protein was studied by NMR spectroscopy and X-ray crystallography [G. M. Clore et al., Science, 265: 386 (1994); W. Lee et al., Nature Structural Biol., 1: 877 (1994); G. M. Clore et al., Nature Structural Biol., 2: 321 (1995); P. D. Jeffrey et al., Science, 267: 1498 (1995)].

Using these techniques, wtp53 [SEQ ID NO: 2] was found to have a structure which consists of a B-strand, a tight turn and an a-helix. Four identical subunits assemble as a dimer of dimers (Fig. 1). In the primary dimer, the f3-strands form an antiparallel 13-sheet and the a-helices also pack antiparallel. Two primary dimers form a tetramer by packing their a-helices at an 81 ° angle.

According to the method of the invention, the residues of the hydrophobic core were distinguished into large and small hydrophobic residues. The center of the hydrophobic core of the domain is formed primarily by residues Phe341 and Leu344, Phe341 is positioned at the interface of the two p53 monomers that form the primary dimers, whereas Leu344 forms the interface between the primary dimers [Fig. 1; SEQ ID NO: 2].

Mutants of p53 were then generated according to the method of the invention and assayed for a switch in conformation. Plasmids encoding mutant p53 proteins were generated by PCR-directed mutagenesis of pGEMhump53wtB, as described in Waterman et al., EMBO J., 14: 512-519 (1995) which is incorporated by reference herein. The names of the mutants indicate the hydrophobic residues at positions 341 and 344 [SEQ ID NO: 3], respectively, using the single letter amino acid code. For example, p53FL is wild-type p53 [SEQ ID NO: 2]. Additional mutants made include: p53A344, Ala344 [SEQ ID NO: 3]; p53IF, Ile341 and Phe344 [SEQ ID NO: 3]; p53IY, Ile341 and Tyr344; p53KIY, Lys340, Ile341 and Tyr344 [SEQ ID NO: 4]; and p53K340, Lys340 [SEQ ID NO: 4].

Once generated, DNA binding of the mutants was assayed using 32P-labeled oligonucleotide BC. S10 and in vitro translated p53 (Waterman et al., cited

above)]. DNA binding is an indirect measure of the function of the p53 oligomerization domain. This assay was performed to screen the mutants whose function is similar to wtp53, as a preliminary to assaying for conformational switch.

Table 1 shows the effect of amino acid substitutions targeting residues 341 and 344 of human p53 on the subunit stoichiometry of p53 as assayed by the electrophoretic migration of its complexes with DNA. The names of the mutants [SEQ ID NO: 3] indicate the hydrophobic residues at positions 341 and 344, respectively, using the single letter amino acid code. p53FL is wild-type p53 [SEQ ID NO: 2].

Table 1 Protein Subunit Stoichiometry p53VI......... tetramer p53VL......... tetramer p53VF......... dimer p53VW dimer p53II tetramer p53IL......... tetramer p53IF......... dimer p53IW......... dimer p53LI......... tetramer p53LL......... tetramer p53LF......... dimer/tetramer mixture p53LW......... dimer p53FI......... tetramer p53FL......... tetramer p53FF......... tetramer p53FW......... dimer

A. DNA Binding Assay Full-length p53 proteins with Phe341 substituted for Val, Ile or Leu, and Leu344 for Ile, Phe or Trp, and all combinations thereof, retain their ability to bind DNA (Table 1) [SEQ ID NO: 3]. Since monomeric full-length p53 does not bind DNA [M. J. F. Waterman et al., Cancer Res., 56: 158 (1996; T. D. Halazonetis et al., EMBO J., 12: 5057 (1993); P. Hainaut et al., Oncogene, 9: 299 (1994)], these substitutions do not denature the oligomerization domain. Some complexes of the mutant proteins with DNA migrate in gel electrophoresis faster than those with the wild-type protein, a behavior that in general correlated with the side chain size of residue 344 being greater than that of residue 341 (Table 1). Electrophoretic migration on native gels depends on molecular size [J. L. Hedrick and A. J. Smith, Archs. Biochem. Biophys., 126: 155 (1968)], suggesting an effect of the substitutions on oligomerization stoichiometry. Indeed, the DNA complex of p53IF [SEQ ID NO: 3] comigrated with that of p53 Ala344, a known dimeric mutant [J. L. F. Waterman and T. D. Halazonetis, EMBO J, 14: 512-519 (1995)]. Furthermore, DNA binding of p53IF [SEQ ID NO: 3] was unaffected by substitutions of Met 340 with Lys and/or of Phe344 with Tyr. Since these residues are at the dimer-dimer interface in wild-type p53 (Fig. 1), tolerance to the substitutions in p53IF [SEQ ID NO: 3] is consistent with it not having a dimer-dimer interface. In contrast, substitution of Met340 with Lys dissociated wild-type p53 into dimers.

B. Assay for Stoichiometry Since DNA binding is only an indirect measure of the function of the p53 oligomerization domain, certain mutants whose complexes with DNA migrated faster than wild-type p53 were selected for further study. Wild-type p53 (p53wt; SEQ ID NO: 2) and these mutant p53 oligomerization domains [SEQ ID NOS: 3 and 4] were expressed in E coli, purified to homogeneity and assayed for subunit stoichiometry by glutaraldehyde crosslinking using the techniques described below.

1. Expressiorr and Purification A PCR fragment with 5'Bam HI and 3'Hind III restriction sites encoding a methionine and amino acids 304-363 of human wild-type p53 [SEQ ID NO: 2] was cloned in the vector pTST [S. P. Eisenberg et al., Nature, 343: 341 (1990)]. Equivalent plasmids were constructed for the mutants. Proteins were expressed in E. coli (BL21) grown at 30°C in labeled or isotopically labeled 'SN-or'SN-and'3C-labeled minimal media (Isotec, Miamisburg, OH). After overnight induction with IPTG [Sigma, St. Louis, MO, the cells were pelleted, stirred on ice for 20 min. in glycerol, 0.7% v/v Triton-X and 0.4% v/v 13-mercaptoethanol, and then for another 15 min. in lysis buffer (10 mM Tris [pH 8. 0], 500 mM NaCl, 5 mM EDTA, 1 mM DTT, 0.6 mg/ml lysozyme) containing Pefabloc SC (Pentapharm, Basel, Switzerland) and pepstatin (Sigma, St. Louis, MO). 120 U/ml DNase I (Sigma) was added and stirring continued for another 30 min. The lysate was cleared by centrifugation at 300,000 g for 70 min., adjusted to 1. 5 M ammonium sulfate and cleared again by centrifugation at 300,000 for 15 min. p53 was purified in three steps.

In a first step, the p53 was run on a 50 ml Phenyl-Sepharose column (Pharmacia, Piscataway, NJ) equilibrated with 1.5 M ammonium sulfate, 50 mM sodium phosphate [pH 7.0], 5 mM EDTA and eluted by decreasing salt concentration on a GradiFrac system (Pharmacia). In a second step, the eluate from the first step was run on an 8 ml Phenyl-Superose column (Pharmacia) equilibrated with 1.7 M ammonium sulfate, 50 mM sodium phosphate [pH 7.0], 5mM EDTA and eluted by decreasing salt concentration on a SMART system (Pharmacia). In a third step, the eluate from the second step was run on a 1 ml Resource S column (Pharmacia) equilibrated with 50 mM sodium phosphate [pH 7.0], 50 mM NaCl, 0.01 mM EDTA and eluted by increasing NaCI concentration on the SMART system. The eluate from the third step, purified p53, was adjusted to 50 mM sodium phosphate [pH 7.0], 50 mM NaCl, 0.1 mM deuterated EDTA (Isotec), 0.1 mM sodium azide by gel filtration on a Sephadex G-50 column (NAP, Pharmacia) and concentrated to 0.25 ml by ultrafiltration (Amicon, Beverly, MA) to a final concentration of about 2 mM

2. Glcrtaraldehyde Crosslislking Mutants p53Ile341Tyr344 (p53IY) [SEQ ID NO: 3] and p53Lys340Ile341 Tyr344 (p53KIY; SEQ ID NO : 4) were studied, because they function like p53IF [SEQ ID NO: 3] in the DNA binding assay, but are less hydrophobic and therefore less likely to aggregate non-specifically in concentrated samples.

Concentrations (0.1 mM protein) of the p53 oligomerization domains, purified as described above, were incubated with 0.1% v/v glutaraldehyde (Sigma, St. Louis, MO) for 15 min. at 37°C in a buffer containing 200 mM sodium phosphate and 50 mM Tris-HCl [pH 7.5] and immediately analyzed by SDS-PAGE.

Table 2 illustrates subunit stoichiometry of p53wt [SEQ ID NO: 2], p53IY (Ile341 and Tyr344) [SEQ ID NO: 2] and p53KIY (Ile341 and Tyr344) [SEQ ID NO: 4] oligomerization domains as determined by glutaraldehyde (G1.) crosslinking and SDS-gel electrophoresis.

TABLE 2 Protein Subunit Stoichiometry p53wt Tetramer p53IY Dimer p53KIY Dimer The wild-type domain crosslinked into tetramers, confirming its known tetrameric subunit stoichiometry. In contrast, p53IY [SEQ ID NO: 3] and p53KIY [SEQ ID NO: 4] formed only dimers. These dimers were quite stable, since the efficiency of crosslinking was undiminished at temperatures as high as 55°C.

Thus, the oligomerization stoichiometries of p53wt [SEQ ID NO: 2] and the p53 mutants [SEQ ID NOS: 3 and 4] were found to correlate with the relative sizes of their hydrophobic side chains at positions 341 and 344.

Example 2-NMR Structure Determination Multidimensional solution NMR spectroscopy was performed on uniformly 15N-and 15N, l3C-labeled samples to determine the structural basis for the switch in oligomerization stoichiometry observed in the mutant proteins as follows.

At high concentration and temperature, p53IY [SEQ ID NO: 3] aggregated in less than one hour. In contrast, p53KIY [SEQ ID NO: 4] (and p53wt; SEQ ID NO: 2) could be maintained at 40°C at high concentrations for months without aggregation. Detailed structural analysis of p53 KIY was then pursued.

All experiments were performed at 40°C on two Bruker DMX instruments operating at 750 and 600 MHz and the spectra processed using FELIX (Biosym Inc., San Diego, CA). Backbone sequential assignments were determined from HNCA, HNCOCA, CBCACONH and HNCACB 3D triple ('H,"C,"N) resonance and TOCSY-HMQC and NOESY-HMQC 3D double ('H,"N) resonance experiments [D. Marion et al., Biochemistry, 28: 6150 (1989); S. Grzesiek and A.

Bax, J. Magn. Reson., 96: 432 (1992); S. Grzesiek and A. Bax, J. Am. Chem. Soc., 114: 6291 (1992); M. Wittekind and L. Mueller, J. Magn. Reson., 101B : 201 (1993); F. Lohr and H. Ruterjans, J. Biomolec NMR, 5: 25 (1995)]. Side chain assignments were made using 2D COSY, 2D TOCSY, 3D TOCSY-HMQC and 3D HCCH-TOCSY experiments [R. R. Ernst et al., Principles of Nuclear Magnetic Resonance in One and Two Dimensions (Oxford University Press, New York, 1986); A. Bax et al., J. Magn. Reson., 88: 425 (1990)]. NOE restraints were obtained from 2D NOESY (unlabeled sample in D20), 3D"N-edited NOESY-HMQC ("N-labeled sample) and'3C-edited HSQC-NOESY (l5N/l3C-labeled sample) experiments [A.

Majumdar and E. R. P. Zuiderweg, J. Magn. Reson., 102B: 242 (1993)]. Inter-and intrasubunit NOEs were differentiated with l2C-filtered/l3C-edited and l3C-filtered/l3C- edited HMQC-NOESY-HMQC experiments performed on an equilibrated 1: 1 mixture of 12C-and 13 C-labeled samples [W. Lee et al., FEBS Letters, 350: 87 (1994)].

Coupling constants were obtained from HMQC-J, HNHA and HNCA-J measurements

and hydrogen bonds from proton-deuterium exchange [K. Wuthrich, NMR of proteins and nucleic acids (Wiley, New York, 1986), G. Vuister and A. Bax, J. Am. Chem.

Soc., 115: 7777 (1993)].

The backbone'H and"N amide resonance frequencies of residues Gly325, Leu330, Gln331, Arg333, Arg337, Phe338, Glu339, Glu343, Asn345, Glu346, Ala347 and Lys351 [corresponding to SEQ ID NO: 2] differed by more than 0.4 ppm in the proton frequency and/or by more than 1.2 ppm in the nitrogen frequency between p53KIY [SEQ ID NO: 4] and p53wt [SEQ ID NO: 2] (Table 3).

Such large chemical shift differences are suggestive of changes in three-dimensional structure [K. Wuthrich, cited above], especially since some residues, such as Gln331 and Arg333 are 10 and 7A, respectively, from the nearest substituted residue in the established p53wt structure [G. M. Clore et a/., Science, 265: 386 (1994); W. Lee et al., Nature Structural Biol., 1 : 877 (1994); G. M. Clore et al., Nature Structural Biol., 2: 321 (1995); P. D. Jeffrey et al., Science, 267 : 1498 (1995).

Using amide resonance frequencies (two-dimensional HSQC spectra) as monitors of three-dimensional structure [K. Wuthrich, cited above], verification that this pattern of chemical shift changes was due to the amino acids at positions 341 and 344 was obtained by studying p53IF [SEQ ID NO: 3], p53Lys340Ile341Phe344 (p53KIF; SEQ ID NO: 4), p53K340 [SEQ ID NO: 4] and p53Phe344 (p53FF ; SEQ ID NO: 3) proteins. The first three mutant proteins were examined as crude bacterial lysates [A. M. Gronenborn and G. M. Clore, Protein Science, 5: 174 (1996)], whereas p53FF [SEQ ID NO: 3] was purified. A number of well-resolved resonances in a small region of the HSQC spectra, corresponding to the 13-strand, indicated that the shifts in resonance frequencies characteristic of p53KIY [SEQ ID NO: 4] occurred in p53KIF and p53IF [SEQ ID NO: 3] (mutants with a larger hydrophobic side chain at position 344 relative to 341), but not in p53K340 [SEQ ID NO: 4] or in p53FF [SEQ ID NO: 3] (Table 3) which did not differ from p53wt [SEQ ID NO: 2].

Table 3 illustrates shifts in the amide resonance frequencies of Gln331 and Arg333 by amino acid substitutions targeting Phe341 and Leu344 related to

p53wt [SEQ ID NO: 2]. Proteins are labelled as in Tables 1 and 2. P53K340 [SEQ ID NO: 4] has Lys at position 340. P53KIF [SEQ ID NO: 4] has Lys340, live341 and Phe344.

Table 3 Shift > 1. 5 ppm Shift > 0.5 ppm in the 15N amide in the'H amide Protein frequency of Gln331 frequency of Arg333 p53wt No No p53KIY Yes Yes p53KIF Yes Yes p53IF Yes Not determined p53K340 No No p53FF No No While proteins with either Tyr or Phe at position 344 yield similar spectra [SEQ ID NO: 3], the substitution of Met340 for Lys [SEQ ID NO: 4], useful in preventing aggregation, does not affect the spectra. A double amino acid substitution involving residues 341 and 344 was required to elicit changes in resonance frequencies, since p53FF [SEQ ID NO: 3], which contains a substitution of only Leu344 for Phe, displays HSQC spectra very similar to p53wt [SEQ ID NO: 2] (Table 3). Notably, in the DNA binding assays, p53FF [SEQ ID NO: 3] and p53wt [SEQ ID NO: 2] are tetramers, whereas p53IF [SEQ ID NO: 3], p53KIY [SEQ ID NO: 4], and p53KIF [SEQ ID NO: 4] are dimers (Table 2).

To monitor intersubunit NOEs an equilibrated 1 : 1 mixture Of 15N-and 15N, 13C-labeled p53KIY polypeptides [SEQ ID NO: 3] was subjected to a'2C- filtered,'3C-edited HMQC-NOESY-HMQC experiment. The intersubunit contacts involving residues Thr329 and Leu330 were consistent with an antiparallel B-sheet, as in p53wt [SEQ ID NO: 2]. However, the short-range contacts involving residues Ile341, Ala347, Leu348 and Leu350 suggested parallel packing of the regions

corresponding to residues 341 to 350 [SEQ ID NO: 2]. This was inconsistent with antiparallel packing of the a-helices as observed in the primary dimer of p53wt [SEQ ID NO: 2] (Fig. 1).

The surprising magnitude of structural change was confirmed in the well-defined three-dimensional structure for p53KIY [SEQ ID NO: 4] (Table 4) calculated from experimentally derived distance, dihedral angle and hydrogen bond restraints as follows. The structures were folded from both random and distance geometry embedded starting points using simulated annealing minimization, a total of 678 experimental distance, dihedral angle and hydrogen bond restraints for residues 327-353 [SEQ ID NOS: 2 and 3] and a non-crystallographic symmetry term [A. T.

Brunger, X-PLOR. A system for X-ray crstallography and NMR (Yale University Press, New Haven and London, 1992)]. NOE intensities were classified as strong, medium and weak, corresponding to distance restraints of 1.8-3.2,1.8-4.0 and 1.8-5.0 A, respectively. Dihedral angles were restrained to-14060° or to 6055'four 3JHNHa measurements of >8. 5 or <5 Hz, respectively. Hydrogen bond restraints were incorporated as two NOEs restraining O-NH to 1.7-2.3 A and N-O to 2.8-3.3 A. Pseudo atom restraints were used whenever stereospecific assignments could not be made. The average structure was calculated from thirty simulated annealing structures and was refined using restrained minimization and a repulsive term to stimulate the van der Waal's potential [Brunger, cited above]. Geometry was evaluated with PROCHECK and PROMOTIF [R. A. Laskowski et al., J. Appl. Cryst., 26: 283 (1993)].

The results are provided in Table 4. In this table, <SA> is the ensemble of the 30 final simulated annealing structures and (SA) r is the refined average structure. The number of restraints is given in parentheses and applies to the entire dimer. The Lennard-Jones Energy was calculated using the CHARMM22 force field [B. R. Brooks et al, J. Comput. Chem., 4: 187 (1983)].

Table 4-Structure statistics RMSD from Experimental Distance Restraints (A)' <SA> (SA) r Intrasubunit (384) Intraresidue (122) 0.001+0.0003 0. 000 Sequential (120) 0.0530.012 0.097 Short Range (124) 0.031+0.013 0.035 Long Range (18) 0. 064+0. 020 0.061 Intersubunit (200) 0.044+0.007 0.077 Hydrogen Bonds (38) 0. 073+0. 011 0.100 RMSD from Experimental Angle Restraints () Dihedral Angles (56) 0.62+0.21 0.87 RMSD from Idealized Geometry Bonds (A) 0.0028+0.0001 0.0062 Angles (°) 0 55+0. 026 0.69 Impropers (°) 0. 47+0. 078 0.66 Lennard-Jones Energy (kcal/mol)-407+23-437 Atomic RMSD from Average (A) Backbone 0.51 All Atoms 1.41 Core Sidechains 1.05 'None of the simulated annealing structures was found to have distance violations greater than 0.3 A or dihedral angle violations greater than 3 °

Example 3-Three-Dimensional Structure The p53KIY oligomerization domain [SEQ ID NO: 4] is a dimer with two-fold cyclic symmetry. Each subunit consists of three secondary structure elements: a B-strand forms an antiparallel 8-sheet and the two a-helices pack parallel to each other (Fig. 2). Parallel packing of the a-helices is stabilized by hydrophobic interactions substantially involving Tyr344, which interacts with lie341 of the same subunit and Tyr344 of the other (Fig. 2).

The structural switch between p53KIY [SEQ ID NO: 4] and p53wt [SEQ ID NO: 2] can be evaluated by comparing their structures (Figs. 1 and 2). Such comparison reveals differences in the orientation of the secondary structure elements.

The interhelical angle changes from 155'ion the primary dimer of p53wt to 83'ion p53KIY [SEQ ID NO: 4], similar to the 81 ° angle with which the a-helices pack across primary dimers in p53wt.

The switch between two well-defined structures observed here for the p53 oligomerization domain is remarkable. p53wt [SEQ ID NO: 2] and p53KIY [SEQ ID NO: 4] differ by only three amino acid substitutions. The changes in the sizes of the side chains at positions 341 and 344 appear to be necessary and sufficient for the structural switch.

The effect of changing the side chain sizes of residues 341 and 344 on protein three-dimensional structure may be attributed to the hydrophobic effect [W.

Kauzmann, Adv. Protein Chem., 14 : 1 (1959); C. Chothia, Nature, 248: 338 (1974); F. M. Richards, Annu. Rev. Biophys. Bioeng 6 : 151 (1977); K. A. Dill, Biochemistry 24: 1501 (1985); K. A. Dill, Biochemistry, 29: 7133 (1990); J. R. Livingstone et al., Biochemistry, 30 : 4237 (1991) ; K. A. Sharp et al., Biochemistry, 30 : 9686 (1991) ; C.

N. Pace, J. Mol. Biol., 226: 29 (1992). In p53wt, Phe341 is protected from exposure to solvent by being in the protein interior, where it interacts with multiple amino acids.

In p53KIY [SEQ ID NO: 3], the decrease in side chain size of residue 341 results in fewer interactions being required to shield its surface from solvent, while the increase in side chain size of residue 344 requires new interactions. The switch in three dimensional structure between p53wt [SEQ ID NO: 2] and p53KIY [SEQ ID NO: 4]

can be explained in terms of the loss of contacts that bury the side chain of residue 341 and gain in contacts that bury the side chain of residue 344. Specifically in p53wt, the tip of the Phe341 ring lies in a hydrophobic pocket formed by the side chains of Leu344, Asn345 and Leu348, all from the other subunit (Fig. 3A). In p53KIY, Tyr344 not Ile341, interacts with Tyr344, Asn345 and Leu348, from the other subunit (Fig. 3B). The intersubunit interaction between Leu348 and Phe341 in p53wt [SEQ ID NO: 2] stabilizes antiparallel packing of the a-helices, whereas in p53KIY [SEQ ID NO: 4] the intersubunit interaction between Leu348 and Tyr344 stabilizes parallel packing.

The change in subunit stoichiometry between the p53wt [SEQ ID NO: 2] and p53KIY [SEQ ID NO: 4] oligomerization domains is probably secondary to the altered packing of the a-helices. In p53wt, residues Leu344 form a hydrophobic patch for assembly of two primary dimers into a tetramer (Fig. 1). In p53KIY [SEQ ID NO: 4], Tyr 344 is involved in parallel packing of the a-helices and does not allow two dimers to form a tetramer.

All references referred to above are incorporated by reference herein.

Numerous modifications and variations of the present invention are included in the above-identified specification and are expected to be obvious to one of skill in the art.

Such modifications and alterations to the compositions and processes of the present invention are believed to be encompassed in the scope of the claims appended hereto.

SEQUENCE LISTING (1) GENERAL INFORMATION: (i) APPLICANT: Wistar Institute of Anatomy &, Biology Halazonetis, Thanos D.

(ii) TITLE OF INVENTION: Methods for Altering Three-Dimensional Protein Structure and Compositions Produced Thereby (iii) NUMBER OF SEQUENCES: 4 (iv) CORRESPONDENCE ADDRESS: (A) ADDRESSEE: Howson and Howson (B) STREET: Spring House Corporate Cntr., PO Box 457 (C) CITY: Spring House (D) STATE: Pennsylvania (E) COUNTRY: USA (F) ZIP: 19477 (v) COMPUTER READABLE FORM: (A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS (D) SOFTWARE: PatentIn Release #1. 0, Version #1. 30 (vi) CURRENT APPLICATION DATA: (A) APPLICATION NUMBER: WO (B) FILING DATE: (C) CLASSIFICATION: (vii) PRIOR APPLICATION DATA: (A) APPLICATION NUMBER: US 60/035,458 (B) FILING DATE: 17-JAN-1997 (viii) ATTORNEY/AGENT INFORMATION: (A) NAME: Kodroff, Cathy A.

(B) REGISTRATION NUMBER: 33,980 (C) REFERENCE/DOCKET NUMBER: WST74APCT (ix) TELECOMMUNICATION INFORMATION: (A) TELEPHONE: 215-540-9200 (B) TELEFAX: 215-540-5818 (2) INFORMATION FOR SEQ ID NO : 1 : (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 1317 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic) (ix) FEATURE: (A) NAME/KEY: CDS (B) LOCATION: 136.. 1314 (xi) SEQUENCE DESCRIPTION: SEQ ID NO : 1 : GTCTAGAGCC ACCGTCCAGG GAGCAGGTAG CTGCTGGGCT CCGGGGACAC TTTGCGTTCG 60 GGCTGGGAGC GTGCTTTCCA CGACGGTGAC ACGCTTCCCT GGATTGGCAG CCAGACTGCC 120 TTCCGGGTCA CTGCC ATG GAG GAG CCG CAG TCA GAT CCT AGC GTC GAG CCC 171 Met Glu Glu Pro Gln Ser Asp Pro Ser Val Glu Pro 1 5 10 CCT CTG AGT CAG GAA ACA TTT TCA GAC CTA TGG AAA CTA CTT CCT GAA 219 Pro Leu Ser Gln Glu Thr Phe Ser Asp Leu Trp Lys Leu Leu Pro Glu 15 20 25 AAC AAC GTT CTG TCC CCC TTG CCG TCC CAA GCA ATG GAT GAT TTG ATG 267 Asn Asn Val Leu Ser Pro Leu Pro Ser Gln Ala Met Asp Asp Leu Met 30 35 40 CTG TCC CCG GAC GAT ATT GAA CAA TGG TTC ACT GAA GAC CCA GGT CCA 315 Leu Ser Pro Asp Asp Ile Glu Gln Trp Phe Thr Glu Asp Pro Gly Pro 45 50 55 60 GAT GAA GCT CCC AGA ATG CCA GAG GCT GCT CCC CCC GTG GCC CCT GCA 363 Asp Glu Ala Pro Arg Met Pro Glu Ala Ala Pro Pro Val Ala Pro Ala 65 70 75 CCA GCA GCT CCT ACA CCG GCG GCC CCT GCA CCA GCC CCC TCC TGG CCC 411 Pro Ala Ala Pro Thr Pro Ala Ala Pro Ala Pro Ala Pro Ser Trp Pro 80 85 90 CTG TCA TCT TCT GTC CCT TCC CAG AAA ACC TAC CAG GGC AGC TAC GGT 459 Leu Ser Ser Ser Val Pro Ser Gln Lys Thr Tyr Gln Gly Ser Tyr Gly 95 100 105 TTC CGT CTG GGC TTC TTG CAT TCT GGG ACA GCC AAG TCT GTA ACT TGC 507 Phe Arg Leu Gly Phe Leu His Ser Gly Thr Ala Lys Ser Val Thr Cys 110 115 120 ACG TAC TCC CCT GCC CTC AAC AAG ATG TTT TGC CAA CTG GCC AAG ACC 555 Thr Tyr Ser Pro Ala Leu Asn Lys Met Phe Cys Gln Leu Ala Lys Thr 125 130 135 140 TGC CCT GTG CAG CTG TGG GTT GAT TCC ACA CCC CCG CCC GGC ACC CGC 603 Cys Pro Val Gln Leu Trp Val Asp Ser Thr Pro Pro Pro Gly Thr Arg 145 150 155 GTC CGC GCC ATG GCC ATC TAC AAG CAG TCA CAG CAC ATG ACG GAG GTT 651 Val Arg Ala Met Ala Ile Tyr Lys Gln Ser Gln His Met Thr Glu Val 160 165 170 GTG AGG CGC TGC CCC CAC CAT GAG CGC TGC TCA GAT AGC GAT GGT CTG 699 Val Arg Arg Cys Pro His His Glu Arg Cys Ser Asp Ser Asp Gly Leu 175 180 185 GCC CCT CCT CAG CAT CTT ATC CGA GTG GAA GGA AAT TTG CGT GTG GAG 747 Ala Pro Pro Gln His Leu Ile Arg Val Glu Gly Asn Leu Arg Val Glu 190 195 200 TAT TTG GAT GAC AGA AAC ACT TTT CGA CAT AGT GTG GTG GTG CCC TAT 795 Tyr Leu Asp Asp Arg Asn Thr Phe Arg His Ser Val Val Val Pro Tyr 205 210 215 220 GAG CCG CCT GAG GTT GGC TCT GAC TGT ACC ACC ATC CAC TAC AAC TAC 843 Glu Pro Pro Glu Val Gly Ser Asp Cys Thr Thr Ile His Tyr Asn Tyr 225 230 235 ATG TGT AAC AGT TCC TGC ATG GGC GGC ATG AAC CGG AGA CCC ATC CTC 891 Met Cys Asn Ser Ser Cys Met Gly Gly Met Asn Arg Arg Pro Ile Leu 240 245 250 ACC ATC ATC ACA CTG GAA GAC TCC AGT GGT AAT CTA CTG GGA CGG AAC 939 Thr Ile Ile Thr Leu Glu Asp Ser Ser Gly Asn Leu Leu Gly Arg Asn 255 260 265 AGC TTT GAG GTG CGT GTT TGT GCC TGT CCT GGG AGA GAC CGG CGC ACA 987 Ser Phe Glu Val Arg Val Cys Ala Cys Pro Gly Arg Asp Arg Arg Thr 270 275 280 GAG GAA GAG AAT CTC CGC AAG AAA GGG GAG CCT CAC CAC GAG CTG CCC 1035 Glu Glu Glu Asn Leu Arg Lys Lys Gly Glu Pro His His Glu Leu Pro 285 290 295 300 CCA GGG AGC ACT AAG CGA GCA CTG CCC AAC AAC ACC AGC TCC TCT CCC 1083 Pro Gly Ser Thr Lys Arg Ala Leu Pro Asn Asn Thr Ser Ser Ser Pro 305 310 315 CAG CCA AAG AAG AAA CCA CTG GAT GGA GAA TAT TTC ACC CTT CAG ATC 1131 Gln Pro Lys Lys Lys Pro Leu Asp Gly Glu Tyr Phe Thr Leu Gln Ile 320 325 330 CGT GGG CGT GAG CGC TTC GAG ATG TTC CGA GAG CTG AAT GAG GCC TTG 1179 Arg Gly Arg Glu Arg Phe Glu Met Phe Arg Glu Leu Asn Glu Ala Leu 335 340 345 GAA CTC AAG GAT GCC CAG GCT GGG AAG GAG CCA GGG GGG AGC AGG GCT 1227 Glu Leu Lys Asp Ala Gln Ala Gly Lys Glu Pro Gly Gly Ser Arg Ala 350 355 360 CAC TCC AGC CAC CTG AAG TCC AAA AAG GGT CAG TCT ACC TCC CGC CAT 1275 His Ser Ser His Leu Lys Ser Lys Lys Gly Gln Ser Thr Ser Arg His 365 370 375 380 AAA AAA CTC ATG TTC AAG ACA GAA GGG CCT GAC TCA GAC TGA 1317 Lys Lys Leu Met Phe Lys Thr Glu Gly Pro Asp Ser Asp 385 390 (2) INFORMATION FOR SEQ ID NO : 2: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 393 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO : 2: Met Glu Glu Pro Gln Ser Asp Pro Ser Val Glu Pro Pro Leu Ser Gln 1 5 10 15 Glu Thr Phe Ser Asp Leu Trp Lys Leu Leu Pro Glu Asn Asn Val Leu 20 25 30 Ser Pro Leu Pro Ser Gln Ala Met Asp Asp Leu Met Leu Ser Pro Asp 35 40 45 Asp Ile Glu Gln Trp Phe Thr Glu Asp Pro Gly Pro Asp Glu Ala Pro 50 55 60 Arg Met Pro Glu Ala Ala Pro Pro Val Ala Pro Ala Pro Ala Ala Pro 65 70 75 80 Thr Pro Ala Ala Pro Ala Pro Ala Pro Ser Trp Pro Leu Ser Ser Ser 85 90 95 Val Pro Ser Gln Lys Thr Tyr Gln Gly Ser Tyr Gly Phe Arg Leu Gly 100 105 110 Phe Leu His Ser Gly Thr Ala Lys Ser Val Thr Cys Thr Tyr Ser Pro 115 120 125 Ala Leu Asn Lys Met Phe Cys Gln Leu Ala Lys Thr Cys Pro Val Gln 130 135 140 Leu Trp Val Asp Ser Thr Pro Pro Pro Gly Thr Arg Val Arg Ala Met 145 150 155 160 Ala Ile Tyr Lys Gln Ser Gln His Met Thr Glu Val Val Arg Arg Cys 165 170 175 Pro His His Glu Arg Cys Ser Asp Ser Asp Gly Leu Ala Pro Pro Gln 180 185 190 His Leu Ile Arg Val Glu Gly Asn Leu Arg Val Glu Tyr Leu Asp Asp 195 200 205 Arg Asn Thr Phe Arg His Ser Val Val Val Pro Tyr Glu Pro Pro Glu 210 215 220 Val Gly Ser Asp Cys Thr Thr Ile His Tyr Asn Tyr Met Cys Asn Ser 225 230 235 240 Ser Cys Met Gly Gly Met Asn Arg Arg Pro Ile Leu Thr Ile Ile Thr 245 250 255 Leu Glu Asp Ser Ser Gly Asn Leu Leu Gly Arg Asn Ser Phe Glu Val 260 265 270 Arg Val Cys Ala Cys Pro Gly Arg Asp Arg Arg Thr Glu Glu Glu Asn 275 280 285 Leu Arg Lys Lys Gly Glu Pro His His Glu Leu Pro Pro Gly Ser Thr 290 295 300 Lys Arg Ala Leu Pro Asn Asn Thr Ser Ser Ser Pro Gln Pro Lys Lys 305 310 315 320 Lys Pro Leu Asp Gly Glu Tyr Phe Thr Leu Gln Ile Arg Gly Arg Glu 325 330 335 Arg Phe Glu Met Phe Arg Glu Leu Asn Glu Ala Leu Glu Leu Lys Asp 340 345 350 Ala Gln Ala Gly Lys Glu Pro Gly Gly Ser Arg Ala His Ser Ser His 355 360 365 Leu Lys Ser Lys Lys Gly Gln Ser Thr Ser Arg His Lys Lys Leu Met 370 375 380 Phe Lys Thr Glu Gly Pro Asp Ser Asp 385 390 (2) INFORMATION FOR SEQ ID NO : 3: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 393 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (ix) FEATURE: (A) NAME/KEY: Modified-site (B) LOCATION: 341 (D) OTHER INFORMATION:/note="Amino acid in position 341 can be Ile, Leu, Phe or Val" (ix) FEATURE: (A) NAME/KEY: Modified-site (B) LOCATION: 344 (D) OTHER INFORMATION:/note="Amino acid in position 344 can be Ala, Ile, Leu, Phe, Trp or Tyr" (xi) SEQUENCE DESCRIPTION: SEQ ID NO : 3: Met Glu Glu Pro Gln Ser Asp Pro Ser Val Glu Pro Pro Leu Ser Gln 1 5 10 15 Glu Thr Phe Ser Asp Leu Trp Lys Leu Leu Pro Glu Asn Asn Val Leu 20 25 30 Ser Pro Leu Pro Ser Gln Ala Met Asp Asp Leu Met Leu Ser Pro Asp 35 40 45 Asp Ile Glu Gln Trp Phe Thr Glu Asp Pro Gly Pro Asp Glu Ala Pro 50 55 60 Arg Met Pro Glu Ala Ala Pro Pro Val Ala Pro Ala Pro Ala Ala Pro 65 70 75 80 Thr Pro Ala Ala Pro Ala Pro Ala Pro Ser Trp Pro Leu Ser Ser Ser 85 90 95 Val Pro Ser Gln Lys Thr Tyr Gln Gly Ser Tyr Gly Phe Arg Leu Gly 100 105 110 Phe Leu His Ser Gly Thr Ala Lys Ser Val Thr Cys Thr Tyr Ser Pro 115 120 125 Ala Leu Asn Lys Met Phe Cys Gln Leu Ala Lys Thr Cys Pro Val Gln 130 135 140 Leu Trp Val Asp Ser Thr Pro Pro Pro Gly Thr Arg Val Arg Ala Met 145 150 155 160 Ala Ile Tyr Lys Gln Ser Gln His Met Thr Glu Val Val Arg Arg Cys 165 170 175 Pro His His Glu Arg Cys Ser Asp Ser Asp Gly Leu Ala Pro Pro Gln 180 185 190 His Leu Ile Arg Val Glu Gly Asn Leu Arg Val Glu Tyr Leu Asp Asp 195 200 205 Arg Asn Thr Phe Arg His Ser Val Val Val Pro Tyr Glu Pro Pro Glu 210 215 220 Val Gly Ser Asp Cys Thr Thr Ile His Tyr Asn Tyr Met Cys Asn Ser 225 230 235 240 Ser Cys Met Gly Gly Met Asn Arg Arg Pro Ile Leu Thr Ile Ile Thr 245 250 255 Leu Glu Asp Ser Ser Gly Asn Leu Leu Gly Arg Asn Ser Phe Glu Val 260 265 270 Arg Val Cys Ala Cys Pro Gly Arg Asp Arg Arg Thr Glu Glu Glu Asn 275 280 285 Leu Arg Lys Lys Gly Glu Pro His His Glu Leu Pro Pro Gly Ser Thr 290 295 300 Lys Arg Ala Leu Pro Asn Asn Thr Ser Ser Ser Pro Gln Pro Lys Lys 305 310 315 320 Lys Pro Leu Asp Gly Glu Tyr Phe Thr Leu Gln Ile Arg Gly Arg Glu 325 330 335 Arg Phe Glu Met Xaa Arg Glu Xaa Asn Glu Ala Leu Glu Leu Lys Asp 340 345 350 Ala Gln Ala Gly Lys Glu Pro Gly Gly Ser Arg Ala His Ser Ser His 355 360 365 Leu Lys Ser Lys Lys Gly Gln Ser Thr Ser Arg His Lys Lys Leu Met 370 375 380 Phe Lys Thr Glu Gly Pro Asp Ser Asp 385 390 (2) INFORMATION FOR SEQ ID NO : 4: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 393 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (ix) FEATURE: (A) NAME/KEY: Modified-site (B) LOCATION: 341 (D) OTHER INFORMATION:/note="Amino acid in position 341 can be Ile, Leu, Phe or Val" (ix) FEATURE: (A) NAME/KEY: Modified-site (B) LOCATION: 344 (D) OTHER INFORMATION:/note="Amino acid in position 344 can be Ala, Ile, Leu, Phe, Trp or Tyr" (xi) SEQUENCE DESCRIPTION: SEQ ID NO : 4: Met Glu Glu Pro Gln Ser Asp Pro Ser Val Glu Pro Pro Leu Ser Gln 1 5 10 15 Glu Thr Phe Ser Asp Leu Trp Lys Leu Leu Pro Glu Asn Asn Val Leu 20 25 30 Ser Pro Leu Pro Ser Gln Ala Met Asp Asp Leu Met Leu Ser Pro Asp 35 40 45 Asp Ile Glu Gln Trp Phe Thr Glu Asp Pro Gly Pro Asp Glu Ala Pro 50 55 60 Arg Met Pro Glu Ala Ala Pro Pro Val Ala Pro Ala Pro Ala Ala Pro 65 70 75 80 Thr Pro Ala Ala Pro Ala Pro Ala Pro Ser Trp Pro Leu Ser Ser Ser 85 90 95 Val Pro Ser Gln Lys Thr Tyr Gln Gly Ser Tyr Gly Phe Arg Leu Gly 100 105 110 Phe Leu His Ser Gly Thr Ala Lys Ser Val Thr Cys Thr Tyr Ser Pro 115 120 125 Ala Leu Asn Lys Met Phe Cys Gln Leu Ala Lys Thr Cys Pro Val Gln 130 135 140 Leu Trp Val Asp Ser Thr Pro Pro Pro Gly Thr Arg Val Arg Ala Met 145 150 155 160 Ala Ile Tyr Lys Gln Ser Gln His Met Thr Glu Val Val Arg Arg Cys 165 170 175 Pro His His Glu Arg Cys Ser Asp Ser Asp Gly Leu Ala Pro Pro Gln 180 185 190 His Leu Ile Arg Val Glu Gly Asn Leu Arg Val Glu Tyr Leu Asp Asp 195 200 205 Arg Asn Thr Phe Arg His Ser Val Val Val Pro Tyr Glu Pro Pro Glu 210 215 220 Val Gly Ser Asp Cys Thr Thr Ile His Tyr Asn Tyr Met Cys Asn Ser 225 230 235 240 Ser Cys Met Gly Gly Met Asn Arg Arg Pro Ile Leu Thr Ile Ile Thr 245 250 255 Leu Glu Asp Ser Ser Gly Asn Leu Leu Gly Arg Asn Ser Phe Glu Val 260 265 270 Arg Val Cys Ala Cys Pro Gly Arg Asp Arg Arg Thr Glu Glu Glu Asn 275 280 285 Leu Arg Lys Lys Gly Glu Pro His His Glu Leu Pro Pro Gly Ser Thr 290 295 300 Lys Arg Ala Leu Pro Asn Asn Thr Ser Ser Ser Pro Gln Pro Lys Lys 305 310 315 320 Lys Pro Leu Asp Gly Glu Tyr Phe Thr Leu Gln Ile Arg Gly Arg Glu 325 330 335 Arg Phe Glu Lys Xaa Arg Glu Xaa Asn Glu Ala Leu Glu Leu Lys Asp 340 345 350 Ala Gln Ala Gly Lys Glu Pro Gly Gly Ser Arg Ala His Ser Ser His 355 360 365 Leu Lys Ser Lys Lys Gly Gln Ser Thr Ser Arg His Lys Lys Leu Met 370 375 380 Phe Lys Thr Glu Gly Pro Asp Ser Asp 385 390