Imported: 10 Mar '17 | Published: 27 Nov '08
USPTO - Utility Patents
The invention relates method of disrupting the interface between Gs and tubulin. The disruption of this interaction is a mechanism for treatment of mood and anxiety disorders and this interaction may be used to screen for and design therapeutics for mood and anxiety disorders.
Priority is claimed to U.S. Provisional Application No. 60/678,050 filed May 5, 2005, which is incorporated herein by reference in its entirety.
Certain of the studies described in the present application were conducted with the support of government funding in the form of a grant from the National Institutes of Mental Health, Grant No. MH039595 and MH00699. The United States government has certain rights in the invention.
1. Field of Invention
The invention relates to the disruption of the interface between Gs(alpha) and tubulin. The disruption of this interaction is a mechanism for the treatment of mood and anxiety disorders. The invention also provides for methods of screening for and the design of therapeutics for mood and anxiety disorders, such as depression.
2. Related Technology
Affective disorders are characterized by changes in mood as the primary clinical manifestation. Major depression is one of the most common mental illnesses and is often under diagnosed and frequently undertreated, or treated inappropriately. Major depression is characterized by feelings of intense sadness and despair, mental slowing and loss of concentration, pessimistic worry, agitation, and self-deprecation. Physical changes usually occur that include insomnia, anorexia and weight loss (or overeating) decreased energy and libido, and disruption of the normal circadian rhythms of activity, body temperature, and many endocrine functions. Many as 10-15% of individuals with this disorder display suicidal behavior during their lifetime.
Certain G protein a subunits, components of the G-protein Coupled Receptor (GPCR) signal transduction system, have been shown to form tight complexes with the cytoskeletal protein, such as tubulin. Microtubules, an essential component of the cytoskeleton, are composed of tubulin dimers where each dimer consists of an and monomer. One GTP binds to the nonexchangable site (N-site) on -tubulin and another GTP binds at the exchangeable site (E-site) on -tubulin. GTP hydrolysis occurs at the E-site when another dimer binds to the growing microtubule at the positive end (Carlier, Curr. Opin. Cell. Biol. 3: 12-17, 1991; Nogales, Annu. Rev. Biophys. Biomol. Struct. 30: 397-420, 2001). Over the past few years, significant progress has been made in the structural determination of the tubulin dimer. The domains on tubulin where drugs such as taxanes, colchine and vinblastine bind have been revealed. Much less information exists on where the microtubule associated proteins (MAPs) bind tubulin. However, many of the sites have been proposed to be on the C-terminus of tubulin. As the structure/function of tubulin dimers and microtubules is deciphered, more novel protein-protein interactions with tubulin will be determined (Nogales (2001), Annu. Rev. Biophys. Biomol. Struct. 30, 397-420).
G proteins are heterotrimeric structures composed of , , and subunits. Upon agonist binding to membrane receptors, the G subunit is activated by the exchange of GDP for GTP leading to the extracellular message being passed to the intracellular side (Lambright et al., Nature 369: 621-628, 1994). Activated G subunit interacts with effector proteins and allows G to interact with effectors as well. Recently, it has become apparent that G and G proteins interact with a vast array of other cellular proteins that can affect the G protein activation/deactivation cycle (Blumer et al., Receptors Channels 9, 195-204, 2003; Hepler, Mol. Pharmacol. 64: 547-549, 2003). Although distinct in structure and other properties from other G protein regulators, tubulin has long been known to interact with certain G proteins (Rasenick et al., Nature 294: 560-562, 1981; Wang et al., J. Biol. Chem. 265: 1239-1242, 1990). Of the G family of proteins, the inhibitory G protein subunit of adenylyl cyclase (Gi1) and the stimulatory G protein subunit of adenylyl cyclase (Gs) bind with a high affinity to tubulin while other G subunits (e.g., the subunit of the retinal G protein transducin; Gt) show no measurable tubulin binding (Wang et al., J. Biol. Chem. 265: 1239-1242, 1990).
The protein-protein interaction between one of these G proteins, Gs and tubulin has direct implications to the therapeutic mechanism by which antidepressants (and perhaps other drugs for mood and anxiety disorders) exert their effect. Studies suggest that chronic antidepressant treatment moves Gs from a subcellular region enriched in tubulin to one where Gs becomes less associated with tubulin. In these regions, Gs also engages in a more facile activation of adenylyl cyclase. Elucidation of the binding sites between G subunits and tubulin dimers will provide insight into this complex and interaction and the biochemical mechanism of antidepressant therapy.
Antidepressant therapies are present in many diverse forms, including tricyclic compounds, monoamine oxidase inhibitors, selective serotonin reuptake inhibitors (SSRIs), atypical antidepressants, and electroconvulsive treatment. There remains a need for the identification and development of new antidepressant therapies, as well as methods for screening for novel antidepressant agents.
It is known that tubulin interacts with many different proteins (Nogales, Annu. Rev. Biophys. Biomol. Struct. 30: 397-420,2001). Structural information on these protein interactions has been limited except in a few cases (Nogales, Annu. Rev. Biophys. Biomol. Struct. 30: 397-420, 2001; Gigant et al., Cell 102: 809-816, 2000). Many of these proteins, in particular Tau proteins, appear to interact with tubulin through the acidic C-terminal of tubulin (Nogales, Annu. Rev. Biophys. Biomol. Struct. 30: 397-420, 2001, 2001). This model described herein suggests that a binding site for G at the plus end of a microtubule which is quite different from that of MAPs. As seen in FIGS. 3-5, the nucleotide-binding pocket and the surrounding residues of -tubulin comprise a majority of the Gs interacting surface. Further, the Gs molecule completely encases the nucleotide-binding pocket of -tubulin. Gs appears to be the first protein identified to associate with -tubulin at the GTP binding site.
In one embodiment, the invention provides for methods of disrupting the complex of Gs and tubulin in a cell comprising inhibiting the interaction of Gs and tubulin within said cell by contacting tubulin with a molecule that inhibits the interaction of Gs and tubulin, wherein the molecule binds to tubulin within at least one region selected from the group consisting of the nucleotide binding site, H1 region or H2 region of tubulin.
In another embodiment, the invention provides for methods of disrupting the complex of Gs and tubulin in a cell comprising inhibiting the interaction of Gs and tubulin within said cell by contacting Gs with a molecule that inhibits the interaction of G and tubulin, wherein the molecule binds to Gs within at least one region selected from the group consisting of the 3-5 region, 4-6 region, 2-4 region or the amino terminus of Gs.
The invention also provides for compositions comprising a molecule that inhibits the interaction of Gs and tubulin. These molecules include those that bind tubulin within at least one region selected from the group consisting of the nucleotide binding site, H1 region or H2 region of tubulin. These molecules also include those that bind Gs within at least one region selected from the group consisting of 3-5 region, 4-6 region, 2-4 region or the amino terminus of Gs.
The molecules that inhibit the interaction of Gs and tubulin are peptides, small molecules, antibodies, including single-chained antibodies, and monoclonal antibodies or peptidomimetics. Peptides of the invention include peptide comprising an amino acid sequence of SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 8, SEQ ID NO: 9, SEQ ID NO: 10, SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 13 or SEQ ID NO: 14.
In a further embodiment, the invention provides for methods of treating a mood or anxiety disorder comprising administering a composition comprising a molecule that inhibits the interaction of Gs and tubulin.
In another embodiment, the invention provides for methods of identifying modulators of the interaction of Gs and tubulin comprising contacting a cell expressing Gs and tubulin with a candidate compound, and monitoring said cell for modulation of Gs binding to tubulin, wherein a candidate compound that reduces the binding of Gs to tubulin is an inhibitor of the interaction of G and tubulin, and a candidate compound that increase the binding of Gs to tubulin is a agonist of the interaction of Gs and tubulin.
Despite several decades of studies, the mechanism of antidepressant action has not been clearly established. One of the most widely known biochemical effects of antidepressant treatment is an alteration in the density and/or sensitivity of several neurotransmitter receptor systems (Sulser, Adv. Biochem. Psychopharmacol., 39:249-261; 1984). However, these effects do not fully explain the clinical efficacy of all antidepressants, mainly because of the dissociation between the time course of the change in the receptor numbers and their clinical time course (Rasenick et al., J. Clin. Psychiatry, 57:49-55; 1996).
Many studies searching for a common mechanism of antidepressant action have focused on postreceptor neuronal cell signaling processes as potential targets of such action (Menkes et al. Science, 219:65-76, 1983; Ozawa et al., Mol. Pharmacol., 36:803-808, 1989; Duman et al., Arch. Gen. Psychiatry, 54:597-606, 1997; Takahashi et al., J. Neurosci., 19:610-616, 1999). Much of this previous work has focused on the downstream effects of antidepressant action, particularly those involving cAMP (Perez et al., Eur. J. Pharmacol., 172:305-316, 1989; Perez et al., Neuropsychopharmacology, 4:57-64, 1991; Duman et al., Arch. Gen. Psychiatry, 54:597-606, 1997; Takahashi et al., J. Neurosci., 19:610-616, 1999).
Much of the current thinking about G protein-coupled receptors is based on the idea of freely mobile receptors, G proteins, and effectors in which the specificity of their interaction is derived from the three-dimensional structure of the sites of protein-protein interactions. However, recent evidence indicates that an organized interaction of receptors, G proteins, and effectors with significant limitations on lateral mobility (Kuo et al., Science, 260:232-234; 1993). Furthermore, these membrane proteins are associated with tubulin or other cytoskeletal proteins (Carlson et al., Mol. Pharmacol., 30:463-468, 1986; Rasenick et al., Adv. Second Messenger Phosphoprotein Res., 22:381-386, 1990; Wang et al., Biochemistry, 30:10957-10965, 1991), which restrict distribution and mobility of G proteins to a surprising degree (Neubig, FASEB, 8:939-946; 1994). The presence of a well organized network of cytoskeletal elements and the components of neurotransmitter and hormonal G protein-mediated signal transduction systems may play an important role in achieving this function.
Tubulin-G interaction has been shown to induce changes in the GTP/GDP binding and kinetics in both G and tubulin. G proteins binding to tubulin activate the GTPase activity of tubulin, destabilizing the microtubules (Roychowdhury et al., J. Biol. Chem. 274: 13485-13490, 1999). Conversely, G proteins can be activated in a receptor independent mechanism in which a direct transfer of GTP (transactivation) from the E-site on tubulin to the G subunit occurs (Rasenick et al., Methods Enzymol. 390: 389-403, 2004). In the case of Gs, this receptor-independent activation of Gs subunits increases the coupling of Gs to adenylyl cyclase (Yan et al., J Neurochem. 76: 182-190, 2001). Studies using chimeric proteins of Gi1 and Gt to disrupt tubulin-G interaction, demonstrated that residues 237-270 of Gi1 (which corresponds to residues 253-293 in Gs) are crucial for the transactivation of G by tubulin and also play a role in modulating microtubule organization and elongation of cellular processes. Elucidation of the binding sites between G subunits and tubulin dimers provides insight into this complex and novel interaction.
Crystallographic studies have provided an excellent method to determine protein-protein structures. However, tubulin has been difficult to study by crystallographic approaches (Nogales et al., Nature 391: 199-203, 1998; Gigant et al., Cell 102: 809-816, 2000; Mandelkow et al., Nature 287: 595-599, 1980). Molecular docking programs along with verification through biochemical assays, as described herein, are another approach for determining protein-protein structures.
The model of Gs and tubulin interaction described herein was developed using a combination of biochemical and molecular docking techniques. Overlapping peptides were covalently attached via the primary amino acid Gs sequence to a membrane and tubulin binding to specific spots on the membrane was determined (Frank, J. Immunol. Methods 267: 13-26, 2002). This study identified potential high affinity sites important for tubulin-Gs interaction. (see Example 2). In addition, protein-protein docking algorithms were used to generate and refine a model for the interacting facets of these molecules (Chen, et al., Proteins 52: 68-73, 2003; Comeau et al., Bioinformatics 20: 45-50, 2004). (see Example 3)
The relative position of -tubulin to a fixed Gs protein is shown in FIG. 3a. This binding position of tubulin is within the GTPase domain of Gs. The relative position of Gs to a fixed -tubulin protien is shown in FIG. 3b. The binding position of G to -tubulin is within the exchangable nucleotide-binding site of -tubulin and the H1 (-helix 1) and H2 (-helix 2) regions of -tubulin. In addition, common contact regions between Gs and -tubulin are within the 3-5 (switch III), and the 4-6 loop of Gs, the 2-4 (switch II) region of Gs and the amino terminal of Gs.
The model of Gs and tubulin interaction described herein is the first report of a structural models of the Gs-tubulin complex and suggests that tubulin interacts with Gs predominantly in the GTPase domain, more precisely with regions essential to adenylyl cyclase activation (2-4 and 3-5) (Tesmer et al., Science 278: 1907-1916, 1997). This model also suggests that Gs binds to tubulin such that it surrounds the nucleotide-binding site of -tubulin, in a region of tubulin normally involved in docking other tubulin molecules during microtubule polymerization. These structures reveal how tubulin might transactivate Gs and how Gs can activate tubulin GTPase.
The structural interaction between tubulin and Gs described herein has implications for the function of each protein. The domains on Gs (2-4 and 3-5) that are essential to the binding and activation of adenylyl cyclase (Tesmer et al., Science 278: 1907-1916, 1997) are also important for the interaction with tubulin. These observations indicate how Gs-tubulin interaction may alter the interaction of Gs with adenylyl cyclase (Wang et al., J. Biol. Chem. 265: 1239-1242, 1990; Yan et al., J Neurochem. 76: 182-190, 2001; Roychowdhury et al., J. Biol. Chem. 274: 13485-13490, 1999; Rasenick, et al., Methods Enzymol. 390: 389-403, 2004).. Further, these observations provide a structural basis to the finding that Gs interacts with tubulin at the exchangeable site on -tubulin activating the GTPase of tubulin and increasing dynamics of microtubules (Roychowdhury et al., J. Biol. Chem. 274: 13485-13490, 1999). Gs appears to surround the exposed GTP on the -tubulin, which includes the region of the GTP cap of microtubules.
Gs facilitates GTP hydrolysis on tubulin, which leads to microtubule depolymerization by increased GTPase activation on tubulin (Roychowdhury et al., J. Biol. Chem. 274: 13485-13490, 1999). This is consistent with the structural models described herein. The experimental and theoretical analyses described herein provides the first proposed structural model for the Gs-tubulin complex.
Disruption of the tubulin-Gs complex is contemplated to be useful as a therapeutic mechanism for treating and preventing mood and anxiety disorders. It is known that antidepressant therapy induces a shift in the subcellular localization of Gs from a region enriched with tubulin to a region less associated with tubulin. Thus, molecules that induce this shift in Gs subcellular localization or those that disrupt the interaction of tubulin with Gs are contemplated as therapeutics for preventing or ameliorating mood and anxiety disorders, such as depression.
The invention contemplates as therapeutics for mood and anxiety disorders, any molecule that disrupts or inhibits the formation of the Gs-tubulin complex. These therapeutic molecules include peptides, small molecules, antibodies and fusion or chimeric proteins. Any of the therapeutic compositions described below can be used alone or in combination with each other. Further, the present invention also contemplates the use of the following compositions in combination with standard treatments presently being used for the treatment of mood and anxiety disorders, such as known antidepressants, such as tricyclic compounds (e.g. amitriptyline, clomipramine, amitriptyline, amitriptyline, maprotiline, desipramine, nortryptyline, desipramine, doxepin, trimipramine, imipramine, protriptyline), monoamine oxidase inhibitors (e.g. phenelzine, tranylcypromine), selective serotonin reuptake inhibitors (SSRIs) (e.g. citalopram, escitalopram oxalate, fluvoxamine, paroxetine, fluoxetine, sertraline), atypical antidepressants and electroconvulsive treatment.
The present invention provides peptides that may be used to disrupt the Gs-tubulin complex. The invention particularly provides peptides that specifically bind within the Gs binding site on tubulin or the tubulin binding site on Gs. Exemplary peptides that bind tubulin within the Gs binding site are those set out in Table 1.
The peptides, set out in Table 1, were designed using the immobilized-peptide array technique described in Example 2. The invention contemplates other peptides that bind Gs within the GTPase domain, including the switch regions of Gs (switch I, switch II and switch III). For example, the invention contemplates peptides that bind to 3-5 region of Gs or the 2-4 (switch II region) of Gs. Other exemplary peptides include peptides that bind to Ala18-Lys34 of the Gs amino acid sequence or peptides that bind to Asn66-Phe68 of the Gs amino acid sequence or peptides that bind to amino acids 253-293 of the Gs amino acid sequence are contemplated. The invention also contemplates peptides that bind within the relative position of Gs as shown in FIG. 3a.
Peptides that bind to Gs and thereby inhibit its binding to tubulin are also contemplated as therapeutic molecules of the invention. The invention also contemplates peptides that bind to the amino terminus of Gs. For example, the invention contemplates peptides that bind near the nucleotide binding site on tubulin, H1 region of tubulin and the H2 region of tubulin. The invention also contemplates peptides that bind within the relative position of -tubulin as shown in FIG. 3b.
In discussing the sequences of the peptides of the invention, the present application employs the conventional abbreviations for the amino acids as follows: Alanine, Ala, A; Arginine, Arg, R; Asparagine, Asn, N; Aspartic acid, Asp, D; Cysteine, Cys, C; Glutamine, Gln, Q; Glutamic Acid, Glu, E; Glycine, Gly, G; Histidine, His, H; Isoleucine, Ile, I; Leucine, Leu, L; Lysine, Lys, K; Methionine, Met, M; Phenylalanine, Phe, F; Proline, Pro, P; Serine, Ser, S; Threonine, Thr, T; Tryptophan, Trp, W; Tyrosine, Tyr, Y; Valine, Val, V; Aspartic acid or Asparagine, Asx, B; Glutamic acid or Glutamine, Glx, Z; Norleucine, Nle; Acetyl-glycine (Ac)G; Any amino acid, Xaa, X. Additional modified amino acids known to those of skill in the art also may be used.
The peptides of the invention may be tested for their ability to bind to tubulin of Gs using the SPOT membrane assays as described in detail in Example 1. Another method to determine the interactions between tubulin and G proteins as well as the binding of peptides to either tubulin or G proteins and the effect of those peptides on the interaction between tubulin and G proteins is surface plasmon resonance.
The peptide of the present invention may be any length of amino acids so long as the amino acids are of a sufficient length to interfere with the interaction of Gs and tubulin. Preferably, the novel peptide inhibitors of the Gs-tubulin interaction are at least about five amino acids in length, in certain embodiments the novel peptides of the present invention may comprise a contiguous amino acid sequence of about 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 45, 50, or more amino acids.
In considering the particular amino acid to be positioned at any of the positions of the peptide, it may be useful to consider the hydropathic index of amino acids at each of the positions in a peptide known to be an effective inhibitor of the Gs-tubulin interaction, and substitute a given amino acid with one of a similar hydropathic index. It is accepted that the relative hydropathic character of the amino acid contributes to the secondary structure of a resultant protein or peptide, which in turn defines the interaction of that protein with other molecules, for example, enzymes, substrates, receptors, DNA, antibodies, antigens, and the like. Each amino acid has been assigned a hydropathic index on the basis of their hydrophobicity and charge characteristics (Kyte Doolittle, J. Mol. Biol., 157(1):105-132, 1982, incorporated herein by reference). Generally, amino acids may be substituted by other amino acids that have a similar hydropathic index or score and still result in a protein with similar biological activity i.e., still obtain a biological functionally equivalent protein or peptide. This is known as a conservative amino acid substitution. In the context of the peptides of the present invention, a biologically functionally equivalent protein or peptide will be one which still retains its ability to be an antagonist of the Gs binding to tubulin or an antagonist of tubulin binding Gs.
In addition, the substitution of like amino acids can be made effectively on the basis of hydrophilicity. U.S. Pat. No. 4,554,101, incorporated herein by reference, states that the greatest local average hydrophilicity of a protein, as governed by the hydrophilicity of its adjacent amino acids, correlates with a biological property of the protein. As such, an amino acid can be substituted for another having a similar hydrophilicity value and still obtain a biologically equivalent and immunologically equivalent protein.
In preferred aspects of the invention, peptides are synthesized according to methods known to those of skill in the art (Carter, et al., Biotechnology, 10(5): p. 509-13, 1992; Chen, et al., in Peptides: Wave of the FutureProceedings of the 17th American Peptide Symposium, ed. M. Lebl and R. Houghten, Editors., Mayflower Scientific Ltd.: Kingswinford, UK. pp 206-207 and pp 318-319, 2001. In addition, short peptides sequences may be prepared by chemical synthesis using standard means. Particularly convenient are solid phase techniques (see, e.g., Erikson et al., The Proteins (1976) v. 2, Academic Press, New York, p. 255). Automated solid phase synthesizers are commercially available. In addition, modifications in the sequence are easily made by substitution, addition or omission of appropriate residues. The peptides of the present invention can also be produced by recombinant techniques. The coding sequence for peptides of this length can easily be synthesized by chemical techniques, e.g., the phosphotriester method described in Matteucci et al., J Am. Chem. Soc., 103: 3185, 1981.
In addition to the novel peptide inhibitors described above, the present invention further contemplates the generation terminal additions, also called fusion proteins or fusion polypeptides, of the peptides described above or identified according to the present invention. This fusion polypeptide generally has all or a substantial portion of the native molecule (i.e., the peptide inhibitors discussed above), linked at the N- and/or C-terminus, to all or a portion of a second or third polypeptide. It is contemplated that the fusion polypeptide may be produced by recombinant protein production or by automated peptide synthesis.
General principles for designing and making fusion proteins are well known to those of skill in the art. For example, fusions typically employ leader sequences from other species to permit the recombinant expression of a protein or peptide in a heterologous host. Another useful fusion includes the addition of an immunologically active domain, such as an antibody epitope, to facilitate purification of the fusion polypeptide. Inclusion of a cleavage site at or near the fusion junction will facilitate removal of the extraneous polypeptide after purification. The recombinant production of these fusions is described in further detail elsewhere in the specification. Other useful fusions include linking of functional domains, such as active sites from enzymes, glycosylation domains, cellular targeting signals or transmembrane regions.
There are various commercially available fusion protein expression systems that may be used to provide a tagged sequence in this context of the present invention. Particularly useful systems include but are not limited to the glutathione S-transferase (GST) system (Pharmacia, Piscataway, N.J.), the maltose binding protein system (NEB, Beverley, Mass.), the FLAG system (IBI, New Haven, Conn.), and the 6xHis system (Qiagen, Chatsworth, Calif.). These systems are capable of producing recombinant polypeptides bearing only a small number of additional amino acids, which are unlikely to affect the biologically relevant activity of the recombinant fusion protein. For example, both the FLAG system and the 6xHis system add only short sequences, both of which are known to be poorly antigenic and which do not adversely affect folding of the polypeptide to its native conformation. Another N-terminal fusion that is contemplated to be useful is the fusion of a Met-Lys dipeptide at the N-terminal region of the protein or peptides.
In addition to creating fusion polypeptides, it is contemplated that the fusion proteins or the peptide inhibitors may be further modified to incorporate, for example, a label or other detectable moiety.
Preferred peptide will comprise internally quenched labels that result in increased detectability after cleavage of the peptide inhibitors. The peptide inhibitors may be modified to have attached a paired fluorophore and quencher including but not limited to 7-amino-4-methyl coumarin and dinitrophenol, respectively. Other paired fluorophores and quenchers include bodipytetramethylrhodamine and QSY-5 (Molecular Probes, hIc.). In a variant of this assay, biotin or another suitable tag may be placed on one end of the peptide to anchor the peptide to a substrate assay plate and a fluorophore may be placed at the other end of the peptide. Useful fluorophores include those listed above as well as Europium labels such as W8044 (EGg Wallac, Inc.).
Further, the peptides may be labeled using labels well known to those of skill in the art, e.g., biotin labels are particularly contemplated. The use of such labels is well known to those of skill in the art and is described in, e.g., U.S. Pat. No. 3,817,837; U.S. Pat. No. 3,850,752; U.S. Pat. No. 3,996,345 and U.S. Pat. No. 4,277,437. Other labels that will be useful include but are not limited to radioactive labels, fluorescent labels and chemiluminescent labels. U.S. Patents concerning use of such labels include for example U.S. Pat. No. 3,817,837; U.S. Pat. No. 3,850,752; U.S. Pat. No. 3,939,350 and U.S. Pat. No. 3,996,345. Any of the peptides of the present invention may comprise one two or more of any of these labels.
Disruption of the Gs-tubulin complex can also be accomplished through the use of an organochemical composition (i.e., a small molecule inhibitor) that interferes with the Gs binding to tubulin or tubulin binding to Gs, by use of an antibody that blocks an Gs active site or the Gs binding site on tubulin or tubulin binding Gs, or by use of a molecule that mimics the tubulin target (tubulin or Gs).
With respect to small molecule inhibitors such compounds may be identified through standard screening assays. In particular, small molecules of the invention may be designed or developed based on the peptides set out in Table 1. Various candidate substances can be contacted with Gs followed by further determination of the ability of treated Gs to bind to tubulin. An agent that inhibits such binding will be a useful for blocking the Gs-tubulin interaction. Alternatively, small molecules that bind to tubulin and block binding to Gs are also contemplated.
The present invention provides for antibodies and antibody fragments that bind to tubulin or Gs and antagonize the Gs-tubulin interaction. The invention also provides for antibodies that bind to the tubulin or Gs and induce a conformational change that prevents Gs-tubulin interaction. The antibodies may be polyclonal including monospecific polyclonal, monoclonal (mAbs), recombinant, chimeric, humanized such as CDR-grafted, human, single chain, and/or bispecific, as well as fragments, variants or derivatives thereof. Antibody fragments include those portions of the antibody which bind to an epitope on tubulin or Gs. Examples of such fragments include Fab and F(ab) fragments generated by enzymatic cleavage of full-length antibodies. Other binding fragments include those generated by recombinant DNA techniques, such as the expression of recombinant plasmids containing nucleic acid sequences encoding antibody variable regions. The methods by which antibodies are generated are well known to those of skill in the art.
A particularly useful antibody for disrupting the Gs-tubulin interaction is a single chain antibody. Methods for the production of single-chain antibodies are well known to those of skill in the art. The skilled artisan is referred to U.S. Pat. No. 5,359,046, (incorporated herein by reference) for such methods. A single chain antibody, preferred for the present invention, is created by fusing together the variable domains of the heavy and light chains using a short peptide linker, thereby reconstituting an antigen binding site on a single molecule.
Single-chain antibody variable fragments (Fvs) in which the C-terminus of one variable domain is tethered to the N-terminus of the other via a 15 to 25 amino acid peptide or linker, have been developed without significantly disrupting antigen binding or specificity of the binding. These Fvs lack the constant regions (Fc) present in the heavy and light chains of the native antibody.
Polyclonal antibodies of the invention generally are produced in animals (e.g., rabbits or mice) by means of multiple subcutaneous or intraperitoneal injections of the target protein of interest and an adjuvant. It may be useful to conjugate a target peptide to a carrier protein that is immunogenic in the species to be immunized, such as keyhole limpet heocyanin, serum, albumin, bovine thyroglobulin, or soybean trypsin inhibitor. Also, aggregating agents such as alum are used to enhance the immune response. After immunization, the animals are bled and the serum is assayed for antibody titer.
Monoclonal antibodies are produced using any method which provides for the production of antibody molecules by continuous cell lines in culture. Examples of suitable methods for preparing monoclonal antibodies include the hybridoma methods of Kohler et al., Nature, 256:495-497 (1975) and the human B-cell hybridoma method, Kozbor, J. Immunol., 133:3001 (1984); Brodeur et al., Monoclonal Antibody Production Techniques and Applications, pp. 51-63 (Marcel Dekker, Inc., New York, 1987). Also provided by the invention are hybridoma cell lines which produce monoclonal antibodies reactive with target peptides of the invention.
With respect to inhibitors that mimic tubulin targets, the use of mimetics provides one example of custom designed molecules. Such molecules may be small molecule inhibitors that specifically inhibit Gs binding to tubulin or tubulin binding to Gs. Such molecules may be sterically similar to the actual target compounds, at least in key portions of the target's structure and or organochemical in structure. Alternatively these inhibitors may be peptidyl compounds, these are called peptidomimetics. Peptide mimetics are peptide-containing molecules which mimic elements of protein secondary structure. The underlying rationale behind the use of peptide mimetics is that the peptide backbone of proteins exists chiefly to orient amino acid side chains in such a way as to facilitate molecular interactions, such as those of ligand and receptor. An exemplary peptide mimetic of the present invention would, when administered to a subject, bind Gs in a manner analogous to the tubulin domain binding to wild-type Gs.
Successful applications of the peptide mimetic concept have thus far focused on mimetics of -turns within proteins, which are known to be highly antigenic. Likely -turn structures within an antigen of the invention can be predicted by computer-based algorithms as discussed above. Once the component amino acids of the turn are determined, mimetics can be constructed to achieve a similar spatial orientation of the essential elements of the amino acid side chains.
In a preferred embodiment, the present invention will provide an agent that binds competitively to tubulin at the Gs binding site. In a more preferred embodiment, the agent will have an even greater affinity for tubulin than Gs does. Affinity for tubulin can be determined in vitro by performing kinetic studies on binding rates.
In a preferred embodiment, the present invention will provide an agent that binds competitively to Gs at the tubulin binding site. In a more preferred embodiment, the agent will have an even greater affinity for Gs than tubulin does. Affinity for Gs can be determined in vitro by performing kinetic studies on binding rates.
Other compounds may be developed based on computer modeling and predicted higher order structure, of the heterotrimeric complex of Gs molecule and tubulin, such as the sites identified in FIGS. 3 and 3b. This approach has proved successful in developing inhibitors for a number of receptor/ligand interactions. The detailed structural information of the complex can guide one to build specific small chemical entities that interact with this site. One such a structure-based approach is to use nuclear magnetic resonance (NMR) spectroscopy and a linked-fragment approach to identify chemical leads for development of specific ligands for therapeutic targets (Hajduk et al., Science, 278, 498-499, 1997; Moore, Curr Opin in Biotech., 10, 54-58, 1999; Pellecchia et al., Nature Reviews Drug Discovery, 1, 211-219, 2002). The novel ligands are chemical entities constructed from building blocks identified from NMR-based screening and optimized for binding to a target protein.
The NMR-based chemical compound screening has significant advantages that make it preferable even over the newest methods of high-throughout screening of natural products or combinatorial chemical libraries (Fejzo et al., Chem. Biol. 6, 755-769, 1999; Hajduk et al., J Am Chem Soc 119, 5818-5827, 1997; Hajduk et al., Bioorganic Medicinal Chemistry Letters 9, 2403-2406, 1999; Pellecchia et al., J Biomol NMR 22, 165-173, 2002). These unique advantages include (i) structure-based and selective screening for specific sites on a target protein; (ii) rapid and reliable screening of weak binding ligands; (iii) a large virtual library of small chemical compounds; and (iv) independent optimization of individual chemical fragments. The resulting linked chemical compounds with high affinity and selectivity are then subject to detailed structure-based analysis of their interactions with the target protein using a combination of NMR and computational modeling techniques. Refinement, chemical diversification through various chemical linkages and selectivity enhancement are achieved at this stage.
For a detailed description of methods for identification of small molecule inhibitors those of skill in the art are referred to WO 01/51521, which describes the three-dimensional structure of a complex between phosphotyrosine binding domain of Suc1-associated neurotrophic factor target protein and the SNT binding site of fibroblast growth factor receptor. Rational drug design predicated on the three-dimensional structure of this interaction is described in detail. It is contemplated that the techniques therein may be used for rational drug design to identify agents that can inhibit the deleterious effects of Gs binding to tubulin. For example such a method would involve identifying a compound that destabilizes the Gs-tubulin complex and would involve obtaining a set of atomic coordinates that define the three dimensional structure of a Gs-tubulin complex. These coordinates are determined using a complex which comprises an tubulin protein interacting with a Gs protein (Gs-tubulin complex). The next step involves performing rational drug design with the atomic coordinates to select a drug that interferes with the Gs-tubulin complex at a given site, such as the binding site. This rational drug design is preferably performed in conjunction with computer modeling. Upon selection of the candidate drug, the candidate is contacted with a Gs-tubulin complex comprising a full length or fragment of tubulin protein and a full length or fragment of a Gs protein. The stability of the Gs-tubulin complex is monitored in the presence and absence of the candidate substance to identify a potential therapeutic agent which destabilizes the complex. Similar methods may be performed to identify a compound which inhibits the formation of the complex. Such methods are described in detail in WO 01/51521.
Methods of Screening for Compounds that Disrupt the Gs-Tubulin Complex
The present invention also contemplates screening for compounds that disrupt the interaction of Gs-tubulin complex. These compounds are contemplated to be potential antidepressant agents or therapeutics for mood or anxiety disorders.
This realization affords the ability to create cellular, organ and organismal systems which mimic these diseases, which provide an ideal setting in which to test various compounds for therapeutic activity. Particularly preferred compounds exhibit antidepressant effects by disrupting the interaction of Gs-tubulin or induce a shift in the subcellular localization of Gs from a tubulin-rich location to a region of less tubulin. In the screening assays of the present invention, the candidate compound may first be screened for basic biochemical activitye.g., binding to a target molecule and then tested for its ability to induce antidepressant effects, at the cellular, tissue or whole animal level.
The present invention provides methods of screening for compounds that disrupt the Gs-tubulin interaction. It is contemplated that this screening techniques will prove useful in the identification of compounds that induce anti-depressant effects.
In these embodiment, the present invention is directed to a method for determining the ability of a candidate compound to disrupt the Gs-tubulin interaction, generally including the steps of:
To identify a candidate compound as being capable of disrupting the Gs-tubulin interaction or exhibiting antidepressant activity in the assay above, one would measure or determine various characteristics of the cell, for example, an increase in adenylyl cyclase, activity, particularly outside of lipid rafts, would indicate a test compound that is capable of disrupting the Gs-tubulin interaction. One would add the candidate compound to the cell and determine the response in the presence of the candidate compound. A candidate substance which modulates any of these characteristics is indicative of a candidate substance having modulatory activity. In the in vivo screening assays of the present invention, the compound is added to the cells, over period of time and in various dosages, and desired cellular response is measured.
As used herein the term candidate compound refers to any molecule that may potentially act as an inhibitor of the Gs-tubulin complex. The candidate compound may be a protein or fragment thereof, a small molecule inhibitor, peptidomimetics or antibody. It may prove to be the case that the most useful pharmacological compounds will be compounds that are structurally related to other known antidepressant agents. Rational drug design includes not only comparisons with known antidepressant agents, but predictions relating to the structure of target molecules. Particularly useful compounds for use in rational drug design are those that will disrupt the interaction of Gs with tubulin.
The goal of rational drug design is to produce structural analogs of biologically active polypeptides or target compounds. By creating such analogs, it is possible to fashion drugs which are more active or stable than the natural molecules, which have different susceptibility to alteration or which may affect the function of various other molecules. In one approach, one would generate a three-dimensional structure for a molecule like tubulin, or a fragment thereof. This could be accomplished by x-ray crystallography, computer modeling or by a combination of both approaches.
It also is possible to use antibodies to ascertain the structure of a target compound or inhibitor. In principle, this approach yields a pharmacore upon which subsequent drug design can be based. It is possible to bypass protein crystallography altogether by generating anti-idiotypic antibodies to a functional, pharmacologically active antibody. As a mirror image of a mirror image, the binding site of anti-idiotype would be expected to be an analog of the original antigen. The anti-idiotype could then be used to identify and isolate peptides from banks of chemically- or biologically-produced peptides. Selected peptides would then serve as the pharmacore. Anti-idiotypes may be generated using the methods described herein for producing antibodies, using an antibody as the antigen.
On the other hand, one may simply acquire, from various commercial sources, small molecule libraries that are believed to meet the basic criteria for useful drugs in an effort to brute force the identification of useful compounds. Screening of such libraries, including combinatorially generated libraries (e.g., peptide libraries), is a rapid and efficient way to screen large number of related (and unrelated) compounds for activity. Combinatorial approaches also lend themselves to rapid evolution of potential drugs by the creation of second, third and fourth generation compounds molded of active, but otherwise undesirable compounds.
Candidate compounds may include fragments or, parts of naturally-occurring compounds or may be found as active combinations of known compounds which are otherwise inactive. It is proposed that compounds isolated from natural sources, such as animals, bacteria, fungi, plant sources, including leaves and bark, and marine samples may be assayed as candidates for the presence of potentially useful pharmaceutical agents. It will be understood that the pharmaceutical agents to be screened could also be derived or synthesized from chemical compositions or man-made compounds. Thus, it is understood that the candidate substance identified by the present invention may be polypeptide, polynucleotide, small molecule inhibitors or any other compounds that may be designed through rational drug design starting from known antidepressants.
Effective amounts in certain circumstances are those amounts effective to disrupt the Gs-tubulin interaction in a cell. Compounds that achieve significant appropriate changes in activity will be used.
B. In vitro Assays
A quick, inexpensive and easy assay to run is a binding assay. Binding of a molecule to a target may, in and of itself, be inhibitory, due to steric, allosteric or charge-charge interactions. This can be performed in solution or on a solid phase and can be utilized as a first round screen to rapidly eliminate certain compounds before moving into more sophisticated screening assays. In one embodiment of this kind, the screening of compounds that bind to tubulin or fragment thereof or microtubules is provided. In another embodiment, the screening of compounds that bind to Gs or fragments thereof is provided.
The target may be either free in solution, fixed to support, such as a membrane, expressed in or on the surface of a cell. Either the target or the compound may be labeled, thereby permitting determining of binding. In another embodiment, the assay may measure the inhibition of binding of a target to a natural or artificial substrate or binding partner (such as tubulin and Gs). Competitive binding assays can be performed in which one of the agents (Gs, for example) is labeled. Usually, the target will be the labeled species, decreasing the chance that the labeling will interfere with the binding moiety's function. One may measure the amount of free label versus bound label to determine binding or inhibition of binding.
A technique for high throughput screening of compounds is described in WO 94/03564. Large numbers of small peptide test compounds are synthesized on a solid substrate, such as plastic pins or some other surface. The peptide test compounds are reacted with, for example, tubulin and washed. Bound polypeptide is detected by various methods.
Purified target, such as tubulin or Gs, can be coated directly onto plates for use in the aforementioned drug screening techniques. However, non-neutralizing antibodies to the polypeptide can be used to immobilize the polypeptide to a solid phase. Also, fusion proteins containing a reactive region (preferably a terminal region) may be used to link an active region to a solid phase.
C. In cyto Assays
Various cell lines can be utilized for screening of candidate substances. For example, rat glioma cells, such as C6-2B cells or human neuroblastoma cells (SK N SH), can be used to study various functional attributes of candidate compounds. In such assays, the compound would be formulated appropriately, given its biochemical nature, and contacted with a target cell. In addition, cells may be selected for assays of the invention for their endogenous Gs-coupled receptors or receptors. Further, the assays of the invention may be carried out with cells that are co-transfected with nucleic acids encoding GFP-Gs (green fluorescent protein-Gs).
Depending on the assay, culture may be required. As discussed above, the cell may then be examined by virtue of a number of different physiologic assays. Alternatively, molecular analysis may be performed in which the function of Gs and related pathway may be explored. This involves assays such as those for protein expression, enzyme function, substrate utilization, mRNA expression (including differential display of whole cell or polyA RNA) and others. For cell-free assays, the Gs-tubulin interaction can be assessed by using a solid-phase binding assay such as the SPOT assay described in Example 1.
D. In vivo Assays
The present invention particularly contemplates the use of various animal models. Treatment of these animals with test compounds will involve the administration of the compound, in an appropriate form, to the animal. Administration will be by any route that could be utilized for clinical or non-clinical purposes, including but not limited to oral, nasal, buccal, or even topical. Alternatively, administration may be by intratracheal instillation, bronchial instillation, intradermal, subcutaneous, intramuscular, intraperitoneal or intravenous injection. Specifically contemplated are systemic intravenous injection, regional administration via blood or lymph supply.
Determining the effectiveness of a compound in vivo may involve a variety of different criteria. Such behaviors exhibited by the animal include, but are not limited to, reduced sleep disruption, reduced weight loss, improved reward insensitivity, reduced attention deficits, increased sexual activity, increased time of struggling in the forced swim test, increased sucrose preference and increased social interaction. Additional assays for measuring the effectiveness of antidepressant activity are provided in Willner, Trends Pharmacol Sci. 12(4):131-6, 1991 and Dekeyne, Therapie. 60(5):477-84, 2005. It also is possible to perform histologic studies on tissues from these animals, or to examine the molecular and morphological state of the cells.
In the sections above, the present invention describes various novel compositions for disruption of the Gs-tubulin complex, also described are assays for identifying additional composition. It is contemplated that therapeutic compositions of the present invention will be useful in the intervention of various disease states such as for example, mood disorders such as depression and bipolar disorder, anxiety disorders such as phobia disorder, panic disorders, stress disorders and obsessive-compulsive disorders, and addiction to abusive drugs such as cocaine or opiates. Such agents may be used either alone or in combination with other therapeutic agents presently being used to control these disorders. In order to be used in such therapeutic indications, it will be preferable to prepare the compositions of the invention in pharmaceutically acceptable formats.
Also, it should be understood that it may well be that purified compositions that disrupts the Gs-tubulin complex may be routinely prepared into pharmaceutically acceptable forms of the proteins once they are isolated from the media and/or cellular compositions described above. Generally, this will entail preparing compositions that are essentially free of pyrogens, as well as other impurities that could be harmful to humans or animals.
One will generally desire to employ appropriate salts and buffers to render the compositions stable and allow for uptake by target cells. Buffers also will be employed when recombinant cells or nucleic acids are introduced into a subject. The phrase pharmaceutically or pharmacologically acceptable refer to molecular entities and compositions that do not produce adverse, allergic, or other untoward reactions when administered to an animal or a human. As used herein, pharmaceutically acceptable carrier includes any and all solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents and the like. The use of such media and agents for pharmaceutically active substances is well known in the art. Except insofar as any conventional media or agent is incompatible with the therapeutic compositions produced by the present invention, its use in therapeutic compositions is contemplated. Supplementary active ingredients also can be incorporated into the compositions.
The compositions of the present invention include classic pharmaceutical preparations. Administration of these compositions according to the present invention will be via any common route so long as the target tissue is available via that route. The pharmaceutical compositions may be introduced into the subject by any conventional method, e.g., by intravenous, intradermal, intramusclar, intramammary, intraperitoneal, intrathecal, intraocular, retrobulbar, intrapulmonary (e.g., term release); by oral, sublingual, nasal, anal, vaginal, or transdermal delivery, or by surgical implantation at a particular site, e.g., embedded under the splenic capsule, brain, or in the cornea. The treatment may consist of a single dose or a plurality of doses over a period of time.
The compositions produced using the present invention may be prepared for administration as solutions of free base or pharmacologically acceptable salts in water suitably mixed with a surfactant, such as hydroxypropylcellulose. Dispersions also can be prepared in glycerol, liquid polyethylene glycols, and mixtures thereof and in oils. Under ordinary conditions of storage and use, these preparations contain a preservative to prevent the growth of microorganisms.
The pharmaceutical forms suitable for injectable use include sterile aqueous solutions or dispersions and sterile powders for the extemporaneous preparation of sterile injectable solutions or dispersions. In all cases the form must be sterile and must be fluid to the extent that easy syringability exists. It must be stable under the conditions of manufacture and storage and must be preserved against the contaminating action of microorganisms, such as bacteria and fungi. The carrier can be a solvent or dispersion medium containing, for example, water, ethanol, polyol (for example, glycerol, propylene glycol, and liquid polyethylene glycol, and the like), suitable mixtures thereof, and vegetable oils. The proper fluidity can be maintained, for example, by the use of a coating, such as lecithin, by the maintenance of the required particle size in the case of dispersion and by the use of surfactants. The prevention of the action of microorganisms can be brought about by various antibacterial an antifungal agents, for example, parabens, chlorobutanol, phenol, sorbic acid, thimerosal, and the like. In many cases, it will be preferable to include isotonic agents, for example, sugars or sodium chloride. Prolonged absorption of the injectable compositions can be brought about by the use in the compositions of agents delaying absorption, for example, aluminum monostearate and gelatin.
Sterile injectable solutions are prepared by incorporating the active compounds in the required amount in the appropriate solvent with various of the other ingredients enumerated above, as required, followed by filtered sterilization. Generally, dispersions are prepared by incorporating the various sterilized active ingredients into a sterile vehicle which contains the basic dispersion medium and the required other ingredients from those enumerated above. In the case of sterile powders for the preparation of sterile injectable solutions, the preferred methods of preparation are vacuum-drying and freeze-drying techniques which yield a powder of the active ingredient plus any additional desired ingredient from a previously sterile-filtered solution thereof.
As used herein, pharmaceutically acceptable carrier includes any and all solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents and the like. The use of such media and agents for pharmaceutical active substances is well known in the art. Except insofar as any conventional media or agent is incompatible with the active ingredient, its use in the therapeutic compositions is contemplated. Supplementary active ingredients also can be incorporated into the compositions.
For oral administration the compositions produced by the present invention may be incorporated with excipients and used in the form of non-ingestible mouthwashes and dentifrices. A mouthwash may be prepared incorporating the active ingredient in the required amount in an appropriate solvent, such as a sodium borate solution (Dobell's Solution). Alternatively, the active ingredient may be incorporated into an antiseptic wash containing sodium borate, glycerin and potassium bicarbonate. The active ingredient may also be dispersed in dentifrices, including: gels, pastes, powders and slurries. The active ingredient may be added in a therapeutically effective amount to a paste dentifrice that may include water, binders, abrasives, flavoring agents, foaming agents, and humectants.
The compositions of the present invention may be formulated in a neutral or salt form. Pharmaceutically-acceptable salts include the acid addition salts (formed with the free amino groups of the protein) and which are formed with inorganic acids such as, for example, hydrochloric or phosphoric acids, or such organic acids as acetic, oxalic, tartaric, mandelic, and the like. Salts formed with the free carboxyl groups also can be derived from inorganic bases such as, for example, sodium, potassium, ammonium, calcium, or ferric hydroxides, and such organic bases as isopropylamine, trimethylamine, histidine, procaine and the like.
Upon formulation, solutions will be administered in a manner compatible with the dosage formulation and in such amount as is therapeutically effective. The formulations are easily administered in a variety of dosage forms such as injectable solutions, drug release capsules and the like. For parenteral administration in an aqueous solution, for example, the solution should be suitably buffered if necessary and the liquid diluent first rendered isotonic with sufficient saline or glucose. These particular aqueous solutions are especially suitable for intravenous, intramuscular, subcutaneous and intraperitoneal administration.
Unit dose is defined as a discrete amount of a therapeutic composition dispersed in a suitable carrier. For example, parenteral administration may be carried out with an initial bolus followed by continuous infusion to maintain therapeutic circulating levels of drug product. Those of ordinary skill in the art will readily optimize effective dosages and administration regimens as determined by good medical practice and the clinical condition of the individual patient.
The frequency of dosing will depend on the pharmacokinetic parameters of the agents and the routes of administration. The optimal pharmaceutical formulation will be determined by one of skill in the art depending on the route of administration and the desired dosage. See for example Remington's Pharmaceutical Sciences, 18th Ed. (1990, Mack Publ. Co, Easton Pa. 18042) pp 1435-1712, incorporated herein by reference. Such formulations may influence the physical state, stability, rate of in vivo release and rate of in vivo clearance of the administered agents. Depending on the route of administration, a suitable dose may be calculated according to body weight, body surface areas or organ size. Further refinement of the calculations necessary to determine the appropriate treatment dose is routinely made by those of ordinary skill in the art without undue experimentation, especially in light of the dosage information and assays disclosed herein as well as the pharmacokinetic data observed in animals or human clinical trials.
Appropriate dosages may be ascertained through the use of established assays for determining blood levels in conjunction with relevant dose-response data. The final dosage regimen will be determined by the attending physician, considering factors which modify the action of drugs, e.g., the drug's specific activity, severity of the damage and the responsiveness of the patient, the age, condition, body weight, sex and diet of the patient, the severity of any infection, time of administration and other clinical factors. As studies are conducted, farther information will emerge regarding appropriate dosage levels and duration of treatment for specific diseases and conditions.
It will be appreciated that the pharmaceutical compositions and treatment methods employing such compositions may be useful in fields of human medicine and veterinary medicine. Thus the subject to be treated may be a mammal, preferably human or other animal. For veterinary purposes, subjects include for example, farm animals including cows, sheep, pigs, horses and goats, companion animals such as dogs and cats, exotic and/or zoo animals, laboratory animals including mice rats, rabbits, guinea pigs and hamsters; and poultry such as chickens, turkey, ducks and geese.
In view of the foregoing discussion and by way of illustration of the invention, the examples describe methods for identifying peptides that disrupt the interaction of Gs and tubulin, describes the characterization of the Gs and tubulin binding sites and describes the heterotrimeric complex comprising Gs and tubulin.
Peptides were synthesized on to a cellulose membrane with PEG spacer (812 cm2) (AIMS Scientific Products, Braunschweig, Germany) via the C-terminal amino acid in sequential spots by the use of a SPOT synthesis kit (SIGMA genosys, St. Louis, Mo.) (Frank, J. Immunol. Methods 267: 13-26, 2002; Frank, Tetrahedron 48: 9217-9232, 1992). The peptides corresponded to the amino acid sequence of Gs (Genbank accession number P04895, homo sapiens, Gs long form; SEQ ID NO: 15), which was divided into overlapping peptides (12 amino acids in length with 7 amino acid overlap between sequential peptide, 73 total spots). Spot 1 corresponded to amino acids 1-12 and Spot 2 corresponded to residues 6-17 in the primary amino acid sequence (SEQ ID NO: 15), etc.
Another SPOT membrane was created based on the results of SPOT membrane 1, to compare Gs-peptides that bound tubulin to Gt-peptides, because Gt has been shown to not bind tubulin. The sequence of the Gs-peptides that were found to bind tubulin (with an approximate 5 amino acid extension added on both sides) was used as well as the corresponding amino acid sequence of Gt. For instance, the Gs sequence of SPOTS 4-11 in FIG. 1 was taken, the 5 amino acids from the Gs sequence toward the N-terminal as well as C-terminal were added to the both ends of the sequence, and then the sequence was divided into overlapping 15 amino acid length peptides. These peptides were divided into overlapping peptides (15 amino acids in length with 10 amino acid overlap between sequential peptide, 70 total spots). Spot 1 corresponded to the 1st 15 amino acids from Gs that bound tubulin, spot 2 corresponded to sequence aligned amino acids from Gt, spot 3 corresponded to a peptide shifted 5 amino acids toward the C-terminal end of that sequence from Gs and spot 4 corresponded to the sequence aligned amino acids from Gt, etc. Certain regions of Gs lacked corresponding regions in Gt. For these regions of Gt, the corresponding amino acids of Gs in these Gt-peptides were substituted.
Membranes were blocked with TBS-containing 0.1% Tween-20 (TBS-T) with 2.5% milk for 1 hour, washed with TBS-T and incubated overnight at 4 C. with 150 nM tubulin in RIPA buffer (10 mM Tris-Cl, pH 7.4, 1% Titron-X-100, 1% Sodium Deoxycholate, 1% SDS and 500 mM NaCl). This assay was also carried out in the above buffer in the absence of SDS under nonreducing conditions. Next, the membranes were washed 3 with RIPA buffer and incubated with anti -tubulin antibody (Sigma, St. Louis, Mo.), followed by the horseradish peroxidase conjugated secondary antibody (1 hour each at room temperature in the RIPA buffer containing 1% milk) and developed with enhanced chemiluminescence western blotting detection reagents (Amersham Biosciences).
For stripping the membranes, a previously described procedure was modified (14). Before reuse, the membrane was blocked and probed with the antibodies to verify that residual peptide-bound protein was stripped. Control experiments verified no specific binding of the primary antibody, secondary antibody or tubulin with the peptides after stripping the membranes. To further control for specific-binding, tubulin was incubated with equimolar Gs (150 nM) at 37 C. for 1 hour to form protein complexes and next, the complexes were incubated with the SPOT membrane (SPOT membrane 1) as above. This prior incubation with Gs prevented tubulin from binding to the immobilized peptides on SPOT membrane 1.
Because the crystal structure of Gs was determined as a dimer (Sunahara et al., Science 278: 1943-1947, 1997), one of the Gs molecules in the dimer was deleted along with all of its corresponding ligands (the PDB file is 1AZT). In the remaining Gs molecule, the Mg2+ and PO44 molecules were deleted and GTPS was retained. For the structure of the tubulin dimer (Nogales et al., Cell 96: 79-88, 1999), the -tubulin subunit was removed along with its corresponding nucleotide and taxol (the PDB file is 1TUB). For the remaining subunit, GDP was retained and taxol was removed.
The docking algorithm, ZDOCK2.3 (Chen, et al., Proteins 52: 68-73, 2003) was used first for the unbound protein-protein docking where 2000 predictions were generated using -tubulin as receptor and Gs as ligand. ZDOCK was downloaded to a Linux system. Parameters were added to uniCHARM for GTPS (GSP, as named in the original GsPDB). The parameters used were from the uniCHARM file for GNP and the sulfur in GTPS was used from the sulfur in CYS. ZDOCK uses a fast Fourier Transform algorithm. The protein-protein interface was evaluated by shape complementarity, desolvation energy, and electrostatics.
The 2000 complexes generated from ZDOCK were then submitted to ClusPro, http://nrc.bu.edu/cluster/ (Comeau et al., Bioinformatics 20: 45-50, 2004). ClusPro calculates pair-wise RMSD values to find neighbour complexes within 9 of another complex. These complexes were then clustered and the top 30 clusters were returned for further evaluation. To minimize the side chain clashes, the ranked complexes in the clusters were subjected to a minimization using CHARMM (Comeau et al., Bioinformatics 20: 45-50, 2004). These clusters were then ranked according to population in each cluster. The parameters for ClusPro were set in the advanced options section for filtering and clustering, with a radius of 9 , the electrostatic hits at 1500, and a return cluster output of 30. The representative complex for each particular cluster was the complex that is most centrally located in the array of complexes.
Further characterization of the top five complexes was performed by energy minimization of the ZDOCK/ClusPro-derived complexes. Each of the top 5 complexes was further examined for an additional 5000 cycles with the SANDER package within AMBER7 and a minimization energy score was determined. The buried surface area (BSA) of each complex was determined within the GRASP program, as to determine which complex has the largest contact area.
Visualization of protein structures was performed via GRASP (Nicholls et al., Proteins 11: 281-296, 1991) or SYBYL 6.9 (Tripos, Inc, St. Louis, Mo.) software.
Ovine brain tubulin was prepared as previously described (Shelanski et al., Proc. Natl. Acad. Sci. USA. 70: 765-768, 1973). Briefly, the brains were obtained at a local slaughterhouse from freshly-killed animals and were placed on ice upon removal from animals. Purity of the prepared tubulin, as determined by SDS gel electrophoresis, was always greater than 95%. Nucleotide replacement on tubulin was performed as before (Popova et al., J. Neurosci. 20: 2774-2782, 2000), using charcoal to strip bound nucleotide. Protein concentrations were determined by Coomasie Blue binding (BioRad Protein Assay) with Bovine Serum Albumin as a standard (Bradford, Anal. Biochem. 72: 248-254, 1976).
The SPOT technique was used to determine the specific domains on Gs that bind to tubulin (Frank, J. Immunol. Methods 267: 13-26, 2002). To carry out this analysis, peptides corresponding to the primary Gs sequence were covalently attached to a cellulose-based membrane. The membrane was probed with tubulin in the GDP or GTP bound states. Tubulin-GDP interacted with 9 peptides and tubulin-GTP interacted with 6 peptides. Probing was carried out with binding a tubulin specific antibody, therefore control experiments were performed to determine whether the primary or secondary antibody interact non-specifically with the peptides. No non-specific binding was detected on the SPOT membrane. After each experiment examining tubulin binding to the SPOT membrane, the membrane was stripped and probed with primary antibody followed by secondary antibody. Control experiments indicated no residual tubulin or antibody binding to the immobilized peptides.
Tubulin binds with a high affinity (KD 130 nM) to Gs and Gi, but does not bind to the photoreceptor G protein, Gt(Wang et al., J. Biol. Chem. 265: 1239-1242, 1990). FIG. 1a shows the amino acids of the Gs-peptide that exhibited binding to tubulin along with the corresponding amino acids from Gi and Gt. In spots 4-11, the amino terminal region of Gs, there are an additional 4 amino acids in Gs and Gi1 that are not found in Gt. Other interesting domains that interacted with tubulin, the 2-4 and 3-5 regions, are known to be important in the interaction with adenylyl cyclase (Tesmer et al., Science 278: 1907-1916, 1997). Furthermore, in the 3-5 region, there is a tryptophan in Gs (TRP281) and Gi1 that corresponds to a tyrosine in Gt. This residue is located on the protein surface where solvent-exposed hydrophobic residues often contribute to protein-protein interactions (FIGS. 1 and 2).
The structure of G proteins includes two domains: a GTPase domain and an -helical domain (Lambright et al., Nature 369: 621-628, 1994). Data from peptide binding (SPOT studies) suggest that the primary tubulin binding sites on Gs are localized to the GTPase domain (FIG. 2). The GTPase domain of Gs includes the switch regions (Lambright et al., Nature 369: 621-628, 1994): switch I (F-2); switch II (3-2-4), and switch III (4-3), which are important for adenylyl cyclase activation and are structurally altered upon exchange of GDP for GTP (17). Two domains of Gs that bound tubulin on the SPOT membrane, 2-4 (spots 46-47, residues 240-256) and 3-5 (spots 54-55, residues 280-296), are included in these regions.
To further test the binding of tubulin to the Gs-peptides, the binding of these Gs-peptides was compared to the binding of Gt-Peptides, using a separate peptide-array membrane (see Example 1) because Gt is known to not bind tubulin despite significant sequence and structural similarities to Gs. Peptides from Gs that had enhanced binding as compared to the corresponding Gt peptide are shown (FIG. 1b). As before, this membrane was probed with tubulin in both the GDP and GTP bound state. Since we were probing for binding with antibodies, control experiments were also performed which demonstrated no non-specific binding of the primary or secondary antibodies or residual tubulin binding to the membrane following stripping (see Example 1). Distinct regions in the amino terminus and the switch II and III regions of Gs bind tubulin but not the corresponding region of Gt providing novel leads into which domains of Gt may cause it to not bind tubulin (see FIG. 1b). These data also reconfirmed the binding shown with the first SPOT membrane even though the length of the peptides were longer in this membrane (15 vs. 12 amino acids), in which all the Gs-peptides probed in this membrane bound to tubulin in one or the other (or both) nucleotide states of tubulin.
To carry out binding analysis by Surface Plasmon Resonance (SPR), quantitative analysis of peptide-tubulin interactions were performed on a BIAcore advance system (Pharmacia Biosensor AB, Uppsala, Sweden). Tubulin was immobilized in Hepes buffer pH 8.0 (50 mM Hepes, 0.1M NaCl, 1 mMEDTA, 1 mM DTT) at a flow rate of 10 L/min on sensor chip CM4 according to the manufacturer's instructions. The immobilization level was approx. 300 RU (resonance units). Different concentrations of synthetic peptides (25-100 M in Hepes buffer) were injected onto the flow cell at a flow rate of 30 L/min. Sensograms were subtracted from buffer blank injections. The surface was regenerated by injection of 1M NaCl, 1% tritronX-100 in Hepes buffer (pH8.0). Sensograms were analzyed by the BIAevaluation 4.1 program (Pharmacia Biosensor AB).
To further analyze the interaction between Gs and tubulin, two protein-protein docking programs (ZDOCK (Chen, et al., Proteins 52: 68-73, 2003) and ClusPro (Comeau et al., Bioinformatics 20: 45-50, 2004) were used to generate and analyze protein complexes. These docking algorithms do not include conformational changes induced by the interactions; however, these programs have been successful in predicting near-native structural complexes for other systems (Chen, et al., Proteins 52: 68-73, 2003; Comeau et al., Bioinformatics 20: 45-50, 2004; Camacho et al., Proteins 52: 92-97, 2003). Using the structures of Gs-GTPS (Sunahara et al., Science 278: 1943-1947, 1997) and -tubulin-GDP (Nogales et al., Cell 96: 79-88, 1999), 2000 potential complexes by ZDOCK were generated and were clustered with ClusPro. ClusPro created and ranked 30 clusters with the cluster ranking based on the number of similar complexes in each cluster. The complex that is most centrally located in each cluster was used to represent that cluster (Comeau et al., Bioinformatics 20: 45-50, 2004). In the docking programs, only the structure of the -tubulin subunit was used, based on the assumption that G proteins interact only with this subunit because this subunit contains the exchangeable nucleotide that is likely involved in the transactivation mechanism (Wang et al., J. Biol. Chem. 265: 1239-1242, 1990; Rasenick et al., Methods Enzymol. 390: 389-403, 2004; Roychowdhury et al., J. Biol. Chem. 274:13485-13490, 1999).
After obtaining the final 30 complexes, the number of amino acids on Gs that were in the interface of this protein-protein interaction were analyzed and were also predicted from the SPOT membrane (Table 2). Twenty-eight of the 30 complexes included some residues on Gs that were in the protein interface that were also determined by this direct binding technique. Fourteen of the 30 complexes had over thirty percent of the residues in the interface predicted by the direct binding results of the SPOT membrane.
In Table 2, the number in parentheses in the left column indicates the number of ZDOCK-generated complexes that made up that particular cluster as determined by ClusPro. The numerator corresponds to number of residues in Gs (middle column) or -tubulin (right column) within 5 of the other protein that were detected from the SPOT membrane data (middle column) or predicted from a hypothetical -tubulin interface (right column), respectively. The denominator corresponds to the total number of residues within the interface, as determined from that ZDOCK-generated complex. The residues that were less than 5 from another residue on the other protein were considered to be in the protein interface.
In FIG. 3A, the relative position of -tubulin (for the top thirty complexes) to a fixed Gs (represented as a whole molecule) is shown. Most of the potential -tubulins cluster near the GTPase domain of Gs. This indicates a favorable orientation of -tubulin predicted by the docking programs to be located in the GTPase domain of Gs. Further, the regions predicted by the SPOT membrane are highlighted on Gs in FIG. 3A indicating many of the predicted interfaces of the 30 complexes are in this region.
The interactions that occur between tubulin dimers within microtubules (called interdimer interactions) have been well defined and occur at several residues surrounding the nucleotide on -tubulin (27 amino acids have been defined) (Nogales et al., Cell 96: 79-88, 1999). It was hypothesized that Gs binds close to the nucleotide binding site on -tubulin and in a similar region to which the -tubulin of another tubulin dimer would bind. Modeling of the top thirty complexes (FIG. 3B) with the relative position of Gs, represented by circles, to a fixed -tubulin, indicated that a majority of these docked Gs molecules were located around the exchangeable nucleotide-binding site of -tubulin. This is consistent with the above hypothesis. The residues in the interface of the top 30 complexes were analyzed to test this hypothesis. This analysis determined that most of the complexes did indeed contain these -tubulin residues in the interface (FIG. 3B. and Table 3). For 18 of the 30 complexes, over 40% of the residues on -tubulin in the interface were predicted by the above assumption. Of the top 10 complexes, only one (complex 9) did not contain any of these residues.
The top five complexes from the ZDOCK/ClusPro anaylsis were farther analyzed by energy minimization cycles (5000) and the buried surface area (BSA) of each of these complexes was then determined (Table 3). The energy values indicate that complex 2 has the lowest energy, followed by both complexes 1 and 3 being slightly higher in energy than 2, with 4 and 5 showing the highest energies of the five complexes. Complex 2 also has the most buried surface area at the interface, perhaps suggesting a better interaction of the two molecules.
In Table 3, domain interactions between the two proteins were determined by identifying residues that were less than 4 from another residue on the other protein and were considered to be potential protein contacts. Based on these residue-residue contacts, the domains that defined these residues are shown. In parentheses, the residue number or the range of residues that contributed to the residue-residue contacts are indicated. For -tubulin, cc-helices are listed as H1-H12 and -sheets are listed as B1-B10 (20).
FIG. 4 shows an -carbon alignment of the top five complexes, with a fixed Gs and the relative position of corresponding -tubulin for the top 5 complexes. As apparent, the top four complexes are similar in the location that the -tubulin interacts with the Gs. Complex 5, is in a different location on the Gs, and is significantly higher in energy than the other four complexes, and therefore, presumably, an unfavorable complex. Complex 4 is significantly higher in energy than complexes 1-3, but has a number of similar regions of interactions with the top three complexes (Table 2). The top complexes have a number of common contact regions between Gs and -tubulin, in particular, in the regions of 3-5 and 4-6 loops of Gs (Table 3)
Prior to this study, chimeric proteins comprised of Gs, Gi1 and Gt were used to define the domains on G proteins that are important in the interaction with tubulin. One study (Popova et al., J. Biol. Chem. 269: 21748-21754, 1994) demonstrated that the amino terminus (residues 1-63) of Gs may be important for tubulin to activate adenylyl cyclase. The data provided herein indicated that the amino terminus plays a role in the binding to tubulin. For the top five complexes, only complex 4 (Table 2) includes a significant portion of the amino terminus of Gs in the interface, complex 2 includes a small portion of the amino terminus of Gs in the interface. The simplest reason for seeing less amino terminal involvement than expected in the top five complexes is that the structure of Gs used in the molecular docking studies was missing portions of the amino terminus. This absent region of the amino terminus included the following residues predicted by the SPOT technique to be important: ALA18-LYS34 and ASN66-PHE68 of SEQ ID NO: 15. Therefore, a significant portion of the amino terminus of the Gs was not available for analysis by the protein-protein docking programs.
In another study (Chen et al., J. Biol. Chem. 278: 15285-15290, 2003), it was demonstrated that one of the major Gi1-tubulin interacting domains was between residues 237-270 of Gi1 (which corresponds to residues 253-293 in Gs (SEQ ID NO: 15), regions 4-3-5). However, this was not the only region involved in binding of the two molecules (Chen et al., J. Biol. Chem. 278: 15285-15290, 2003). It is assumed that Gs and Gi1 complex with tubulin at a similar interface, regions 3-5 of the G protein family appear important for interaction with tubulin. As seen in Table 2, the 3-5 region of Gs is included in the interface of 4 of the 5 top . complexes indicating that the predicted models agree with the data from both SPOT membrane and Gt-Gi1 chimeras (Chen et al., J. Biol. Chem. 278: 15285-15290, 2003). Further, the switch II region (2-4) is in the interface of complex 4 (and complex 2), which is another domain predicted from the binding data and studies with chimeric G proteins (Chen et al., J. Biol. Chem. 278: 15285-15290, 2003).
Previous studies on G protein-tubulin interaction showed that the certain G protein heterotrimers could form a complex with tubulin (Wang et al., Biochemistry 30: 10957-10965, 1991). However, the deduced structure of Gi112 (Wall et al., Cell 83: 1047-1058, 1995) suggests that most of the switch II region (including the following domains, 2-4) as well as the amino terminus of Gi1 are involved in the interface with the 12 subunits. This interface on the G subunit has many similarities to our proposed interface on Gs to tubulin, which could be interpreted to mean that G and tubulin could not bind contemporaneously to G. A report (Ford et al., Science 280: 1271-1274, 1998) showing interactions of these domains on G with effector and G has data consistent with the possibility that both G or effector (and, by inference, tubulin) can bind to G at the same time (although this was not the conclusion of the authors). The data described herein showed some overlap between sites of G for G and effector. In fact, while there was overlap, this observation, coupled with the movable nature of the switch II and switch III regions, gives rise to the possibility that G might enjoy contemporaneous-association with G and effector (or tubulin). It is also noteworthy that some recent reports suggest that physical dissociation of G and G might not be a prerequisite of G activation (Bunemann et al., Proc. Natl. Acad. Sci. U S A. 100: 16077-16082, 2003; Gales et al., Nat Methods 2: 177-184, 2005).
Comparison of the interacting surfaces of Gs and -tubulin for the top five complexes in Table 2 indicated similar interactions. However, in FIG. 5, complex 2 is shown, because this model has the lowest energy as seen in Table 2. In this model, the switch II region (2-4) and the amino terminal of Gs both contain contacts to the H1 and H2 regions of -tubulin (FIG. 5). Interestingly, the amino terminal of Gs (seen directly under the switch II region, FIG. 5) extends directly into the nucleotide-binding pocket of -tubulin.