Apoptosis or programmed cell death is a physiologic process that plays a central role in normal development and tissue homeostasis. Many factors interact in complex pathways to lead to cell death or cell survival.

A. NF-KB 1. NF-KB in immune and irtflammatory responses NF-KB transcription factors are central coordinating regulators of innate and adaptive immune responses. A signature characteristic of NF-KB is its rapid translocation from cytoplasm to nucleus in response to a large array of extra-cellular signals, among which tumor necrosis factor (TNFa) stands out as one of the most potent. NF-KB dimers generally lie dormant in the cytoplasm of unstimulated cells, retained there by inhibitory proteins known as IKBs, and can be activated rapidly by signals that induce the sequential phosphorylation and proteolytic degradation ofIxBs. Removal of the inhibitor allows NF-KB to migrate into the cell nucleus and rapidly induce coordinate sets of defense-related genes, such as those encoding numerous cytokines, growth factors, chemokines, adhesion molecules and immune receptors. In evolutionary terms, the association between cellular defense genes and NF-KB dates as far back as half a billion years ago, because it is found in both vertebrates and invertebrates. While in the latter organisms, NF-KB factors are mainly activated by Toll receptors to induce innate defense mechanisms. In vertebrates, these factors are also widely utilized by B and T lymphocytes to mount cellular and tumoral responses to antigens.

Evidence exists for these crucial roles of NF-KB in immune and inflammatory responses. This transcription factor also plays a crucial role in widespread human diseases, including autoimmune and chronic inflammatory conditions such as asthma, rheumatoid arthritis, and inflammatory bowel disease. Indeed, the anti-inflammatory and immunosuppressive agents that are most widely used to treat these conditions such as glucocorticoids, aspirin, and gold salts, work primarily by suppressing NF-KB.

TNFa is arguably the most potent pro-inflammatory cytokine and one of the strongest activators of NF-KB. In turn, NF-KB is a potent inducer of TNFa, and this mutual regulation between the cytokine and the transcription factor is the basis for the establishment of a positive feedback loop, which plays a central role in the pathogenesis of septic shock and chronic inflammatory conditions such as rheumatoid arthritis (RA) and inflammatory bowel disease (IBD). Indeed, the standard therapeutic approach in the treatment of these latter disorders consists of the administration of high doses of NF-KB blockers such as aspirin and glucocorticoids, and the inhibition of TNFa by the use of neutralizing antibodies represents an effective new tool in the treatment of these conditions. However, chronic treatment with NF-KB inhibitors has considerable side effects, including immunosuppressive effects, and due to the onset of the host immune response, patients rapidly become refractory to the beneficial effects of anti-TNFa neutralizing antibodies.

2. NF-KB and the control of apoptosis In addition to coordinating immune and inflammatory responses, the NF-KB/Rel group of transcription factors controls apoptosis. Apoptosis, that is, programmed cell death (PCD), is a physiologic process that plays a central role in normal development and tissue homeostasis. The hallmark of apoptosis is the active participation of the cell in its own destruction through the execution of an intrinsic suicide program. The key event in this process is the activation by proteolytic cleavage of caspases, a family of evolutionarily conserved proteases. One pathway of caspase activation, or"intrinsic"pathway, is triggered by Bcl-2 family members such as Bax and Bak in response to developmental or environmental cues such as genotoxic agents. The other pathway is initiated by the triggering of"death receptors" (DRs) such as TNF-receptor 1 (TNF-R1), Fas (CD95), and TRAIL-R1 and R2, and depends on the ligand-induced recruitment of adaptor molecules such as TRADD and FADD to these receptors, resulting in caspase activation.

The deregulation of the delicate mechanisms that control cell death can cause serious diseases in humans, including autoimmune disorders and cancer. Indeed, disturbances of apoptosis are just as important to the pathogenesis of cancer as abnormalities in the regulation of the cell cycle. The inactivation of the physiologic apoptotic mechanism also allows tumor cells to escape anti-cancer treatment. This is because chemotherapeutic agents, as well as radiation, ultimately use the apoptotic pathways to kill cancer cells.

Evidence including analyses of various knockout models-shows that activation of NF-KB is required to antagonize killing cells by numerous apoptotic triggers, including TNFa and TRAIL. Indeed, most cells are completely refractory to TNFa, cytotoxicity, unless NF-KB activation or protein synthesis is blocked. Remarkably, the potent pro-survival effects of NF-KB serve a wide range of physiologic processes, including B lymphopoiesis, B-and T- cell co stimulation, bone morphogenesis, and mitogenic responses. The anti-apoptotic function of NF-KB is also crucial to ontogenesis and chemo-and radio-resistance in cancer, as well as to several other pathological conditions.

There is strong evidence to suggest that JNK is involved in the apoptotic response to TRAIL. First, the apoptotic mechanisms triggered by TRAIL-Rs are similar to those activated by TNF-R1. Second, as with TNF-R1, ligand engagement of TRAIL-Rs leads to potent activation of both JNK and NF-KB. Thirdly, killing by TRAIL is blocked by this activation of NF-KB. Nevertheless, the role of JNK in apoptosis by TRAIL has not been yet formally demonstrated.

Of note, the triggering of TRAIL-Rs has recently received wide attention as a powerful new tool for the treatment of certain cancers, and clinical trials involving the administration of TRAIL are currently underway. This is largely because, unlike normal cells, tumor cells are highly susceptible to TRAIL-induced killing. The selectivity of the cytotoxic effects of TRAIL for tumor cells is due, at least in part, to the presence on normal cells of so-called"decoy receptors", inactive receptors that effectively associate with TRAIL, thereby preventing it from binding to the signal-transuding DRs, TRAIL-R1 and R2. Decoy receptors are instead expressed at low levels on most cancer cells. Moreover, unlike with FasL and TNFa, systemic administration of TRAIL induces only minor side effects, and overall, is well-tolerated by patients.

Cytoprotection by NF-KB involves activation of pro-survival genes. However, despite intense investigation, the bases for the NF-KB protective function during oncogenic transformation, cancer chemotherapy, and TNFa stimulation remain poorly understood. With regard to TNF-Rs, protection by NF-KB has been linked to the induction of Bcl-2 family members, Bol-l and Al/Bfl-l, XIAP, and the simultaneous upregulation of TRAF1/2 and c- IAP1/2. However, TRAF2, c-IAPl, BCl-XL, and XIAP are not significantly induced by TNFa in various cell types and are found at near-normal levels in several NF-xB deficient cells.

Moreover, Bcl-2 family members, XIAP, or the combination of TRAFs and c-IAPs can only partly inhibit PCD in NF-KB null cells. In addition, expression of TRAF1 and Al/Bfl-l is restricted to certain tissues, and many cell types express TRAF1 in the absence of TRAF2, a factor needed to recruit TRAF1 to TNF-R1. Other putative NF-KB targets, including A20 and IEX-1L, are unable to protect NF-KB deficient cells or were recently questioned to have anti- apoptotic activity. Hence, these genes cannot fully explain the protective activity of NF-KB.

3. NF-KB in oncogenesis and cancer therapy resistance NF-KB plays a pivotal role in oncogenesis. Genes encoding members of the NF-KB group, such as p52/plO0, Rel, and RelA and the IKB-like protein Bcl-3, are frequently rearranged or amplified in human lymphomas and leukemias. Inactivating mutations of IKBa are found in Hodgkin's lymphoma (HL). NF-KB is also linked to cancer independently of mutations or chromosomal translocation events. Indeed, NF-KB is activated by most viral and cellular oncogene products, including HTLV-I Tax, EBV EBNA2 and LMP-1, SV40 large-T, adenovirus ElA, Bcr-Abl, Her-2/Neu, and oncogenic variants of Ras. Although NF-KB participates in several aspects of oncogenesis, including cancer cell proliferation, the suppression of differentiation, and tumor invasiveness, direct evidence from both in vivo and in vitro models indicates that its control of apoptosis is crucial to cancer development. In the early stages of cancer, NF-KB is required to suppress apoptosis associated with transformation by oncogenes. For instance, upon expression of Bcr-Abl or oncogenic variants of Ras-one of the most frequently mutated oncogenes in human tumors-inhibition of NF- KB leads to an apoptotic response rather than to cellular transformation. Tumorigenesis driven by EBV is also inhibited by IKBaM-a super-active form of the NF-KB inhibitor, IKBa. In addition, NF-KB is essential for maintaining survival of a growing list of late stage tumors, including HL, diffuse large B cell lymphoma (DLBCL), multiple myeloma, and a highly invasive, estrogen receptor (ER) in breast cancer. Both primary tissues and cell line models of these malignancies exhibit constitutively high NF-KB activity. Inhibition of this aberrant activity by IKBaM or various other means induces death of these cancerous cells. In ER breast tumors, NF-KB activity is often sustained by PI-3K and Aktl kinases, activated by over-expression of Her-2/Neu receptors. Constitutive activation of this Her-2/Neu/PI- 3K/Aktl/NF-KB pathway has been associated with the hormone-independent growth and survival of these tumors, as well as with their well-known resistance to anti-cancer treatment and their poor prognosis. Due to activation of this pathway cancer cells also become resistant to TNF-R and Fas triggering, which helps them to evade immune surveillance.

Indeed, even in those cancers that do not contain constitutively active NF-KB, activation of the transcription factors by ionizing radiation or chemotherapeutic drugs (e. g. daunorubicin and etoposide) can blunt the ability of cancer therapy to kill tumor cells. In fact, certain tumors can be eliminated in mice with CPT-11 systemic treatment and adenoviral delivery of IKBaM.

B. JNK 1. Roles of JNK in apoptosis The c-Jun-N-terminal kinases (JNK1/2/3) are the downstream components of one of the three major groups of mitogen-activated protein kinase (MAPK) cascades found in mammalian cells, with the other two consisting of the extracellular signal-regulated kinases (ERK1/2) and the p38 protein kinases (p38a1 (3/y/8). Each group of kinases is part of a three- module cascade that include a MAPK (JNKs, ERKs, and p38s), which is activated by phosphorylation by a MAPK kinase (MAPKK), which in turn is activated by phosphorylation by a MAPKK kinase (MAPKKK). Whereas activation of ERK has been primarily associated with cell growth and survival, by and large, activation of JNK and p38 have been linked to the induction of apoptosis. Using many cell types, it was shown that persistent activation of JNK induces cell death, and that the blockade of JNK activation by dominant-negative (DN) inhibitors prevents killing by an array of apoptotic stimuli. The role of JNK in apoptosis is also documented by the analyses of mice with targeted disruptions of jnk genes. Mouse embryonic fibroblasts (MEFs) lacking both JNK1 and JNK2 are completely resistant to apoptosis by various stress stimuli, including genotoxic agents, UV radiation, and anisomycin, andjnk3-1-neurons exhibit a severe defect in the apoptotic response to excitotoxins. Moreover, JNK2 was shown to be required for anti-CD3-induced apoptosis in immature thymocytes.

However, while the role of JNK in stress-induced apoptosis is well established, its role in killing by DRs such as TNF-R1, Fas, and TRAIL-Rs has remained elusive. Some initial studies have suggested that JNK is not a critical mediator of DR-induced killing. This was largely based on the observation that, during challenge with TNFa, inhibition of JNK activation by DN mutants of MEKKl-an upstream activator of JNK had no effect on cell survival. In support of this view, it was also noted that despite their resistance to stress- induced apoptosis, JNK null fibroblasts remain sensitive to killing by Fas. In contrast, another early study using DN variants of the JNK kinase, MKK4/SEK1, had instead indicated an important role for JNK in pro-apoptotic signaling by TNF-R.

2. Roles of JNK in cancer JNK is potently activated by several chemotherapy drugs and oncogene products such as Bcr-Abl, Her-2/Neu, Src, and oncogenic Ras. Hence, cancer cells must adopt mechanisms to suppress JNK-mediated apoptosis induced by these agents. Indeed, non-redundant components of the JNK pathway (e. g. JNKK1/MKK4) have been identified as candidate tumor suppressors, and the well-characterized tumor suppressor BRCA1 is a potent activator of JNK and depends on JNK to induce death. Some of the biologic functions of JNK are mediated by phosphorylation of the c-Jun oncoprotein at S63 and S73, which stimulates c-Jun transcriptional activity. However, the effects of c-Jun on cellular transformation appear to be largely independent of its activation by JNK. Indeed, knock-in studies have shown that the JNK phospho-acceptor sites of c-Jun are dispensable for transformation by oncogenes, in vitro. Likewise, some of the activities of JNK in transformation and apoptosis, as well as in cell proliferation, are not mediated by c-Jun phosphorylation. For instance, while mutations of the JNK phosphorylation sites of c-Jun can recapitulate the effects of JNK3 ablation in neuronal apoptosis-which is dependent on transcriptional events-JNK-mediated apoptosis in MEFs does not require new gene induction by c-Jun. Moreover, JNK also activates JunB and JunD, which act as tumor suppressors, both in vitro and in vivo. Other studies have shown that inhibition of JNK in Ras-transformed cells has no effect on anchorage- independent growth or tissue invasiveness. Hence, JNK and c-Jun have independent functions in apoptosis and oncogenesis, and JNK is not required for transformation by oncogenes in some circumstances, but may instead contribute to suppress tumorigenesis.

Indeed, the inhibition of JNK might represent a mechanism by which NF-KB promotes oncogenesis and cancer chemoresistance.

C. Gadd45 1. Biologic fufzctions of (add45 proteihs gadd45/3 (also known as Mydl l 8) is one of three members of the gadd45 family of inducible genes, also including gadd45a (gadd45) and gadd45y (oig37leY6/grpl 7). Gadd45 proteins are regulated primarily at the transcriptional level and have been implicated in several biological functions, including G2/M cell cycle checkpoints and DNA repair. These functions were characterized with Gadd45a and were linked to the ability of this factor to bind to PCNA, core histones, Cdc2 kinase, and p21. Despite sequence similarity to Gadd45a, Gadd45p exhibits somewhat distinct biologic activities, as for instance, it does not appear to participate in negative growth control in most cells. Over-expression of Gadd45 proteins has also been linked to apoptosis in some systems. However, it is not clear that this is a physiologic activity, because in many other systems induction of endogenous Gadd45 proteins is associated with cytoprotection, and expression of exogenous polypeptides does not induce death. Finally, Gadd45 proteins have been shown to associate with MEKK4/MTK1 and have been proposed to be initiators of JNK and p38 signaling. Other reports have concluded that expression of these proteins does not induce JNK or p38 in various cell lines, and that the endogenous products make no contribution to the activation of these kinases by stress. The ability of Gadd45 proteins to bind to MEKK4 supports the existence of a link between these proteins and kinases in the MAPK pathways. Studies using T cell systems, have implicated Gadd45y in the activation of both JNK and p38, and Gadd45ß in the regulation of p38 during cytokines responses.

Although the prior studies have helped elucidate many important cellular processes, additional understanding remains needed, particularly with respect to the cellular pathways responsible for controlling apoptosis. For example, the manner in which NF-KB controls apoptosis has remained unclear. Elucidation of the critical pathways responsible for modulation of apoptosis is necessary in order to develop new therapeutics capable of treating a variety of diseases that are associated with aberrant levels of apoptosis.

Inhibitors of NF-KB are routinely used in combination with standard anti-cancer agents to treat cancer patients, such as patients with HL or multiple myeloma. Yet, therapeutic inhibitors (e. g. glucocorticoids) only achieve partial inhibition of NF-KB and exhibit considerable side effects, which limits their use in humans. A better therapeutic approach might be to employ agents that block, rather than NF-KB, its downstream anti- apoptotic effectors in cancer cells. However, despite intense investigation, these effectors remain unknown.

SUMMARY OF THE INVENTION The JNK pathway was found to be a focus for control of pathways leading to programmed cell death.

The present invention is based on the following: 1) in addition to playing a role in stress-induced apoptosis, JNK activation is necessary for efficient killing by TNF-R1, as well as by other DRs such as Fas and TRAIL-Rs ; 2) the inhibition of the JNK cascade represents a pivotal protective mechanism by NF-xB against TNFa-induced cytotoxicity; 3) suppression of JNK activation might represent a general protective mechanism by NF-KB and is likely to mediate the potent effects of NF-KB during oncogenesis and cancer chemoresistance; 4) inhibition of JNK activation and cytoprotection by NF-KB involve the transcriptional activation of gade45, ; 5) Gadd45 (3 protein blocks JNK signaling by binding to and inhibiting JNKK2/MKK7-a specific and non-redundant activator of JNK. With regard to this latter finding, the Gadd45 (3-interaction domains of JNKK2 and the JNKK2-binding surface of Gadd45 (3 were identified. This facilitates the isolation of cell-permeable peptides and small molecules that are able to interfere with the ability of Gadd45ß, and thereby of NF-KB, to block JNK activation and prevent apoptosis.

A method for modulating pathways leading to programmed cell death, includes the steps of : (a) selecting a target withing the JNK pathway; and (b) interferring with said target to either upregulate or downregulate the JNK pathway.

A way to interfere is: (a) obtaining an agent that is sufficient to block the suppression of JNK activation by Gadd45 proteins; and (b) contacting the cell with said agent to increase the percent of cells that undergo programmed cell death.

The agent may be an antisense molecule to a gadd45, gene sequence or fragments thereof, a small interfering RNA molecule (siRNA), a ribozyme molecule, a cell-permeable peptide fused to JNKK2 that effectively competes with the binding site of Gadd45ß, a small inorganic molecule or a peptide mimetic that mimics the functions of a Gadd45 protein.

Another way to interfere is: (a) obtaining a molecule that suppresses JNK signaling by interacting with a Gadd45-binding region on JNKK2 ; and (b) contacting a cell with the molecule to protect the cell from programmed cell death.

Using a cDNA to interfere includes: (a) obtaining a cDNA molecule that encodes a full length and portions of a Gadd45 protein; (b) transfecting the cell with the cDNA molecule; and (c) providing conditions for expression of the cDNA in the cell so that JNKK2 is bound and unavailable to activate the JNK pathway that induce programmed cell death.

The cDNA molecule may encode a fragment of Gadd45 protein that is sufficient to suppress JNK signaling, a peptide that corresponds to amino acids 69-113 of Gadd45 p.

The programmed cell death may be induced by TNFa, Fas, TRAIL or a genotoxic agent such as deunorubicin or cisplatinum.

A method to identify agents that modulate JNK signaling includes the steps of : (a) determining whether the agent binds to Gadd45p ; and (b) assaying for activity of the bound Gadd45p to determine the effect on JNK signalling.

A method for obtaining a mimetic that is sufficient to suppress JNK activation by interacting with JNKK2, includes the steps of : (a) designing the mimetic to mimic the function of Gadd45 protein; (b) contacting the mimetic to a system that comprises the JNK pathway; and (c) determining whether there is suppression of JNK signalling.

A method for screening and identifying an agent that modulates JNK pathway in vitro, includes the steps of : (a) obtaining a target component of the pathway; (b) exposing the cell to the agent; and (c) determining the ability of the agent to modulate JNK activity.

Suitable agents include peptides, peptide mimetics, peptide-like molecules, mutant proteins, cDNAs, antisense oligonucleotides or constructs, lipids, carbohydrates, and synthetic or natural chemical compounds.

A method for screening and identifying an agent that modulates JNK activity in vivo, includes the steps of : (a) obtaining a candidate agent; (b) administering the agent to a non-human animal; and (c) determining the level of JNK activity compared to JNK activity in animals not receiving the agent.

A method for identifying an agent that prevents Gadd45 (3 from blocking apoptosis, includes the steps of : (a) containing cells that express high levels of Gadd45 (3 which are protected against TNFa-induced apoptosis with the TNFa ; (b) comparing apoptosis in the cells in (a) with control cells exposed to the agent but not to TNFa ; and (c) inferring from differences in apoptosis in treated versus control cells, whether the agent prevents Gadd45 (3 from blocking apoptosis.

A method for screening for a modulator of the JNK pathway includes the steps of : (a) obtaining a candidate modulator of the JNK pathway, wherein the candidate is potentially any agent capable of modulating a component of the JNK pathway, including peptides, mutant proteins, cDNAs, anti-sense oligonucleotides or constructs, synthetic or natural chemical compounds; (b) administering the candidate agent to a cancer cell; (c) determining the ability of the candidate substance to modulate the JNK pathway, including either upregulation or downregulation of the JNK pathway and assaying the levels of up or down regulation.

A method of treating degenerative disorders and other conditions caused by effects of apoptosis in affected cells, includes the steps of : (a) obtaining a molecule that interferes with the activation of JNK pathways; and (b) contacting the affected cells with the molecule.

A method of aiding the immune system to kill cancer cells by augmenting JNK signaling, includes the steps of : (a) obtaining an inhibitor to block JNK signaling; and (b) contacting the cancer cells with the inhibitor.

The inhibitor may block activation of JNKK2 by Gadd45ß.

A method for transactivating a gadd45 3 promoter, includes the steps of : (a) binding NF-1cB complexes to promoter elements of gadd45ß ; and (b) assaying for gad45, gene expression.

A method for treating cancer, includes the steps of : (a) increasing JNK activity by inhibiting Gadd45p function ; and (b) administering inhibitors that interfere with Gadd45 (3 function.

Chemotherapeutic agents may also be used.

A method to determine agents that interfere with binding between Gadd45 protein and JNKK2, includes the steps of : (a) obtaining an agent that binds to Gadd45 protein; (b) contacting a cell with the agent under conditions that would induce transit JNK activation; and (c) comparing cells contacted with the agent to cells not contacted with the agent to determine if the JNK pathway is activated.

Compositions of this invention include: A nucleotide sequence having Gene Bank Acc. # AF441860 that functions as a gadd45ß promoter.

A nucleotide sequence that is an element of the promoter at amino acid positions selected from the group consisting of positions-447/-438 (-Kp-l),-426/-417 (-KP-2),-377/-368 (K, B-3) according to FIG. 8.

A molecule comprising a region of Gadd45p, characterized by the amino acid sequence from positions 60-114 of the full length of Gadd45ß protein.

A molecule comprising a binding region of JNKK2 characterized by the amino acid sequence from positions 132-156-231-244 of full length JNKK2 BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 shows Gadd45ß antagonizes TNFR-induced apoptosis in NF-#B null cells.

FIG 1A : Gadd45ß as well as Gadd45a and Gadd45y (left) rescue RelA-/-MEFs, TNFa- induced killing. Plasmids were used as indicated. Cells were treated with CHS (0.1 ug/ml or CHX plus TNFa (100 units/ml) and harvested at the indicated time points. Each column represents the percentage of GB+ live cells in TNFa treated cultures relative to the cultures treated with CHX alone. Values are the means of three independent experiments. The Figure indicates that Gadd45a, Gadd45p and Gadd45y have anti-apoptotic activity against TNFa.

FIG. 1B: NF-KB null 3DO cells are sensitive to TNFa. Cell lines harboring I#ßαM or neo plasmids were treated with TNFa (300 units/ml) and harvested at 14 hours. Columns depict percentages of live cells as determined by PI staining. Western blots show levels of IKßaM protein (bottom panels). FIG. 1C: 3DO IKßaM-Gadd45ß cells are protected from TNFa killing. Cells are indicated. Cells were treated with TNFa (25 units/ml) or left untreated and harvested at the indicated time points. Each value represents the mean of three independent experiments and expresses the percentages of live cells in treated cultures relatively to controls (left). PI staining profiles of representative clones after an 8-hour incubation with or without TNFa (right panel, TNFa and US. respectively). FIG. 1D : Protection correlates with levels of Gadd45B of the 8-hr. time point experiment shown in (C) with the addition of two MB-Gadd45B lines. Western blots are as indicated (lower panels), . FIG. 1E : Gadd45B functions downstream of NF-xB complexes. EMSA with extracts of untreated and TNFa- treated 3DO cells. Composition of the KB-binding complexes was assessed by using supershifting antibodies. FIG. IF shows Gadd45B is essential to antagonize TNFa-induced apoptosis. 3DO lines harboring anti-sense Gadd45B (AS-Gadd45B) or empty (Hygro) plasmids were treated with CHX (0. 1 ug/ml) plus or minus TNFa (1000 units/ml) and analyzed at 14 hours by nuclear PI staining. Low concentration of CHX was used to lower the threshold of apoptosis. Each column value represents the mean of three independent experiments and was calculated as described in FIG. 1C.

FIGS 2A-2D shows Gadd45B is a transcriptional target of NF-KB. FIG. 2A: Northern blots with RNA from untreated and TNFa (1000 u/ml) treated RelA-/-and +/+ MEF. Probes are as indicated. FIG. 2B-2D: 3 DO IKßaM cells and controls were treated with TNFa (1000 u/ml). PMA (50g/ml) plus ionomycin (luM) or daunorubicin (0.5 pM), respectively and analyzed as in FIG. 2A.

FIGS. 3A-3E shows Gadd45 (3 prevents caspase activation inNF-KB null cells. FIG 3A: Gadd45-dependent blockade of caspase activity. 3DO lines were treated with TNFa (50 units/ml) and harvested at the indicated time points for the measurement of caspase activity by in vitro fluorometric assay. Values express fluorescence units obtained after subtracting the background. FIG. 3B: Gadd45a inhibits TNFa-induced processing of Bid and pro- caspases. Cell were treated as described in FIG 2A. Closed and open arrowheads indicate unprocessed and processed proteins, respectively. FIG. 3C: Gadd450 completely abrogates TNFa-induced mitochondrial depolarization in NFKB-null cells. 3DO lines and the TNFa treatment were as described in FIG. 3A and B. Each value represents the mean of three independent experiments and expresses the percentage of JC-1+ cells in each culture. FIG.

3D-# : Gadd45 (3 inhibits cisplatinum-and daunorubicin-induced toxicity. Independently generated IKBaM-Gadd45ß and-Hygro clones were treated for 24 hr with (concentration) 0. 025, uM cisplatinum (FIG. 3D) or with 0.025 RM daunorubicin (FIG. 3E) as indicated.

Values represent percentages of live cells as assessed by nuclear PI staining and were calculated as described in FIG. 1C.

FIG. 4 shows Gadd450 is a physiologic inhibitor of JNK signaling. FIG. 4A: Western blots showing kinetics of JNK activation by TNFa (1000 U/ml) in IKBaM-Hygro and I#BαM-Gadd45ß 3DO clones. Similar results were obtained with four additional IKBaM--Gadd45ß and three IKBaM--Hygro clones. FIG. 4B: Western blots showing ERK, p38, and JNK phosphorylation in 3DO clones treated with TNFa for 5 minutes. FIG. 4D: Western blots (top and middle) and kinase assays (bottom) showing JNK activation in anti- sense-Gadd45 (3 and Hygro clones treated with TNFa as in (A). FIG. 4C: JNK activation by hydrogen peroxide (H202, 600, uM) and sorbitol (0.3M) in IKBaM-Hygro and IKBaM- Gadd45p clones. Treatments were for 30 minutes.

FIG. 5A-E shows the inhibition of JNK represents a protective mechanism by NF-xB.

FIG. 5A : Kinetics of JNK activation by TNKa (1000 U/ml) in 3DO-IKBaM and 3DO-Neo clones. Western blots with antibodies specific for phosphorylated (P) or total JNK (top and middle, respectively) and JNK kinase assays (bottom). Similar results were obtained with two additional IKBaM and five Neo clones. FIG. 5B : Western blots (top and middle) and kinase assays (bottom) showing JNK activation in RelA-/-and +/+ MEFs treated as in (A).

FIG. 5C : Western blots (top and middle) and kinase assays (bottom) showing JNK activation in parental 3DO cells treated with TNFa (1000 U/ml), TNFa plus CHX (10 llg/ml), or CHX alone. CHX treatments were carried out for 30 minutes in addition to the indicated time.

FIG. 5D: Survival of transfected RelA-/- MEFs following treatment with TNFα (1000 U/ml) plus CHX (0.1µg/ml) for 10 hours. Plasmids were transfected as indicated along with pEGFP (Clontech). FIG. 5E : Survival of 3D0-IoBaM cells pretreated with MAPK inhibitors for 30 minutes and then incubated with either TNFa (25 U/ml) or PBS for an additional 12 hours.

Inhibitors (Calbiochem) and concentrations are as indicated. In (D) and (E), values represent the mean of three independent experiments.

FIG. 6 shows gadd45ß expression is strongly induced by RelA, but not by Rel or p50.

Northern blots showing expression of gadd45ß transcripts in HtTA-1 cells and HtTA-p50, HtTA-p50, HtTA-RelA, and HtTA-CCR43 cell clones maintained in the presence (0 hours) or absence of tetracycline for the times shown. Cell lines, times after tetracycline withdrawal, and 32P-labeled probes specific to gadd45/3, ikba, relA, p50, rel, or control gapdh cDNAs, are as indicated. The tetracycline-inducible nf-kb transgenes are boxed. Transcripts from the endogenous pl O5 gene and p50 transgene are indicated.

FIG. 7 shows gadd45ß expression correlates with NF-KB activity in B cell lines.

Northern blots showing constitutive and inducible expression of gadd45ß in 70Z/3 pre-B cells and WEHI-231 B cells (lanes 1-5 and 5-5, respectively). Cells were either left untreated (lanes 1,6, and 11) or treated with LPS (40, ug/ml) or PMA (100 ng/ml) and harvested for RNA preparation at the indicated time points. Shown are two different exposures of blots hybridized with a 32P-labeled probe specific to the mouse gadd45p cDNA (top panel, short exposure; middle panel, long exposure). As a loading control, blots were re-probed with gapdh (bottom panel).

FIG. 8 shows the sequence of the proximal region of the murine gadd45ß promoter.

Strong matches for transcription factor binding sites are underlined and cognate DNA- binding factors are indicated. Positions where murine and human sequences are identical, within DNA stretches of high homology, are highlighted in gray. Within these stretches, gaps introduced for alignment are marked with dashes. KB binding sites that are conserved in the human promoter are boxed. A previously identified transcription start site is indicated by an asterisk and transcribed nucleotides are italicized. Numbers on the left indicate the base pair position relative to the transcription start site. It also shows the sequence of the proximal region of the murine gadd45 3 promoter. To understand the regulation of Gadd45 (3 by NF- KB, the murine gadd45, 8 promoter was cloned. A BAC library clone containing the gadd45ß gene was isolated, digested with XhoI, and subcloned into pBS. The 7384 b XhoI fragment containing gadd45, 8 was completely sequenced (accession number: AF441860), and portions were found to match sequences previously deposited in GeneBank (accession numbers: AC073816, AC073701, and AC091518). This fragment harbored the genomic DNA region spanning from-5. 4 kb upstream of a previously identified transcription start site to near the end of the fourth exon of gadd45ß. A TATA box was located at position-56 to-60 relative to the transcription start site. The gadd45, 8 promoter also exhibited several NF-KB-binding elements. Three strong KB sites were found in the proximal promoter region at positions- 377/-368,-426/-417, and-447/-438; whereas a weaker site was located at position-1159/- 1150 and four other matches mapped further upstream at positions-2751/-2742,-4525/-4516, -4890/-4881, and-5251/-5242 (gene bank accession number AF441860). Three KB consensus sites within the first exon of gadd45ß (+271+36, +71/+80, and +171/+180). The promoter also contained a Spl motif (-890/-881) and several putative binding sites for other transcription factors, including heat shock factor (HSF) 1 and 2, Ets, Stat, AP1, N-Myc, MyoD, CREB, and C/EBP.

To identify conserved regulatory elements, the 5.4 kb murine DNA sequence located immediately upstream of the gadd45 3 transcription start site was aligned with the corresponding human sequence, previously deposited by the Joint Genome Initiative (accession number: AC005624). The-1477/-1197 and-466/-300 regions of murine gadd45ß were highly similar to portions of the human promoter, suggesting that these regions contain important regulatory elements (highlighted in gray are identical nucleotides within regions of high homology). A less well-conserved region was identified downstream of position-183 to the beginning of the first intron. Additional shorter stretches of homology were also identified. No significant similarity was found upstream of position-2285. The homology region at-466/-300 contained three KB sites (referred to as KB-1, KB-2, and KB-3), which unlike the other KB sites present throughout the gadd45ß promoter, were conserved among the two species. These findings suggest that these KB sites may play an important role in the regulation of gadd45} S, perhaps accounting for the induction of gadd45ß by NF-#B.

FIG. 9 shows the murine gadd45ß promoter is strongly transactivated by RelA. (A) Schematic representation of CAT reporter gene constructs driven by various portions of the murine gade45, promoter. Numbers indicate the nucleotide position at the ends of the promoter fragment contained in each CAT construct. The conserved KB-1, KB-2, and KB-3 sites are shown as empty boxes, whereas the TATA box and the CAT coding sequence are depicted as filled and gray boxes, whereas the TATA box and the CAT coding sequence are depicted as filled and gray boxes, respectively. (B) Rel-A-dependent transactivation of the gade45, promoter. NTera-2 cells were cotransfected with individual gadd45ß-CAT reporter plasmids (6 u. g) alone or together with 0.3, 1, or 3 Rg of Pmt2t-RelA, as indicated. Shown in the absolute CAT activity detected in each cellular extract and expressed as counts per minute (c. p. m. ). Each column represents the mean of three independent experiments after normalization to the protein concentration of the cellular extracts. The total amount of transfected DNA was kept constant throughout by adding appropriate amounts of insert-less pMT2T. Each reporter construct transfected into Ntera-2 cells with comparable efficiency, as determined by the cotransfection of 1 llg of pEGFP (encoding green fluorescent protein; GFP; Contech), and flow cytometric analysis aimed to assess percentages of GFP cells and GFP expression levels (data not shown).

FIG. 10 shows the gadd45ß promoter contains three functional KB elements. (A) Schematic representation of wild-type and mutated-592/+23-gadd45ß-CAT reporter constructs. The KB-1, KB-2, and KB-3 binding sites, the TATA box, and the CAT gene are indicated as in Figure 9A. Mutated KB sites are crossed. (B) KB-1, KB-2, and KB-3 are each required for the efficient transactivation of the gadd45ß promoter by RelA. Ntera-2 cells were cotransfected with wild-type or mutated-592/+23-gadd45, 8-CAT reporter constructs alone or together with 0. 3, 1, or 3, ug pMT2T-RelA, as indicated. Shown is the relative CAT activity (fold induction) over the activity observed with transfection of the reporter plasmid alone. Each column represents the mean of three independent experiments after normalization to the protein concentration of the cellular extracts. Empty pMT2T vectors were used to keep the amount of transfected DNA constant throughout. pEGFP was used to control the transfection efficiencies of CAT plasmids, as described in Figure 9B.

FIG. 11 shows KB elements from the gazez promoter are sufficient for RelA- dependent transactivation. Ntera cells were cotransfected with A56-KB-1/2-CAT, A56-KB-3- CAT, or A56-KB-M-CAT reporter constructs alone or together with 0.3 or 1, ug of RelA expression plasmids, as indicated. As in Figure 1 OB, columns show the relative CAT activity (fold induction) observed after normalization to the protein concentration of the cellular extracts and represent the mean of three independent experiments. Insert-less pMT2T plasmids were used to adjust for total amount of transfected DNA.

FIG. 12 shows gadd45ß promoter #B sites bind to NF-KB complexes in vitro. (A) EMSA showing binding of p/50p5 and p50/RelA complexes to KB-1, KB-2, and KB-3 (lanes 9-12, 5-8, and 1-4, respectively). Whole cell extracts were prepared from NTera-2 cells transfected with pMT2T-p50 (9 u ; lanes 1-3,5-7, and 11-12) or pMT2T-p50 (3gag) plus pMT2T-RelA (6µg ; lanes 4,8, and 12). Various amounts of cell extracts (0. 1 p, l, lanes 3,7, and 11; 0. 3 jj, l, lanes 2,6, and 10; or 1, ul, lanes 1,4, 5,8, 9, and 12) were incubated in vitro with 32P-labeled KB-1, KB-2, or KB-3 probes, as indicated, and the protein-DNA complexes were separated by EMSA. NF-KB-DNA binding complexes are indicated. (B) Supershift analysis of DNA-binding NF-KB complexes. KB sites were incubated with 1 ul of the same extracts used in (A) or of extracts from NTera-2 cells transfected with insert-less pMT2T (lanes 1-3,10-12, and 19-21). Samples were loaded into gels either directly or after preincubation with antibodies directed against human p50 or RelA, as indicated. Transfected plasmids and antibodies were as shown. DNA-binding NF-KB complexes, supershifted complexes, and non-specific (n. s. ) bands are labeled. (C) shows gadd45, KB sites bind to endogenous NF-KB complexes in vitro. To determine whether gadd45ß-KB elements can bind to endogenous NF-KB complexes, whole cell extracts were obtained from untreated and lypopolysaccharide (LPS)-treated WEHI-231 cells. Cells were treated with 40, ug/ml LPS (Escherichia coli serotype 0111 : B4) for 2 hours, and 2 gel of whole cell extracts were incubated, in vitro, with 32P-labeled gadd45ß-KB probes. Probes, antibodies against individual NF-KB subunits, predominant DNA-binding complexes, supershifted complexes, and non-specific (n. s. ) bands are as labeled. All three gadd45/j-KB sites bound to both constitutively active and LPS-induced NF-KB complexes (lanes 1-3,9-11, and 17-19). KB-3 bound avidly to a slowly-migrating NF-KB complex, which was supershifted only by the anti- Rel antibody (lanes 4-8). This antibody also retarded the migration of the slower dimers binding to KB-2 and, much more loosely, to KB-1, but had no effect on the faster-migrating complex detected with these probes (lanes 15 and 23, respectively). The slower complex interacting with KB-1 and KB-2 also contained large amounts of p50 and smaller quantities of p52 and RelA (lanes 12-14 and 20-22, RelA was barely detectable with KB-1). The faster complex was instead almost completely supershifted by the anti-p50 antibody (lanes 12 and 20), and the residual DNA-binding activity reacted with the anti-p52 antibody (lanes 13 and 21; bottom band). With each probe, RelB dimers contributed to the KB-binding activity only marginally. Specificity of the DNA-binding complexes was confirmed by competitive binding reactions using unlabeled competitor oligonucleotides. Thus, the faster complex binding to KB-1 and KB-2 was predominantly composed of p50 homodimers and contained significant amounts of p52/p52 dimers, whereas the slower one was made up of p50/Rel heterodimers and, to a lesser extent, p52/Rel, Rel/Rel, and RelA-containing dimers.

Conversely, KB-3 only bound to Rel homodimers. Consistent with observations made with transfected NTera-2 cells, KB-1 exhibited a clear preference for p50 and p52 homodimers, while KB-2 preferentially bound to Rel-and RelA-containing complexes. Overall, KB-3 yielded the strongest NF-KB-specific signal, whereas KB-1 yielded the weakest one.

Interestingly, the in vitro binding properties of the DNA probes did not seem to reflect the relative importance of individual KB sites to promoter transactivation in vivo. Nevertheless, the findings do demonstrate that each of the functionally relevant aB elements of the gadd45, promoter can bind to NF-KB complexes, thereby providing the basis for the dependence of gadd45ß expression on NF-KB.

FIG. 13 shows Gadd45p expression protects BJAB cells against Fas-and TRAIL-R- induced apoptosis. To determine whether Gadd45 (3 activity extended to DRs other than TNF- Rs, stable HA-Gadd45p and Neo control clones were generated in BJAB B cell lymphomas, which are highly sensitive to killing by both Fas and TRAIL-Rs. As shown by propidium iodide (PI) staining assays, unlike Neo clones, BJAB clones expressing Gadd45 (3 were dramatically protected against apoptosis induced either (B) by agonistic anti-Fas antibodies (APO-1, 1, ug/ml, 16 hours) or (A) by recombinant (r) TRAIL (100 ng/ml, 16 hours). In each case, cell survival correlated with high levels of HA-Gadd45p proteins, as shown by Western blots with anti-HA antibodies (bottom panels). Interestingly, with Fas, protection by Gadd45p was nearly complete, even at 24 hours.

FIG. 14 shows the inhibition of JNK activation protects BJAB cells from Fas induced apoptosis. Parental BJAB cells were treated for 16 hours with anti-APOl antibodies (1 ug/ml), in the presence or absence of increasing concentrations of the specific JNK blocker SP600125 (Calbiochem), and apoptosis was monitored by PI staining assays. While BJAB cells were highly sensitive to apoptosis induced by Fas triggering, the suppression of JNK activation dramatically rescued these cells from death, and the extent of cytoprotection correlated with the concentration of SP600125. The data indicate that, unlike what was previously reported with MEFs (i. e. with ASK1-and JNK-deficient MEFs), in B cell lymphomas, and perhaps in other cells, JNK signaling plays a pivotal role in the apoptotic response to Fas ligation. This is consistent with findings that, in these cells, killing by Fas is also blocked by expression of Gadd45ß (FIG. 13B). Thus, JNK might be required for Fas- induced apoptosis in type 2 cells (such as BJAB cells), which unlike type 1 cells (e. g. MEFs), require mitochondrial amplification of the apoptotic signal to activate caspases.

FIG. 15 shows JNK is required for efficient killing by TNFa. In FIG. 5D and 5E, we have shown that the inhibition of JNK by either expression of DN-MKK7 or high doses of the pharmacological blocker SB202190 rescues NF-KB null cells from TNFa-induced killing.

Together with the data shown in FIG. 5A-C, these findings indicate that the inhibition of the JNK cascade represents a pivotal protective mechanism by NF-KB. They also suggest that the JNK cascade plays an important role in the apoptotic response to the cytokine. Thus, to directly link JNK activation to killing by TNF-R1, the sensitivity of JNK1 and JNK2 was tested in double knockout fibroblasts (provided by Dr. Roger Davis) to apoptosis by TNFa.

Indeed, as shown in FIG. 15A, mutant cells were dramatically protected against combined cytotoxic treatment with TNFa (1,000 U/ml) and CHX (filled columns) for 18 hours, whereas wild-type fibroblasts remained susceptible to this treatment (empty columns). JNK kinase assays confirmed the inability of knockout cells to activate JNK following TNFa stimulation (left panels). The defect in the apoptotic response of JNK null cells to TNFa plus CHX was not a developmental defect, because cytokine sensitivity was promptly restored by viral transduction of MIGRl-JNKK2-JNKl, expressing constitutively active JNK1 (FIG.

15B; see also left panel, JNK kinase assays). Thus, together with the data shown in FIG. 5A- E, these latter findings with JNK null cells indicate that JNK (but not p38 or ERK) is essential for PCD by TNF-R, and confirm that a mechanism by which NF-KB protects cells is the down-regulation of the JNK cascade by means of Gadd45 FIG. 16 shows Gadd45 (3 is a potential effector of NF-KB functions in oncogenesis.

Constitutive NF-KB activation is crucial to maintain viability of certain late stage tumors such as ER breast tumors. Remarkably, as shown by Northern blots, gadd45, was expressed at constitutively high levels in ER breast cancer cell lines-which depend on NF-KB for their survival-but not in control lines or in less invasive, ER breast cancer cells. Of interest, in these cells, gadd45 3 expression correlated with NF-KB activity. Hence, as with the control of TNFa-induced apoptosis, the induction of gade45/3 might represent a mechanism by which NF-KB promotes cancer cell survival, and thereby oncogenesis. Thus, Gadd45p might be a novel target for anti-cancer therapy.

FIG. 17 shows the suppression of JNK represents a mechanism by which NF-lcB promotes oncogenesis. The ER-breast cancer cell lines, BT-20 and MDA-MD-231, are well- characterized model systems of NF-KB-dependent tumorigenesis, as these lines contain constitutively nuclear NF-#B activity and depend on this activity for their survival. In these cells the inhibition of NF-KB activity by well-characterized pharmacological blockers such as prostaglandin Al (PGA1, 100 pM), CAPE (50 llg/ml), or parthenolide (2. 5, ug/ml) induced apoptosis rapidly, as judged by light microscopy. All NF-aB blockers were purchased from Biomol and concentrations were as indicated. Treatments were carried out for 20 (PGA1), 4 (parthenolide), or 17 hours (CAPE). Apoptosis was scored morphologically and is graphically represented as follows: ++++, 76-100% live cells; +++, 51-75% live cells; ++, 26-50% live cells; +, 1-25% live cells;-, 0% live cells. Remarkably, concomitant treatment with the JNK inhibitor SP600125 dramatically rescued breast tumor cells from the cytotoxicity induced by the inhibition of NF-KB, indicating that the suppression of JNK by NF-xB plays an important role in oncogenesis.

FIG. 18 is a schematic representation of TNF-Rl-induced pathways modulating apoptosis. The blocking of the NF-#B-dependent pathway by either a RelA knockout mutation, expression of IKBaM proteins or anti-sense gadd45ß plasmids, or treatment with CHX leads to sustained JNK activation and apoptosis. Conversely, the blocking of TNFa- induced JNK activation by either JNK or ASK1 null mutations, expression of DN-MKK7 proteins, or treatment with well characterized pharmacological blockers promotes cell survival, even in the absence of NF-KB. The blocking of the JNK cascade by NF-KB involves the transcriptional activation of gadd45ß. Gadd45ß blocks this cascade by direct binding to and inhibition of MKK7/JNKK2, a specific and non-redundant activator of JNK. Thus, MKK7 and its physiologic inhibitor Gadd45ß, are crucial molecular targets for modulating JNK activation, and consequently apoptosis.

FIG. 19 shows physical interaction between Gadd45) 3 and kinases in the JNK pathway, in vivo. Gadd associates with MEKK4. However, because this MAPKKK is not activated by DRs, no further examination was made of the functional consequences of this interaction. Thus, to begin to investigate the mechanisms by which Gadd45 (3 blunts JNK activation by TNF-R, the ability ofGadd45 (3 to physically interact with additional kinases in the JNK pathway was examined, focusing on those MAPKKKs, MAPKKs, and MAPKs that had been previously reported to be induced by TNF-Rs. HA-tagged kinases were transiently expressed in 293 cells, in the presence or absence of FLAG-Gadd45ß, and cell lysates were analyzed by co-immunoprecipitation (IP) with anti-FLAG antibody-coated beads followed by Western blot with anti-HA antibodies. These assays confirmed the ability of Gadd45ß to bind to MEKK4. These co-IP assays demonstrated that Gadd45 (3 can also associate with ASK1, but not with other TRAF2-interacting MAPKKKs such as MEKK1, GCK, and GCKR, or additional MAPKKKs that were tested (e. g. MEKK3). Notably, Gadd45 (3 also interacted with JNKK2/MKK7, but not with the other JNK kinase, JNKK1/MKK4, or with any of the other MAPKKs and MAPKs under examination, including the two p38-specific activators MKK3b and MKK6, and the ERK kinase MEK1. Similar findings were obtained using anti-HA antibodies for IPs and anti-FLAG antibodies for Western blots. Indeed, the ability to bind to JNKK2, the dominant JNK kinase induced by TNF-R, as well as to ASK1, a kinase required for sustained JNK activation and apoptosis by TNFa, may represent the basis for the control of JNK signaling by Gadd45ß. The interaction with JNKK2 might also explain the specificity of the inhibitory effects of Gadd45ß on the JNK pathway.

FIG. 20 shows physical interaction between Gadd45 (3 and kinases in the JNK pathway, in vitro. To confirm the above interactions, in vitro, GST pull-down experiments were performed. pBluescript (pBS) plasmids encoding full-length (FL) human ASK1, MEKK4, JNKK1, and JNKK2, or polypeptides derived from the amino-or carboxy-terminal portions of ASK1 (i. e. N-ASK1, spanning from amino acids 1 to 756, and C-ASK1, spanning from amino acids 648 to 1375) were transcribed and translated in vitro using the TNT coupled retyculocyte lysate system (Promega) in the presence of 35S-methionine. 5 p1 of each translation mix were incubated, in vitro, with sepharose-4B beads that had been coated with either purified glutathione-S-transferase (GST) polypeptides or GST-Gadd45 (3 proteins. The latter proteins contained FL murine Gadd45ß directly fused to GST. Binding assays were performed according to standard procedures, and 35S-labeled proteins that bound to beads, as well as 2 J. l of each in vitro translation mix (input), were then resolved by SDS polyacrylamide gel electrophoresis. Asterisks indicate the intact translated products. As shown in FIG. 20, FL-JNKK2 strongly associated with GST-Gadd45 p, but not with GST, indicating that JNKK2 and Gadd45ß also interacted in vitro, and that their interaction was specific. Additional experiments using recombinant JNKK2 and Gadd45ß have demonstrated that this interaction is mediated by direct protein-protein contact. Consistent with in vivo findings, GST-Gadd45p also associated with ASK1, N-ASK1, C-ASK1, and MEKK4-albeit less avidly than with JNKK2-and weakly with JNKK1. Thus, GST pull-down experiments confirmed the strong interaction between Gadd45ß and JNKK2 observed in vivo, as well as the weaker interactions of Gadd45 3 with other kinases in the JNK pathway. These assays also uncovered a weak association between Gadd45 P and JNKK1.

FIG. 21 shows Gadd45) 3 inhibits JNKK2 activity in vitro. Next, the functional consequences, in vitro, of the physical interactions of Gadd45ß with kinases in the JNK pathway was assessed. Murine and human, full-length Gadd45p proteins were purified from E. coli as GST-Gadd45p and Hiss-tagged Gadd45ß, respectively, according to standard procedures. Prior to employing these proteins in in vitro assays, purity of all recombinant polypeptides was assured by >98%, by performing Coomassie blue staining of SDS polyacrylamide gels. Then, the ability of these proteins, as well as of control GST and His6- EF3 proteins, to inhibit kinases in the JNK pathways was monitored in vitro. FLAG-tagged JNKK2, JNKK1, MKK3, and ASK1 were immunoprecipitated from transiently transfected 293 cells using anti-FLAG antibodies and pre-incubated for 10 minutes with increasing concentrations of recombinant proteins, prior to the addition of specific kinase substrates (i. e.

GST-JNK1 with JNKK1 and JNKK2 ; GST-p38y with MKK3 ; GST-JNNK1 or GST-JNKK2 with ASK1). Remarkably, both GST-Gadd45ß and His6-Gadd45p effectively suppressed JNKK2 activity, in vitro, even at the lowest concentrations that were tested, whereas control polypeptides had no effect on kinase activity (FIG. 21A). In the presence of the highest concentrations of Gadd45ß proteins, JNKK2 activity was virtually completely blocked. These findings indicate that, upon binding to Gadd45ß, JNKK2 is effectively inactivated.

Conversely, neither GST-Gadd45 (3 nor HiS6-Gadd45 p had significant effects on the ability of the other kinases (i. e. JNKK1, MKK3, and ASK1) to phosphorylate their physiologic substrates, in vitro, indicating that Gadd45p is a specific inhibitor of JNKK2. Of interest, Gadd45p also inhibited JNKK2 auto-phosphorylation.

FIG. 22A-B shows Gadd45 (3 inhibits JNKK2 activity in vivo. The ability ofGadd45 (3 to inhibit JNKK2 was confirmed in vivo, in 3DO cells. In these cells, over-expression of Gadd45(3 blocks JNK activation by various stimuli, and the blocking of this activation is specific, because Gadd45 (3 does not affect either the p38 or the ERK pathway. These findings suggest that Gadd45 (3 inhibits JNK signaling downstream of the MAPKKK module.

Kinase assays were performed according to procedures known to those of skill in the art using extracts from unstimulated and TNFa-stimulated 3DO cells, commercial antibodies that specifically recognize endogenous kinases, and GST-JNK1 (with JNKK2) or myelin basic protein (MBP; with ASK1) substrates (FIG. 22A). Activity ofJNKKl and MKK3/6 was instead assayed by using antibodies directed against phosphorylated (P) JNKK1 or MKK3/6 (FIG. 22B) -the active forms of these kinases. In agreement with the in vitro data, these assays demonstrated that, in 3DO cells, Gadd45 (3 expression is able to completely block JNKK2 activation by TNFa (FIG. 22A). This expression also partly suppressed JNKK1 activation, but did not have significant inhibitory effects on MKK3/6-the specific activators of p38-or ASK1 (FIG. 22A-B).

Hence, Gadd45 (3 is a potent blocker of JNKK2-a specific activator of JNK and an essential component of the TNF-R pathway of JNK activation. Of interest, this inhibition of JNKK2 is sufficient on its own to account for the effects ofGadd45 (3 on MAPK signaling, and explains the specificity of these effects for the JNK pathway. Together, the data indicate that Gadd45 [3 suppresses JNK activation, and thereby apoptosis, induced by TNFa and stress stimuli by direct targeting of JNKK2. Since Gadd45p is able to bind to and inhibit JNKK2 activity in vitro (FIGS. 20 and 21), Gadd45 (3 likely blocks this kinase directly, either by inducing conformational changes or steric hindrances that impede kinase activity. These findings identify JNKK2/MKK7 as an important molecular target of Gadd45 (3 in the JNK cascade. Under certain circumstances, Gadd45p may also inhibit JNKK1, albeit more weakly than JNKK2. Because ASK1 is essential for sustained activation of JNK and apoptosis by TNF-Rs, it is possible that the interaction between Gadd45ß and this MAPKKK is also relevant to JNK induction by these receptors.

FIG. 23A-B shows that two distinct polypeptide regions in the kinase domain of JNKK2 are essential for the interaction with Gadd45p. By performing GST pull-down assays with GST-and GST-Gadd45ß-coated beads, the regions of JNKK2 that are involved in the interaction with Gadd450 were determined. pBS plasmids encoding various amino-terminal truncations of JNKK2 were translated in vitro in the presence of 35S-metionine, and binding of these peptides to GST-Gadd45 (3 was assayed as described herein (FIG. 23A, Top), JNKK2 (1-401; FL), JNKK2 (63-401), JNKK2 (91-401), and JNKK2 (132-401) polypeptides strongly interacted with Gadd45p, in vitro, indicating that the amino acid region spanning between residue 1 and 131 is dispensable for the JNKK2 association with Gadd45ß.

However, shorter JNKK2 truncations-namely JNKK2 (157-401), JNKK2 (176-401), and JNKK2 (2-31-401)-interacted with Gadd45ß more weakly, indicating that the amino acid region between 133 and 156 is critical for strong binding to Gadd45p. Further deletions extending beyond residue 244 completely abrogated the ability of the kinase to associate with Gadd45 P, suggesting that the 231-244 region of JNKK2 also contributes to binding to Gadd45ß.

To confirm these findings, carboxy-terminal deletions of JNKK2 were generated, by programming retyculo-lysate reactions with pBS-JNKK2 templates that had been linearized with appropriate restriction enzymes (FIG. 23B, Bottom). Binding assays with these truncations were performed as described herein. Digestions of pBS-JNKK2 (FL) with SacII (FL), PpuMI, or NotI did not significantly affect the ability of JNKK2 to interact with Gadd45 ß, indicating that amino acids 266 to 401 are dispensable for binding to this factor.

Conversely, digestions with XcmI or BsgI, generating JNKK2 (1-197) and JNKK2 (1-186) polypeptides, respectively, partly inhibited binding to Gadd45p. Moreover, cleavage with BspEI, BspHI, or PflMI, generating shorter amino terminal polypeptides, completely abrogated this binding. Together these findings indicate that the polypeptide regions spanning from amino acids 139 to 186 and 198 to 265 and are both essential for strong association of JNKK2 with Gadd45ß. The interaction of JNKK2 with Gadd45ß was mapped primarily to two polypeptides spanning between JNKK2 residue 132 and 156 and between residue 231 and 244. JNKK2 might also contact Gadd45p through additional amino acid regions.

The finding that Gadd45ß directly contacts two distinct amino acid regions within the catalytic domain of JNKK2 provides important mechanistic insights into the basis for the inhibitory effects of Gadd45ß on JNKK2. These regions of JNKK2 shares no homology within MEKK4, suggesting that Gadd45ß contacts these kinases through distinct surfaces.

Since it is not known to have enzymatic activity (e. g. phosphatase or proteolytic activity), and its binding to JNKK2 is sufficient to inhibit kinase function, in vitro, Gadd45ß might block JNKK2 through direct interference with the catalytic domain, either by causing conformational changes or steric hindrances that inhibit kinase activity or access to substrates. With regard to this, the 133-156 peptide region includes amino acid K149-a critical residue for kinase activity-thereby providing a possible mechanism for the potent inhibition of JNKK2 by Gadd45ß.

FIG. 24A-B shows the Gadd45ß amino acid region spanning from residue 69 to 104 is essential for interaction with JNKK2. To identify the region of Gadd45ß that mediated the association with JNKK2, GST pull-down experiments were performed. Assays were performed using standard protocols and GST-JNKK2-or GST-coated beads. pBS plasmids encoding progressively shorter amino-terminal deletions of Gadd45ß were translated in vitro and labeled with 35S-metionine (FIG. 24A). Murine Gadd45p (l-160 ; FL), Gadd45p (41-160), Gadd45p (60-160), and Gadd45ß (69-160) polypeptides strongly interacted with JNKK2, whereas Gadd45ß (87-160) bound to the kinase only weakly. In contrast, Gadd45p (l 14-160) was unable to associate with JNKK2.

To confirm these findings, a series of carboxy-terminal Gadd45ß truncations were generated by programming in vitro transcription/translation reactions with appropriately linearized pBS-Gadd45 ? plasmids (FIG. 24B). Although digestion of pBS-Gadd45/3 with NgoMI did not affect Gadd45p binding to JNKK2, digestions with SphI and EcoRV, generating Gadd45p (l-95) and Gadd45p (l-68), respectively, progressively impaired Gadd45ß affinity for JNKK2. Indeed, the latter polypeptides were unable to associate with JNKK2. Together the data indicate that the Gadd45ß polypeptide spanning from residue 69 to 104 is required for the interaction with JNKK2. Interestingly, this polypeptide region is outlined in the Gadd45 decision test required for the protective activity of Gadd45ß against TNFa.

FIG. 25 show the amino acid region spanning between residue 69 and 113 is essential for the ability of Gadd45 ß to suppress TNFa-induced apoptosis. By performing mutational analyses, the domain of Gadd45 ß that is required for the blocking of TNFa-induced killing was mapped to the 69-113 amino acid region. Upon expression in RelA''cells, GFP- Gadd45 P (69-160) and GFP-Gadd45ß(1-113) exhibited anti-apoptotic activity against TNFa that was comparable to that of full-length GFP-Gadd45p. hi contrast, in these assays, GFP proteins fused to Gadd45p (87-160) or Gadd45p (l-86) had only modest protective effects.

Shorter truncations had virtually no effect on cell survival, indicating that the Gadd45ß region spanning between amino acids 69 and 113 is essential for cytoprotection, and that the adjacent 60-68 region contributes only modestly to this activity.

This amino acid region contains the domain of Gadd that is also essential for the interaction with JNKK2. This is consistent with the notion that the protective activity of Gadd45 P is linked to its ability to bind to JNKK2 and suppress JNK activation.

DETAILED DESCRIPTION OF THE INVENTION The JNK pathway was found to be a focus for control of pathways leading to programmed cell death.

The present invention facilitates development of new methods and compositions for ameliorating of diseases. Indeed, the observation that the suppression of JNK represents a central protective mechanism by NF-KB suggests that apoptosis of unwanted self-reactive lymphocytes and other pro-inflammatory cells (e. g. macrophages) at the site of inflammation - where there are high levels of TNFa-may be augmented by interfering with the ability of NF-KB to shut down JNK activation. Potential means for achieving this interference include, for instance, using blockers of Gadd45p.

Like Fas, TNF-R1 is also involved in host immune surveillance mechanisms. Thus, in another aspect of the invention, the agents might provide a powerful new adjuvant in cancer therapy.

Conversely, an enhancement of cell survival by the down-modulation of JNK will have beneficial effects in degenerative disorders and immunodeficiencies, conditions that are generally characterized by exaggerated cell death.

The invention allows design of agents to modulate the JNK pathway e. g. cell permeable, fusion peptides (such as TAT-fusion peptides) encompassing the amino acid regions of JNKK2 that come into direct contact with Gadd45ß. It is expected that these peptides will effectively compete with endogenous Gadd45p proteins for binding to JNKK2.

In addition, these findings allow design of biochemical assays for the screening of libraries of small molecules and the identification of compounds that are capable to interfere with the ability of Gadd45 P to associate with JNKK2. It is anticipated that both these peptides and these small molecules are able to prevent the ability of Gadd45(3, and thereby of NF-KB, to shut down JNK activation, and ultimately, to block apoptosis. These compounds are useful in the treatment of human diseases, including chronic inflammatory and autoimmune conditions and certain types of cancer.

The new molecular targets for modulating the anti-apoptotic activity of NF-KB, are useful in the treatment of certain human diseases. The application of these findings appears to pertain to the treatment of two broadly-defined classes of human pathologies: a) immunological disorders such as autoimmune and chronic inflammatory conditions, as well as immunodeficiencies; b) certain malignancies, in particular those that depend on NF-KB for their survival-such as breast cancer, HL, multiple myeloma, and DLBCL.

A question was whether JNK played a role in TNF-R-induced apoptosis. Confirming findings in NF-icB-deficient cells, evidence presented herein now conclusively demonstrated that JNK activation is obligatory not only for stress-induced apoptosis, but also for efficient killing by TNFa. It was shown that fibroblasts lacking ASK1-an essential component of the TNF-R pathway signaling to JNK (and p38)-are resistant to killing by TNFa. Foremost, JNK1 and JNK2 double knockout MEFs exhibit a profound-albeit not absolute-defect in the apoptotic response to combined cytotoxic treatment with TNFa and cycloheximide.

Moreover, it was shown that the TNFa homolog of Drosophila, Eiger, completely depends on JNK to induce death, whereas it does not require the caspase-8 homolog, DREDD. Thus, the connection to JNK appears to be a vestigial remnant of a primordial apoptotic mechanism engaged by TNFa, which only later in evolution begun to exploit the FADD-dependent pathway to activate caspases.

How can then the early observations with DN-MEKK1 be reconciled with these more recent findings? Most likely, the key lies in the kinetics of JNK induction by TNF-Rs. Indeed, apoptosis has been associated with persistent, but not transient JNK activity. This view is supported by the recent discovery that JNK activation is apoptogenic on its own-elegantly demonstrated by the use of MKK7-JNK fusion proteins, which result in constitutively active JNK in the absence of extrinsic cell stimulation. Unlike UV and other forms of stress, TNFa causes only transient induction of JNK, and in fact, this induction normally occurs without significant cell death, which explains why JNK inhibition by DN-MEKK1 mutants has no effect on cell survival. JNK pro-apoptotic activity is instead unmasked when the kinase is allowed to signal chronically, for instance by the inhibition of NF-KB.

The exact mechanism by which JNK promotes apoptosis is not known. While in some circumstances JNK-mediated killing involves modulation of gene expression, during challenge with stress or TNFa, the targets of JNK pro-apoptotic signaling appear to be already present in the cell. Killing by MKK7-JNK proteins was shown to require Bax-like factors of the Bcl-2 group ; however, it is not clear that these factors are direct targets of JNK, or that they mediate JNK cytotoxicity during TNF-R signaling.

I. Activation of the JNK cascade is required for efficient killing by DRs (TNF-R1, Fas, and TRAIL-Rs), and the suppression of this cascade is crucial to the protective activity of NF-KB.

A. TNF-Rs-induced apoptosis.

The JNK and NF-KB pathways-almost invariably co-activated by cytokines and stress-are intimately linked. The blocking of NF-KB activation by either the ablation of the NF-KB subunit RelA or expression of the IKBaM super-inhibitor hampers the normal shut down of JNK induction by TNF-R (FIGS. 5A and 5B). Indeed, the down-regulation of the JNK cascade by NF-KB is needed for suppression of TNFa-induced apoptosis, as shown by the finding that inhibition of JNK signaling by various means rescues NF-KB-deficient cells from TNFa-induced apoptosis (FIGS. 5D and 5E). In cells lacking NF-KB, JNK activation remains sustained even after protective treatment with caspase inhibitors, indicating that the effects of NF-KB on the JNK pathway are not a secondary consequence of caspase inhibition.

Thus, NF-KB complexes are true blockers of JNK activation. These findings define a novel protective mechanism by NF-KB and establish a critical role for JNK (and not for p38 or ERK) in the apoptotic response to TNFa (see FIG. 18).

B. Fas-induced apoptosis.

Although ASKI-/-and JNK null fibroblasts are protected against the cytotoxic effects of TNFa, these cells retain normal sensitivity to Fas-induced apoptosis, which highlights a fundamental difference between the apoptotic mechanisms triggered by Fas and TNF-R.

Nevertheless, in certain cells (e. g. B cell lymphomas), JNK is also involved in the apoptotic response to Fas triggering. Indeed, the suppression of JNK by various means, including the specific pharmacological blocker SP600125, rescues BJAB cells from Fas-induced cytotoxicity (FIG. 14). Consistent with this observation, in these cells, killing by Fas is also almost completely blocked by over-expression of Gadd450 (FIG. 13B). Together, these findings indicate that JNK is required for Fas-induced apoptosis in some circumstance, for instance in type 2 cells (e. g. BJAB cells), which require mitochondrial amplification of the apoptotic signal to activate caspases and undergo death.

Like TNF-Rs, Fas plays an important role in the host immune surveillance against cancerous cells. Of interest, due to the presence of constitutively high NF-KB activity, certain tumor cells are able to evade these immune surveillance mechanisms. Thus, an augmentation of JNK signaling-achieved by blocking the JNK inhibitory activity of Gadd45 P, or more broadly of NF-KB-aids the immune system to dispose of tumor cells efficiently.

Fas is also critical for lymphocyte homeostasis. Indeed, mutations in this receptor or its ligand, FasL, prevent elimination of self-reactive lymphocytes, leading to the onset of autoimmune disease. Thus, for the treatment of certain autoimmune disorders, the inhibitory activity of Gadd45 (3 on JNK may serve as a suitable target.

C. TRAIL-R-induced apoptosis.

Gadd45) 3 also blocks TRAIL-R-involved in apoptosis (FIG. 1A), suggesting that JNK plays an important role in the apoptotic response to the triggering of this DR. The finding that JNK is required for apoptosis by DRs may be exploited for cancer therapy. For example, the sensitivity of cancer cells to TRAIL-induced killing by adjuvant treatment is enhanced with agents that up-regulate JNK activation. This can be achieved by interfering with the ability of Gadd45 (3 or NF-icB to block TRAIL-induced JNK activation. This finding may also provide a mechanism for the synergistic effects of combined anti-cancer treatment because JNK activation by genotoxic chemotherapeutic drugs may lower the threshold for DR- induced killing.

II. The suppression of JNK represents a mechanism by which NF-KB promotes oncogenesis and cancer chemoresistance.

In addition to antagonizing DR-induced killing, the protective activity of NF-oB is crucial to oncogenesis and chemo-and radio-resistance in cancer. However, the bases for this protective activity is poorly understood. It is possible that the targeting of the JNK cascade represents a general anti-apoptotic mechanism by NF-xB, and indeed, there is evidence that the relevance of this targeting by NF-xB extends to both tumorigenesis and resistance of tumor cells to anti-cancer therapy. During malignant transformation, cancer cells must adopt mechanisms to suppress JNK-mediated apoptosis induced by oncogenes, and at least in some cases, this suppression of apoptotic JNK signaling might involve NF-KB.

Indeed, while NF-KB activation is required to block transformation-associated apoptosis, non-redundant components of the JNK cascade such as MKK4 and BRCA1 have been identified as tumor suppressors.

Well-characterized model systems of NF-KB-dependent tumorigenesis, including such as breast cancer cells provide insight into mechanism of action. Breast cancer cell lines such as MDA-MD-231 and BT-20, which are known to depend on NF-KB for their survival, can be rescued from apoptosis induced by NF-KB inhibition by protective treatment with the JNK blocker SP600125 (FIG. 17). Thus, in these tumor cells, the ablation of JNK can overcome the requirement for NF-KB, suggesting that cytotoxicity by NF-KB inactivation is associated with an hyper-activation of the JNK pathway, and indicates a role for this pathway in tumor suppression. Gadd45p mediates the protective effects of NF-KB during oncogenesis and cancer chemoresistance, and is a novel target for anti-cancer therapy.

With regard to chemoresistance in cancer, apoptosis by genotoxic stress-a desirable effect of certain anti-cancer drugs (e. g. daunorubicin, etopopside, and cisplatinum)-requires JNK activation, whereas it is antagonized by NF-KB. Thus, the inhibition of JNK is a mechanism by which NF-KB promotes tumor chemoresistance. Indeed, blockers of NF-KB are routinely used to treat cancer patients such as patients with HL and have been used successfully to treat otherwise recalcitrant malignancies such as multiple myeloma. However, these blockers (e. g. glucocorticoids and proteosome inhibitors) can only achieve a partial inhibition of NF-KB, and when used chronically, exhibit considerable side effects, including immune suppressive effects, which limit their use in humans. Hence, as discussed with DRs, in the treatment of certain malignancies, it is beneficial to employ, rather than NF-KB- targeting agents, therapeutic agents aimed at blocking the anti-apoptotic activity of NF-KB.

For instance, a highly effective approach in cancer therapy may be the use of pharmacological compounds that specifically interfere with the ability of NF-KB to suppress JNK activation. These compounds not only enhance JNK-mediated killing of tumor cells, but allow uncoupling of the anti-apoptotic and pro-inflammatory functions of the transcription factor. Thus, unlike global blockers of NF-KB, such compounds lack immunosuppressive effects, and thereby represent a promising new tool in cancer therapy. A suitable therapeutic target is Gadd45p itself, because this factor is capable of inhibiting apoptosis by chemotherapeutic drugs (FIGS. 3D and 3E), and its induction by these drugs depends on NF- KB (FIG. 2D). With regard to this, the identification of the precise mechanisms by which Gadd45(3 and NF-KB block the JNK cascade (i. e. the testing of JNKK2) opens up new avenues for therapeutic intervention in certain types of cancer, in particular in those that depend on NF-KB, including tumors driven by oncogenic Ras, Bcr-Abl, or EBV-encoded oncogenes, as well as late stage tumors such as HL, DLBCL, multiple myeloma, and breast cancers.

III. Gadd45ß mediates the inhibition of the JNK cascade by NF-KB.

A. Gadd45ß mediates the protective effects of NF-KB against DR- induced apoptosis.

Cytoprotection by NF-KB involves activation of a program of gene expression. Pro- survival genes that mediate this important function of NF-KB were isolated. In addition to gaining a better understanding of the molecular basis for cancer, the identification of these genes provides new targets for cancer therapy. Using a functional screen in NF-KB/RelA null cells, Gadd45 ß was identified as a pivotal mediator of the protective activity of NF-KB against TNFa-induced killing. gadd45ß is upregulated rapidly by the cytokines through a mechanism that requires NF-KB (FIGS. 2A and 2B), is essential to antagonize TNFa-induced killing (FIG. 1F), and blocks apoptosis in NF-#B null cells (FIGS. 1A, 1C, 1D, 3A and 3B).

Cytoprotection by Gadd45p involves the inhibition of the JNK pathway (FIGS. 4A, 4C and 4D), and this inhibition is central to the control of apoptosis by NF-KB (FIGS. 5A, 5B, 5D and 5E). Expression of Gadd45ß in cells lacking NF-KB completely abrogates the JNK activation response to TNFa, and inhibition of endogenous proteins by anti-sense gadd45 i hinders the termination of this response (FIG. 4D). Gadd45ß also suppresses the caspase- independent phase of JNK induction by TNFα, and hence, is a bonafide inhibitor of the JNK cascade (FIG. 4A and 4C). There may be additional NF-KB-inducible blockers of JNK signaling.

Activation of gadd45ß by NF-#B was shown to depend on three conserved KB elements located at positions-447/-438 (KB-1),-426/-417 (KB-2), and-377/-368 (KB-3) of the gadd45ß promoter (FIGS. 8, 9A, 9B, 10A, 10B, and 11). Each of these sites binds to NF- KB complexes in vitro and is required for optimal promoter transactivation (FIGS. 12A, 12B, and 12C). Together, the data establish that Gadd45ß is a novel anti-apoptotic factor, a physiologic inhibitor of JNK activation, and a direct transcriptional target of NF-KB. Hence, Gadd45ß mediates the targeting of the JNK cascade and cytoprotection by NF-KB.

The protective activity of Gadd45ß extends to DRs other than TNF-Rs, including Fas and TRAIL-Rs. Expression of Gadd45 ß dramatically protected BJAB cells from apoptosis induced by the triggering of either one of these DRs, whereas death was effectively induced in control cells (FIGS. 13B and 13A, respectively). Remarkably, in the case of Fas, protection by Gadd45p was nearly complete. Similar to TNF-R1, the protective activity of Gadd45p against killing by Fas, and perhaps by TRAIL-Rs, appears to involve the inhibition of the JNK cascade (FIGS. 13A, 13B and 14). Thus, Gadd45p is a new target for modulating DR- induced apoptosis in various human disorders.

B. Gadd45ß is a potential effector of the protective activity of NF-#B during oncogenesis and cancer chemoresistance.

The protective genes that are activated by NF-#B during oncogenesis and cancer chemoresistance are not known. Because it mediates JNK inhibition and cytoprotection by NF-KB, Gadd45ß is a candidate. Indeed, as with the control of DR-induced apoptosis, the induction of gadd45ß represents a means by which NF-#B promotes cancer cell survival. In 3DO tumor cells, Gadd45p expression antagonized killing by cisplatinum and daunorubicin (FIG. 3D and 3E) -two genotoxic drugs that are widely-used in anti-cancer therapy. Thus, Gadd45p blocks both the DR and intrinsic pathways of caspase activation found in mammalian cells. Since apoptosis by genotoxic agents requires JNK, this latter protective activity of Gadd45ß might also be explained by the inhibition of the JNK cascade. In 3DO cells, gadd45@ expression was strongly induced by treatment with either daunorubicin or cisplatinum, and this induction was almost completely abolished by the IKBaM super- repressor (FIG. 2D), indicating that gadd45@ activation by these drugs depends on NF-idB.

Hence, Gadd45ß may block the efficacy of anti-tumor therapy, suggesting that it contributes to NF-KB-dependent chemoresistance in cancer patients, and that it represents a new therapeutic target.

Given the role of JNK in tumor suppression and the ability of Gadd45ß to block JNK activation, Gadd45ß also is a candidate to mediate NF-xB functions in tumorigenesis.

Indeed, expression patterns suggest that Gadd45p may contribute to NF-1cB-dependent survival in certain late stage tumors, including ER breast cancer and HL cells. In cancer cells, but not in control cells such as less invasive, ER+ breast cancers, gadd45, 8 is expressed at constitutively high levels (FIG. 16), and these levels correlate with NF-KB activity.

C. Identification of the mechanisms by which Gadd45ß blocks JNK activation: the targeting of JNKK2/MKK7 Neither Gadd45ß nor NF-KB affect the ERK or p38 cascades (FIG. 4C), suggesting that these factors block JNK signaling downstream of the MAPKKK module. Consistent with this notion, the MAPKK, JNKK2/MKK7-a specific activator of JNK and an essential component of the TNF-R pathway of JNK activation were identified as the molecular target of Gadd45 P in the JNK cascade.

Gadd45p was previously shown to associate with MEKK4. However, since this MAPKKK is not activated by DRs, the functional consequences of this interaction were not further examined. Thus, to begin to investigate the mechanisms by which Gadd45 [3 controls JNK induction by TNF-R, Gadd45ß was examined for the ability to physically interact with additional kinases, focusing on those MAPKKKs, MAPKKs, and MAPKs that have been reported to be induced by TNF-Rs. Co-immunoprecipitation assays confirmed the ability of Gadd45 P to bind to MEKK4 (FIG. 19). These assays also showed that Gadd45ß is able to associate with ASK1, but not with other TRAF2-interacting MAPKKKs such as MEKK1, GCK, and GCKR, or additional MAPKKK that were tested (e. g. MEKK3) (FIG. 19).

Notably, Gadd45ß also interacted with JNKK2/MKK7, but not with the other JNK kinase, JNKK1/MKK4, or with any of the other MAPKKs and MAPKs under examination, including the two p38-specific activators MKK3b and MKK6, and the ERK kinase MEK1 (FIG. 19).

In vitro GST pull-down experiments have confirmed a strong and direct interaction between Gadd45p and JNKK2, as well as a much weaker interaction with ASK1 (FIG. 20). They also uncovered a very weak association between Gadd45ß and JNKK1 (FIG. 20).

Gadd45ß is a potent inhibitor of JNKK2 activity. This has been shown both in in vitro assays (FIG. 22A), using recombinant Gadd45ß proteins, and in in vivo assays, using lysates of 3DO clones (FIG. 22A). The effects of Gadd45ß on JNKK2 activity are specific, because even when used at high concentrations, this factor is unable to inhibit either JNKK1, MKK3b, or-despite its ability to bind to it-ASK1 (FIGS. 21B, 21C, 22A and 22B). This inhibition of JNKK2 is sufficient on its own to account for the effects of Gadd45 (3 on MAPK signaling, and explains the specificity of these effects for the JNK pathway. Together, the data indicate that Gadd45ß suppresses JNK activation, and thereby apoptosis, induced by TNFa and stress stimuli by directly targeting JNKK2 (FIGS. 21A and 22A). Consistent with the notion that it mediates the effects of NF-KB on the JNK cascade, Gadd45p and NF-xB have similar effects on MAPK activation by TNFa, in vivo (FIG. 4C). Because ASK1 is essential for sustained activation of JNK and apoptosis by TNF-Rs, it is possible that the interaction between Gadd45ß and this MAPKKK is also relevant to JNK induction by these receptors.

By performing GST pull-down experiments using either GST-Gadd45 (3 or GST- JNKK2 and several N-and C-terminal deletion mutants of JNKK2 and Gadd45ß, respectively, the kinase-binding surfaces (s) of Gadd45p (FIGS. 24A and 24B) and the Gadd45p-binding domains of JNKK2 (FIGS. 23A and 23B) were identified. Gadd45ß directly contacts two distinct amino acid regions within the catalytic domain of JNKK2 (FIGS. 23A and 23B), which provides important mechanistic insights into the basis for the inhibitory effects of Gadd45p on JNKK2. These regions of JNKK2 share no homology within MEKK4, suggesting that Gadd45p contacts these kinases through distinct surfaces.

Since it is not known to have enzymatic activity (e. g. phosphatase or proteolytic activity), and its binding to JNKK2 is sufficient to inhibit kinase function, in vitro (FIG. 21A), Gadd45ß might block JNKK2 through direct interference with the catalytic domain, either by causing conformational changes or steric hindrances that inhibit kinase activity or access to substrates.

By performing mutational analyses, the domain of Gadd45p that is crucial for the blocking of TNFα-induced killing was mapped (FIG. 25). Cytoprotection assays in RelA-/- cells have shown that GFP-Gadd45 (3 (69-160) and GFP-Gadd45ß(1-113) exhibit anti- apoptotic activity against TNFa that is comparable to that of full-length GFP-Gadd45ß while GFP proteins fused to Gadd45ß (87-160) or Gadd45R (l-86) have only modest protective effects. Shorter truncations have virtually no effect on cell survival (FIG. 25), indicating that the Gadd45ß region spanning between amino acids 69 and 113 is essential for cytoprotection, and that the adjacent 60-68 region contributes modestly to this activity.

This same amino acid region containing Gadd45ß domain (69-104) that is essential for the Gadd45ß interaction with JNKK2 (FIG. 24A and 24B). This is consistent with the notion that the protective activity of Gadd45ß is linked to its ability to bind to JNKK2 and suppress JNK activation. Of interest, these findings now allow the design of cell permeable, TAT-fusion peptides encompassing the amino acid regions of JNKK2 that come into direct contact with Gadd45ß. It is expected that these peptides can effectively compete with endogenous Gadd45 P proteins for binding to JNKK2. In addition, these findings allow to design biochemical assays for screening libraries of small molecules and identifying compounds that are capable of interfering with the ability ofGadd45p to associate with JNKK2. It is anticipated that both these peptides and these small molecules will be able to prevent the ability of Gadd45 (3, and thereby of NF-KB, to shut down JNK activation, and ultimately, to block apoptosis. As discussed throughout this summary, these compounds might find useful application in the treatment of human diseases, including chronic inflammatory and autoimmune conditions and certain types of cancer.

EXAMPLES The following examples are included to demonstrate embodiments of the invention.

It should be appreciated by those of skill in the art that techniques disclosed in the examples which follow represent techniques discovered by the inventor to function well in the practice of the invention. However, those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention.

Example 1: Identification of Gadd45p as novel antagonist of TNFR-induced apoptosis Functional complementation of RelA-/-fibroblasts which rapidly undergo apoptosis when treated with TNFa (Beg and Baltimore, 1996), was achieved by transfection of cDNA expression libraries derived from TNFa-activated, wild-type fibroblasts. A total of four consecutive cycles of library transfection, cytotoxic treatment with TNFa, and episomal DNA extraction were completed, starting from more than 4 x 106 independent plasmids.

After selection,-200 random clones were analyzed in transient transfection assays, with 71 (35%) found to significantly protect RelA-null cells from TNFa-induced death.

Among these were cDNAs encoding murine RelA, cFLIP, and dominant negative (DN) forms of FADD, which had been enriched during the selection process, with RelA representing 3.6% of the newly-isolated library. Thus, the library abounded in known regulators of TNFR-triggered apoptosis (Budihardjo et al., 1999).

One of the cDNAs that scored positive in cytoprotection assays encoded full-length Gadd45ß, a factor that had not been previously implicated in cellular responses to TNFa.

Gadd45ß inserts had been enriched 82 folds after two cycles of selection, reaching an absolute frequency of 0.41%. The above experiment shows that Gadd45ß is a novel putative anti-apoptotic factor.

To confirm the above findings, pEGFP-Gadd45ß, pEGFP-RelA, or insert-less pEGFP constructs were tested in transient transfection assays in RelA-/-fibroblasts. Whereas cells expressing control GFP proteins were, as expected, highly susceptible to TNFa-induced death, whereas in contrast, cells that had received pEGFP-Gadd45p were dramatically protected form apoptosis-exhibiting a survival rate of almost 60% after an 8-hour treatment versus 13% in control cultures (FIG. 1A). As shown previously, with pEGFP-RelA the cell rescue was virtually complete (Beg and Baltimore, 1996).

To determine whether the activity of Gadd450 was cell type-specific an additional cellular model of NK-KB deficiency was generated, where 3DO T cell hybridomas were forced to stably express IKBaM, a variant of the IxBa inhibitor that effectively blocks the nuclear translocation of NF-xB (Van Antwerp et al., 1996).

In the presence of the repressor, 3DO cells became highly sensitive to TNFa-induced killing, as shown by nuclear propidium iodide (PI) staining, with the degree of the toxicity correlating with IKBaM protein levels (FIG. 1B, lower panels). Neo control cells retained instead, full resistance to the cytokine. Next, constructs expressing full-length Gadd450, or empty control vectors (Hygro) were stably introduced into the 3DO-IKBaM-25 line, which exhibited the highest levels of IKBaM (FIG. 1B). Although each of 11 IKBaM-Hygro clones tested remained highly susceptible to TNFa, clones expressing Gadd45p became resistant to apoptosis, with the rates of survival of 31 independent IKBaM-Gadd45p clones correlating with Gadd45 (3 protein levels (FIGS. 1C and 1D, representative lines expressing high and low levels of Gadd45ß and IKBaM-Hygro controls). The protective effects of Gadd45ß were most dramatic at early time points, when viability of some M3aM-Gadd45p lines was comparable to that of Neo clones (FIGS. 1C and 1D, 8 hours). In the IlcBaM-Gadd45ß-33 line, expressing high amounts of Gadd45ß, the frequency of cell death was only ~15% higher than in Neo controls even at 24 hours (FIG. 1C). Thus, Gadd45ß is sufficient to temporarily compensate for the lack of NF-xB.

Further, IKBaM-Gadd45ß cells retained protein levels of IKBaM that were similar or higher than those detected in sensitive IlcBaM clones (FIG. 1D, lower panels) and that were sufficient to completely block NF-#B activation by TNFa, as judged by electrophoretic mobility shift assays (EMSAs ; FIG. 1E). Hence, as also seen in RelA-/- cells, Gadd45ß blocks apoptotic pathways by acting downstream of NF-KB complexes.

Example 2: Gadd45 is a physiologic target of NF-#B Gadd can be induced by cytokines such as IL-6, IL-18, and TGFß, as well as by genotoxic stress (Zhang et al., 1999 ; Yang et al., 2001; Wang et al., 1999b). Because the NF-#B anti-apoptotic function involves gene activation, whether Gadd45ß was also modulated by TNFa was determined. As shown in FIG. 2A, cytokine treatment determined a strong and rapid upregulation of Gadd45 (3 transcripts in wild-type mouse embryo fibroblasts (MEF). In contrast, in cells lacking RelA, gene induction was severely impaired, particularly at early time points (FIG. 2A, compare +/+ and-/-lanes at 0.5 hours). In these cells, induction was also delayed and mirrored the pattern of expression ofIkpaM a known target of NH-KB (Ghosh et al., 1998), suggesting that the modest induction was likely due to NF-#B family members other than RelA (i. e. , Rel). Gadd45a was not activated by TNFa, while Gadd45y was modestly upregulated in both cell types.

Analogously, Gadd was induced by TNFa in parental and Neo 3DO cells, but not in the IKBaM lines (FIG. 2B), with modest activation seen only in IKBaM-6 cells, which expressed low levels of the repressor (see FIG. 1B). In Neo clones, Gadd45 P was also induced by daunorubicin or PMA plus ionomycin (P/I ; FIG. 2D and 2C, respectively), treatments that are known to activate NF-#B (Wang et al., 1996). Again, gene induction was virtually abrogated by IKBaM. Gadd45a was unaffected by TNFa treatment, but was upregulated by daunorubicin or P/I, albeit independently of NF-KB (FIG. 2B, C, D); whereas Gadd45y was marginally induced by the cytokine only in some lines (FIG. 2B). nficbl was used as a positive control of NF-#B-dependent gene expression (Ghosh et al., 1998).

The results establish that gad45, is a novel TNFa-inducible gene and a physiologic target of NF-KB. The inspection of the gadd45ß promoter revealed the presence of 3 KB binding sites. EMSAs and mutational analyses confirmed that each of these sites was required for optimal transcriptional activation indicating that gade45, is also a direct target of NF-#B. These finding are consistent with a role of gadd45ß as a physiologic modulator of the cellular response to TNFα.

Example 3: Endogenous Gadd45p is required for survival of TNFa Gadd45ß is a downstream target of NK-#B and exogenous Gadd45ß can partially substitute for the transcription factor during the response to TNFa. However, it could be argued that since experiments were carried out in overexpression, cytoprotection might not represent a physiologic function of Gadd45ß. To address this critical issue, 3DO clones stably expressing Gadd45ß in anti-sense orientation were generated. The inhibition of constitutive Gadd45p expression in these clone led to a slight redistribution in the cell cycle, reducing the fraction of cells residing in G2, which might underline previously proposed roles of Gadd45 proteins in G2/M checkpoints (Wang et al., 1999c). Despite their ability to activate NF-KB, cells expressing high levels of anti-sense Gadd45ß (AS-Gadd45ß) exhibited a marked susceptibility to the killing by TNFa plus sub-optimal concentrations of CHX (FIG.

1F). In contrast, control lines carrying empty vectors (AS-Hygro) remained resistant to the treatment (FIG. 1F). As with the alterations of the cell cycle, cytotoxicity correlated with high levels of anti-sense mRNA. The data indicate that, under normal circumstances, endogenous Gadd450 is required to antagonize TNFR-induced apoptosis, and suggest that the sensitivity of NF-KB-null cells to cytokine killing is due, at least in part, to the inability of these cells to activate its expression.

Example 4: Gadd45p effectively blocks apoptotic pathways in NF null cells A question was whether expression of Gadd45p affected caspase activation. In NF-K- deficient cells, caspase-8 activity was detected as early as 4 hours after TNFa treatment, as assessed by the ability of 3DO extracts to proteolyze caspase-8-specific substrates in vitro (FIG. 3A, IKBaM and I#BαM-Hygro). This coincided with the marked activation of downstream caspases such as caspase-9, -2,-6, and-3/7. As previously reported, this cascade of events, including the activation of procaspase-8, was completely blocked by NF-KB (Neo; Wang et al., 1998). The cytokine-induced activation of both initiator and executioner caspases was also suppressed in IKBaM-Gadd45ß-10 cells expressing high levels of Gadd45ß (FIG. 3A). Although very low caspase-3/7 activity was detected in these latter cells by 6 hours (bottom, middle panel), the significance of this finding is not clear since there was no sign of the processing of either caspase-3 or-7 in Western blots (FIG. 3B). Indeed, in IKBaM-Gadd45ß and Neo cells, the cleavage of other procaspases, as well as of Bid, was also completely inhibited, despite the presence of normal levels of protein proforms in these cells (FIG. 3B). Proteolysis was specific because other proteins, including 0-action, were not degraded in the cell extracts. Thus, Gadd45p abrogates TNFa-induced pathways of caspase activation in NF-KB-null cells.

To further define the Gadd45p-dependent blockade of apoptotic pathways, mitochondrial functions were analyzed. In IlcBaM and IrBaM-Hygro clones, TNFa induced a drop of the mitochondrial Avm, as measured by the use of the fluorescent dye JC-1. JC-1+ cells began to appear in significant numbers 4 hours after cytokine treatment, reaching-80% by 6 hours (FIG. 3C). Thus in NF-KB-null 3DO cells, the triggering of mitochondrial events and the activation of initiator and executioner caspases occur with similar kinetics. The ability of Bcl-2 to protect IKBaM cells against TNFa-induced killing indicates that, in these cells, caspase activation depends on mitochondrial-amplification mechanisms (Budihardjo et al, 1999). In IKBaM-Gadd45ß-10 as well as in Neo cells, mitochondrial depolarization was completely blocked (FIG. 3A). Inhibition was nearly complete also in IKBaM-Gadd45ß-5 cells, where low caspase activity was observed (FIG. 3A). These findings track the protective activity ofGadd45 (3 to mitochondria, suggesting that the blockade of caspase activation primarily depends on the ability of Gadd45ß to completely suppress mitochondrial amplification mechanisms. As shown in FIGS. 3D and 3E, Gadd45p was able to protect cells against cisplatinum and daunorubicin, suggesting that it might block apoptotic pathways in mitochondria. Consistent with this possibility, expression of this factor also protected cells against apoptosis by the genotoxic agents cisplatinum and daunorubicin (FIGS. 3D and 3E, respectively). Because Gadd45 (3 does not appear to localize to mitochondria, it most likely suppresses mitochondrial events indirectly, by inhibiting pathways that target the organelle.

Example 5: Gadd45p is a specific inhibitor of JNK activation A question explored was whether Gadd450 affected MAPK pathways, which play an important role in the control of cell death (Chang and Karin, 2001). In IKBaM-Hygro clones, TNFa induced a strong and rapid activation of JNK, as shown by Western blots with anti- phospho-JNK antibodies and JNK kinase assays (FIGS. 4A and 5A, left panels). Activation peaked at 5 minutes, to then fade, stabilizing at sustained levels by 40 minutes. The specific signals rose again at 160 minutes due to caspase activation (FIGS. 4A and 5A). Unlike the early induction, this effect could be prevented by treating cells with the caspase inhibitor zVAD-fmk. In IKBaM-Gadd45ß cells, JNK activation by TNFa was dramatically impaired at each time point, despite the presence of normal levels of JNK proteins in these cells (FIG.

4A, right panels). Gadd450 also suppressed the activation of JNK by stimuli other than TNFa, including sorbitol and hydrogen peroxide (FIG. 4B). The blockade, nevertheless, was specific, because the presence of Gadd45 P did not affect either ERK or p38 activation (FIG.

4C). The anti-sense inhibition of endogenous Gadd45p led to a prolonged activation of JNK following TNFR triggering (FIG. 4D, AS-Gadd45ß and Hygro), indicating that this factor, as well as other factors (see down-regulation in AS-Gadd45ß cells) is required for the efficient termination of this pathway. The slightly augmented induction seen at 10 minutes in AS- Gadd45ß cells showed that constitutively expressed Gadd450 may also contribute to the inhibition of JNK (see FIG. 2, basal levels of Gadd45p). Gadd450 is a novel physiological inhibitor of JNK activation. Given the ability of JNK to trigger apoptotic pathways in mitochondria, these observations may offer a mechanism for the protective activity of Gadd45ß.

Example 6: Inhibition of the JNK pathway as a novel protective mechanism by NF-#B Down-regulation of JNK represents a physiologic function of NF-xB. Whereas in Neo cells, JNK activation returned to near-basal levels 40 minutes after cytokine treatment, in MBaM as well as in IKBaM-Hygro cells, despite down-modulation, JNK signaling remained sustained throughout the time course (FIG. 7A; see also FIG. 5A). Qualitatively similar results were obtained with RelA-deficient MEF where, unlike what is seen in wild-type fibroblasts, TNFa-induced JNK persisted at detectable levels even at the latest time points (FIG. 5B). Thus, as with Gadd45p, NF-KB complexes are required for the efficient termination of the JNK pathway following TNFR triggering thus establishing a link between the NF-KB and JNK pathways.

CHX treatment also impaired the down-regulation to TNFa-induced JNK (FIG. 5C), indicating that, in 3DO cells, this process requires newly-induced and/or rapidly turned-over factors. Although in some systems, CHX has been reported to induce a modest activation of JNK (Liu et al., 1996), in 3DO cells as well as in other cells, this agent alone had no effect on this pathway (FIG. 5C ; Guo et al., 1998). The findings indicate that the NF-wB-dependent inhibition of JNK is most likely a transcriptional event. This function indicates the involvement of the activation of Gadd45ß, because this factor depends on the NF-KB for its expression (FIG. 2) and plays an essential role in the down-regulation of TNFR-induced JNK (FIG. 4D).

With two distinct NF-wB-null systems, CXH-treated cells, as well as AS-Gadd45 (3 cells, persistent JNK activation correlated with cytotoxicity, whereas with IKBaM-Gadd45p cells, JNK suppression correlated with cytoprotection. To directly assess whether MAPK cascades play a role in the TNFa-induced apoptotic response of NF-KB-null cells, plasmids expressing catalytically inactive mutants of JNKK1 (MKK4 ; SEK1) or JNKK2 (MKK7), each of which blocks JNK activation (Lin et al., 1995), or of MKK3b, which blocks p38 (Huang et al., 1997), or empty vectors were transiently transfected along with pEGFP into RelA-/-cells. Remarkably, whereas the inhibition of p38 had no impact on cell survival, the suppression of JNK by DN-JNKK2 dramatically rescued RelA-null cells from TNFa-induced killing (FIG. 5D). JNKK1 is not primarily activated by proinflammatory cytokines (Davis, 2000), which may explain why JNKK1 mutants had no effect in this system. Similar findings were obtained in 3DO-IKBaM cells, where MAPK pathways were inhibited by well-characterized pharmacological agents. Whereas, PD98059 and low concentrations of SB202190 (5aM and lower), which specifically inhibit ERK and p38, respectively, could not antagonize TNFa cytotoxicity, high concentrations of SB202190 (50, uM), which blocks both p38 and JNK (Jacinto et al., 1998), dramatically enhanced cell survival (FIG. 5E). The data indicate that JNK, but not p38 (or ERK), transduces critical apoptotic signals triggered by TNFR and that NF-KB complexes protect cells, at least in part, by prompting the down- regulation of JNK pathways.

Example 7: gadd45p is induced by the ectopic expression of RelA, but not Rel or p50 The activation of gadd45ß by cytokines or stress requires NF-KB, as is described herein because induction in abolished either by RelA-null mutations or by the expression of IlcBaM, a variant of the I#Bα inhibitor that blocks that nuclear translocation of NF-#B (Van Antwerp et al., 1996). To determine whether NF-KB is also sufficient to upregulate gadd45fl and, if so, to define which NF-KB family members are most relevant to gene regulation, HeLa-derived HtTA-RelA, HtTA-CCR43, and HtTA-p50 cell lines, which express RelA, Rel, and p50, respectively, were used under control of a teracyclin-regulated promoter (FIG.

6). These cell systems were employed because they allow NF-KB complexes to localize to the nucleus independently of extracellular signals, which can concomitantly activate transcription factors of the NF-KB.

As shown in FIG. 6, the withdrawal of tetracycline prompted a strong increase of gadd45ß mRNA levels in HtTA-RelA cells, with kinetics of induction mirroring those of relA, as well as ixba and p105, two known targets of NF-KB. As previously reported, RelA expression induced toxicity in these cells (gadph mRNA levels) lead to underestimation of the extent of gadd45ß induction. Conversely, gadd45ß was only marginally induced in HtTA-CCR43 cells, which conditionally express high levels of Rel. ixba and p105 were instead significantly activated in these cells, albeit to a lesser extent than in the HtTA-RelA line, indicating that tetracycline withdrawal yielded functional Rel-containing complexes. As it might have been expected, the induction of p50, and NF-#B subunit that lacks a defined activation domain, did not affect endogenous levels of either gadd45ß, i#bα, or p105. The withdrawal of tetracycline did not affect gadd45ß (or relA, rel, or p105) levels in HtTA control cells, indicating the gadd45ß induction in HtTA-RelA cells was due to the activation of NF-KB complexes.

Kinetics of induction of NF-KB subunits were confirmed by Western blot analyses.

Hence gadd45ß expression is dramatically and specifically upregulated upon ectopic expression of the transcriptionally active NF-KB subunit RelA, but not of p50 or Rel (FIG. 6).

These findings are consistent with the observations with RelA-null fibroblasts described above and underscore the importance of RelA in the activation of gadd45ß. Formally, however, these studies do not rule out the possibility that RelA may activate gadd45ß indirectly, for instance by activating other transcription factors.

Example 8: gadd45ß expression correlates with NF-KB activity in B cell lines NF-KB plays a critical role in B lynphopoiesis and is required for survival of mature B cells. Thus, constitutive and inducible expression of gadd45ß were examined in B cell model systems that had been well-characterized from the stand point of NF-KB. Indeed, gadd45ß mRNA levels correlated with nuclear NF-KB activity in these cells (FIG. 7). Whereas gadd45ß kanscripts could be readily seen in unstimulated WEHI-231 B cells, which exhibit constitutively nuclear NF-KB, mRNA levels were below detection in 70Z/3 pre-B cells, which contain instead the classical inducible form of the transcription factor. In both cell types, expression was dramatically augmented by LPS (see longer exposure for 70Z/3 cells) and, in WEH-231 cells, also by PMA, two agents that are known to activate NF-KB in these cells. Thus gadd45ß may mediate some of the important functions executed by NF-KB in B lymphocytes.

Example 9: The gadd45ß promoter contains several putative KB elements To begin to understand the regulation of gadd45ß expression by NF-KB, the muring gadd45ß promoter was cloned. A BAC clone containing the gadd45fl gene was isolated from a 129SV mouse genomic library, digested with XhoI, and subcloned into pBS vector. The 7384 bp Xhol fragment containing gadd45ß was completely sequenced, and portions were found to match sequences previously deposited in GeneBank (accession numbers AC073816, AC073701, and AC091518) (see also FIG. 8). The fragment harbored the genomic DNA region spanning from-5. 4 kbp upstream of a previously identified transcription start site to near the end of the 4th exon of gadd45ß. Next, the TRANSFAC database was used to identify putative transcription factor-binding elements. A TATAA box was found to be located at position-56 to-60 relative to the transcription start site (FIG. 10). The gade45/3 promoter also exhibited several KB elements, some of which were recently noted by others. Three strong KB sites were found in the proximal promoter region at positions-377/-368,-426/-417, and-447/-438 (FIG. 8) ; whereas a weaker site was located as position-4516,-4890/-4881, and-5251/-5242 (FIG. 8). Three KB consensus sites were also noted with the first exon of gadd45ß (+27/+36, +71/+80, and +171/+180). The promoter also contained an Spl motif (- 890/-881) and several putative binding sties for other transcription factors, including heat shock factor (HSF) 1 and 2, Ets, Stat, AP1, N-Myc, MyoD, CREB, and C/EBP (FIG. 8).

To identify conserved regulatory elements, the 5.4 kbp murine DNA sequence immediately upstream of the gaeld45/ transcription start site was aligned with corresponding human sequence, previously deposited by the Joint Genome Initiative (accession number AC005624). As shown in FIG. 8, DNA regions spanning from position-1477 to-1197 and from-466 to-300 of the murine gadd45ß promoter were highly similar to portions of the human promoter (highlighted in gray are identical nucleotides within regions of homology), suggesting that these regions contain important regulatory elements. A less well-conserved regions was identified downstream of position-183 up to the beginning of the first intron.

Additional shorter stretches of homology were also identified (see FIG. 8). No significant similarity was found upstream of position-2285. Of interest, the-466/-300 homology region contained three KB sites (hereafter referred to as KB 1, #B2, and KB3), which unlike the other aB sites present throughout the gadd45ß promoter, were conserved among the two species.

These findings suggest that these KB sites play an important role in the regulation of gade45/3, perhaps accounting for the induction of gadd45ß by NF-#B.

Example 10: NF-KB regulates the gadd45p promoter through three proximal leB elements To determine the functional significance of the KB sites present in the gadd45/3 promoter, a series of CAT reporter constructs were generated where CAT gene expression is driven by various portions of this promoter (FIG. 9A). Each CAT construct was transfected alone or along with increasing amounts of RelA expression plasmids into NTera-2 embryo carcinoma cells, and CAT activity measured in cell lysates by liquid scintillation counting (FIG. 9B). RelA was chosen for these experiments because of its relevance to the regulation of gadd45ß expression as compared to other NF-lcB subunits (see FIG. 6). As shown in FIG.

9B, the-5407/+23-gadd45ß-CATT reporter vector was dramatically transactivated by RelA in a dose-dependent manner, exhibiting an approximately 340-fold induction relative to the induction seen in the absence of RelA with the highest amount of pMT2T-RelA.

Qualitatively similar, RelA-dependent effects were seen with the-3465/+23-gadd45ß-and- 592/+23-gadd45ß-CAT constructs, which contained distal truncations of the gadd45ß promoter. The relatively lower constructs, which contained distal truncations of the gadd45ß promoter. The relatively lower basal and RelA-dependent CAT activity observed with the- 5407/+23- gadd45ß-CAT, may have been due, at least in part, to the lack of a proximal 329 bp regulatory region, which also contained the TATA box, in the former constructs (FIG. 9A and 9B). Importantly, even in the presence of this region, deletions extending proximally to position-592 completely abolished the ability of RelA to activate the CAT gene (FIG. 9B, see -265/+23- gadd45ß- and -103/+23- gadd45ß-CAT constructs). Similar findings were obtained with analogous reporter constructs containing an additional 116 b promoter fragment downstream of position +23. Whereas analogously to-592/+23-gadd45ß-CAT,- 592/+139-gadd45ß-CAT was highly response to RelA,-265/+139-J-CAT was not transactivated even by the highest amounts of pMT2T-RelA. It should be noted that this reporter construct failed to respond to RelA despite retaining two putative KB binding elements at position +27/+36 and +71/+80 (see FIG. 8, SEQ ID NO: 35). Together, the findings indicate that relevant NF-#B/RelA responsive elements in the murine gade45/3 promoter reside between position-592 and +23. They also imply that the KB sites contained in the first exon, as well as the distal KB sites, may not significantly contribute to the regulation of gadd45ß by NF-#B. Similar conclusions were obtained in experiments employing Jurkat or HeLa cells where NF-#B was activated by PMA plus ionomycin treatment.

As shown in FIG. 8, the-592/+23 region of the gadd45fl promoter contains three conserved KB binding sties, namely KBl, icB2, and KB3. To test the functional significance of these KB elements, each of these sites were mutated in the context of -592/+23-gadd45ß- CAT (FIG. 10A), which contained the minimal promoter region that can be transactivated by RelA. Mutant reporter constructs were transfected alone or along with increasing amounts of PMT2T-RelA in NTera-2 cells and CAT activity measured as described for FIG. 9B. As shown in FIG. 10B, the deletion of each KB site significantly impaired the ability of RelA to transactivate the-592/+23-gadd45ß-CAT construct, with the most dramatic effect seen with the mutation of KB 1, resulting in a #70% inhibition of CAT activity (compare-592/+23- gadd45ß-CAT and #B-1M-gadd45ß-CAT). Of interest, the simultaneous mutation of KB1 and KB2 impaired CAT induction by approximately 90%, in the presence of the highest amount of transfected RelA plasmids (FIG. 10B) (see #B-1/2M-gadd45ß-CAT). Dramatic effects were also seen when the input levels of RelA were reduced to 1 llg or 0.3 jig (-eight- and-five-fold reduction, respectively, as compared to the wild-type promoter). The residual CAT activity observed with the latter mutant construct was most likely due to the presence of an intack #B3 site. Qualitatively similar results were obtained with the transfection of RelA plus p50, or Rel expression constructs. It was concluded that the gade45&num promoter contains three functional KB elements in its proximal region and that each is required for optimal transcriptional activation of NF-KB.

To determine whether these sites were sufficient to drive NF-KB-dependent transcription the A56-KB-1/2-, A56-1B-3-, and A56-KB-M-CAT, reporter constructs were constructed, where one copy of the gadd45ßKB sites or of a mutated site, respectively, were cloned into A56-CAT to drive expression of the CAT gene (FIG. 11). Each A56-CAT construct was then transfected alone or in combination with increasing amounts of RelA expression plasmids into Ntera2 cells and CAT activity measured as before. As shown in FIG. 11, the presence of either KB-1 plus KB-2, or #B-3 dramatically enhanced the responsiveness of A56-CAT to RelA. As it might have been expected from the fact that it harbored two, rather than one, oB sites, A56-KB-1/2-CAT was induced more efficiently than oB3, particularly with the highest amount of pMT2T-RelA. Low, albeit significant, RelA- dependent CAT activity was also noted with A56-KB-M-CAT, as well as empty A56-CAT vectors, suggesting that A56-CAT contains cryptic KB sites (FIG. 11). Hence, either the KB-1 plus KB-2, or KB-3 cis-acting elements are sufficient to confer promoter responsiveness to NF-#B.

Example 11: The KB-1, KB-2, and #B-3 elements bind to NF-KB in vitro To assess the ability of KB elements in the gadd45ß promoter to interact with NFKB complexes, EMSAs were performed. Oligonucleotides containing the sequence of KB-1, KB- 2, or #B-3 were radiolabeled and independently incubated with extracts of NTera-2 cells transfected before hand with pMT2T-p50, pMT2T-p50 plus pMT2T-RelA, or empty pMT2T plasmids, and DNA-binding complexes separated by polyacrylamide gel electrophoresis (FIG. 12A). The incubation of each KB probe with various amounts of extract from cells expressing only p50 generated a single DNA-binding complex comigrating with p50 homodimers (FIG 12A, lanes 1-3,5-7, and 9-11). Conversely, extracts from cells expressing both p50 and RelA gave rise to two specific bands: one exhibiting the same mobility of p50/p50 dimers and the other comigrating with p50/RelA heterodimers (lanes 4, 8, and 12).

As shown previously, extracts from mock-transfected NTera2 cells did not generate any specific signal in EMSAs (FIG. 12B). Specificity of each complex was confirmed by competition assays where, in addition to the radiolabeled probe, extracts were incubated with a 50-fold excess of wild-type or mutated cold KB probes. Thus, each of the functionally relevant KB elements in the gade45, promoter can bind to NF-KB complexes in vitro.

To confirm the composition of the DNA binding complexes, supershift assays were performed by incubating the cell extracts with polyclonal antibodies raised against human p50 or RelA. Anti-p50 antibodies completely supershifted the specific complex seen with extracts of cells expressing p50 (FIG. 12B, lanes 5,14, and 23), as well as the two complexes detected with extracts of cells expressing both p50 and RelA (lanes 8,17, and 26).

Conversely, the antibody directed against RelA only retarded migration of the slower complex seen upon concomitant expression of p50 and RelA (lanes 9,18, 27) and did not affect mobility of the faster DNA-binding complex (lanes 6,9, 15,18, 24, and 27).

It should be noted that the gadd45ß-1B sites exhibited apparently distinct in vitro binding affinities for NF-KB complexes (see discussion below). Indeed, with p50/RelA heterodimers, KB-2 and KB-3 yielded significantly stronger signals as compared with KB-1 (FIG. 12B). Conversely, KB-2 gave rise to the strongest signal with p50 homodimers, whereas icB-3 appeared to associate with this complex most poorly in vitro (FIG. 12B).

Judging from the amounts of p50/p50 and p50/RelA complexes visualized on the gel, the presence of the antibodies (especially the anti-RelA antibody) may have stabilized NF-KB- DNA interactions (FIG. 12B). Neither antibody gave rise to any band when incubated with the radiolabeled probe in the absence of cell extract. The specificity of the supershifted bands was further demonstrated by competitive binding reactions with unlabeled competitor oligonucleotides. Hence, consistent with migration patterns (FIG. 14A), the faster complex is predominantly composed of p50 homodimers, whereas the lower one is predominantly composed of p50/RelA heterodimers. These data are consistent with those obtained with the CAT assays and demonstrate that each of the relevant KB elements of the gade45&num promoter can specifically bind to p50/p50 and p50/RelA, NFKB complexes, in vitro, thereby providing the basis for the dependence of gadd45ß expression on NF-KB. Hence, gade453 is a novel direct target of NFKB.

MATERIALS AND METHODS 1. Library preparation and enrichment cDNA was prepared from TNFa-treated NIH-3T3 cells and directionally inserted into the pLTP vector (Vito et al., 1996). For the enrichment, RelA-/-cells were seeded into 1.5 x 106/plate in 100 mm plates and 24 hours later used for transfection by of the spheroplasts fusion method. A total of 4.5 x 106 library clones were transfected for the first cycle. After a 21-hours treatment with TNFa (100 units/ml) and CHX (0.25 pg/ml), adherent cells were harvested for the extraction of episomal DNA and lysed in 10 mM EDTA, 0.6% SDS for the extraction of episomal DNA after amplification, the library was used for the next cycle of selection. A total of 4 cycles were completed.

2. Constructs IKBaM was excised from pCMX-IKBaM (Van Antwerp et al., 1996) and ligated into the EcoRI site of pcDNA3-Neo (Invitrogen). Full length human RelA was PCR-amplified from BS-RelA (Franzoso et al., 1992) and inserted into the BamHI site of pEGFP-Cl (Clontech). Gadd45p, Gadd45a and Gadd45y cDNAs were amplified by PCR for the pLTP library and cloned into the XhoI site and pcDNA 3.1-Hygro (Invitrogen) in both orientations.

To generate pEGFP-Gadd45p, Gadd45p was excised from pCDNA Hygro with XhoI-XbaI and ligated with the linker 5'-CTAGAGGAACGCGGAAGTGGTGGAAGTGGTGGA-3' (SEQ ID NO: 13) into the XbaI-BamHI sites of pEGFP-Nl. pcDNA-Gadd45a was digested with EcoRI-XhoI and ligated with XhoI-BamHI opened pEGFP-Cl and the linker 5'- GTACAAGGGAAGTGGTGGAAGTGTGGAATGACTTTGGAGG-3' (SEQ ID NO: 14). pEGFP-Nl-Gadd45y was generated by introducing the BspEI-XhoI fragment of pCDNA- Hygro-Gadd45y along with the adapter 5'-ATTGCGTGGCCAGGATACAGTT-3' (SEQ ID NO: 15) into pEGFP-Cl-Gadd45a, where Gadd45a was excised by EcoRI-SalI. All constructs were checked by sequencing. pSRa3 plasmids expressing DN-JNKK1 (S257A, T261A), DN-JNKK2 (K149M, S271A, T275A) and MKK3bDN (S128A, T222A) were previously described (Lin et al., 1995 ; Huang et al., 1997).

3. Anti Sense Constructs of gadd45P Modulators of the JNK pathway, such as Gadd45p, can be modulated by molecules that directly affect RNA transcripts encoding the respective functional polypeptide.

Antisense and ribozyme molecules are examples of such inhibitors that target a particular sequence to achieve a reduction, elimination or inhibition of a particular polypeptide, such as a Gadd45 sequence or fragments thereof (SEQ ID NO: 1, SEQ ID NO: 3, SEQ ID NO: 5, SEQ ID NO: 7, SEQ ID NO: 9 or SEQ ID NO: 11).

Antisense methodology takes advantage of the fact that nucleic acids tend to pair with ''complementary''sequences. Antisense constructs specifically form a part of the current invention, for example, in order to modulate the JNK pathway. In one embodiment of the invention, antisense constructs comprising a Gadd45 nucleic acid are envisioned, including antisense constructs comprising nucleic acid sequence of SEQ ID NO: 1, SEQ ID NO: 3, SEQ ID NO: 5, SEQ ID NO: 7, SEQ ID NO: 9, SEQ ID NO: 11 and SEQ ID NOS: 35-41 in antisense orientation, as well as portions of fragments thereof.

By complementary, it is meant that polynucleotides are those which are capable of base-pairing according to the standard Watson-Crick complementarily rules. That is, the larger purines will base pair with the smaller pyrimidines to form combinations of guanine paired with cytosine (G: C) and adenine paired with either thymine (A: T) in the case of DNA, or adenine paired with uracil (A: U) in the case of RNA. Inclusion of less common bases such as inosine, 5-methylcytosine, 6-methyladenine, hypoxanthine and others in hybridizing sequences doe not interfere with pairing.

Targeting double-stranded (ds) DNA with polynucleotides leads to triple-helix formation ; targeting RNA will lead to double-helix formation. Antisense polynucleotides, when introduced into a target cell, specifically bind to their target polynucleotide and interfere with transcription, RNA processing, transport, translation and/or stability. Antisense RNA constructs, or DNA encoding such antisense RNAs, may be employed to inhibit gene transcription or translation of both within a host cell, either in vitro or in vivo, such as within a host animal, including a human subject.

Antisense constructs, including synthetic anti-sense oligonucleotides, may be designed to bind to the promoter and other control regions, exons, introns or even exon-intron boundaries of a gene. It is contemplated that the most effective antisense constructs may include regions complementary to intron/exon splice junctions. Thus, antisense constructs with complementarily to regions within 50-200 bases of an intron-exon splice junction may be used. It has been observed that some exon sequences can be included in the construct without seriously affecting the target selectivity thereof. The amount of exonic material included will vary depending on the particular exon and intron sequences used. One can readily test whether too much exon DNA is included simply by testing the constructs in vitro to determine whether normal cellular function is affected or whether the expression of related genes having complementary sequences is affected.

It may be advantageous to combine portions of genomic DNA with cDNA or synthetic sequences to generate specific constructs. For example, where an intron is desired in the ultimate construct, a genomic clone will need to be used. The cDNA or a synthesized polynucleotide may provide more convenient restriction sites for the remaining portion of the construct and, therefore, would be used for the rest of the sequence.

4. Cell Lines, transfections and treat7nents MEF and 3DO cells were cultured in 10% Fetal bovine serum-supplemented DMEM and RPMI, respectively. Transient transfections in RelA-/-MEF were performed by Superfect according to the manufacturer's instructions (Qiagen). After cytotoxic treatment with CHX (Sigma) plus or minus TNFa (Peprotech), adherent cells were counted and analyzed by FCM (FACSort, Becton Dickinson) to assess numbers of live GFP+ cells. To generate 3DO stable lines, transfections were carried out by electroporatoration (BTX) and clones were grown in appropriate selection media containing Geneticin (Gibco) and/or Hygromycin (Invitrogen). For the assessment of apoptosis, 2DO cells were stained with PI (Sigma) and analyzed by FCM, as previously described (Nicoletti et aL, 1991).

Daunorubicin, PMA, Ionomycin, hydrogen peroxide, and sorbitol were from Sigma; Cisplatin (platinol AQ) was from VHAplus, PD98059 and SB202190 were from Calbiochem.

5. Northern Blots, Western blots, EMSAs, and kinase assays Northern blots were performed by standard procedures using 6llg of total RNA. The EMSAs with the palindromic probes and the preparation of whole cell extracts were as previously described (Franzoso et al., 1992). For western blots, cell extracts were prepared either in a modified lysis buffer (50mM Tris, pH 7.4, 100 mM NaCl, 50 mM NaF, 1mM NaVo4, 30 mM pyrophosphate, 0.5% NP-40, and protease inhibitors (FIG. 1B; Boehringer Mannheim), in Triton X-100 buffer (FIG. 4A; Medema et al., 1997) or in a lysis buffer containing 1% NP-40 350mM NaCI, 20 MM HEPES (pH 8.0), 20% glycerol, 1mM MgCl2, 0.1 mM EGTA, 0.5 mM DTT, 1 mM Na3V04, 50 mM NaF and protease inhibitors. Each time, equal amounts of proteins (ranging between 15 and 50 Rg) were loaded and Western blots prepared according to standard procedures. Reactions were visualized by ECL (Amersham). Antibodies were as follows: IKBa, Bid, and (3-actin from Santa Cruz Biotechnology ; caspase-6, -7 and-9, phospho and total-p38, phosph and total-ERK, phospho and total-JNK from Cell Signaling Technology ; caspase-8 from Alexis; Caspase-2 and-3 from R&D systems. The Gadd45ß-specific antibody was generated against an N- Terminal peptide. Kinase assays were performed with recombinant GST-cjun and anti-JNK antibodies (Pharmingen), (Lin et al., 1995).

6. Measurement of caspase activity and mitochondrial transmembrane potential For caspase in vitro assays, cells were lysed in Triton X-100 buffer and lysates incubated in 40, M of the following amino trifluromethyl coumarin (ATC) -labeled caspase- specific peptides (Bachem) : xVDVAD (caspase 2), zDEVD (caspases 3/7), xVEID (caspase 6), xIETD (caspase 8), and Ac-LEHD (caspase 9). Assays were carried out as previously described (Stegh et al., 2000) and specific activities were determined using a fluorescence plate reader. Mitochondrial transmembrane potential was measured by means of the fluorescent dye JC-1 (Molecular Probes, Inc.) as previously described (Scaffidi et al., 1999).

After TNFa treatment, cells were incubated with 1. 25 ug/ml of the dye for 10 min at 37°C in the dark, washed once with PBS and analyzed by FCM.

7. Therapeutic Application of the Invention The current invention provides methods and compositions for the modulation of the JNK pathway, and thereby, apoptosis. In one embodiment of the invention, the modulation can be carried out by modulation of Gadd45ß and other Gadd45 proteins or genes.

Alternatively, therapy may be directed to another component of the JNK pathway, for example, JNK1, JNK2, JNK3, MAPKKK (Mitogen Activated Protein Kinase Kinase Kinase): GCK, GCKR, ASK1/MAPKKK5, ASK2/MAPKKK6, DLKIMUK/ZPK, LZK, MEKK1, MEKK2, MEKK3, MEKK4/MTK1, MLK1, MLK2/MST, MLK3/SPRK/PTK1, TAK1, Tpl-2/Cot. MAPKK (Mitogen Activated Protein Kinase Kinase): MKK4/SEKl/SERKl/SKKl/JNKK1, MKK7/SEK2/SKK4/JNKK2. MAPK (Mitogen Activated Kinase): JNK1/SAPK/SAPK1 c, JNK2/SAPKa/SAPKl a, JNK3/SAPK (3/SAPKlb/p49F 12.

Further, there are numerous phosphatases, scaffold proteins, including JIPl/IB 1, JIP2/IB2, JIP3/JSAP and other activating and inhibitory cofactors, which are also important in modulating JNK signaling and may be modulated in accordance with the invention. The invention may find therapeutic uses for potentially any condition that can be affected by an increase or decrease in apoptosis. The invention is significant because many diseases are associated with an inhibition or increase of apoptosis. Conditions that are associated with an inhibition of apoptosis include cancer; autoimmune disorders such as systemic lupus erythemaosus and immune-mediated glomerulonephritis; and viral infections such as Herpesviruses, Poxviruses and Adenoviruses. The invention therefore provides therapies to treat these, and other conditions associated with the inhibition of apoptosis, which comprise administration of a JNK pathway modulator that increases apoptosis. As upregulation of Gadd45 blocks apoptosis, diseases caused by inhibition of apoptosis will benefit from therapies aimed to increase JNK activation, for example via inhibition of Gadd45. one example of a way such inhibition could be achieved is by administration of an antisense Gadd45 nucleic acid.

The invention may find particular use for the modulation of apoptosis, and particularly the increase of apoptosis, for the treatment of cancer. In these instances, treatments comprising a combination of one or more other therapies may be desired. For example, a modulator of the JNK pathway might be highly beneficial when used in combination with conventional chemo-or radio-therapies. A wide variety of cancer therapies, known to one of skill in the art, may be used individually or in combination with the modulators of the JNK pathway provided herein. Combination therapy can be used in order to increase the effectiveness of a therapy using an agent capable of modulating a gene or protein involved in the JNK pathway. Such modulators of the JNK pathway may include sense or antisense nucleic acids.

One example of a combination therapy is radiation therapy followed by gene therapy with a nucleic acid sequence of a protein capable of modulating the JNK pathway, such as a sense or antisense Gadd45) 3 nucleic acid sequence. Alternatively, one can use the JNK modulator based anti-cancer therapy in conjunction with surgery and/or chemotherapy, and/or immunotherapy, and/or other gene therapy, and/or local heat therapy. Thus, one can use one or several of the standard cancer therapies existing in the art in addition with the JNK modulator-based therapies of the present invention.

The other cancer therapy may precede or follow a JNK pathway modulator-based therapy by intervals ranging from minutes to days to weeks. In embodiments where other cancer therapy and a Gadd45p inhibitor-based therapy are administered together, one would generally ensure that a significant period of time did not expire between the time of each delivery. In such instances, it is contemplated that one would administer to a patient both modalities without about 12-24 hours of each other and, more preferably, within about 6-12 hours of each other, with a delay time of only about 12 hours being most preferred. In some situations, it may be desirable to extend the time period for treatment significantly, however, where several days (2,3, 4,5, 6 or 7) to several weeks (1, 2,3, 4,5, 6,7 or 8) lapse between the respective administrations.

It also is conceivable that more than one administration of either another cancer therapy and a Gadd45p inhibitor-based therapy will be required to achieve complete cancer cure. Various combinations may be employed, where the other cancer therapy is"A"and a JNK pathway modulator-based therapy treatment, including treatment with a Gadd45 inhibitor, is"B", as exemplified below: A/B/A B/A/B B/B/A A/A/B B/A/A A/B/B B/B/B/A B/B/A/B A/A/B/B A/B/A/B A/B/B/A B/B/A/A/B/AB/A B/A/A/B B/B/B/A A/A/A/B B/A/A/A A/B/A/A A/A/B/A A/B/B/B B/A/B/B B/B/A/B Other combinations also are contemplated. A description of some common therapeutic agents is provided below.

8. Chemotherapeutic Agents In the case of cancer treatments, another class of agents for use in combination therapy are chemotherapeutic agents. These agents are capable of selectively and deleteriously affecting tumor cells. Agents that cause DNA damage comprise one type of chemotherapeutic agents. For example, agents that directly cross-link DNA, agents that intercalate into DNA, and agents that lead to chromosomal and mitotic aberrations by affecting nucleic acid synthesis. Some examples of chemotherapeutic agents include antibiotic chemotherapeutics such as Doxorubicin, Daunorubucin, Mitomycin (also known as mutamycin and/or mitomycin-C), Actinomycine D (Dactinomycine), Bleomycin, Plicomycin.

Plant alkaloids such as Taxol, Vincristine, Vinblastine. Miscellaneous agents such as Cisplatin, VP16, Tumor Necrosis Factor. Alkylating Agents such as, Carmustine, Melphalan (also known as alkeran, L-phenylalanine mustard, phenylalanine mustard, L-PAM, or L- sarcolysin, is a phenylalanine derivative of nitrogen mustard), Cyclophosphamide, Chlorambucil, Busulfan (also known as myleran), Lomustine. And other agents for example, Cisplatin (CDDP), Carboplatin, Procarbazine, Mechlorethamine, Camptothecin, Ifosfamide, Nitrosurea, Etoposide (VP16), Tamoxifen, Raloxifene, Estrogen Receptor Binding Agents, Gemcitabien, Mavelbine, Farnesyl-protein transferase inhibitors, Transplatinum, 5- Fluorouracil, and Methotrexate, Temaxolomide (an aqueous form of DTIC), or any analog or derivative variant of the foregoing. a. Cisplatinum Agents that directly cross-link nucleic acids, specifically DNA, are envisaged to facilitate DNA damage leading to a synergistic, anti-neoplastic combination with a mutant oncolytic virus. Cisplatinum agents such as cisplatin, and other DNA alkylating agents may be used. Cisplatinum has been widely used to treat cancer, with efficacious doses used in clinical applications of 20 mg/m2 for 5 days every three weeks for a total of three courses.

Cisplatin is not absorbed orally and must therefore be delivered via injection intravenously, subcutaneously, intratumorally or intraperitoneally. b. Dauno7^ubicin Daunorubicin hydrochloride, 5,12-Naphthacenedione, (8S-cis)-8-acetyl-10-[(3-amino- 2,3, 6-trideoxy-a-L-lyxo-hexanopyranosyl) oxy] -7,8, 9, 10-tetrahydro-6, 8, 11-trihydroxy-10- methoxy-, hydrochloride; also termed cerubidine and available from Wyeth. Daunorubicin intercalates into DNA, blocked DNA-directed RNA polymerase and inhibits DNA synthesis.

It can prevent cell division in doses that do not interfere with nucleic acid synthesis.

In combination with other drugs it is included in the first-choice chemotherapy of acute myelocytic leukemia in adults (for induction of remission), acute lymphocytic leukemia and the acute phase of chronic myelocytic leukemia. Oral absorption is poor, and it must be given intravenously. The half-life of distribution is 45 minutes and of elimination, about 19 hr. the half-life of its active metabolite, daunorubicinol, is about 27 hr. daunorubicin is metabolized mostly in the liver and also secreted into the bile (ca 40%). Dosage must be reduced in liver or renal insufficiencies.

Suitable doses are (base equivalent), intravenous adult, younger than 60 yr. 45 mg/m2/day (30 mg/m2 for patients older than 60 yr. ) for 1,2 or 3 days every 3 or 4 wk or 0.8 mg/kg/day for 3 to 6 days every 3 or 4 wk; no more than 550 mg/m2 should be given in a lifetime, except only 450 mg/m2 if there has been chest irradiation; children, 25 mg/m2 once a week unless the age is less than 2 yr. or the body surface less than 0.5 m, in which case the weight-based adult schedule is used. It is available in injectable dosage forms (base equivalent) 20 mg (as the base equivalent to 21.4 mg of the hydrochloride). Exemplary doses may be 10 mg/m2, 20 mg/m2, 30 mg/m2, 50 mg/m2, 100 mg/m2, 150 mg/m2, 175 mg/m2, 200 mg/m2, 225 mg/m2, 250 mg/m2, 275 mg/m2, 300 mg/m, 350 mg/m2, 400 mg/m2, 425 mg/m2, 450 mg/m2, 475 mg/m2, 500 mg/m2. Of course, all of these dosages are exemplary, and any dosage in-between these points is also expected to be of use in the invention.

9. Immunotherapy In accordance with the invention, immunotherapy could be used in combination with a modulator of the JNK pathway in therapeutic applications. Alternatively, immunotherapy could be used to modulate apoptosis via the JNK pathway. For example, anti-Gadd45 (3 antibodies or antibodies to another component of the JNK pathway could be used to disrupt the function of the target molecule, thereby inhibiting Gadd45 and increasing apoptosis.

Alternatively, antibodies can be used to target delivery of a modulator of the JNK pathway to a cell in need thereof. For example, the immune effector may be an antibody specific for some marker on the surface of a tumor cell. Common tumor markers include carcinoembryonic antigen, prostate specific antigen, urinary tumor associate antigen, fetal antigen, tyrosinse (97), gp68, TAG-72, HMFG, Sialyl Lewis Antigen, MucA, MucB, PLAP, estrogen receptor, laminin receptor, erb B and pi 55.

In an embodiment of the invention the antibody may be an anti-Gadd45 [3 antibody.

The antibody alone may serve as an effector of therapy or it may recruit other cells to actually effect cell killing. The antibody also may be conjugated to a drug or toxin (chemotherapeutic, radionuclide, ricin A chain, cholera toxin, pertussis toxin, etc.) and serve merely as a targeting agent. Alternatively, the effector may be a lymphocyte carrying a surface molecule that interacts, either directly or indirectly, with a target in a tumor cell, for example Gadd45ß. Various effector cells include cytotoxic T cells and NK cells. These effectors cause cell death and apoptosis. The apoptotic cancer cells are scavenged by reticuloendothelial cells including dendritic cells and macrophages and presented to the immune system to generate anti-tumor immunity (Rovere et al., 1999; Steinman et al., 1999).

Immune stimulating molecules may be provided as immune therapy: for example, cytokines such as IL-2, IL-4, IL-12, GM-CSF, gamma-IFN, chemokines such as MIP-1, MCP-1, IL-8 and growth factors such as FLT ligand. Combining immune stimulating molecules, either as proteins or using gene delivery in combination with Gadd45 inhibitor will enhance anti-tumor effects. This may comprise: (i) Passive Immunotherapy which includes: injection of antibodies alone; injection of antibodies coupled to toxins or chemotherapeutic agents; injection of antibodies coupled to radioactive isotopes; injection of anti-idiotype antibodies; and finally, purging of tumor cells in bone marrow; and/or (ii) Active Immunotherapy wherein an antigenic peptide, polypeptide or protein, or an autologous or allogenic tumor cell composition or"vaccine"is administered, generally with a distinct bacterial adjuvant (Ravindranath & Morton, 1991; Morton & Ravindranath, 1996; Morton et al., 1992; Mitchell et al., 1990; Mitchell et al., 1993) and/or (iii) Adoptive Immunotherapy wherein the patient's circulating lymphocytes, or tumor infiltrated lymphocyltes, are isolated in vitro, activated by lymphokines such as IL-2 or transduced with genes for tumor necrosis, and readministered (Rosenberg et al., 1998 ; 1989).

10. Gene therapy In yet another embodiment, therapy in accordance with the invention may comprise gene therapy, in which one or more therapeutic polynucleotide is administered to a patient in need thereof. This can comprise administration of a nucleic acid that is a modulator of the JNK pathway, and may also comprise administration of any other therapeutic nucleotide in combination with a modulator of the JNK pathway. One embodiment of cancer therapy in accordance with the invention comprises administering a nucleic acid sequence that is an inhibitor of Gadd45p, such as a nucleic acid encoding a Gadd45 P inhibitor polypeptide or an antisense Gadd45ß sequence. Delivery of a vector encoding a JNK inhibitor polypeptide or comprising an antisense JNK pathway modulator in conjunction with other therapies, including gene therapy, will have a combined anti-hyperproliferative effect on target tissues.

A variety of proteins are envisioned by the inventors as targets for gene therapy in conjunction with a modulator of the JNK pathway, some of which are described below.

11. Clinical Protocol A clinical protocol has been described herein to facilitate the treatment of cancer using a modulator of the JNK pathway, such as an inhibitor of a Gadd45 protein, including the activity or expression thereof by a Gadd45 gene. The protocol could similarly be used for other conditions associated with a decrease in apoptosis. Alternatively, the protocol could be used to assess treatments associated with increased apoptosis by replacing the inhibitor of Gadd45 with an activator of Gadd45.

12. Therapeutic kits Therapeutic kits comprising a modulator of the JNK pathway are also described herein. Such kits will generally contain, in suitable container means, a pharmaceutically acceptable formulation of at least one modulator of the JNK pathway. The kits also may contain other pharmaceutically acceptable formulations, such as those containing components to target the modulator of the JNK pathway to distinct regions of a patient or cell type where treatment is needed, or any one or more of a range of drugs which may work in concert with the modulator of the JNK pathway, for example, chemotherapeutic agents.

The kits may have a single container means that contains the modulator of the JNK pathway, with or without any additional components, or they may have distinct container means for each desired agent. When the components of the kit are provided in one or more liquid solutions, the liquid solution is an aqueous solution, with a sterile aqueous solution being particularly preferred. However, the components of the kit may be provided as dried powder (s). When reagents or components are provided as a dry powder, the powder can be reconstituted by the addition of a suitable solvent. It is envisioned that the solvent also may be provided in another container means. The container means of the kit will generally include at least one vial, test tube, flask, bottle, syringe or other container means, into which the monoterpene/triterpene glycoside, and any other desired agent, may be placed and, preferably, suitably aliquoted. Where additional components are included, the kit will also generally contain a second vial or other container into which these are placed, enabling the administration of separated designated doses. The kits also may comprise a second/third container means for containing a sterile, pharmaceutically acceptable buffer or other diluent.

The kits also may contain a means by which to administer the modulators of the JNK pathway to an animal or patient, e. g., one or more needles or syringes, or even an eye dropper, pipette, or other such like apparatus, from which the formulation may be injected into the animal or applied to a diseased area of the body. The kits of the present invention will also typically include a means for containing the vials, or such like, and other component, in close confinement for commercial sale, such as, e. g., injection or blow-molded plastic containers into which the desired vials and other apparatus are placed and retained.

13. Gadd45 Compositions Certain aspects of the current invention involve modulators of Gadd45. In one embodiment of the invention, the modulators may Gadd45 or other genes or proteins. In particular embodiments of the invention, the inhibitor is an antisense construct. An antisense construct may comprise a full length coding sequence in antisense orientation and may also comprise one or more anti-sense oligonucleotides that may or may not comprise a part of the coding sequence. Potential modulators of the JNK pathway, including modulators of Gadd45 (3, may include synthetic peptides, which, for instance, could be fused to peptides derived from the Drosophila Antennapedia or HIV TAT proteins to allow free migration through biological membranes ; dominant negative acting mutant proteins, including constructs encoding these proteins; as well as natural and synthetic chemical compounds and the like. Modulators in accordance with the invention may also upregulate Gadd45, for example, by causing the overexpression of a Gadd45 protein. Similarly, nucleic acids encoding Gadd45 can be delivered to a target cell to increase Gadd45. The nucleic acid sequences encoding Gadd45 may be operably linked to a heterologous promoter that may cause overexpression of the Gadd45.

Exemplary Gadd45 gene can be obtained from Genbank Accession No. NM-015675 for the human cDNA, NP 056490.1 for the human protein, NM-008655 for the mouse cDNA and NM-032681. 1 for the mouse protein (SEQ ID NOS: 1-4, respectively). Similarly, for Gadd45a nucleotide and protein sequences the Genbank Accession NOS. are: NM-001924 for the human cDNA ; NP-001915 for the human protein ; NM-007836 for the mouse cDNA and NP-031862.1 for the mouse protein (SEQ ID NOS: 5-8, respectively). For Gadd45y nucleotide and protein sequences the Genbank Accession Nos. are: NM-006705 for the human cDNA, NP-006696.1 for the human protein, NM-011817 for the mouse cDNA and NP-035947.1 for the mouse protein (SEQ ID NOS: 9-12, respectively). Also forming part of the invention are contiguous stretches of nucleic acids of SEQ ID NO: 1, SEQ ID NO: 3, SEQ ID NO: 5, SEQ ID NO: 7, SEQ ID NO: 9, SEQ ID NO: 11, and SEQ ID NO: 35, including about 25, about 50, about 75, about 100, about 150, about 200, about 300, about 400, about 55, about 750, about 100, about 1250 and about 1500 or more contiguous nucleic acids of these sequences. The binding sites of the Gadd45 promoter sequence of SEQ ID NO: 35, including the core binding sites of kB-1, kB-2 and kB-3, given by SEQ ID NO: 36, SEQ ID NO: 38 and SEQ ID NO: 40, also form part of the invention. In further embodiments, the binding sites may have the nucleic acid sequences of SEQ ID NO: 37, SEQ ID NO: 39 and SEQ ID NO: 41. Any of these sequences may be used in the methods and compositions described herein.

Further specifically contemplated by the inventors are arrays comprising any of the foregoing sequences bound to a solid support. Proteins of Gadd45 and other components of the JNK pathway may also be used to produce arrays. In certain embodiments of the invention, the Gadd45 proteins comprise the polypeptide sequence of SEQ ID NO: 2, SEQ ID NO: 4, SEQ ID NO: 6, SEQ ID NO: 8, SEQ ID NO: 10 and SEQ ID NO: 12, including portions thereof comprising about 5,10, 15,20, 25,30, 40,50, 60 or more contiguous amino acids of these sequences.

14. Ribozymes The use of ribozymes specific to a component in the JNK pathway including Gadd45 (3 specific ribozymes, is also a part of the invention. The following information is provided in order to complement the earlier section and to assist those of skill in the art in this endeavor.

Ribozymes are RNA-protein complexes that cleave nucleic acids in the site-specific fashion. Ribozymes have specific catalytic domains that possess endonuclease activity (Kim and Cech, 1987; Gerlack et al., 1987 ; Forster and Symons, 1987). For example, a large number of ribozymes accelerate phosphoester transfer reactions with a high degree of specificity, often cleaving only one of several phosphoesters in an oligonucleotide substrate (Cech et al., 1981 ; Michel and Westhof, 1990; Reinhold-Hurek and Shub, 1992). This specificity has been attributed to the requirement that the substrate bind via specific base- pairing interactions to the internal guide sequence ("IGS") of the ribozyme prior to chemical reaction.

15. Proteins a. Encoded Proteins Protein encoded by the respective gene can be expressed in any number of different recombinant DNA expression systems to generate large amounts of the polypeptide product, which can then be purified and used to vaccinate animals to generate antisera with which further studies may be conducted. In one embodiment of the invention, a nucleic acid that inhibits a Gadd45 gene product or the expression thereof can be inserted into an appropriate expression system. Such a nucleic acid may encode an inhibitor of Gadd45, including a dominant negative mutant protein, and may also comprise an antisense Gad45 nucleic acid.

The antisense sequence may comprise a full length coding sequence in antisense orientation and may also comprise one or more anti-sense oligonucleotides that may or may not comprise a part of the coding sequence. Potential modulators of the JNK pathway, including modulators of Gadd45) 3, may include synthetic peptides, which, for instance, could be fused to peptides derived from a Drosophila Antennapedia or HIV TAT proteins to allow free migration through biological membranes; dominant negative acting mutant proteins, including constructs encoding these proteins ; as well as natural and synthetic chemical compounds and the like.

Examples of other expression systems known to the skilled practitioner in the art include bacteria such as E. coli, yeast such as Pichia pastoris, baculovirus, and mammalian expression fragments of the gene encoding portions of polypeptide can be produced. b. Mimetics Another method for the preparation of the polypeptides according to the invention is the use of peptide mimetics. Mimetics are peptide-containing molecules which mimic elements of protein secondary structure. See, for example, Johnson et al.,"Peptide Turn Mimetics"in BIOTECHNOLOGYAND PHARMACY, Pezzuto et al., Eds. , Chapman and Hall, New York (1993). 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 antibody and antigen. A peptide mimic is expected to permit molecular interactions similar to the natural molecule.

16. Pharmaceutical Formulations and Delivery In an embodiment of the present invention, a method of treatment for a cancer by the delivery of an expression construct comprising a Gadd45 inhibitor nucleic acid is contemplated. A"Gadd45 inhibitor nucleic acid"may comprise a coding sequence of an inhibitor of Gadd45, including polypeptides, anti-sense oligonucleotides and dominant negative mutants. Similarly, other types of inhibitors, including natural or synthetic chemical and other types of agents may be administered. The pharmaceutical formulations may be used to treat any disease associated with aberrant apoptosis levels.

An effective amount of the pharmaceutical composition, generally, is defined as that amount of sufficient to detectably and repeatedly to ameliorate, reduce, minimize or limit the extent of the disease or its symptoms. More rigorous definitions may apply, including elimination, eradication or cure of the disease.

17. Methods of discovering modulators of the JNKpathway An aspect of the invention comprises methods of screening for any one or more properties of Gadd45, including the inhibition of JNK or apoptosis. The modulators may act at either the protein level, for example, by inhibiting a polypeptide involved in the JNK pathway, or may act at the nucleic acid level by modulating the expression of such a polypeptide. Alternatively, such a modulator could affect the chemical modification of a molecule in the JNK pathway, such as the phosphorylation of the molecule. The screening assays may be both for agents that modulate the JNK pathway to increase apoptosis as well as those that act to decrease apoptosis. In screening assays for polypeptide activity, the candidate substance may first be screened for basic biochemical activity--e. g., binding to a target molecule and then tested for its ability to regulate expression, at the cellular, tissue or whole animal level. The assays may be used to detect levels of Gadd45 protein or mRNA or to detect levels of protein or nucleic acids of another participant in the JNK pathway.

Exemplary procedures for such screening are set forth below. In all of the methods presented below, the agents to be tested could be either a library of small molecules (i. e. , chemical compounds), peptides (e. g. , phage display), or other types of molecules. a. Screeningfor agents that bind Gadd45ß in vitro 96 well plates are coated with the agents to be tested according to standard procedures (see Section VI, above). Unbound agent is washed away, prior to incubating the plates with recombinant Gadd45ß proteins. After, additional washings, binding of Gadd45P to the plate is assessed by detection of the bound Gadd45, for example, using anti-Gadd45ß antibodies and methodologies routinely used for immunodetection (e. g. ELISA). b. Screeningfor agents that inhibit binding of Gadd453 to its molecular target in the JNK pathway In certain embodiments, the present invention provides methods of screening and identifying an agent that modulates the JNK pathway, for example, that inhibits or upregulates Gadd45p. Compounds that inhibit Gadd45 can effectively block the inhibition of apoptosis, thus making cells more susceptible to apoptosis. This is typically achieved by obtaining the target polypeptide, such as a Gadd45 protein, and contacting the protein with candidate agents followed by assays for any change in activity.

Candidate compounds can include fragments or parts of naturally-occurring compounds or may be only found as active combinations of known compounds which are otherwise inactive. In a preferred embodiment, the candidate compounds are small molecules. Alternatively, 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.

Recombinant Gadd45) 3 protein is coated onto 96 well plates and unbound protein is removed by extensive washings. The agents to be tested are then added to the plates along with recombinant Gadd45 p-interacting protein. Alternatively, agents are added either before or after the addition of the second protein. After extensive washing, binding of Gadd45 (3 to the Gadd45 p-interacting protein is assessed, for example, by using an antibody directed against the latter polypeptide and methodologies routinely used for immunodetection (ELISA, etc. ). In some cases, it might be preferable to coat plates with recombinant Gadd45 p-interacting protein and assess interaction with Gadd45 (3 by using an anti-Gadd45p antibody. The goal is to identify agents that disrupt the association between Gadd45ß and its partner polypeptide. c. Screeningfor agents that preve7gt the ability of Gadd45ß to block apoptosis NF-KB-deficient cell lines expressing high levels of Gadd45 (3 are protected against TNFa-induced apoptosis. Cells (e. g., 3DO-IKBaM-Gadd45ß clones) are grown in 96 well plates, exposed to the agents tested, and then treated with TNFa. Apoptosis is measured using standard methodologies, for example, colorimetric MTS assays, PI staining, etc.

Controls are treated with the agents in the absence of TNFa. In additional controls, TNFa- sensitive NF-KB-null cells (e. g., 3DO-IKBaM cells), as well as TNFa-resistant NF-KB- competent cells (e. g., 3DO-Neo) are exposed to the agents to be tested in the presence or absence of TNFa. The goal is to identify agents that induce apoptosis in TNFa-treated 3DO- IoBaM-Gadd45 (3, with animal toxicity in untreated cells and no effect on TNFa-induced apoptosis in 3DO-IKBaM or 3DO-Neo cells. Agents that fit these criteria are likely to affect Gadd45 (3 function, either directly or indirectly. d. Screeningfor agents that prevent the ability of Gadd45ß to block JNK activation Cell lines, treatments, and agents are as in c. However, rather than the apoptosis, JNK activation by TNFa is assessed. A potential complication of this approach is that it might require much larger numbers of cells and reagents. Thus, this type of screening might not be most useful as a secondary screen for agents isolated, for example, with other methods. e. In vitro Assays The present embodiment of this invention contemplates the use of a method for screening and identifying an agent that modulates the JNK pathway. A quick, inexpensive and easy assay to run is a binding assay. Binding of a molecule to a target may, in and of itself, by 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. The target may be either free in solution, fixed to a support, express in or on the surface of a cell.

Examples of supports include nitrocellulose, a column or a gel. Either the target or the compound may be labeled, thereby permitting determining of binding. In another embodiment, the assay may measure the enhancement of binding of a target to a natural or artificial substrate or binding partner. 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 84/03564. In high throughput screening, large numbers of candidate inhibitory test compounds, which may be small molecules, natural substrates and ligands, or may be fragments or structural or functional mimetics thereof, are synthesized on a solid substrate, such as plastic pins or some other surface. Alternatively, purified target molecules can be coated directly onto plates or supports for u se in drug screening techniques. Also, fusion proteins containing a reactive region (preferably a terminal region) may be used to link an active region of an enzyme to a solid phase, or support. The test compounds are reacted with the target molecule, such as Gadd45 (3, and bound test compound is detected by various methods (see, e. g. , Coligan et al., Current Protocols in Immunology 1 (2): Chapter 5,1991).

Examples of small molecules that may be screened including small organic molecules, peptides and peptide-like molecules, nucleic acids, polypeptides, peptidomimetics, carbohydrates, lipids or other organic (carbon-containing) or inorganic molecules. Many pharmaceutical companies have extensive libraries of chemical and/or biological mixtures, often fungal, bacterial, or algal extracts, which can be screened with any of the assays of the invention to identify compounds that modulate the JNK pathway.

Further, in drug discovery, for example, proteins have been fused with antibody Fc portions for the purpose of high-throughput screening assays to identify potential modulators of new polypeptide targets. See, D. Bennett et al., Journal of Molecular Recognition, 8: 52-58 (1995) and K. Johanson et al., The Journal of Biological Chemistry, 270, (16): 9459-9471 (1995).

In certain embodiments of the invention, assays comprise binding a Gadd45 protein, coding sequence or promoter nucleic acid sequence to a support, exposing the Gadd45 (3 to a candidate inhibitory agent capable of binding the Gadd45 (3 nucleic acid. The binding can be assayed by any standard means in the art, such as using radioactivity, immunologic detection, fluorescence, gel electrophoresis or colorimetry means. Still further, assays may be carried out using whole cells for inhibitors of Gadd 450 through the identification of compounds capable of initiating a Gadd45ß-dependent blockade of apoptosis (see, e. g, Examples 8-11, below). f. hz vivo Assays The present invention particularly contemplates the use of various transgenic animals, such as mice. Transgenic animals may be generated with constructs that permit the use of modulators to regulate the signaling pathway that lead to apoptosis.

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 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. g In cyto assays The present invention also contemplates the screening of compounds for their ability to modulate the JNK pathway in cells. Various cell lines can be utilized for such screening assays, including cells specifically engineered for this purpose. Depending on the assay, culture may be required. The cell is examined using any of a number of different assays for screening for apoptosis or JNK activation in cells.

In particular embodiments of the present invention, screening may generally include the steps of : (a) obtaining a candidate modulator of the JNK pathway, wherein the candidate is potentially any agent capable of modulating a component of the JNK pathway, including peptides, mutant proteins, cDNAs, anti-sense oligonucleotides or constructs, synthetic or natural chemical compounds, etc.; (b) admixing the candidate agent with a cancer cell; (c) determining the ability of the candidate substance to modulate the JNK pathway, including either upregulation or downregulation of the JNK pathway and assaying the levels up or down regulation.

The levels up or down regulation will determine the extent to which apoptosis is occurring in cells and the extent to which the cells are, for example, receptive to cancer therapy. In order to detect the levels of modulation, immunodetection assays such as ELISA may be considered.

18. Methods of AssessingModulators of ApoptoticPathways Involving Gadd45ß In vitro and In vivo After suitable modulators of Gadd45ß are identified, these agents may be used in accordance with the invention to increase or decrease Gadd45ß activity either in vitro and/or in vivo.

Upon identification of the molecular target (s) of Gadd45p in the JNK pathway, agents are tested for the capability of disrupting physical interaction between Gadd45p and the Gadd45p-interacting protein (s). This can be assessed by employing methodologies commonly used in the art to detect protein-protein interactions, including immunoprecipitation, GST pull-down, yeast or mammalian two-hybrid system, and the like.

For these studies, proteins can be produced with various systems, including ill vitro transcription translation, bacterial or eukaryotic expression systems, and similar systems.

Candidate agents are also assessed for their ability to affect the Gadd45p-dependent inhibition of JNK or apoptosis. This can be tested by using either cell lines that stably express Gadd45ß (e. g. 3DC-I M-Gadd45ß) or cell lines transiently transfected with Gadd45p expression constructs, such as HeLa, 293, and others. Cells are treated with the agents and the ability of Gadd45ß to inhibit apoptosis or JNK activation induced by various triggers (e. g., TNFa) tested by using standard methodologies. In parallel, control experiments are performed using cell lines that do not express Gadd45ß.

As an extension of the previous study, animal models are used. For instance, transgenic mice expressing Gadd45ß or mice injected with cell lines (e. g., cancer cells) expressing high levels of Gadd45ß are used, either because they naturally express high levels of Gadd45ß or because they have been engineered to do so (e. g, transfected cells). Animals are then treated with the agents to be tested and apoptosis and/or JNK activation induced by various triggers is analyzed using standard methodologies. These studies will also allow an assessment of the potential toxicity of these agents.

19. Methods of Treating Cancer with Modulators of Apoptotic Pathways Involving Gadd45/3 This method provides a means for obtaining potentially any agent capable of inhibiting Gadd45ß either by way of interference with the function of Gadd45ß prótein, or with the expression of the protein in cells. Inhibitors may include: naturally-occurring or synthetic chemical compounds, particularly those isolated as described herein, anti-sense constructs or oligonucleotides, Gadd45p mutant proteins (i. e. , dominant negative mutants), mutant or wild type forms of proteins that interfere with Gadd45p expression or function, anti-Gadd45ß antibodies, cDNAs that encode any of the above mentioned proteins, ribozymes, synthetic peptides and the like. a. In vitro Methods i) Cancer cells expressing high levels of Gadd45p, such as various breast cancer cell lines, are treated with candidate agent and apoptosis is measured by conventional methods (e. g., MTS assays, PI staining, caspase activation, etc. ). The goal is to determine whether the inhibition of constitutive Gadd45ß expression or function by these agents is able to induce apoptosis in cancer cells. ii) In separate studies, concomitantly with the agents to be tested, cells are treated with TNFa or the ligands of other"death receptors" (DR) (e. g., Fas ligand binding to Fas, or TRAIL binding to both TRAIL-R1 and-R2). The goal of these studies is to assess whether the inhibition ofGadd45p renders cancer cells more susceptible to DR- induced apoptosis. iii) In other studies, cancer cells are treated with agents that inhibit Gadd450 expression or function in combination with conventional chemotherapy agents or radiation. DNA damaging agents are important candidates for these studies. However, any chemotherapeutic agent could be used. The goal is to determine whether the inhibition of Gadd45) 3 renders cancer cells more susceptible to apoptosis induced by chemotherapy or radiation. b. I7l vivo Methods The methods described above are used in animal models. The agents to be tested are used, for instance, in transgenic mice expressing Gadd450 or mice injected with tumor cells expressing high levels of Gadd45p, either because they naturally express high levels of Gadd45 (3 or because they have been engineered to do so (e. g., transfected cells). Of particular interest for these studies, are cell lines that can form tumors in mice. The effects of Gadd450 inhibitors are assessed, either alone or in conjunction with ligands of DRs (e. g.

TNFa and TRAIL), chemotherapy agents, or radiation on tumor viability. These assays also allow determination of potential toxicity of a particular means of Gadd45ß inhibition or combinatorial therapy in the animal.

20. Regulation of the gadd45p Promoter by NF-KB oB binding sites were identified in the sadd45, promoter. The presence of functional KB sites in the gadd45ß promoter indicates a direct participation of NF-KB complexes in the regulation of Gadd45ß, thereby providing an important protective mechanism by NF-KB.

21. Isolation and Analysis of the gadd45ß Promoter A BAC clone containing the murine gadd45ß gene was isolated from a 129 SB mouse genomic library (mouse ES I library; Research Genetics), digested with Xho I, and ligated into the XhoI site of pBluescript II SK- (pBS; Stratagene). A pBS plasmid harboring the 7384 bp Xho I fragment of gadd45ß (pBS-014D) was subsequently isolated and completely sequenced by automated sequencing at the University of Chicago sequencing facility. The TRANSFAC database (Heinemeyer et al., 1999) was used to identify putative transcription factor-binding DNA elements, whereas the BLAST engine (Tatusova et al., 1999) was used for the comparative analysis with the human promoter.

22. Plasmids The pMT2T, pMT2T-p50, and pMT2T-RelA expression plasmids were described previously (Franzoso et al., 1992). To generate the gadd45ß-CAT reporter constructs, portions of the gade45/3 promoter were amplified from pBS-014D by polymerase chain reaction (PCR) using the following primers: 5'- GGATAACGCGTCACCGTCCTCAAACTTACCAAACGTTTA-3' (SEQ ID NO: 6) and 5'- GGATGGATATCCGAAATTAATCCAAGAAGACAGAGATGAAC-3' (SEQ ID NO: 17) (-5921+23-gadd45ß, MluI and EcoRV sites incorporated into sense and anti-sense primers, respectively, are underlined); 5'- GGATAACGCGTTAGAGCTCTCTGGCTTTTCTAGCTGTC-3' (SEQ ID NO: 18) and 5'- GGATGGATATCCGAAATTAATCCAAGAAGACAGAGATGAAC-3' (SEQ ID NO: 19) (-2651+23-gadd45ß) ; 5'-GGATAACGCGTAAAGCGCATGCCTCCAGTGGCCACG- 3' (SEQ ID NO: 20) and 5'- GGATGGATATCCGAAATTAATCCAAGAAGACAGAGATGAAC-3' (SEQ ID NO: 21) (-1031+23-gadd45ß) ; 5'-GGATAACGCGTCACCGTCCTCAAACTTACCAAACGTTTA-3' (SEQ ID NO: 22) and 5'- GGATGGATATCCAAGAGGCAAAAAAACCTTCCCGTGCGA-3' (SEQ ID NO: 23) (- 5921+139-gadd45ß) ; 5'-GGATAACGCGTTAGAGCTCTCTGGCTTTTCTAGCTGTC-3' (SEQ ID NO: 24) and 5'- GGATGGATATCCAAGAGGCAAAAAAACCTTCCCGTGCGA-3' (SEQ ID NO: 25) (- 265/+139-gadd45ß). PCR products were digested with Mlul and EcoRV and ligated into the Mlul and SmaI sites of the promoterless pCAT3-Basic vector (Promega) to drive ligated into the MluI and SmaI sites of the promoterless pCAT2-Basic vector (Promega) to drive expression of the chloramphenicol acetyl-transferase (CAT) gene. All inserts were confirmed by sequencing. To generate-5407/+23-gadd45ß-CAT and -3465/+23-gadd45ß-CAT, pBS- 014D was digested with XhoI or EcoNI, respectively, subjected to Klenow filling, and further digested with BssHII. The resulting 5039 bp XhoI-BssHII and 3097 bp EcoNI-BssH II fragments were then independently inserted between a filled-in Mlul site and the BssHII site of -592/+23-gadd45ß-CAT. The two latter constructs contained the gadd45ß promoter fragment spanning from either-5407 or-3465 to-368 directly joined to the-38/+23 fragment. Both reporter plasmids contained intact KB-1, KB-2, and #B-3 sites (see FIG. 10).

#B-1M-gadd45ß-CAT, #B-2M-gadd45ß-CAT, and #B-3M-gadd45ß-CAT were obtained by site-directed mutagenesis of the-592+23-gadd45ß-CAT plasmid using the QuikChange kit (Stratagene) according to the manufacturer's instructions. The following base substitution were introduced: 5'-TAGGGACTCTCC-2' (SEQ ID NO: 26) to 5'- AATATTCTCTCC-3' (SEQ ID NO: 27) (#B-1M-gadd45ß-CAT ; KB sites and their mutated counterparts are underlined; mutated nucleotides are in bold); 5'-GGGGATTCCA-3' (SEQ ID NO: 28) to 5'-ATCGATTCCA-3' (SEQ ID NO: 29) (KB-2M-gadd45ß-CAT) ; and 5'- GGAAACCCCG-3' (SEQ ID NO: 30) to 5'-GGAAATATTG-3' (SEQ ID NO: 31) (KB- 3M-gadd45ß-CAT). #B-1/2-gadd45ß-CAT, containing mutated KB-1 and KB-2 sites, was derived from KB-2M-gadd45ß-CAT by site-directed mutagenesis of KB-1, as described above. With all constructs, the-592/+23 promoter fragment, including mutated KB elements, and the pCAT-3-Basic region spanning from the SmaI cloning site to the end of the CAT poly-adenylation signal were confirmed by sequencing.

A56-KB-1/2-CAT, A56-KB-3-CAT, and A56-KB-M-CAT reporter plasmids were constructed by inserting wild-type or mutated oligonucleotides derived from the mouse gadd45, promoter into A56-CAT between the BglII and XhoI sites, located immediately upstream of a minimal mouse c-fos promoter. The oligonucleotides used were: 5'- GATCTCTAGGGACTCTCCGGGGACAGCGAGGGGATTCCAGACC-3' (SEQ ID NO: 32) (KB-1/2-CAT ; KB-1 and KB-2 sites are underlined, respectively); 5'- GATCTGAATTCGCTGGAAACCCCGCAC-3' (SEQ ID NO: 33) (KB-3-CAT ; KB-3 is underlined); and 5'-GATCTGAATTCTACTTACTCTCAAGAC-3' (SEQ ID NO: 34) (KB- M-CAT).

23. Transfections, CAT assays, and Electrophoretic Mobiliy Shift Assays (EMSAs) Calcium phosphate-mediate transient transfection of NTera-2 cells and CAT assays, involving scintillation vial counting, were performed as reported previously (Franzoso et al., 1992,1993). EMSA, supershifting analysis, and antibodies directed against N-terminal peptides of human p50 and RelA were as described previously (Franzoso et al., 1992). Whole cell extracts from transfected NTera-2 cells were prepared by repeated freeze-thawing in buffer C (20 mM HEPES [pH 7.9], 0.2 MM EDTA; 0.5 mM MgCl2, 0.5 M NaCl, 25% glycerol, and a cocktail of protease inhibitors [Boehringer Mannheim] ), followed by ultracentrifugation, as previously described.

24. Generation and treatments of BJAB clones and Oropidium iodide staining assays To generate stable clones, BJAB cells were transfected with pcDNA-HA-Gadd45 (3 or empty pcDNA-HA plamids (Invitrogen), and 24 hours later, subjected to selection in G418 (Cellgro; 4 mg/ml). Resistant clones where expanded and HA-Gadd45ß expression was assessed by Western blotting using anti-HA antibodies or, to control for loading, anti-p-actin antibodies.

Clones expressing high levels of HA-Gadd45ß and control HA clones (also referred to as Neo clones) were then seeded in 12-well plates and left untreated or treated with the agonistic anti-Fas antibody APO-1 (1 J. gel ; Alexis) or recombinant TRAIL (100 ng/ml; Alexis). At the times indicated, cells were harvested, washed twice in PBS and incubated overnight at 4°C in a solution containing 0. 1% Na citrate (pH 7.4), 50 J. g/ml propidium iodide (PI; Sigma), and 0. 1% Triton X-100. Cells were then examined by flow cytometry (FCM) in both the FL-2 and FL-3 channels, and cells with DNA content lesser than 2N (sub- Gl fraction) were scored as apoptotic.

For the protective treatment with the JNK blocker SP600125 (Calbiochem), BJAB cells were left untreated or pretreated for 30 minutes with various concentrations of the blocker, as indicated, and then incubated for an additional 16 hours with the agonistic anti- Fas antibody APO-1 (1 ßg/ml). Apoptosis was scored in PI assays as described herein.

25. Treatrnents, viral tranduction, and JNK kinase assays witl2 JNK nullfibroblasts JNK null fibroblast-containing the simultaneous deletion of the jnkl and jnk2 genes- along with appropriate control fibroblasts, were obtained from Dr. Roger Davis (University of Massachusetts). For cytotoxicity experiments, knockout and wild-type cells were seeded at a density of 10,000 cells/well in 48-well plates, and 24 hours later, treated with TNFa alone (1,000 U/ml) or together with increasing concentrations of cycloheximide (CHX). Apoptosis was monitored after a 8-hour treatment by using the cell death detection ELISA kit (Boehringer-Roche) according to the manufacturer's instructions. Briefly, after lysing the cells directly in the wells, free nucleosomes in cell lysates were quantified by ELISA using a biotinylated anti-histone antibody. Experiments were carried out in triplicate.

The MIGR1 retroviral vector was obtained from Dr. Harinder Singh (University of Chicago). MIGRl-JNKK2-JNKl, expressing constitutively active JNK1, was generated by excising the HindIII-BglII fragment of JNKK2-JNK1 from pSRa-JNKK2-JNKl (obtained from Dr. Anning Lin, University of Chicago), and after filling-in this fragment by Klenow's reaction, inserting it into the filled-in XhoI site of MIGR1. High-titer retroviral preparations were obtained from Phoenix cells that had been transfected with MIGR1 or MIGR1-JNKK2- JNK1. For viral transduction, mutant fibroblasts were seeded at 100,000/well in 6-well plates and incubated overnight with 4 ml viral preparation and 1 ml complete DMEM medium in 5 llg/ml polybrene. Cells were then washed with complete medium, and 48 hours later, used for cytotoxic assays.

For JNK kinase assays, cells were left untreated or treated with TNFa (1,000 U/ml) for 10 minutes, and lysates were prepared in a buffer containing 20 mM HEPES (pH 8.0), 350 mM NaCl, 20% glycerol, 1% NP-40, 1 mM MgCl2, 0.2 mM EGTA, 1 mM DTT, 1 mM Na3V04, 50 mM NaF, and protease inhibitors. JNK was immunoprecipitated from cell lysates by using a commercial anti-JNK antibody (BD Pharmingen) and kinase assays were performed as described for FIGS. 6 and 7 using GST-c-Jun substrates.

26. Treatment of WEHI-231 cells and Electrophoretic Mobility Shift Assays WEHI-231 cells were cultured in 10% FBS-supplemented RPMI medium according to the recommendations of the American Type Culture Collection (ATCC). For electrophoretic mobility shift assays (EMSAs), cells were treated with 40 p. g/ml lypopolysaccharide (LPS; Escherichia coli serotype 0111 : B4), and harvested at the times indicated. Cell lysates were prepared by repeated freeze-thawing in buffer C (20 mM HEPES [pH 7.9], 0.2 mM EDTA, 0.5 mM DTT, 1.5 mM MgCl2, 0.42 M NaCl, 25% glycerol, and protease inhibitors) followed by ultracentrifugation. For in vitro DNA binding assays, 2 1ll cell extracts were incubated for 20 minutes with radiolabeled probes derived from each of the three KB sites found in the murine gadd45p promoter. Incubations were carried out in buffer D (20 mM HEPES [pH 7.9], 20% glycerol, 100 mM KC1, 0.2 mM EDTA, 0.5 mM DTT, 0.5 mM PMSF) containing 1 p-g/ml polydI-dC and 0.1 pLg/ml BSA, and DNA-binding complexes were resolved by polyacrilamide gel electrophoresis. For supershifts, extracts were pre- incubated for 10 minutes with 1 all of antibodies reacting with individual NF-KB subunits.

27. Treatments of BT-20 and MDA-MD-231 cells Breast cancer cell lines were cultured in complete DMEM medium supplemented with 10% FCS and seeded at 100,000/well in 12-well plates. After 24 hours, cultures were left untreated or pre-treated for 1 hour with the indicated concentrations of the SP600125 inhibitor (Calbiochem), after which the NF-KB inhibitors prostaglandin Al, CAPE, or parthenolide (Biomol) were added as shown in FIG. 20. At the indicated times, cell death was scored morphologically by light microscopy.

28. Co-immunoprecipitations with 293 cell lysates 293 cells were transfected by the calcium phosphate method with 15 pg pcDNA-HA plasmids expressing either full-length (FL) human MEKK1, MEKK3, GCK, GCKR, ASK1, MKK7/JNKK2, and JNK3, or murine MEKK4 and MKK4/JNKK1 along with 15 lug pcDNA-FLAG-Gadd45ß-expressing FL murine Gadd45 P-or empty pcDNA-FLAG vectors. pcDNA vectors (Invitrogen). 24 hours after transfection, cells were harvested, and cell lysates were prepared by resuspending cell pellets in CO-IP buffer (40 mM TRIS [pH 7.4], 150 mM NaCl, 1% NP-40,5 mM EGTA, 20 mM NaF, 1 mM Na3VO4, and protease inhibitors) and subjecting them to ultracentrifugation.

For co-immunoprecipitations (co-IP), 200 llg cell lysate were incubated with anti- FLAG (M2)-coated beads (Sigma) in CO-IP buffer for 4 hours at 4°C. After incubation, beads were washed 4 times and loaded onto SDS-polyacrylamide gels, and Western bots were performed by using anti-HA antibodies (Santa Cruz).

29. GSTfusion proteins constructions and GSTpull-dowya assays Murine Gadd45p and human JNKK2 were cloned into the EcoRi and BamHI sites of the pGEX-3X and pGEX-2T bacterial expression vectors (both from Amersham), respectively. These constructs and the pGEX-3X vector an without insert were introduced into E. coli BL21 cells in order to express GST-Gadd45p, GST-JNKK2, and GST proteins.

Following induction with 1 mM IPTG, cells were lysed by sonication in PBS and then precipitated with glutathione-sepharose beads (Sigma) in the presence of 1% Triton X-100, and washed 4 times in the same buffer.

In vitro transcription and translation reactions were carried out by using the TNT coupled reticulocyte lysate system (Promega) according to the manufacturer's instructions in the presence of [35S] methionine. To prime in vitro reactions, cDNAs were cloned into the pBluescript (pBS) SK-plasmid (Stratagene). FL murine MEKK4 was cloned into the SpeI and EcoRI sites of pBS and was transcribed with the T3 polymerase; FL human JNKK2, FL murine JNKK1, and FL human ASK1, were cloned into the XbaI-EcoRI, NotI-EcoRI, and XbaI-ApaI sites of pBS, respectively, and were transcribed by using the T7 polymerase. pBS- C-ASK1-encoding amino acids 648-1375 of human ASK1-was derived from pBS-FL- ASK1 by excision of the Earl and XbaI fragment of ASK1 and insertion of the following oligonucleotide linker: 5'-CGCCACCATGGAGATGGTGAACACCAT-3'. N-ASK1- encoding the 1-756 amino acid fragment of ASK1-was obtained by priming the in vitro transcription/translation reaction with pBS-FL-ASK1 digested with PpuMI. pBS plasmids expressing N-terminal deletions of human JNKK2 were generated by digestion of pBS-FL-JNKK2 with BamHI and appropriate restriction enzymes cleaving within the coding sequence of JNKK2 and replacement of the excised fragments with an oligonucleotide containing (5'to 3'): a BamHI site, a Kozak sequence, an initiator ATG, and a nucleotide sequence encoding between 7 and 13 residues of JNKK2. resulting pBS plasmids encoded the carboxy-terminal amino acidic portion of JNKK2 that is indicated in FIG. 28. To generate JNKK2 C-terminal deletions, pBS-FL-JNKK2 was linearized with SacII, PpuMI, NotI, XcmI, BsgI, BspEI, BspHI, or PflMI, prior to be used to prime in vitro transcription/translation reactions. The resulting polypeptide products contain the amino- terminal amino acidic sequence of JNKK2 that is indicated in FIG. 28.

To generate Gadd45ß polypeptides, in vtro reactions were primed with pBS-GFP- Gadd45p plasmids, encoding green fluorescent protein (GFP) directly fused to FL or truncated Gadd45ß. To obtain these plasmids, pBS-Gadd45ß (FL), pBS-Gadd45p (41-160), pBS-Gadd45 (3 (60-160), pBS-Gadd45ß (69-160), pBS-Gadd45ß (87-160), and pBS- Gadd45ß (113-160)-encoding the corresponding amino acid residues of murine Gadd45p were generated-by cloning appropriate gadd45, cDNA fragments into the XhoI and HindIII sites of pBS SK-. These plasmids, encoding either FL or truncated Gadd45ß, were then opened with Kpnl and XhoI, and the excised DNA fragments were replaced with the KpnI-BsrGI fragment of pEGFP-Nl (Clontech ; containing the GFP-coding sequence) directly joined to the following oligonucleotide linker: 5'- GTACAAGGGTATGGCTATGTCAATGGGAGGTAG-3'. These constructs were designated as pBS-GFP-Gadd45ß. Gadd45ß C-terminal deletions were obtained as described for the JNKK2 deletions by using pBS-GFP-Gadd45 (3 (FL) that had been digested with the NgoMI, SphI, or EcoRV restriction enzymes to direct protein synthesis in vitro. These plasmids encoded the 1-134,1-95, and 1-68 amino acid fragments of Gadd45(3, respectively.

All pBS-Gadd45ß constructs were transcribed using the T7 polymerase.

For GST pull-down experiments, 5 1ll of in vitro-translated and radio-labeled proteins were mixed with glutathione beads carrying GST, GST-JNKK2 (only with Gadd45 (3 translation products), or GST-Gadd45 (3 (only with ASK1, MEKK4, JNKK1, and JNKK2 translation products) and incubated for 1 hour at room temperature in a buffer containing 20 mM TRIS, 150 mM NaC, and 0.2% Triton X-100. The beads were then precipitated and washed 4 times with the same buffer, and the material was separated by SDS polyacrylamide gel electrophoresis. Alongside of each pair of GST and GST-JNKK2 or GST-Gadd45 (3 beads were loaded 2 µl of crude in vitro transcription/translation reaction (input).

30. Kinase assays To test the inhibitory effects of recombinant Gadd45 (3 proteins on kinase activity, HEK-293 cells were transfected by using the calcium phosphate method with 1 to 10 pg of pCDNA-FLAG-JNKK2, pCDNA-FLAG-JNKKl, pCDNA-FLAG-MKK3b or pCDNA- FLAG-ASK1, and empty pCDNA-FLAG to 30, ug total DNA. 24 hours later, cells were treated for 20 minutes with human TNFR (1,000 U/ml) or left untreated, harvested, and then lysed in a buffer containing 20 mM HEPES (pH 8.0), 350 mM NaCl, 20% glycerol, 1% NP- 40,1 mM MgCl2, 0.2 mM EGTA, ImM DTT, 1 mM Na3V04, 50 mM NaF, and protease inhibitors, and subjected to ultracentrifugation. Immunoprecipitations were performed using anti-FLAG (M2) -coated beads (Sigma) and 200 ug cell lysates. After immunoprecipitation, beads were washed twice in lysis buffer and twice more in kinase buffer (see below). To assay for kinase activity of immunoprecipitates, beads were pre-incubated for 10 minutes with increasing amounts of recombinant His6-Gadd45p, GST-Gadd45p, or control proteins in 3 0 gel kinase buffer containing 10 M ATP and l OwCi [32P] yATP, and then incubated for 1 additional hour at 30 °C with 1 Rg of the appropriate kinase substrate, as indicated. the following kinase buffers were used: 20 mM HEPES, 20 mM MgCl2, 20 mM ß-glycero- phosphate, 1mM DTT, and 50 uM Na3V04 for JNKK2 ; 20 mM HEPES, 10 mM MgCl2, 20 mM ß-glycero-phosphate, and 0.5 mM DTT for JNKK1 ; 25 mM HEPES, 25 mM MgCl2, 25 mM ß-glycero-phosphate, 0.5 mM DTT, and 50 uM Na3VO4 for MKK3 ; 20 mM TrisHCl, 20 mM MgCl2, 20 mM ß-glycero-phosphate, 1mM DTT, and 50 aM Na3V04 for ASK1.

To assay activity of endogenous kinases, immunoprecipitations were performed by using appropriate commercial antibodies (Santa Cruz) specific for each enzyme and cell lysates obtained from 3DO-IKBaM-Gadd45p and 3DO-IKBaM-Hygro clones prior and after stimulation with TNFa (1,000 U/ml), as indicated. Kinase assays were performed as described above, but without pre-incubating immunoprecipitates with recombinant Gadd45 (3 proteins.

31. Cytoprotection assays in RelA knockout cells andpEGFP- Gadd45/3 constructs Plasmids expressing N-and C-terminal truncations of murine Gadd45ß were obtained by cloning appropriate gadd45, 8 DNA fragments into the XhoI and BamHI sites of pEGFP- N1 (Clontech). These constructs expressed the indicated amino acids of Gadd45ß directly fused to the N-terminus of GFP. For cytoprotection assays, GFP-Gadd45 (3-coding plasmids or empty pEGFP were transfected into RelA-/-cells by using Superfect (Qiagen) according to the manufacturer's instructions, and 24 hours later, cultures were treated with CHX alone (0.1 pg/ml) or CHX plus TNFa (1,000 U/ml). After a 12-hour treatment, live cells adhering to tissue culture plates were counted and examined by FCM to assess GFP positivity. Percent survival values were calculated by extrapolating the total number of live GFP+ cells present in the cultures that had been treated with CHX plus TNFa relative to those treated with CHX alone.

DOCUMENTS CITED The following references, to the extent that they provide exemplary procedural or other details supplementary to those set forth herein, are specifically incorporated herein by reference.

Beg and Baltimore, 1996.

Bennett et al., Journal of Molecular Recognition, 8: 52-58 (1995).

Budihardjo et al, 1999.

Cech et al. (1981). Cell, 27 (3 Pt 2): 487-96.

Chang and Karin (2001). Nature 410 (6824): 37-40.

Coligan et al., Current Protocols in Immunology 1 (2): Chapter 5,1991.

Davis, 2000.

Franzoso et al. (1992). Nature 359 (6393): 339-342.

Gerlach et al. (1987). Nature (London) 328: 802-805.

Ghosh etal. (1998). AnnurevImmunol 16 : 225-260.

Guo et al. (1997). Biol. Chem. 273: 4027-4034.

Heinemenyer et al., 1999.

Huang et al. (1997). Immunity 6: 739-749.

Johanson et al., The Journal of Biological Chemistry, 270, (16): 9459-9471 (1995).

Johnson et al. (1993). In : Biotechnology and Pharmacy, Pezzuto et al., eds. , Chapman and Hall, New York.

Kim and Cech (1987). Proc. Nat'l Acad. Sci USA 84: 8788-8792.

Lin et al. (1995). Science 268: 286-290.

Liu et al. (1996). Cell 87 : 566-576.

Medema et al. (1997). EMBO J. 16: 274-2804.

Michel and Westhof (1990). J. Mol. Biol. 216: 585-610.

Mitchell et al. (1993). Ann N YAcac Sci 690: 153-166.

Mithcell et al. (1990). J. Clin. Oncol. 8 (5): 856-869.

Morton et al. (1992). Ann Surg. 216 (4): 463-482.

Morton et al. (1996). CA Cancer J. Clin 46 (4): 225-244.

Ravidranath and Morton (1991).Intern. Rev. Immunol. 7: 303-329.

Reinhold-Hurek and Shub (1992). Nature 357: 173-176.

Rosenberg et al. (1989). Ann Surg 210 (4): 474-548.

Rosenberg et al. (1988). N. Engl. J. Med. 319: 1676.

Rovere et al. (1999). Arthritis Rheum 42 (7): 1412-1420.

Scaffidi et al. (1999). J. Biol. Chem 274: 22532-22538.

Stegh et al. (2000). Mol. Cell Biol 20 (15): 5665-5679.

Steinman et al. (1999). Hum Immunol 60 (7): 562-567.

Tatsuova et al., 1999.

Van Antwerp et al. (1996). Science 239 (4847) : 1534-1536.

Vito et al., 1996.

Wang et al. (1999). J. Biol. Chem 274: 29599.

WO 84/03564 Yang et al. (2001). Nat. Immunol 2 : 157.

Zhang et al. (2001). Int. J. Oncol. 18: 749.