WO1997033606A1

WO1997033606A1 - PROGRAMMED CELL DEATH AND INTERLEUKIN-1$g(b)

Info

Publication number: WO1997033606A1
Application number: PCT/US1996/003468
Authority: WO
Inventors: Junying Yuan; Robert Friedlander
Original assignee: The General Hospital Corporation
Priority date: 1996-03-15
Filing date: 1996-03-15
Publication date: 1997-09-18
Also published as: JP2000506526A; EP0886525A1; CA2248551A1

Abstract

Methods for modulating programmed cell death are provided. Apoptosis, induced by a variety of stimuli, can be inhibited by blocking IL-1β binding to its type-1 receptor. Additionally, IL-1β had anti-apoptotic activity when added exogenously prior to exposure to apoptotic stimuli. ICE cleavage of pro-IL-1β is an important step in apoptosis, and mature IL-1β may function as a positive or negative mediator of cell death.

Description

Programmed Cell Death and Interleukin-lβ

Background of the Invention

Statement as to Rights to Inventions Made Under Federally-Sponsored Research and Development

Part of the work performed during development of this invention utilized

U.S. Government funds. The U.S. Government has certain rights in this invention.

Field of the Invention

The invention is in the field of molecular biology as related to the control of programmed cell death.

Description of the Background Art

Programmed Cell Death

Apoptosis, also referred to as programmed cell death or regulated cell death, is a process by which organisms eliminate unwanted cells. Such cell death occurs as a normal aspect of animal development as well as in tissue homeostasis and aging (Glucksmann, A., Biol. Rev. Cambridge Philos. Soc. 26:59-86 (1950); Ellis et al., Dev. 772:591-603 (1991); Vaux et al, Cell 76:111-119 (1994)). Programmed cell death can also act to regulate cell number, to facilitate morphogenesis, to remove harmful or otherwise abnormal cells and to eliminate cells that have already performed their function. Additionally, programmed cell death is believed to occur in response to various physiological stresses such as hypoxia or ischemia. The moφhological characteristics of apoptosis include plasma membrane blebbing, condensation of nucleoplasm and cytoplasm and degradation of chromosomal DNA at inter-nucleosomal intervals. (Wyllie, A. H., in Cell Death in Biology and Pathology, Bowen and Lockshin, eds., Chapman and Hall (1981 ), pp. 9-34).

Apoptosis is achieved through an endogenous mechanism of cellular suicide ( yllie, A. H., in Cell Death in Biology and Pathology, Bowen and Lockshin, eds., Chapman and Hall (1 81), pp. 9-34) and occurs when a cell activates its internally encoded suicide program as a result of either internal or external signals. The suicide program is executed through the activation of a carefully regulated genetic program (Wylie, A.H., et al., Int. Rev. Cyt. 68: 251 (1980); Ellis, R.E., et al., Ann. Rev. Cell Bio. 7: 663 (1991)). In many cases,gene expression appears to be required, since cell death can be prevented by inhibitors of RNA or protein synthesis (Cohen et al, J. Immunol. 52:38-42 (1984); Stanisic et al., Invest. Urol. 16:19-22 (1978); Martin et al, J. Cell Biol. 706:829-844 (1988). A genetic pathway of programmed cell death was first identified in the nematode C. elegans. In this worm, the products of ced-3 and ced-4 genes carry out the program of cellular suicide (Yuan & Horvitz, Dev. Bio. 138: 33 (1990)).

Interleukin-1§ Converting Enzyme

The mammalian homologue of the ced-3 gene product is interleukin-lβ converting enzyme (ICE), a cysteine protease responsible for the activation of interleukin-lβ (IL-lβ) (Thornberry, N.A., et al, Nature 356: 768 (1992); Yuan, J., et al, Cell 75: 641 (1993); Miura, M, et al, Cell 75: 653 (1993)). The Ice gene is a member of a family of genes. The mammalian ICE/Ced-3 family now includes at least six members: ICE, ICH-1/NEDD2, CPP32/Yama/Apopain, TX/ICEreIII/ICH-2, ICErellll and MCH2 (Yuan et al, Cell 75:641-652 (1993); Wang et al, Cell 75:739-750 (1994); Kumar et al, Genes Dev. 5:1613-1626 (1994); Femandes-Alnermi et al, J. Biol Chem. 269:30761-30764 (1994);

Tewari, M, et al, Cell 57:801-809 (1995); Nicholson, D., et al, Nature 376:31- 43 (1995); Faucheu, C, et al, J. Biol Chem. 269:3016\ -30764 (1994); Munday, N.A., et al, J. Biol Chem. 270: 15870-15876 (1995); Kamens, j., et al, J. Biol Chem. 270:15250-15256 (1995); Fernandes-Alnermi, et al, Cane. Res 55:2131- 2742 (1994)).

Interleukin-lβ converting enzyme (ICE) is a substrate-specific cysteine protease that cleaves the inactive 31 KD prointerleukin-l β at Asp^{, ,6}-Ala"\ releasing a carboxy-terminal 153 amino-acid peptide to produce the mature 17.5 kD interleukin-lβ (IL-l β) (Kostura et al, Proc. Natl. Acad. Set, USA 56:5227- 5231 (1989); Black et al, FEBSLett. 247:386-390 (1989); Cerretti et al, Science 256:97-100 (1992); Thornberry et al, Nature 356:768-774 (1992)). Since this is member of a family of proteases whose active site cysteine residue is essential for ICE-mediated apoptosis, their proteolytic activity appears critical in mediating cell death (Miura et al, J. Cell 75:653-660 (1993)). IL-lβ is also a cytokine involved in mediating a wide range of biological responses including inflammation, septic shock, wound healing, hematopoiesis and growth of certain leukemias (Dinarello, C.A., Blood 77:1627-1652 (1991); diGiovine et al, Today 77:13 (1990)).

A specific inhibitor of ICE, the crmA gene product of cowpox virus, prevents the proteolytic activation of IL-lβ (Ray et al, Cell 69:597-604 (1992)) and also inhibits host inflammatory response (Ray et al, Cell 69:597-604 ( 1992)). Cowpox virus carrying a deleted crmA gene is unable to suppress the inflammatory response of chick embryos, resulting in a reduction in the number of virus-infected cells and less damage to the host (Palumbo et al, Virology 77:262-273 (1989)). This observation indicates the importance of ICE in bringing about the inflammatory response.

It has also been shown that ICE overexpression induces apoptosis, and that mature IL-lβ is released during cell death (Miura, M., et al , Cell 75: 653

(1993); Miura, M., et al, Proc. Natl Acad. Sci. U.S.A. 92:8318-8322, (1995). The cowpox virus gene product CrmA, a member of the seφin family and an inhibitor of ICE also prevents apoptosis (Miura, M., et al, Cell 75: 653 (1993); Miura, M., et al, Proc. Natl Acad. Sci. U.S.A. (In press); Ray, C.A., et al, Cell 69: 597 (1992); Gagliardini, V., et al. Science 263: 826 (1993); Boudreau, N., et al, Science 267: 891 (1995); Enari, M., et al, Nature 375: 78 (1995); Los, M., et al, Nature 375: 81 (1995)). In addition, the ability of CrmA to inhibit apoptosis correlates with its ability to inhibit mature IL-lβ production. Recent reports indicate that tumor necrosis factor-α (TNF-α) induced apoptosis is mediated through a CrmA-inhibitable pathway suggesting involvement of the

ICE family (Tewary, M., et al, J. Biol. Chem. 270: 3255 (1995); Hsu, H., et al, Cell 81: 495 (1995); Miura, M., et al, Natl Acad. Sci. U.S.A. (In press)).

While the critical role of the ICE family in cell death is well accepted, the function of mature IL-lβ in apoptosis is controversial. IL-lβ has been shown to induce apoptosis in some systems (Onozaki et al, Immun. 735:3962-3968 (1985);

Ankarcrona e/ α/., Exp. Cell Res. 213:\12-\ 11 (1994); Fratelli, M., et al, Blood 55:3532-3637 (1995)), and to prevent it in others (Belizario & Dinarello, Cancer Res. 57:2379-2385 (1991); Strijbos & Rothwell, J. Neurosci. 75:3468-3474 (1995)). Mature IL-lβ has not only been detected in the media of TNF-α treated apoptotic fibroblasts, but also in the media of macrophages undergoing apoptosis following Shigella flexneri infection (Zychlinsky, A., et al, J. Clin. Invest. 94: 1328 (1994)). The detection of mature IL-lβ release during apoptosis provides strong evidence for ICE itself being activated in cell death, since in-vivo ICE is the major (if not the only) protease responsible for the processing of proIL-lβ as demonstrated in ICE deficient mice (Li, P., et al. , Cell 80: 401 ( 1995); (Kuida,

K., et al, Science 267: 2000 (1995)).

Tumor Necrosis Factor

Tumor necrosis factor-α (TNF-α) is a pleiotropic tumoricidal cytokine (Tracey, K.J. et al., Ann. Rev. Cell. Biol. 9:317-343 (1993)). One of the striking functions of TNF-α is to induce apoptosis of transformed cells. In the case of non-transformed cells, TNFα can also induce apoptosis in the presence of metabolic inhibitors (Tracey, K.J., et al, Ann. Rev. Cell. Biol. 9:317-343 (1993). Apoptosis induced by TNF-α is also suppressed by bcl-2. One of the most extensively studied functions of TNF-α is its cytotoxicity on a wide variety of tumor cell lines in vitro (Laster, S. M. et al, J. Immunol. 141:2629-2634 (1988)). However, the mechanism of cell death induced by TNF has been largely unknown. HeLa cells express predominantly p55 TNF receptor which is thought to be responsible for cell death signaling (Englemann, H. et al,

J. Biol. Chem. 265:14497-14504 (1990); Thoma, B. et al, J. Exp. Med. 772:1019-1023 (1990)). Additionally, HeLa cells are readily killed by TNF-α in the presence of the metabolic inhibitor cycloheximide (CHX). The cell death induced by TNF-α/CHX shows DNA fragmentation and cytolysis, which are typical features of apoptosis (White, E. et al, Mol Cell. Biol. 72:2570-2580

(1992)). Expression of adenovirus E1B 19K protein, which is functionally similar to bcl-2, inhibits apoptosis induced by TNF in HeLa cells (White, E. et al, Mol Cell Biol. 72:2570-2580 (1992)).

Summary of the Invention

It has now been found that the IL- 1 β receptor antagonist (IL- 1 Ra) inhibits apoptosis induced by trophic factor deprivation and by hypoxia. In addition, mature IL-lβ itself induces cell death through a pathway independent of CrmA - sensitive gene activity and cooperates with ICE and ICH-1_L in apoptosis. As such, the invention identifies proIL-lβ as the first substrate of any apoptosis inducing gene, whose cleavage product is a downstream mediator of the apoptotic cascade.

The invention is first directed to a method of preventing programmed cell death comprising the step of blocking mIL-β receptor binding. Preferably the mIL-β receptor binding is blocked with IL-1RA. The invention is further directed to a method for inhibiting oncogenic transformation comprising stimulating apoptosis in infected cells. Preferably, the apoptosis is stimulated with IL-lβ and/or TNF-α. The invention is further directed to a method of modulating apoptosis comprising activating the ICE pathway and mIL-l β production.

The invention is further directed to a method of modulating apoptosis comprising priming a cell prior to binding of IL-1 to its receptor. Priming the cell can include inter alia, use of trophic factor deprivation, hypoxia, Gl/S phase arrest. This may be followed by IL-lβ treatment.

The invention is further directed to a method of inhibiting hypoxia- induced cell death using an IL-1 receptor blocker. Preferably the IL-1 receptor blocker is selected from the group consisting of IL-lRa, an anti-IL-1 polyclonal neutralizing antibody and an anti-IL-1 type-1 receptor neutralizing monoclonal antibody.

The invention is further directed to a method of preventing cell death resulting from ICH-1 , comprising use of IL-lRa.

Methods of use are provided. These include, inter alia, methods to either increase or decrease cell death in treating various pathologies, including tumors of specific bodily organs of an animal, including humans. Additionally, one may use the invention to inhibit oncogenic cell transformation, to address complications concerning apoptosis which accompany hypoxia or ischemia in various organs or to screen for agents which affect apoptosis.

Brief Description of the Figures

Figures 1A-IB: Hypoxia-induced apoptosis is inhibited by CrmA, IL-l Ra, anti-IL-1 Ab, anti-IL-1 type-1 receptor antibody, and mature IL-lβ. Figure 1 A: HeLa, and HeLa/CrmA cells incubated for 16 hours under hypoxic conditions with IL-lRa, IL-1 antibody, and IL-1 type-1 receptor antibody. Results are expressed as the average of 4 independent experiments. Error bars indicate s.e.m. Figure IB: IL-lRa blocks ^,25I IL-lβ receptor binding in HeLa cells. Figures 2: IL-lRa extends neuronal survival following trophic factor deprivation. Results are expressed as the average of 3 independent experiments. Error bars indicate S.E.M.

Figures 3A- 31: Apoptosis induced by TNF-α and mature IL-l β is mediated by an IL-lRa inhibitable pathway. Figure 3 A: Percent cell death in

L929 cells treated with TNF-α alone (*) and TNF-α plus IL-lRa (•) . Figure 3B: Percent cell death in HU arrested, TNF-α (Symbols are the same as in figure 3A). Figure 3C: Percent cell death in IL- treated HeLa cells (*), HeLa /Crm (■), and HeLa cells treated with IL-lRa (•). Results are expressed as the average of 3 independent experiments. Error bars indicate S.E.M. Phase contrast and fluorescent photomicrographs of: Figure 3D: HU arrested cells, Figure 3E: treated with TNF-α or Figure 3F: IL-lβ; and stained with Hoechst dye (Figures: 3G-3I) showing condensed and fragmented nuclei.

Figures 4A - 4D: Ice Requires mature IL-lβ extracellular receptor binding for the induction of apoptosis in COS cells. Percentage of cell death

(Fig. 4A), and X-gal staining of COS cells 36 hours following transfection with Ice (Fig. 4B), Ice and pro-IL-lβ (Fig. 4C), Ice treated with mature IL-lβ (Fig. 4D). Results are expressed as the average of 3 independent experiments. Error bars indicate s.e.m.

Figures 5A - 5G: Immunofluorescence of COS cells transiently transfected with proIL-lβ. (Figs. 5A-5B), Ice (Figs. 5C-5D), or proIL-lβ and Ice (Figs 5E-5G). COS cells transfected with proIL-lβ. and immunostained anti-human polyclonal IL-1 antibody and a secondary RITC coupled antibody is alive as demonstrated by their nuclear moφhology and moφhologic appearance. Cells transfected with 7ce and immunostained with a anti-human ICE monoclonal antibody and a secondary FITC conjugated antibody appears moφhologically normal, however its nucleus is condensed suggesting initiation of apoptotic pathways, but in the absence of IL-lβ it can not be completed. Coexpression of both Ice and proIL-lβ induces typical apoptotic features (condensed nucleus and round moφhology).

Figures 6A-6B: Figure 6A: Preincubation with exogenous mature IL-lβ

(HeLa/IL-β), inhibits hypoxia-mediated apoptosis in HeLa cells. Figure 6B) ¹²⁵I IL-lβ down-regulates the IL-lβ receptor in HeLa cells.

Figure 7: The cDNA sequence (SEQ ID. No: 1 ) of Ich-1_L and the deduced amino acid sequence (SEQ ID. NO:2) of the lch-1 _L protein product.

Detailed Description

In the description that follows, a number of terms are used extensively. In order to provide a clearer and more consistent understanding of the specification the following defintitions are provided.

Definitions Italicized words such as Ice, ICE or Ich refers to the gene, while "ICE, Ich or ICH" refers to the gene product encoded by the corresponding gene.

Apoptosis should be understood to refer to the process by which organisms eliminate unwanted cells. The process is carefully regulated by a cellular program. Apoptosis may eliminate cells during normal development, aging, tissue homeostasis or following imposition of an external stress such as hypoxia or trophic factor deprivation.

Hypoxia should be understood to refer to a condition where the oxygen concentration available to a cell is decreased relative to normal levels. The most extreme hypoxia would be almost a total lack of oxygen (referred to as anoxia). ICE pathway should be understood to refer to that pathway by which interleukin converting enzyme converts the pro-ILβ to IL-β eventually resulting in programmed cell death.

Blocking IL-1-mediated signal transduction should be understood to refer to using any compound or chemical which blocks the action of IL-1 at the

IL- 1 receptor. The signal transduction may be blocked by an immunoglobulin (such as, a monoclonal or polyclonal antibody or active fragments of such antibody) including for example an anti-IL-1 polyclonal neutralizing antibody or an anti-IL type-1 receptor neutralizing monoclonal antibody. Alternatively, the signal transduction may be blocked by non-immunoglobulin compounds (such as polypeptides, organic compounds, etc.) including for example IL-1 Ra which is a naturally occurring cytokine that binds to the IL-1 receptor. Alternatively, signal transduction may be blocked by any competitive or non-competitive inhibitor of IL-lβ. Trophic factor deprivation should be understood as the removal of factors

(e.g. serum or NGF) which are required for cell survival. Absence of such factors activates the apoptotic pathway.

G/S phase arrest should be understood to be an event which occurs to a cell that causes it to fail to transit from the G, to the S phase of the cell cycle. The transition from G, to S is considered the most critical step of the cell cycle

(Chiarugi et al. Cell. Mol. Biol. Res. 40:603-6\2, 1994).

Modulating apoptosis should be understood to be any action which alters the level of cell death in either a positive or a negative direction. Ways in which to measure such changes are readily known to those of skill in the art, but may include inter alia, trypan blue exclusion, chromium release, specific changes in cell moφhology including plasma membrane blebbing, condensation of nucleoplasm and cytoplasm and degradation of chromosomal DNA at inter- nucleosomal intervals. Additional methods include metabolic assays such as the MTT (3-[4,5-D, methyl-thiazole-yi]-2,5-diphenyltetrazolium bromide; thiazolyl blue) assay or viability measurement by FACS analysis. Priming a cell should be understood to be an event or treatment which the cell undergoes such as trophic factor deprivation, hypoxia or G,/S phase arrest that is required in order for IL-lβ to activate the cell death program. In vivo this may also include any process which makes a cell "ill," e.g. a pathological condition, and thereby ready to be eliminated from the organism.

Ich-1_L and Ice should be understood to be cell death genes. Ich-1_L has the sequence (SEQ. ID.NO. 1 and SEQ. ID.NO. 2) shown in Figure 7. Ich-1_L is a fragment of the Ich-1 gene. The Ich-1 gene is homologous to other cell death genes including, inter alia, nedd2 Ich-1 contains the QACRG sequence characteristic of cell death genes. The sequence of human ICE can be found in

Thornsberr et al, Nature 356:768-774, 1992.

Naturally occurring cell death acts to regulate cell number, to facilitate moφhogenesis, to remove harmful or otherwise abnormal cells and to eliminate cells that have already performed their function. Additionally, programmed cell death is believed to occur in response to physiological stresses such as hypoxia or ischemia.

Acute and chronic disregulation of cell death is believed to lead to a number of major human diseases (Barr et al Biotech. 72:487-493, 1995). These diseases include but are not limited to malignant and pre-malignant conditions, neurological disorder, heart disease, immune system disorders, intestinal disorders, kidney disease and aging

Malignant and pre-malignant conditions may include solid tumors, B cell lymphomas, chronic lymphocytic leukemia, prostate hypertrophy, preneoplastic liver foci and resistance to chemotherapy. Neurological disorders may include stroke, Alzheimer's disease, prion-associated disorder and ataxia telangiectasia.

Heart disease may include ischemic cardiac damage and chemotherapy-induced myocardial suppression. Immune system disorder may include AIDS, type I diabetes, lupus erythematosus, Sjogren's syndrome and glomerulonephritis. Intestinal disorder may include dysentery, inflammatory bowel disease and radiation- and HIV-induced diarrhea. Kidney disease may include polycystic kidney disease and anemia/erythropoiesis. Specific references to these pathophysiological conditions as involving disregulated apoptosis can be found in Barr et al. Id - Table I.

Knowing the genes and substrates involved in the ICE pathway leads to means for intervention of cell death thereby altering apoptosis. Such knowledge can also lead to development of assays for agents which may affect the apoptotic process. Interventions may include, inter alia, agents which affect the activities of the gene products (e.g. agents which block receptors), modulation of the gene product using gene-directed approaches such as anti-sense oligodeoxynucleotide strategies, transcriptional regulation and gene therapy (Kaφ et al, Cancer Res.

54:653-665 (1994)). Therefore, apoptosis should be amenable to therapeutic intervention. In this regard, one may either stimulate or inhibit the process depending upon whether wants to increase or decrease the rate of programmed cell death. Proteolytic cleavage by the ICE family may lead to apoptosis in several ways. One possibility is that cleavage of a large number of proteins destroys the entire cellular machinery. This, however, is unlikely because most proteins appear to remain intact when cells undergo apoptosis (Lazebnik et al, Nature 377:346-347 (1994)). The second possibility is that proteolytic cleavage of one critically important substrate leads to cell death. This also is unlikely because a number of proteins, including pro-IL-lβ ribose polymerase (PARP), Ul-70 kD ribonuclear protein, and nuclear lamin are cleaved during apoptosis (Miura, et al, Proc. Natl. Acad. Sci. 92:8318-8322 (1995); Lazebnik et al, Nature 377:346-347 (1994); Casciola-Rosen ef α/., J. Biol. Chem. 269:30757-30760 (1994); Lazebnik, Y.A., et al, Proc. Natl. Acad. Sci. 92:9042-9046 (1995)). It is not clear (with the exception of pro-IL-lβ), whether the cleavage products of these proteins mediate downstream events of cell death pathways or whether they are merely the end result of apoptosis. The third possibility is that activation of the 7C£ pathway and therefore the ICE family may result in cleavage of several substrates, some being activated (mediating cell death) and others being destroyed (required for cell survival). Activation of the pathway may occur due to events such as trophic factor deprivation, hypoxia, G,/S arrest or TNF-α treatment. The results obtained in the examples of the specification, leads to favoring the last hypothesis because the data indicate that endogenously-produced mature IL-lβ is directly involved in cell death and is the first identified substrate of an apoptosis-inducing gene whose product plays a direct role in mediating the apoptotic cascade. This proposed mechanism, however, should in no way whatsoever be construed as limiting the claims of the invention to operation by such a mechanism.

Additionally, a number of signal transduction mechanisms mediate the biological effect of IL-lβ. Several of these second messengers have been implicated in apoptosis and, following ICE activation, likely mediate cell death following endogenous mature IL-lβ receptor binding. Therefore, blocking receptor binding will modulate apoptosis. IL-lβ induces ceramide production in EL4 thymoma cells (Mathias, S., et al, Science 259:519-522 (1993)). IL-lβ also induces apoptosis in pancreatic Rlm5F cells via a pathway which is dependent on its ability to induce nitric oxide production (Ankarcrona et al, Cell Res. 273:172- 177 (1994)). Both ceramide and nitric oxide are strong candidates for direct mediators of apoptosis (Ankarcrona et al, Cell Res. 213: \12-\11 (1994); Haimovitz-Friedman, A., et al, J. Exp. Med. 750:525-535 (1994)). A recent report showed that NGF deprivation of PC 12 cells, which induces apoptosis, led to a substantial activation of the JNK and p38 MAP kinases (Xia et al, Science 270: 1326-1331 ( 1995)). IL-1 β has been shown to activate the JNK-p38 signaling pathway and NGF withdrawal may induce secretion of IL-lβ which then activates the JNK-p38 pathway and cell death (Raingeaud, J., et al, J. Biol. Chem. 270:7420-7426 (1995)).

Having now generally described the invention, the same will be more readily understood through reference to the following examples which are provided by way of illustration, and are not intended to be limiting of the present invention, unless specified. Examples

The role played by secreted mature IL-lβ in apoptosis induced by trophic factor deprivation of primary dorsal root ganglia (DRG) neurons, and by hypoxia or by TNF-α in L929 and HeLa cells was investigated. The requirement for proIL-lβ in apoptosis induced by ICE and ICH-1_L was also evaluated. The results indicated that endogenously produced mature IL-lβ plays an integral role in these apoptotic models, and since ICE is the major (if not the only) enzyme to process proIL-1 β this provides further evidence for a role of ICE in apoptosis.

Example I Effects of Hypoxia

BCL-2 (B-cell lymphoma-2 gene encoded protein) and p53 have been implicated in hypoxia-mediated apoptosis (Shimizu, S., et al, Nature 374:811- 813 (1995); Jacobson & Raff, Nature 374:814-816 (1995); Graeber, T.G., et al, Nature 379:88-91 (1996)). To investigate if the ICE family is involved in hypoxia-induced apoptosis, it was tested whether CrmA could inhibit this process.

Hypoxia-induced apoptosis was studied as follows. HeLa and

HeLa/CrmA cells (Miura, M., et al, Proc. Natl. Acad. Sci. U.S.A. 92:8318-8322,

1995) were seeded in 35 mm dishes at a density of 6xl0⁴/dish in DMEM/10%

FCS and grown overnight. The medium was then changed and factors were added (IL-lRa, R & D, Minneapolis, MN), IL-1 antibody (Calbiochem, San

Diego, CA), or IL-1 type-1 receptor antibody (R & D, Minneapolis, MN). Dishes were placed in an anaerobic chamber with a BBL GasPack Plus (Becton- Dickenson, USA), which reduced the oxygen concentration to less than 100 p.p.m. within 90 minutes. After 16 hours, cells were removed from the chamber, immediately trypsinized and scored for viability by trypan blue exclusion.

Inhibition of ^,25I IL-lβ binding by IL-lRa which was added for 2 hrs at 37°C. After addition of BSA (1 mg/ml) to the medium, cells were incubated at 4°C for 15 minutes, and then ¹²⁵IL-lβ (lOOng/ml) was added at 4° C for 1 hr. For detection of ^l25IL-l β binding, cells were treated with 50mM glycine-HCl, pH 2.6 for 1 min, and quantitated by γ-counting.

Survival of HeLa cells cultured for 16 hours under hypoxic conditions was 10.1%, compared with 69.0% survival of HeLa cells which stably express

CrmA (HeLa/CrmA) (Fig. la). Thus, CrmA-inhibitable members of the ICE family play an important role in hypoxia-induced apoptosis. To address whether endogenously produced mature IL-lβ plays a role in hypoxia-induced cell death, several methods were employed to prevent IL-1 from binding to its receptor. IL- IRa (a naturally occurring cytokine which binds to the IL-1 receptor, blocking

IL-1 mediated signal transduction) (Dripps, et al, J. Biol. Chem. 266:10331- 10336 (1991); Granowitz, et al, J. Biol. Chem. 266:14147-14150 (1991)), an anti-IL-1 polyclonal neutralizing antibody, and an anti-IL-1 receptor neutralizing monoclonal antibody (the type-1 receptor mediates IL-1 signal transduction) were used. Each of these reagents inhibited hypoxia-induced cell death, suggesting that hypoxia activates an ICE-like, CrmA-inhibitable pathway, and that endogenously produced mature IL-lβ plays a role in hypoxia-induced cell death by binding to the IL-1 type-1 receptor (Fig. la). It was also evaluated and confirmed that IL- lRa indeed blocks ^l25I-IL-l β binding (Fig. lb) (Dripps, et al, J. Biol Chem. 266:10331-10336 (1991); Granowitz, et al, J. Biol. Chem. 266: 14147-14150

(1991)).

Example 2

Apoptosis amd Trophic Factor Deprivation in

Dorsal Root Ganglia

The role in apoptosis of endogenous IL-β was next investigated. Primary dorsal root ganglia (DRG) neurons undergo apoptosis in culture upon NGF withdrawal (Davies, A.M., Development 100: 1019 (1987)). It was previously shown that chicken DRG neuronal death induced by trophic factor deprivation is inhibited by CrmA, suggesting involvement of the ICE family (Gagliardi, V. et al, Science 283:826-828 (1993)). To test if endogenously produced mature IL-lβ, which is produced by neurons in culture (Freidin et al, Proc. Natl Acad. Sci. U.S.A. 59:10440-10443 (1994) plays a role in trophic factor withdrawal- mediated DRG neuronal apoptosis the human IL-1 receptor antagonist (IL-lRa) was used. IL-lRa binds to type I and II IL-1 receptors, blocking the IL-1 signal

(Dripps, D.J, et al, J. Biol. Chem. 266: 10331 (1991); Granowitz, E.V., et al, J. Biol. Chem. 266: 14147 (1991)).

Neuronal trophic factor deprivation was assayed as follows. Post-natal day 1 mouse DRG neurons were isolated, dissociated with trypsin for one hour at 37° C, and plated in a 8 camber poly-lysine/laminin (Sigma, St. Louis,

MO)coated slide. Wells were seeded at approximately 1000 neurons/well (8 wells/mouse). Neurons were cultured in Ham's nutrient F-12 supplemented with 20% FCS (Biowhittaker, Walkesvill, MD), NGF (200 ng/ml) (Sigma, St. Louis, MO), BDNF (100 ng/ml) (Preprotech, Rocky Hill, NJ), glutamine (2mM), and penicillin/streptomycin. The medium was replaced daily with either trophic factor containing medium (TF(+))=20% FCS and NGF (200 ng/ml), or trophic factor deficient medium, TF(-)= serum and NGF-free medium in the presence of saturating concentration of mouse NGF monoclonal antibody (100 ng/ml) (Boehringer Mannheim, Indianapolis, IN), and IL-1RA (100 ng/ml unless otherwise indicated in the text). Healthy neurons were counted under a phase contrast microscope 24 and 48 hours following the media change.

IL-lRa (100 ng/ml) inhibited trophic factor withdrawal-induced apoptosis by 69.2% and 37.8% in 24 and 48 hours respectively (Fig. 2). Inhibition of neuronal apoptosis by IL-lRa was dose dependent (43.5% in 24 hours at a concentration of 40 ng/ml). These results suggest that endogenously produced mature IL-l^β plays a role in neuronal apoptosis following trophic factor withdrawal. However, even though neurons have been shown to produce mature IL-l^β in culture, it can not be excluded that in this mixed cell population, mature IL-l^β is not of non-neuronal origin (Freidin, M., et al, Proc. Natl Acad. Sci. U.S.A. 89: 10440 (1992)). For this reason, several cell line systems were tested to determine if IL-lRa had similar anti-apoptotic properties.

Example 3 TNF-tx and Apoptosis

TNF-α induces apoptosis via a CrmA-inhibitable pathway (Gagliardini,

V., et al, Science 263: 826 (1993); Boudreau, N., et al, Science 267: 891 (1995); Enari, M., et al, Nature 375: 78 (1995); Los, M., et al, Nature 375: 81 (1995); Tewary, M., et al, J. Biol Chem. 270: 3255 (1995); Hsu, H., et al, Cell 81: 495 (1995)). In addition it has been demonstrated that mature IL-lβ is secreted by TNF-α treated cells undergoing apoptosis, suggesting ICE activation during this process (Miura, M„ et al, Proc. Natl. Acad. Sci. U.S.A. 92:8318-8322, 1995). The role of secreted mature IL-lβ plays a role in TNF-α induced apoptosis of L929 and HeLa cells was examined.

HeLa, HeLa/CrmA, and L929 cells were seeded (2xl0⁴) in a 24 well plate and grown overnight in DMEM with 10% FCS. After 12 hours, the cells were washed 3 times with serum free DMEM, and hydroxyurea (HU) (2.5 mM) (Sigma, St. Louis, MO) was added to the HeLa and HeLa/CrmA cells (Meikrantz, W., et al, Proc. Natl. Acad. Sci. U.S.A. 91: 3754 (1994)). After five hours, IL- lRa (40 ng/ml) was added to the appropriate wells, and one hour later either TNF-α or mature IL-lβ were added. Twenty-four hours later, IL-lRa was again added to the appropriate wells, and cell death was evaluated by try pan blue exclusion 60 hours after the initial addition of HU. Each condition was done three times in duplicate and 200 cells counted per well. For the photographs cells were grown on 2 well slides, and for nuclear moφhology determination cells were fixed in 4% paraformaldehyde and incubated with Hoechst dye #33258 (10 μg/ml) (Sigma, St. Louis, MO). IL-lRa protected L929 cells from TNF-α induced death by up to 64.9%, suggesting that secretion and receptor binding of mature IL-lβ is an integral component of TNF-α induced cell death (Fig. 3a).

In addition, hydroxyurea (HU) treated, G,/S phase arrested HeLa cells are induced to undergo programmed cell death by TNF-α (Meikrantz, W., et al,

Proc. Natl Acad. Sci. U.S.A. 91: 3754 (1994)). Under this conditions, IL-lRa also inhibited HeLa cell death by 56.0% (Fig. 3b). HeLa cells induced to die by TNF-a and cyclohexamide were also protected by IL-lRa as well as by three different neutralizing IL-1 antibodies (data not shown). HeLa/CrmA cells were protected from TNF-α induced apoptosis by 59.5%, suggesting that an ICE-like activity is involved in the cell death signaling pathway mediated by this cylokine (Fig. 3b).

Example 4 Mature IL-lβ in Apoptosis

Mature IL-lβ alone does not induce apoptosis of most healthy proliferating cells (including HeLa and L929). To examine if IL-l β would induce cell death in G,/S phase arrested cells, HU treated HeLa cells were exposed to this cytokine.

G,/S phase arrested HeLa cells treated with exogenous mature IL-1 β died in a dose dependent fashion (83.7% at 100 ng/ml), which was inhibited by the addition of IL-lRa (Fig. 3c). HU arrested, mature IL-lfi and TNF-α treated cells underwent typical apoptotic changes of cellular shrinkage, nuclear condensation, and fragmentation (Figure 3d - Figure 3i). It is interesting that HeLa/CrmA cells are not protected from mature IL-lβ as they are from TNF-α killing, suggesting that mature IL-lβ induces the apoptotic cascade distal to ICE, and in HU treated cells this cytokine causes cell death through an ICE-independent pathway (Fig. 3c). This indicates that CrmA is indeed blocking an ICE-like function and that production and secretion of mature IL-lβ is a downstream effector of the apoptotic TNF-α/ ICE cascade. HeLa cells, however, are required to be primed (in this case with HU arrest) to establish the appropriate intracellular milieu to be sensitized to mature IL-lβ induced apoptosis. HU treatment likely mimics intracellular signals which are part of the apoptotic cascade.

Example 5 Pro-IL Processing and Apoptosis

It was next directly investigated whether proIL-lβ processing was required for ICE-mediated apoptosis. For this puφose COS cells were used. These cells are unusual because they are resistant to cell death induced by Ice and Ich-1_L over-expression (Wang, L., et al , Cell 78: 739 (1994)). COS cells were plated (2x 10⁴) in 6-well plates in DMEM with 10% FCS.

After 12 hours the wells were washed with serum and antibiotic free medium, and transfected using lipofectamine with either Ice-lacZ, Ich-l,-lacZ, §actin-lacZ(l μg) or with/>ro7Z- β (0.5 μg) for 3 hours. The sequence for Ich-1_L is shown in Figure 7 and the sequences of human ICE and pro- 7Z--7β are found respectively in Thornsberry et al. Nature 356:768-774 (1992) and in J. Immunol. 737:3644-

3648, 1986. The medium was then removed and DMEM with 10% FCS added. IL-1 Ra (40 ng/ml) was then added to the appropriate wells, and after one hour IL- 1 β ( 100 ng/ml) was added. X-gal reaction was performed 36 hours following the transfection and percentage of round blue (dead) cells were scored (Miura, M., et al, Cell 75: 653 ( 993))

Transfection of Ice or Ich-1_L into Rat-1 cells induces 94.2% and 92.1% cell death respectively within 24 hours (Wang, L., et al, Cell 78: 739 (1994)). In contrast, COS cells transiently expressing Ice-lacZ, Ich- 1 -lacZ or pro- IL-lβ genes for 36 hours, died 9%, 21%, and 6.3% respectively. However, COS cells coexpressing Ice-lacZ and proIL- lβ or Ich- 1 -lacZ and proIL- Iβ, died 51.0% and

57.3%, respectively. In addition, treatment oϊlce-lacZ or Ich-l_L-lacZ transfected cells with extracellular mature IL-lβ or TNF-α efficiently induced cell death Results of treatment of Ice-lacZ transfected cells with IL-lβ are shown in Figure 4. Exogenous mature IL-lβ and TNF-α did not induce apoptosis in COS cells, indicating that ICE and ICH- 1_L have substrates in addition to proIL-lβ required for cell death and that in COS cells, following ICE activation, IL-l β signal transduction is required for the induction of apoptosis. IL-lRa significantly inhibited the death of COS cells expressing Ice-lacZ and proIL-lβ or Ich-l_L-lacZ and proIL-lβ, and of Ice-lacZ or Ich-l,-lacZ in the presence extracellular TNF-α or mature IL-lβ. This indicates a role for mature IL-l β in the induction of apoptosis following ICE family activation.

Dual immunofiuorescence staining (with anti-ICE and anti-IL-1 antibodies) of COS cells cotransfected with Ice and proIL- Iβ indicates that only cells expressing both ICE and proIL-lβ, but not either protein alone undergo apoptosis (Fig. 5). It was consistently noticed that nuclei of cells transfected with Ice are smaller than that of control cells (Fig. 5c). These cells are alive as demonstrated by their flat moφhology and adherence to the plate (Fig. 5d), suggesting that ICE initiates the apoptotic process but requires additional factors (i.e. mature IL-IB or TNF-α) for the complete execution of the cell death pathway.

The method for the dual immunofluorescent staining was as follows. COS cells (1.5xl0⁴) were plated in a poly-lysine coated two chamber slide, and after 12 hours transfected as described above. Cells were fixed after 36 hours with 4% paraformaldehyde (15 min,), blocked with 1% heat inactivated goat serum/2% BSA in PBS (2 hours) and incubated with a rabbit polyclonal IL-1 (l :300)(Calbiochem) and a hybridoma supernatant mouse monoclonal human ICE antibodies (12 hours at 4°C), chambers were washed 3x with PBS, and incubated with a goat anti-mouse FITC-labeled, a goat anti-rabbit RITC-labeled antibodies (1 :200)(Cappel), and Hoechst dye #33258(1 Oμg/ml) for 45min. .Ceils were rinsed 3x with PBS. Slides were examined with an axioplan microscope and photographed with a 40x objective. Example 6 Inhbition of hypoxia-induced Apoptosis

It was determined whether exogenous mature IL-lβ preincubation inhibits cell death in a system where ICE activation, and mature IL-l β receptor binding are important for apoptosis.

Hypoxia was produced as described in Example 1. IL-1 β (100 ng/ml) was added as the cells were placed into the hypoxia chamber (90 min. are required to reach oxygen concentrations of 100 p.p.m.). IL-1 receptor binding assay: HeLa cells (10⁶) were seeded in 10 cm dishes and grown overnight. Media was then exchanged containing 1 mg/ml of BSA and 100 ng/ml of ^l25I IL- 1 β at 4°C for 1 hr. After washing twice with cold medium, the cells were incubated with fresh warm medium at 37°C for 0, 30, 60 and 120 minutes. Cells were then treated as above with glycine and radioactivity scored.

HeLa cells preincubated with exogenous IL-lβwere markedly protected from hypoxia-induced cell death (10.1% vs. 58.7% survival) (Fig. 6a). To explain the inhibition of apoptosis by IL-lβ, it was investigated whether preincubation with exogenous IL-lβ in the system prior to exposure to apoptotic stimuli, down-regulates the IL-1 receptor. Indeed, receptor binding assays demonstrated that exogenous IL-lβ significantly down-regulated the IL-1 receptor (Fig. 6b). Down-regulation of the IL-1 receptor, in part explains the protective role of exogenous IL-lβ when added prior to the induction of apoptosis. The effect of IL-lβ receptor binding on apoptosis is dependent on whether ICE is active (enhancing cell death), or if ICE is inactive (inhibiting cell death, in part by down-regulating the IL-1 receptor). The results presented in the Examples have identified proIL-lβ as the first substrate of an apoptosis inducing gene directly involved in cell death, whose processing, secretion, and extracellular receptor binding play an integral role in the ICE apoptotic cascade. IL-lβ is believed to cause cell death by inducing ceramide and/or nitric oxide production, both of which have been shown to be involved in apoptosis (Mathias, S., et al, Science 259: 519 (1993); Haimovitz- Friedman, et al, J. Exp. Med. 180: 525 (1994); Ankarcrona, M., et al, Exp. Cell Res. 213: 172 (1994)). The results reconfirm that a variety of apoptotic stimuli (trophic factor deprivation, hypoxia, and TNF-α) activate ICE (or another IL-lβ convertase), and that cell death can be inhibited by either blocking ICE activity with CrmA or by blocking IL-lβ receptor binding with IL-lRa.

The fact that IL-1 Ra did not fully inhibit apoptosis likely occurs for the following reasons. Since occupancy of only a few IL-1 receptors (approximately 5 per cell) are necessary for a complete activation of the IL-1 biological response (Dinarello, C.A., FASEB J. 8: 1314 (1994)), IL-lRa, being a competitive inhibitor, does not likely fully displace all the IL-lβ from its receptor, and hence only protecting a portion of cells. Alternatively, following ICE activation, mature IL-lβ might act by enhancing cell death pathways, via the induction of ceramide and/or nitric oxide, and eliminating these signals would result in a delay in apoptosis. Additionally, most cells treated with exogenous mature IL-lβ do not die, suggesting that ICE-family activation, leading to the processing of additional substrates, is a prerequisite for cell death. Clearly, mature IL-lβ can not activate the ICE-family, a characteristic which it differs from TNF-α. However, under conditions where cells are properly primed, mature IL-lβ alone induces cell death, even in the absence of ICE activity as demonstrated by apoptosis induced by mature IL-lβ in G,/S phase arrested HeLa/CrmA cells. In addition, ICH-1_L appears to become activated in COS cells upon exposure to mature IL-lβ or TNF-α. Suφrisingly, ICH-1, induces cell death sensitive to IL-lRa when coexpressed with proIL-l^β, indicating that ICH-1_L either itself or through another ICE-like protease processes proIL-lβ when both are present in high concentrations.

In view of the above results which point to a definite role of ICE in apoptosis, it is interesting that ICE knock-out mice are developmental ly normal (Li, P., et al, Cell 50:401-41 1 (1995); Kuida, K., et al, Science 267:2000-2002 (1995)). To date, the only resistance to apoptosis reported in this mouse is in anti-Fas mediated thymocyte cell death (Kuida, K., et al, Science 267:2000-2002 (1995)). It is not surprising, however, that knocking out only a single member of the ever-growing number of ICE-ced-3 homologies would not produce a striking apoptotic phenotype, considering the redundancy of such an important and terminal process such as cellular suicide. IL-lβ may also be involved in-vivo in the induction of apoptosis in virally infected cells. Several viruses have been identified which express suppressers of either IL-IB and/or of TNF-a activity. Examples other than the cowpox CrmA gene is a TNF-α binding protein expressed by the pox viruses (Smith, C.A., et al , Science 248: 1019 (1990)). The vaccinia and cowpox viruses express a secreted IL-lβ binding protein (Spriggs, M.K., et al, Cell 71: 145 (1992); Alcami &

Smith, Cell 71: 153 (1992)). These viral proteins have been shown to down modulate the immune response, and their deletion diminishes virulence. In addition to immune regulatory effects, these modulators may inhibit apoptosis in infected cells by eliminating the IL-lβ and/or TNF-α signal and thereby allowing the virus to use the cellular machinery for its replication prior to cellular death.

This also suggests a possible mechanism for virally mediated oncogenic transformation through the inhibition apoptosis. Knowing such a mechanism can then lead to methods for killing the oncogenically transformed cells.

Additional relevancy of the present results is that elevated levels of IL-lβ message have been detected in rat models of cerebral ischemia (Lui, T., et al,

Stroke 24: 1746 (1993); Buttini, M., et al, Molec. Brain. Res. 23: 126 (1994)). A separate rat model demonstrated that IL-lRa reduces cerebral infarct size by 50% following ischemia (Relton & Rothwell, Brain Res. Bull. 29: 243 (1992)). In addition, brains of patients with Alzheimer's disease and Down syndrome have elevated levels of IL-IB (Sue, W., et al, Proc. Natl Acad. Sci. U.S.A. 86: 7611

(1989)). These findings suggest that mature IL-lβ is involved in mediating the neuronal cell death pathway under ischemic conditions, and in neurodegenerative diseases. This might be analogous to the notion that a cell needs to be "primed" (in neurons with trophic factor deprivation, in HeLa cells with hypoxia or G,/S phase arrest, and in L929 cells with TNF-a or with IL-lβ) in order for mature IL-1B to activate the cell death program. In-vivo, the "primed" cell idea may translate to an ill cell which is a burden to the organism, and in an example of cellular altruism, the ICE pathway is activated, leading to the production of mature IL-lβ and culminating in cellular suicide. Mature IL-lβ plays a pivotal role in cellular homeostasis. It both modulates the apoptotic cascade and activates the immune system; processes which are respectively involved in the execution and elimination of unwanted cells.

Conclusion

The interleukin-lβ converting enzyme (ICE) family plays an important role in regulating vertebrate cell death. To date, no substrate of any apoptosis inducing gene has been identified which mediates cell death. ProIL- lβ is the only known physiologic substrate of ICE.

A dual functional role for mature IL-lβ in ICE mediated apoptosis was established. It was found that when produced endogenously (i.e., following ICE activation) IL- 1 β mediates cell death, but when provided exogenously IL- 1 β can either stimulate or inhibit cell death. In addition, mature IL-lβ itself induces cell death through a pathway independent of CrmA -sensitive gene activity, and it cooperates with ICE and ICH-1_L in apoptosis.

It was further demonstrated that if IL- 1 β bound to its receptor before exposure to an apoptotic stimulus, it inhibited programmed cell death (by down- regulating the IL-1 receptor); in contrast, if IL-lβ bound after ICE was activated it enhanced cell death. IL-1 receptor antagonist (IL-lRa) inhibits apoptosis induced by trophic factor deprivation in primary neurons, and by hypoxia or TNF-α in fibroblasts. In addition, it was demonstrated that Ice required the co-expression of pro-IL-1 β to induce apoptosis in COS cells. Cell death was inhibited by blocking IL-lβ from binding to its receptor, indicating that following ICE activation, COS cells required IL-lβ signal transduction for the completion of the suicide -24-

program. The results demonstrated that endogenously produced mature IL-lβ plays in integral role in ICE mediated apoptosis. Thus, 1) IL-l β had anti- apoptotic activity when added exogenously prior to exposure to apoptotic stimuli, which was in part due to IL-1 receptor downregulation, 2) ICE cleavage of pro- IL-1 β was an important step in apoptosis and 3) mature IL-lβ may function as a positive or negative mediator of cell death.

These findings identify proIL-lβ as the first substrate of any apoptosis inducing gene, whose cleavage product is a downstream mediator of the apoptotic cascade, and provides further evidence for a role of ICE in apoptosis.

All references mentioned herein are incoφorated by reference in the disclosure. Having now fully described the invention by way of illustration and example for puφoses of clarity and understanding, it will be apparent to those of ordinary skill in the art that certain changes and modification may be made in the disclosed embodiments and such modification are intended to be within the scope of the present invention. As examples, the preferred embodiments constitute only one form of carrying out the claimed invention.

Claims

What Is Claimed Is:

1. A method of preventing programmed cell death comprising the step of blocking IL- 1 β receptor binding.

2. The method of claim 1, wherein IL-lRa is used for receptor blocking.

3. The method of claim 1 wherein cell death is blocked in a vertebrate.

4. The method of claim 2 wherein cell death is blocked in a vertebrate.

5. A method of modulating apoptosis comprising activating the ICE pathway and IL- 1 β production.

6. The method of claim 5, comprising priming a cell prior to activation of the ICE pathway.

7. The method of claim 6 wherein the priming is selected from the group consisting of trophic factor deprivation, hypoxia, G,/S phase arrest.

8. The method of claim 7 wherein cells are treated with IL-lβ following G,/S phase arrest.

9. A method of preventing cell death resulting from ICH-1_L comprising use of IL-lRa.

10. A method of altering the level of hypoxia-induced cell death comprising blocking IL-1 -mediated signal transduction.

11. The method of claim 10 wherein the agent is selected from the group consisting of IL-lRa, an anti-IL-1 polyclonal neutralizing antibody and an anti-IL-1 type-1 receptor neutralizing monoclonal antibody.

12. A method for killing oncogenically transformed cells comprising stimulating apoptosis with IL-lβ orTNF-α.

13. A method for inhibiting hypoxia-induced cell death comprising cell transfection with CrmA.

14. A method of modulating apoptosis by down regulating the IL-1 receptor.