WO1995034307A1

WO1995034307A1 - Ceramide-activated protein kinase and methods of use of effectors

Info

Publication number: WO1995034307A1
Application number: PCT/US1995/007405
Authority: WO
Inventors: Richard N. Kolesnick; David W. Golde
Original assignee: Sloan-Kettering Institute For Cancer Research
Priority date: 1994-06-14
Filing date: 1995-06-07
Publication date: 1995-12-21
Also published as: AU2771395A

Abstract

A purified membrane-bound ceramide-activated protein kinase having a mass of 97 kDa as determined by SDS PAGE has been obtained. This kinase specifically phosphorylates threonine in the sequence, Pro-Leu-Thr-Pro-containing polypeptides. A method for determining if an effector specifically inhibits or stimulates the phosphorylating activity of the kinase is presented. Methods are provided for using said effector that inhibit or stimulate the kinase to treat a subject having inflammatory disorder, to reduce the proliferation of HIV in human subject, or to treat poor stem cell growth. A method for determining if an agent can inhibit the lipopolysaccharide stimulation of the kinase is also presented. Additionally a method of treating a subject suffering from lipopolysaccharide-related disorder with said agent is presented.

Description

CERAMIDE-ACTIVATED PROTEIN KINASE AND METHODS OF USE OF EFFECT

This application is a continuation-in-part of U.S. Serial No. 08/259,700, filed June 14, 1994, which is a 10 ^• continuation-in-part of PCT International Application No. PCT/US93/10952, filed November 12, 1993, and U.S. Serial No. 07/976,378, filed November 13, 1992, the contents of which are hereby incorporated by reference.

15 This invention was made with support under Grant Nos. ROl-CA-42385 and CA-09512-06 from the National Institutes of Health, U.S. Department of Health and Human Services. Accordingly, the United States government has certain rights in the invention.

20

Background of the Invention

Throughout this application, various publications are referenced by Arabic numerals. Full citations for these

25 references may be found at the end of the specification immediately preceding the claims. The disclosure of these publications is hereby incorporated by reference into this application to describe more fully the art to which this invention pertains.

30

Recent investigations have identified a metabolic pathway involving sphingomyelin and derivatives that may be involved in signal transduction [1-8] . This pathway is initiated by the hydrolysis of sphingomyelin to ceramide

35 via the action of a sphingomyelinase^*. Ceramide may then be deacylated to sphingoid bases, putative inhibitors of protein kinase C [9-12] , or phosphorylated to the sphingolipid ceramide 1-phosphate by the action of a recently described calcium-dependent ceramide kinase [4,

40 5, 13] . The biologic role of ceramide 1-phosphate and regulation of the kinase that mediates its synthesis have not yet been determined. This pathway appears specific for ceramide derived from sphingomyelin, as ceramide derived from glycosphingolipids is not converted either to sphingoid bases [14] or to ceramide 1-phosphate [4] .

Recently, Hannun and coworkers [6-8] have provided evidence that this sphingomyelin pathway may be involved in signal transduction. Tumor necrosis factor (TNF) , y interferon, and 1,25-dihydroxyvitamin D₃, factors that induce monocytic differentiation of HL-60 promyelocytic cells, all stimulate sphingomyelin degradation to ceramide as an early event in cellular activation [6-8] . A synthetic ceramide N-acetylsphingosine could replace these agents in induction of monocytic differentiation of these cells. Furthermore, there have also been numerous reports that TNF and IL-1 stimulate a common set of events in diverse biological systems [60] .

Direct evidence for second-messenger function for ceramide has also been shown. Davis and coworkers [15-

17] originally showed that sphingosine induced epidermal growth factor receptor (EGFR) phosphorylation on Thr-669 in A-431 human epidermoid carcinoma cells by a mechanism that did not involve protein kinase C. It was demonstrated that sphingosine was rapidly converted to ceramide by these cells and that ceramide induced identical phosphorylation [18] . These studies were interpreted as evidence that, ceramide had bioeffector properties, and might mediate, in part, the action of exogenous sphingosine. However, prior to the subject invention, no kinase was identified capable of mediating the effects of ceramide as a second messenger.

The subject invention provides a purified ceramide- activated protein kinase which functions as a key element in a sphingomyelin pathway utilizing ceramide as a second messenger. The knowledge that a ceramide-activated protein kinase exists as part of the sphingomyelin pathway enables the treatment of certain disorders by selectively modifying the function of this kinase in appropriate cells. Such disorders where this approach is possible include, by way of example, HIV infection, inflammatory disorders and disorders associated with poor stem cell growth. Accordingly, the subject invention provides methods of treating subjects having such disorders with agents capable of modifying the activity of ceramide-activated protein kinase, and methods of identifying such agents.

Suiimiarv of the Invention

The subject invention provides a purified membrane-bound ceramide-activated protein kinase having an apparent molecular weight of about 97 kD as determined by SDS polyacrylamide gel electrophoresis, which protein kinase is capable of specifically phosphorylating the threonine residue in a Pro-Leu-Thr-Pro-containing polypeptide.

The subject invention also provides a method of determining whether an agent is capable of specifically inhibiting the phosphorylation activity of the ceramide- activated protein kinase of the subject invention which comprises: (a) contacting the protein kinase with a predetermined amount of a polypeptide containing the amino acid sequence Pro-Leu-Thr-Pro, and the agent, under conditions i) which would permit activity of the protein kinase to be linear with respect to time and protein kinase concentration in the absence of the agent, and ii) which would permit the specific phosphorylation by the protein kinase of a predetermined number of the threonine residues in such predetermined amount of the Pro-Leu-Thr- Pro-containing polypeptide in the absence of the agent; ^•(b) quantitatively determining the percentage of such predetermined number of threonine residues which are specifically phosphorylated in the presence of the agent, thereby determining whether the agent is capable of inhibiting the activity of the ceramide-activated protein kinase; and (c) determining whether the agent inhibits the activity of a non-ceramide-activated kinase, so as to determine whether the agent is capable of specifically inhibiting the activity of the ceramide-activated protein kinase.

The subject invention further provides a method of determining whether an agent is capable of specifically stimulating the phosphorylation activity of the ceramide- activated protein kinase of the subject invention which comprises: (a) contacting the protein kinase with a predetermined amount of a polypeptide containing the amino acid sequence Pro-Leu-Thr-Pro, and the agent, under conditions i) which would permit_/activity of the protein kinase to be linear with respect to time and protein kinase concentration in the absence of the agent, and ii) which would permit the specific phosphorylation by the protein kinase of a predetermined number of the threonine residues in such predetermined amount of the Pro-Leu-Thr- Pro-containing polypeptide in the absence of the agent; (b) quantitatively determining the percentage of such predetermined number of threonine residues which are specifically phosphorylated in the presence of the agent, thereby determining whether the agent is capable of stimulating the activity of the ceramide-activated protein kinase; and (c) determining whether the agent stimulates the activity of a non-ceramide-activated kinase, so as to determine whether the agent is capable of specifically stimulating the activity of the ceramide- activated protein kinase.

The subject invention further provides a method of treating a subject having an inflammatory disorder which comprises administering to the subject an agent capable of inhibiting the phosphorylation activity of a ceramide- activated protein kinase of T helper cells and macrophage cells of the subject in an amount effective to inhibit said phosphorylation activity, thereby reducing the inflammation associated with the disorder.

The subject invention further provides a method of treating a human subject infected with HIV so as to reduce the proliferation of HIV in the human subject which comprises administering to the human subject an agent capable of inhibiting the phosphorylation activity of a ceramide-activated protein kinase of the HIV- infected cells of the human subject in an amount effective to inhibit said activity, thereby reducing the proliferation of HIV in the human subject.

The subject invention further provides a method of treating a subject having a disorder associated with poor stem cell growth, which comprises administering to the subject an agent capable of stimulating the phosphorylation activity of a ceramide-activated protein kinase of the stem cells of the subject in an amount effective to stimulate said phosphorylation activity, thereby stimulating stem cell growth.

The subject invention further provides a method of determining whether an agent is capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide-activated protein kinase of the subject which comprises: (a) contacting the protein kinase with a predetermined amount of a polypeptide containing the amino acid sequence Pro- Leu-Thr-Pro, a predetermined amount of lipopolysaccharide, and the agent, under conditions (i) which would permit activity of the protein kinase to be linear with respect to time, lipopolysaccharide concentration and protein kinase concentration in the absence of the agent, and (ii) which would permit the specific phophorylation by the protein kinase of a predetermined number of the threonine residues in such predetermined amount of the Pro-Leu-Thr-Pro-containing polypeptide in the absence of the agent; (b) quantitatively determining the percentage of such predetermined number of threonine residues which are specifically phosphorylated in the presence of the agent, thereby determining whether the agent is capable of inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide-activated protein kinase; and (c) determining whether the agent inhibits the ability of a non-lipopolysaccharide agent to stimulate the phosphorylation activity of the ceramide- activated protein kinase, said non-lipopolysaccharide agent being known to stimulate said activity in the absence of the agent, so as to determine whether the agent is capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide-activated protein kinase.

Finally, the subject invention provides a method of treating a subject suffering from a lipopolysaccharide- related disorder which comprises administering to the subject an agent capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide-activated protein kinase of CD14-positive cells of the subject in an amount effective to specifically inhibit said phosphorylation activity, so as to thereby treat the subject.

Brief Description of the Ficrures

Figure 1

Kinetics of ³²P,- incorporation into the EGFR peptide. Peptide phosphorylation was done in a reaction mixture containing 25 μl of postnuclear supernatant (220 μg of protein) from A-431 cells, 50 μl of EGFR peptide (4 mg/ml in 25 mM Hepes, pH 7.4), and 125 μl of reaction buffer (50 mM Hepes, pH 7.4/20 mM MgCl₂) . The reaction was initiated by addition of 50 μl of [γ-³²P]ATP (150 μM final concentration) and terminated by addition of 50 μl of 0.5 M ATP in 90% (vol/vol) formic acid. Samples were spotted on phosphocellulose paper, washed with 1 M acetic acid/4 mM pyrophosphate, and ³²P incorporation was measured by liquid-scintillation counting, as described [17, 25] . A boiled protein blank was subtracted from each data value. The data (means) represent duplicate samples from one of two similar experiments.

Figure 2

Mg²* concentration-dependence of EGFR peptide phosphorylation. These studies were done as described for Fig. 1, using microsomal membrane (7.5 μg/μl) as the source of kinase activity. Reaction mixtures received various concentrations of Mg²⁺ (0.1-25 mM final concentration) , and reactions were terminated at 2 minutes. Phosphorylated peptide was isolated by HPLC and quantified by Cerenkov counting. The dimensions of velocity (V) are pmol•min^/mg of protein"¹. Data represent values derived from one of three similar experiments.

Figure 3

Identification of phosphorylated EGFR peptide. Reactions were done, and phosphorylated peptide was quantified as for Fig. 2. (Left) HPLC elution profile of samples with or without the EGFR peptide. (Right) Phosphoamino acid analysis of the phosphorylated peptide purified by HPLC. Phosphorylated amino acids (Y, tyrosinase; T, threonine; S, serine) were resolved by one-dimensional thin-layer electrophoresis and identified by ninhydrin staining of carrier phosphoamino acids and autoradiography.

Figure 4

Kinetics of ceramide-induced ³²P_i incorporation into EGFR peptide. Peptide phosphorylation was done as described in Fig. 2 in the absence (o) or presence (•) of 0.5 μM C₈- ceramide. Phosphorylated peptide was resolved by HPLC. Values (means) represent data from three experiments.

Figure 5 Concentration-dependence of ceramide-induced ³²P_i incorporation into EGFR peptide. Peptide phosphorylation reactions were done as described for Fig. 2, for 2 minutes, using various concentrations of ceramide (0.001- 3 μM) . Phosphorylated peptide was resolved by HPLC. Values (means) represent data from three experiments.

Figure 6

Concentration-dependence of sphingosine-induced ³²P_j; incorporation into EGFR peptide. Peptide phosphorylation reactions were done and analyzed as described in Fig. 5, using various concentrations of sphingosine (0.001-3 μM) . Values (means) represent data from two experiments.

Figure 7 Kinetics of TNF-of-induced ³²P₄ incorporation into EGFR peptide. HL-60 cells were resuspended in RPMI 1640 medium (1 X 10⁶ cells per ml) containing 1% FBS for 2 hours before stimulation with TNF-α (30 nM) . At the indicated times, cells were centrifuged at 500 X g for 5 minutes, and the cell pellet was homogenized in buffer, as described. Portions of a microsomal membrane fraction were used in the kinase assay, as described for Fig. 2. Values (means) represent data derived from two separate experiments.

Figure 8 Concentration-dependence of TNF-a-induced ³²P₁ inco-TPoration into EGFR peptide. These studies were done as described for Fig. 7 with various concentrations of TNF-c- for 60 minutes of stimulation. Values (means) represent data derived from two separate experiments.

Figures 9A and 9B

TNF-g- effects on sphingomyelin (A) and ceramide (B) concentrations in a cell-free system. HL-60 cells were grown in RPMI 1640 medium supplemented with 10% bovine calf serum and amino acids [4] . To measure sphingomyelin, cells were resuspended (1 x 10⁶ ml^"1) , labeled for 48 hours in medium with [³H] choline (1 μCi ml" ^x) [57] , in serum-free medium containing bovine insulin (5 μg ml^"1) and human transferrin (5 μg ml^"1) . After 3 hours, cells were resuspended (150 x 10⁶ ml"¹) in homogenization buffer (50 mM NaF, 5 mM EGTA, and 25 mM Hepes, pH 7.4), disrupted at 4°C with 150 strokes of a tight-fitting Dounce homogenizer (Fisher Scientific, Pittsburgh, Pennsylvania) , and centrifuged for 5 minutes (500g) . The nuclei-free supernate was first incubated for 5 minutes . at 4°C with 30 nM human TNF-αr (Genentech, South San Francisco, California) or diluent (50 mM Hepes, pH 7.4) . At time zero, 15 μl of supernate (112 μg per incubation) were added to a reaction mixture containing 30 μl of 25 mM Hepes, pH 7.4, 30 μl of 750 μM ATP, and 75 μl of reaction buffer (50 mM Hepes, pH 7.4 _^and 20 mM MgCl₂) at 22°C. The reaction was terminated with CHC1₃:CH₃OH:HCl (100:100:1, v/v/v) [3, 4, 13] and 150 μl of balanced salt solution (135 mM NaCl, 4.5 mM KCl, 1.5 mM CaCl₂, 0.5 mM MgCl₂, 5.6 mM glucose, and 10 mM Hepes, pH 7.2) containing 20 mM EDTA. Lipids in the organic phase extract were subjected to alkaline methanolysis to remove glycerophospholipids [4] . Sphingomyelin recovery in the nuclei-free supernate was 93% of that in intact cells.

A measure of 10⁶ cell equivalents of supernate contained

50 μg of protein. Sphingomyelin was resolved by thin- layer chromatography (TLC) with CHCl₃:CH₃OH:CH₃COOH:H₂0

(25:15:4:1.5) as solvent, identified by iodine vapor staining and quantified by ^■ liquid scintillation spectrometry [1, 8, 56] . Ceramide was quantified with the diacylglycerol kinase reaction [4, 57] . Values (mean) are derived from triplicate determinations from one experiment representative of three similar studies for sphingomyelin and four similar studies for ceramide.

Figures 1QA and 10B Effect of TNF-of on ceramide-activated protein kinase activity. HL-60 cells were incubated in serum-free medium and homogenized (as in Fig. 9) . After an initial incubation with TNF-cf, 15 μl of nuclei-free supernate

(112 μg per incubation) were added to a reaction mixture containing 30 μl of EGFR peptide (4 mg ml"¹ in 25 mM Hepes, pH 7.4), 30 μl of [γ-P³²] ATP (750 μM, 4000 dpm pmol^"1) , and 75 μl of reaction buffer [40] . The reaction was terminated by adding 30 μl of 0.5 M ATP in 90% formic acid. Phosphorylated peptide was first run on a C₁₈ Sep- Pak cartridge, then resolved by C₁₈ reverse-phase HPLC (Waters, Milford, Massachusetts) , with a linear gradient of acetonitrile. The peptide eluted at 30% acetonitrile, as determined by monitoring Cerenkov radiation in 1-ml fractions. Background activity was subtracted from each point. (A) Kinetics of TNF-α (30 nM) -stimulated EGFR peptide phosphorylation. Values (mean) represent data from four experiments. (B) Concentration dependence of EGFR peptide phosphorylation at 5 minutes of stimulation with TNF-α (0.01 to 30 nM) . Values (mean) represent data derived from duplicate points in two experiments. The SEM of the values in (A) was 18% and the mean range of values in (B) was 3%. Figure 11

Effect of phospholipases on ceramide-activated protein kinase activity. Nuclei-free supernates, prepared as in Fig. 9, were first incubated with TNF-α (3 nM) or added directly to reaction mixtures that contained various phospholipases; sphingomyelinase (SMase) (1 x 10"³ U ml"¹, S. aureus) , phospholipase A₂ (PlA₂) (3.8 x 10"² and 3.8 x 10"¹ U ml^"1, Vipera ruselli) , phospholipase C (PLC) (3.8 x 10^"2 U ml^"1, Bacillus cereus) and phospholipase D (PLD) (3.8 x 10"² U ml^"1, Streptomvces chro ocuscus) . Peptide phosphorylation was measured as in Fig. 10. Control value represents peptide phosphorylation in the absence of phospholipases or TNF-α. Values (mean ± SEM) represent data derived from duplicate samples in three experiments. *P < 0.001 compared to control.

Figures 12A and 12B

IL-lβ effects on sphingomyelin levels in EL4 cells. Time course (A) and dose response (B) . Cells were grown to growth arrest (1-1.5 x 10⁶ cells ml^"1) in DME/Ham's F12 medium (1:1, v/v) containing 10% horse serum and for 48 hours [³H] choline (lμCi ml^"1) . On the day of an experiment, cells were resuspended back into the same medium at 10 x 10⁶ cells ml^"1 and stimulated with 40 ng ml^' ¹ IL-lβ for the indicated times (A) or for 30 minutes with increasing concentrations of IL-lβ (B) . Human IL-lβ may be obtained using methods well known to those skilled in the art. Reactions were terminated with CHC1₃:CH₃OH:HCl

(100:100:1) containing 10 mM EDTA (82) . Lipids in the organic phase extract were dried under N₂ and subjected to mild alkaline hydrolysis (0.1 M methanplic KOH for 1 hour at 37°C) to remove glycerophospholipids. Sphingomyelin was resolved by thin-layer chromatography (TLC) using CHCl₃:CH₃OH:CH₃COOH:H₂0 (60:30:8:5) as solvent, identified by iodine vapor staining, and quantified by liquid scintillation spectrometry. As previously reported, the use of [³H] choline as a measure of sphingomyelin content was validated by simultaneous phospholipid phosphorus measurements [62] . Each value represents the mean ± SEM of triplicate determinations from four experiments in

(A) , and one representative of four similar studies performed in triplicate in (B) .

Figure 13

Effect of IL-lβ on ceramide levels in EL-4 cells. Cells were stimulated as in Fig. 12 with IL-lβ (40 ng ml^"1) and ceramide contained within the organic phase extract quantified enzymatically using the E. coli diacylglycerol kinase reaction [57] . Lipids were resolved by TLC using CHCl₃:CH₃OH:CH₃COOH (65/15/5) as solvent, autoradiographed and quantified by liquid scintillation spectrometry. Each value represents the mean ± SEM of triplicate determinations from 10 experiments.

Figure 14

Effect of IL-lβ on ceramide-activated protein kinase activity. Cells (30 x 10⁶ ml^"1) , handled as in Fig. 12, were stimulated with IL-lβ (10 ng ml^"1) and homogenized at 4°C with a Dounce homogenizer in buffer (25 mM HEPES, pH 7.4, 5 mM EGTA, 50 mM NaF and 10 μg/ml each of leupeptin and soybean trypsin inhibitor) . Homogenates were centrifuged at 500 x g for 5 minutes to remove nuclei and at 200,000 x g for 30 minutes to prepare microsomal membranes. Membranes were resuspended into homogenizing buffer (2.2 μg membrane protein μl"¹) . For assay of kinase activity, the reaction mixture contained 20 μl of microsomal membrane, 40 μl EGFR peptide (4 mg ml"¹ in 25 mM Hepes, pH 7.4) and 100 μl buffer (5J3 mM HEPES, pH 7.4, 20 mM MgCl₂) [40] . Phosphorylation was initiated at 22°C by addition of 40 μl [γ-³²P] ATP (100 μM final concentration) and terminated at the indicated times by addition of 40 μl of 0.5 M ATP in 90% formic acid. Phosphorylated peptide was eluted from a C₁₈ Sep pak cartridge (Millipore) , lyophilized, and resolved by C₁₈ reverse phase HPLC using a linear gradient of acetonitrile. The peptide eluted at 30% acetonitrile as determined by measuring Cerenkov radiation in 1 ml fractions. All assays were performed under conditions determined as linear for time and enzyme concentration. Enzyme activity was determined from the percent conversion of substrate to product and the specific radioactivity of [γ-³²P] ATP. Baseline kinetic analyses revealed a maximum reaction velocity of 12.5 pmol min"¹ mg" ¹ of microsomal membrane protein and Michaelis constants

(K of 70 μM ATP and 0.15 mg/ml for EGFR peptide. For most studies, 100 μM ATP was used to maintain a high ³²P specific radioactivity (4000 dpm pmol^"1) , although qualitatively similar results were obtained with 500 μM ATP. Ceramide and sphingosine (10 nM to 1 μM) enhanced kinase activity to 1.5-2.5 of control. Values (mean ± range) represent duplicate determinations from two experiments.

Figures 15A and 15B

IL-lβ effects on sphingomyelin and ceramide levels (A) and ceramide-activated protein kinase activity (B) in a cell-free system. Nuclei-free supernates, prepared as in Fig. 14, were incubated for 10 minutes at 4°C with IL-lβ (10 ng ml^"1) or diluent (DME:F12 with 10% horse serum) to allow for ligand-receptor interaction. Thereafter, supernates (300 μg incubation^"1 in 25 μl) were added to a reaction mixture (total volume 250 μl) as described in Fig. 14. For studies measuring lipid levels, incubations were stopped by extraction of lipids into an organic phase and resolved as described in Figs. 12 and 13. For studies measuring kinase activity, incubations contained EGFR peptide and [³ P]ATP, and phosphorylated peptide was quantified as described in Fig. 14. Background activity was subtracted from each point. Values (mean) represent data from two experiments for sphingomyelin performed in triplicate, three experiments for ceramide performed in triplicate, and five experiments for ceramide-activated protein kinase activity performed in duplicate.

Figures 16A and 16B Fractionation of protein kinase activity toward the EGFR peptide and PKC activity by anion exchange chromatographv.

Membrane proteins (1 mg) prepared from HL-60 cells were solubilized in 1% Triton X-100 and chromatographed on a 0.2 ml DE52 anion exchange column. The proteins bound to the column were eluted with 1 ml washes of NaCl in a stepwise concentration gradient. Fractions were assayed for ceramide-activated protein kinase activity by phosphorylation of the EGFR peptide and for PKC activity by histone phosphorylation, as described in Experimental Details. Data are representative of three experiments.

Figure 17

Phosphorylation of mvelin basic protein by fractions containing activity toward the EGFR peptide. Portions of membrane proteins (5 ml/assay) prepared from HL-60 cells by fractionation on a DE52 anion exchange column were incubated with MBP (12.5 mg/assay) in the presence of

[γ-³²P]ATP (50 mM final concentration) . Phosphorylated MBP was resolved on a 15% SDS-polyacrylamide gel and autoradiographed as described in Experimental Details. Data are representative of five experiments.

Figure 18 Isolation of the kinase activity toward the EGFR peptide by isoelectric focusing (IEF) . The flow-through fraction of the DE52 column containing kinase activity toward the EGFR peptide was subjected to IEF. The pH was measured and kinase activity of each fraction was analyzed as described in Experimental Details. Lane 1 is the flow-through fraction of the DE52 column mixed with ampholytes (2%) , Triton X-100 (1%) and glycerol (5%) . Lanes 2-20 represent fractions obtained by IEF. The molecular weight of phosphorylated MBP is 18.5 kD. Data represent one of five identical experiments.

Figure 19

Phosphorylation of a 97 kD protein in IEF fractions containing activity toward the EGFR peptide. Fraction 12 obtained by isoelectric focusing as above was incubated with [γ-³²P]ATP (50 mM final concentration) in the presence (+) or absence (-) of MBP (1 mg ml^"1) for 15 min and electrophoresed. The phosphorylated MBP was allowed to run off the gel. Data represent one of three separate experiments.

Figure 20

Renaturation of the kinase activity detected by isoelectric focusing. Intact membrane (125 mg) , cytosol (250 mg) and flow-through fraction (100 mg) of the DE52 column were electrophoresed on a 10% SDS-polyacrylamide gel. The kinase activity was renatured and autophosphorylation was performed as in Experimental Procedures by incubation of the gel for 1 h in a reaction buffer containing [γ-³P]ATP. The gel was then autoradiographed. Data represent one of four experiments.

Figure 21

Enhancement of 97 kD protein kinase activity by ceramide.

HL-60 cells (1 x 10⁶ ml"¹) were treated with C8-ceramide for 15 min at 37°C. Membrane proteins (300 mg/lane) from untreated and ceramide-treated cells_^ were subjected to SDS-polyacrylamide gel electrophoresis. Kinase activity was renatured and autophosphorylation and autoradiography of the gel were performed as in Fig. 20. Lane 1, control; lane 2, 0.03 mM C8-ceramide; lane 3, 0.3 mM C8-ceramide; lane 4, 3 mM C8-ceramide. Data are representative of six experiments. Figure 22

Enhancement of 97 kD protein kinase activity by TNFo*. HL-60 cells (1 x 10⁶ ml^"1) were pre-incubated in serum-free media for 2 h, then treated with TNFo; for 20 min at 37°C. Membrane proteins (170 mg/lane) from untreated and TNF-treated cells were subjected to SDS-polyacrylamide gel electrophoresis. Kinase activity was renatured as in Fig. 20 and autophosphorylation was performed for varying lengths of time. C, Control;- T, TNFα--stimulated cells. Data are representative of 3 experiments.

Figure 23

Enhancement of MBP phosphorylation by renatured kinase from TNF-stimulated cells. Membrane proteins (170 mg/lane) from untreated and TNF-treated cells were subjected to SDS-PAGE. and kinase activity was renatured as in Fig. 22. Autophosphorylation was allowed to proceed for 10 min. Gel slices were then excised from regions corresponding to 97 kD and 75-90 kD, crushed, and incubated for 1 h with 10 ml MBP (5 mg ml^"1) and 40 ml kinase reaction mixture in the presence of [γ-³²P]ATP (50 mM final concentration) . Reactions were terminated by addition of 10 ml Laemmli buffer, boiling for 2 min and centrifugation to precipitate gel particles. The supernatants were electrophoresed and autoradiography was performed as in Fig. 17.

Figure 24

Stimulation of ceramide-activated protein (CAP) kinase by lipid A. On the day of an experiment, HL-60 cells were resuspended (1 x 10⁶ cells ml"¹) into serum-free RPMI 1640 containing 5 μg ml"¹ insulin and transferrin. After 2h, cells were stimulated with lipid A (Escherichia Coli) or diluent (DMSO, <0.01%). Ceramide-activated protein kinase contained within microsomal membranes was detected by renaturation and autophosphorylation. Briefly, membrane proteins (200 μg per lane) were separated by SDS-PAGE (10%) , and the gel was washed with two changes of buffer (50mM Tris, pH 7.4, 5 mM 2-mercaptoethanol) containing 20% 2-propanol at room temperature for 1 h, and once in buffer without 2-propanol for 1 h. Denaturation was accomplished by incubation of the gel in two changes of 6M guanidinium HCl in wash buffer for 1 h each. Renaturation was accomplished by incubation of the gel overnight at 4°C in wash buffer containing 0.04% Tween-20. The gel was then equilibrated for 10 min at room temperature in kinase reaction mixture (25mM HEPES, pH 7.4, 10 mM MgCl₂, 0.5 mM EGTA and 5 mM NaF) and [γ-³²p] ATP (50 μM final concentration; 1000 dpm pmol^"1) was added. Autophosphorylation was terminated by removal of the reaction mixture. The gel was washed with 6 changes of buffer (5% trichloroacetic acid, 1% sodium pyrophosphate) for 2 h and subjected to autoradiography. Top panel - Time course of lipid A activation. Bottom panel - Dose response at 5 minutes of lipid A stimulation. Autoradiograms represent one of three similar studies in both panels A and B.

Figure 25

Enhancement of MBP phosphorylation by renatured kinase from lipid A-stimulated cells. Membrane proteins (200 μg) from untreated and lipid A (30 s to 10 min; 0.5 μM) - treated cells were subjected to SDS-PAGE and kinase activity was renatured as in Fig. 24. Autophosphorylation was allowed to proceed for 10 min as described in Fig. 24 and the gel was washed for 20 min with four changes of 50 mM HEPES buffer, pH 7.4. Gel slices were then excised from regions corresponding to

97kDa, crushed, and incubated for 1 h with 10 μl MBP (5 mg ml"¹) and 40 μl kinase reaction mixture in the presence of [γ^"32P]ATP (50 μM final concentration) . Reactions were terminated by addition of 10 μl Laemmli buffer, boiling for 3 min, and centrifugation to precipitate gel particles. The supernates were subjected to electrophosesis and autoradiography was performed as in Fig. 24. This autoradiogram represents one of. four similar studies.

Figure 26

Activation of ceraminde-activated protein kinase by LPS.

HL-60 cells were handled as described in Fig. 24. Cells

(13 x lO^l^"1) were incubated in serum-free RPMI 1640 for

2 h at 37°C, and LPS (Salmonella typhosa, 50 ng ml^"1) , recombinant LBP (1.7μg ml^"1) [108] or both LPS and LBP were added for the times indicated. Isolation of microsomal membranes and autophosphorylation of CAP kinase were performed as descirbed in Fig. 24. Top panel - LBP- dependent autophosphorylation for 2 min. Bottom panel -Time course of LPS/LBP autophosphorylation.

Figures 27A-27C

CAP kinase phosphorylates recombinant human Raf-1 in vitro and the level of phosphorylation is enhanced by TNF and ceramide.

Fig. 27A - CAP kinase phosphorylates recombinant human Raf-1. Recombinant human Raf-1 bound to antibody- conjugated Sepharose beads was incubated in a reaction buffer containing [g-³²P]ATP with a blank gel piece to measure autophosphorylation (Auto) or with gel slices containing CAP kinase renatured from 4 separate preparations of TNF-stimulated HL-60 cells (CAP kinase- treated) . The data represent one of five similar experiments. Fig. 27B - CAP kinase was renatured from non-stimulated (Control) and TNF-stimulated (TNF) HL-60 cells and used to phosphorylate recombinant Raf-1. The data represent one of four similar experiments. Fig. 27C - CAP kinase was renatured from non-stimulated (Control) , and C8-ceramide- and S. aureus sphingomyelinase-stimulated HL-60 cells and used to phosphorylate recombinant Raf-1. The data represent one of three similar experiments.

Figure 28A-28C

Phosphorylation of recombinant human Raf-1 by CAP kinase in vi tro enhances the kinase activity of Raf-1 towards recombinant human MEK1.

Fig. 28A - Raf-1, phosphorylated by CAP kinase, has enhanced kinase activity toward-MEK1. Fig. 28B - CAP kinase does not-phosphorylate MEK1. Fig. 28C - Reconstitution of the MAP kinase cascade in vi tro .

Figure 29A-29D

Mapping of the Site of Raf-1 phosphorylation by CAP kinase.

Fig. 29A - Reverse-phase HPLC analysis of ³²P-labeled phosphopeptides from a tryptic digest of Raf-1 that had been phosphorylated by CAP kinase. FLAG/Raf-1 was phosphorylated in vitro, subjected to tryptic digestion, and ³²P-labeled Raf-1 tryptic phosphopeptides were resolved using a C₁₈ reverse-phase HPLC column as previously described [144] . The amount of ³²P radioactivity collected in each column fraction is shown as counts per minute (CPM) . Fig. 29B - Edman degradation (left panel) and phosphoamino acid analysis (PAA, right panel) of the tryptic phosphopeptide isolated in HPLC fraction 29

(shown in Fig. 29A) . The phosphopeptide was subjected to automated Edman degradation in a spinning cup sequencer [144] and the amount of ³²P radioactivity released during each cycle of degradation is shown_.._^ The amino acid sequence of the peptide containing threonine268 and threonine269 (underlined) is shown. S, phosphoserine; T, phosphothreonine; Y, phosphotyrosine. Fig. 29C- Phosphorylation by CAP kinase of Raf-1 peptides derived from the site surrounding Thr268 and Thr269. Fig. 29D - Reconstitution of the MAP kinase cascade using wild type and mutant Raf-1.

Figure 30A-30C

TNF stimulates Raf-1 phosphorylation and its kinase activity in vivo.

Fig. 30A - Time course of TNF stimulation of Raf-1 phosphorylation in intact HL-60 cells.

Fig. 3OB - TNF stimulation of intact cells enhances the kinase activity of immunoprecipitated Raf-1 toward MEKl (Top Panel) . For these studies, HL-60 cells were stimulated by TNF for 20 min, Raf-1 was immunoprecipitated and its activity was measured by MEKl phosphorylation in vitro. For MEKl autophosphorylation, Raf-1 immunoprecipitates were omitted from the incubation. Recovery of MEKl was monitored by western blot (Bottom Panel) . Identical results were obtained with cells stimulated for 5 min with TNF. Fig. 30C - Ceramide and sphingomyelinase treatment of HL- 60 cells enhance the kinase activity of Raf-1 toward MEKl (Top Panel) . For these studies, HL-60 cells were stimulated with C8-ceramide or S. aureus sphingomyelinase for 20 min, and Raf-1 activity was measured by MEKl phosphorylation in vitro as above. Recovery of MEKl was monitored by western blot (Bottom Panel) .

Figure 31A and 3IB

Raf-1 complexes with a 97 kDa kinase.

Fig. 31A- Immune complex kinase assay using Raf-1 immunoprecipitates from control and TNF-stimulated HL-60 cells.

Fig. 31B - Western blot using anti-Raf-1 antibody. Figure 32A and 32B

Characterization of the 97 kPa protein as CAP kinase. Fig. 32A - Immune-complex kinase assay using myelin basic protein (MBP) as substrate.

Fig. 32B - Renaturation of the 97 kPa CAP kinase from the Raf-1 immunoprecipitate. Detailed Description of the Invention

Specifically, the subject invention provides a purified membrane-bound ceramide-activated protein kinase having an apparent molecular weight of about 97 kD as determined by SDS polyacrylamide gel electrophoresis, which protein kinase is capable of specifically phosphorylating the threonine residue in a Pro-Leu-Thr-Pro-containing polypeptide.

As used herein, "purified" means free of any other protein kinases. For example, the purified membrane- bound ceramide-activated protein kinase may include the protein kinase, membrane fragments, other non-kinase proteins, and a suitable buffer. Alternatively, the purified membrane-bound ceramide-activated protein kinase may include only the protein kinase bound by a membrane and a suitable buffer.

By way of example, the membrane-bound ceramide-activated protein kinase of the subject invention may be purified by (a) solubilizing the protein kinase from the membrane,

(b) separating the protein kinase from strong anions, and from protein kinase C and MAP kinases by DE52 anion exchange chromatography, (c) performing preparative SDS- gel electrophoresis based on conditions determined from a denaturation/renaturation reaction, (d) performing a high resolution isoelectric focussing using a Rotofor apparatus, (e) performing strong anion exchange chromatography by HPLC, (f) performing hydrophobic column chromatography by HPLC, and (g) performing continuous elution electrophoresis, thereby purifying the protein kinase. The purified protein kinase may then be affixed to a membrane for proper kinase function.

As used herein, "ceramide-activated" means having activity which is accelerated by the presence of ceramide. Specifically, the protein kinase of the subject invention is capable of phosphorylating certain protein substrates (e.g. human epidermal growth factor receptor) if the kinase is membrane-bound, and is in the presence of Mg⁺² and ATP. However, the rate at which the protein kinase phosphorylates its protein substrate is increased by the presence of ceramide.

The purified protein kinase of the subject invention comprises a single peptide chain having an apparent molecular weight of approximately 97 kD as determined by SDS polyacrylamide gel electrophoresis. There are numerous means of determining the molecular weight of a particular protein, some methods yielding slightly differing molecular weights for the same protein. For example, an earlier measurement of the molecular weight of the protein kinase of the subject invention was approximately 95 kD.

The 97 kD molecular weight was determined using a denaturation/renaturation procedure well known to those skilled in the art. Briefly, the method involves running the protein of interest on a denaturing gel having substrate embedded therein, washing the gel, allowing the protein to renature, assaying for protein activity in situ thereby locating the protein on the gel, and comparing the location of the protein on the gel with that of molecular weight markers, thereby determining the molecular weight of the protein.

As used herein, "specifically phosphorylating" means phosphorylating the threonine residue in a Pro-Leu-Thr- Pro-containing polypeptide without phosphorylating other amino acid residues which ordinarily serve as phosphate acceptors (e.g. serine and tyrosine) .

As used herein, "polypeptide" means a single chain of amino acid residues. Accordingly, a Pro-Leu-Thr-Pro- containing polypeptide may be the polypeptide Pro-Leu- Thr-Pro or a larger peptide containing this amino acid sequence.

The subject invention also provides a method of determining whether an agent is capable of specifically inhibiting the phosphorylation activity of the ceramide- activated protein kinase of the subject invention which comprises: (a) contacting the protein kinase with a predetermined amount of a polypeptide containing the amino acid sequence Pro-Leu-Thr-Pro, and the agent, under conditions i) which would permit activity of the protein kinase to be linear with respect to time and protein kinase concentration in the absence of the agent, and ii) which would permit the specific phosphorylation by the protein kinase of a predetermined number of the threonine residues in such predetermined amount of the Pro-Leu-Thr- Pro-containing polypeptide in the absence of the agent; (b) quantitatively determining the percentage of such predetermined number of threonine residues which are specifically phosphorylated in the presence of the agent, thereby determining whether the agent is capable of inhibiting the activity of the ceramide-activated protein kinase; and (c) determining whether the agent inhibits the activity of a non-ceramide-activated kinase, so as to determine whether the agent is capable of specifically inhibiting the activity of the ceramide-activated protein kinase.

As used herein, the term "agent" includes both protein and non-protein moieties. For example, the agent may be a ceramide analog or an antibody directed against a portion of the ceramide-activated protein kinase of the subject invention.

As used herein, "capable of specifically inhibiting" means capable of reducing the phosphorylation activity of the ceramide-activated protein kinase of the subject invention by at least two-fold, but not capable of reducing the phosphorylation activity of a non-ceramide- activated protein kinase. As used herein, a "non- ceramide-activated protein kinase" is a protein kinase whose phosphorylation activity is not altered in the presence of ceramide. An example of a non-ceramide- activated protein kinase is protein kinase C.

As used herein, "phosphorylation activity" means the rate at which a protein kinase phosphorylates its substrate. Accordingly, the phosphorylation activity of the ceramide-activated protein kinase of the subject invention means the rate at which the protein kinase phosphorylates the threonine residue in a Pro-Leu-Thr- Pro-containing polypeptide substrate.

As used herein, conditions which would permit activity of the protein kinase to be linear with respect to time and protein kinase concentration in the absence of the agent are simply conditions in which Michaelis-Menten enzyme kinetics are observed. Specifically, Michaelis-Menten enzyme kinetics are observed when the enzyme concentration is low in comparison with that of the substrate, i.e. the enzyme concentration is rate- limiting, and the enzyme reaction has not yet approached completion.

Quantitatively determining the number of threonine residues which are specifically phosphorylated may be achieved by measuring the kinase reaction rate while Michaelis-Menten kinetics are observed, and from the rate measurement, calculating the number of threonine residues which are specifically phosphorylated. Such methods of calculation are well known to those skilled in the art. An example of the method of the subject invention is provided infra. A rate-limiting amount of membrane-bound ceramide-activated protein kinase is contacted with X μg of polypeptide containing the amino acid sequence Pro- Leu-Thr-Pro, and having Y moles of threonine residues in the Pro-Leu-Thr-Pro sequence, together with an agent under conditions which would permit the phosphorylation of 0.1 x Y moles of threonine residues in the absence of the agent. In the presence of the agent, 0.05 x Y moles of threonine residues are phosphorylated. The agent is shown not to inhibit protein kinase C (a non-ceramide- activated protein kinase) activity using a histone III_S substrate assay well known to those skilled in the art. Accordingly, the agent specifically inhibits the activity of the ceramide-activated protein kinase.

In one embodiment of the subject invention, the polypeptide containing the amino acid sequence Pro-Leu- Thr-Pro is human epidermal growth factor receptor.

The subject invention further provides a method of determining whether an agent is capable of specifically stimulating the phosphorylation activity of the ceramide- activated protein kinase of the subject invention which comprises: (a) contacting the protein kinase with a predetermined amount of a polypeptide containing the amino acid sequence Pro-Le -Thr-Pro, and the agent, under conditions i) which would permit activity of the protein kinase to be linear with respect to time and protein kinase concentration in the absence of the agent, and ii) which would permit the specific phosphorylation by the protein kinase of a predetermined number of the threonine residues in such predetermined amount of the Pro-Leu-Thr- Pro-containing polypeptide in the absence of the agent; (b) quantitatively determining the percentage of such predetermined number of threonine residues which are specifically phosphorylated in the presence of the agent, thereby determining whether the agent is capable of stimulating the activity of the ceramide-activated protein kinase; and (c) determining whether the agent stimulates the activity of a non-ceramide-activated kinase, so as to determine whether the agent is capable of specifically stimulating the activity of the ceramide- activated protein kinase.

As used herein, the term "agent" includes both protein and non-protein moieties. For example, the agent may be a ceramide analog, an antibody directed against a portion of the ceramide-activated protein kinase of the subject invention, tissue necrosis factor ex or interleukin I.

As used herein, "capable of specifically stimulating" means capable of increasing the phosphorylation activity of the ceramide-activated protein kinase of the subject invention by at least two-fold, but not capable of increasing the phosphorylation activity of a non- ceramide-activated protein kinase.

An example of the method of the subject invention is provided infra. A rate-limiting amount of membrane-bound ceramide-activated protein kinase is contacted with X μg of polypeptide containing the amino acid sequence Pro- Leu-Thr-Pro, and having Y moles of threonine residues in the Pro-Leu-Thr-Pro sequence, together with an agent under conditions which would permit the phosphorylation of 0.1 x Y moles of threonine residues in the absence of the agent. In the presence of the agent, 0.2 x Y moles of threonine residues are phosphorylated. The agent is shown not to stimulate protein kinase C (a non-ceramide- activated protein kinase) activity using a histone III_S substrate assay well known to those skilled in the art. Accordingly, the agent specifically stimulates the activity of the ceramide-activated protein kinase.

In the preferred embodiment of the subject invention, the subject is a human. The inflammatory disorder may be rheumatoid arthritis, ulcerative colitis, graft versus host disease, lupus erythematosus or septic shock.

In the practice of the subject invention, the administering of the agent may be effected or performed using any of the various methods known to those of skill in the art. For example, the administration may comprise administering intravenously, intramuscularly or subcutaneously.

Further in the practice of the subject invention, the amount of .agent effective to inhibit the phosphorylation activity of ceramide-activated protein kinase of T helper cells and macrophage cells of the subject means an amount capable of inhibiting the phosphorylation activity by at least two-fold. This amount may be calculated using any of the various methods known to those of skill in the art.

The subject invention further provides a method of treating a human subject infected with HIV so as to reduce the proliferation of HIV in the human subject which comprises administering to the human subject an agent capable of inhibiting the phosphorylation activity of a ceramide-activated protein kinase of HIV-infected cells of the human subject in an amount effective to inhibit said activity, thereby reducing the proliferation of HIV in the human subject.

Further in the practice of the subject invention, the amount of agent effective to inhibit the phosphorylation activity of ceramide-activated protein kinase of the HIV- infected cells of the human subject may be calculated using any of the various methods known to those of skill in the art.

The subject invention further provides a method of determining whether a human subject is infected with HIV which comprises obtaining a sample of cells from the human subject, said cells being susceptible to infection by HIV, contacting the sample of cells with an agent capable of stimulating the phosphorylation activity of a ceramide-activated protein kinase of the cells of the sample in an amount effective to stimulate said phosphorylation activity and thereby stimulating the proliferation of any HIV present in the cells, detecting in the resulting sample the presence of any HIV, the presence of HIV indicating that the human subject is infected with HIV.

As used herein, the "sample" may be obtained from blood or any other bodily fluid known to contain HIV in HIV- infected individuals. The agent capable of stimulating the phosphorylation activity of a ceramide-activated protein kinase may be interleukin-I.

As used herein, detecting the presence of HIV may be performed according to any of the various methods known to those skilled in the art. Such methods include, but are in no way limited to, immunoassays against the HIV coat proteins.

Further in the practice of the subject invention, the amount of agent effective to stimulate the phosphorylation activity of ceramide-activated protein kinase of the cells of the sample means an amount capable of stimulating the phosphorylation activity by at least two-fold. This amount may be calculated using any of the various methods known to those of skill in the art.

The subject invention provides a method of treating a subject having a disorder associated with poor stem cell growth, which comprises administering to the subject an agent capable of stimulating the phosphorylation activity of a ceramide-activated protein kinase of the stem cells of the subject in an amount effective to stimulate said phosphorylation activity, thereby stimulating stem cell growth.

In the preferred embodiment of the subject invention, the subject is a human. Also, in the preferred embodiment of the subject invention, the disorder associated with poor stem cell growth is aplastic anemia. _^

In one embodiment of the subject invention, the agent is interleukin-I. The interleukin-I may be interleukin-Iβ.

Further in the practice of the subject invention, the amount of agent effective to stimulate the phosphorylation activity of ceramide-activated protein kinase of the stem cells of the subject means an amount capable of stimulating the phosphorylation activity by at least two-fold. This amount may be calculated using any of the various methods known to those of skill in the art.

The subject invention further provides a method of determining whether an agent is capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide-activated protein kinase of the subject invention which comprises:

(a) contacting the protein kinase with a predetermined amount of a polypeptide containing the amino acid sequence Pro-Leu-Thr-Pro, a predetermined amount of lipopolysaccharide, and the agent, under conditions (i) which would permit activity of the protein kinase to be linear with respect to time, lipopolysaccharide concentration and protein kinase concentration in the absence of the agent, and (ii) which would permit the specific phosphorylation by the protein kinase of a predetermined number of the threonine residues in such predetermined amount of the Pro-Leu-Thr-Pro-containing polypeptide in the absence of the agent; (b) quantitatively determining the percentage of such predetermined number of threonine residues which are specifically phosphorylated in the presence of the agent, thereby determining whether the agent is capable of inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide-activated protein kinase; and (c) determining_^whether the agent inhibits the ability of a non-lipopolysaccharide agent to stimulate the phosphorylation activity of the ceramide- activated protein kinase, said non-lipopolysaccharide agent being known to stimulate said activity in the absence of the agent, so as to determine whether the agent is capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide-activated protein kinase.

As used herein, "capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide-activated protein kinase" means capable of reducing the ability of lipopolysaccharide to stimulate by at least two-fold, but not capable of reducing the ability of a non- lipopolysaccharide agent to so stimulate. As used herein, a "non-lipopolysaccharide agent" may be, for example, ceramide.

Finally, the subject invention provides a method of treating a subject suffering from a lipopolysaccharide- related disorder which comprises administering to the subject an agent capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide-activated protein kinase of CD14-positive cells of the subject in an amount effective to specifically inhibit said stimulatory ability, so as to thereby treat the subject.

In the preferred embodiment of the subject invention, the subject is a human.

Lipopolysaccharide is also referred to as endotoxin, and lipopolysaccharide-related disorder is also referred to as endotoxin-relateά disorder. As used herein, an endotoxin-related disorder includes, but is not limited to endotoxin-related shock, endotoxin-related disseminated intravascular coagulation, endotoxin-related anemia, endotoxin-related thrombocytopenia, endotoxin- related adult respiratory distress syndrome, endotoxin- related renal failure, endotoxin-related liver disease or hepatitis, SIRS (systemic immune response syndrome) resulting from Gram-negative infection, Gram-negative neonatal sepsis, Gram-negative meningitis, Gram-negative pneumonia, neutropenia and/or leucopenia resulting from Gram-negative infection, hemodynamic shock and endotoxin- related pyresis. Endotoxin-related pyresis is associated with certain surgical procedures, such as trans-urethral resection of the prostate and gingival surgery. The presence of endotoxin may result from infection at any site with a Gram-negative organism, or conditions which may cause ischemia of the gastrointestinal tract, such as hemorrhage, or surgical procedures requiring extracorporeal circulation.

Further in the practice of the subject invention, the amount of agent effective to specifically inhibit the stimulatory ability of lipopolysaccharide means an amount capable of inhibiting the stimulatory ability by at least two-fold. This amount may be calculated using any of the various methods known to those of skill in the art.

As used herein, "CD14-positive cell" means a cell possessing the CD14 receptor on its surface. CD14- positive cells include, by way of example, monocytes and polymorphonuclear leukocytes.

This invention will be better understood by reference to the Experimental Details which follow, but those skilled in the art will readily appreciate that the specific experiments detailed are only illustrative of the invention as described more fully in the claims which follow thereafter.

Experimental Details

I - Characterization of a ceramide-activated protein kinase; Stimulation by tumor necrosis factor o. A. Abstract

Recent investigations have identified a signal- transduction system involving sphingomyelin and derivatives. In this paradigm, sphingomyelin hydrolysis by a sphingomyelinase generates ceramide, which may be converted to the protein kinase C inhibitor sphingosine or to ceramide 1-phosphate. Ceramide may have second- messenger function because it induces epidermal growth factor receptor phosphorylation, presumably on Thr-669 in A-431 cells. The present study describes a kinase that may mediate ceramide action. With a 19-amino acid epidermal growth factor receptor peptide containing Thr- 669, a membrane-bound activity that phosphorylated the peptide was detected in A-431 cells. Activity was linearly related to ATP (0.3-300μM) and peptide concentration (0.02-1 mg/ml), possessed a physiologic pH optimum (pH 7.0-7.4), and was Mg²⁺-dependent. Other cations - Ca²⁺, Mn²⁺, and Zn²⁺ - were ineffective. Natural and synthetic ceramide induced time-and concentration- dependent enhancement of kinase activity. Ceramide (0.5 μM) increased kinase activity 2-fold by 30 s, and activity remained elevated for at least 15 minutes. As little as 0.001 μM ceramide was effective, and 1 μM ceramide induced maximal phosphorylation. Sphingosine was similarly effective. Because tumor necrosis factor (TNF) rapidly induces sphingomyelin hydrolysis to ceramide during monocytic differentiation of HL-60 cells, its effects on kinase activity were assessed. Kinase activity was increased 1.5-fold at 5 minutes and 2-fold at 2 hr in membranes derived from TNF-stimulated cells. The effective concentration range was 3 pM-30 nM TNF. Exogenous ceramide induced a similar effect. In sum, these studies demonstrate the existence of an unusual Mg²⁺-dependent ceramide-activated protein kinase that may mediate some aspects of TNF-α function.

B. Background The present studies were done to identify the kinase that mediated the effect of ceramide on EGFR phosphorylation. The substrate used was a synthetic peptide derived from the amino acid sequence around Thr-669 of the EGFR. These studies demonstrate that A-431 human epidermoid carcinoma cells and HL-60 cells contain a ceramide/sphingosine-activated protein kinase. Further, this kinase is stimulated by TNF-a, which elevates the cellular ceramide level and induces phosphorylation of several proteins [19-24] , including the EGFR, as an early event in cellular activation. These studies provide initial evidence for a sphingolipid-activated, protein kinase-mediated signaling system.

C. Experimental Procedures 1. Materials

Ceramide (type III) , sphingosine, palmitic acid, cholera toxin, hexamethylene bisacetamide, retinoic acid, butyrate, leupeptin, and buffers were from Sigma. Fetal bovine serum (FBS) waε from GIBCO. [γ-³²P] ATP (3000 Ci/mmol; 1 Ci = 37 GBq) was from New England Nuclear. P81 phosphocellulose paper was from Whatman. Liquid scintillation solution (Liquiscint) was from National Diagnostics (Sommerville, NJ) . HPLC grade solvents were from Fisher. The EGFR peptide (amino acids 663-681, NH₂- Glu-Leu-Val-Glu-Pro-Leu-Thr-Pro-Ser-Gly-Glu-Ala-Pro-Asn- Gln-Ala-Leu-Leu-Arg-COOH) was synthesized by using an Applied Biosystems model 431A machine and purified by reverse-phase HPLC. C₈-ceramide (N-octanoylsphingosine ; C₈-cer) and TNF may be prepared according to methods well known to those of skill in the art. TNFα is also commercially available. 2. Cell Culture A-431 human epidermoid carcinoma cells were grown in monolayer culture in a 1:1 mixture of Dulbecco's modified Eagle's medium (DMEM) and Ham's F-12 medium containing 10% FBS and were harvested by trypsmization according to methods well known to those skilled in the art [18] . HL- 60 cells were grown in suspension culture in RPMI 1640 medium containing 10% FBS and supplements, according to methods well known to those skilled in the art [3] . On the day of an experiment HL-60 cells were resuspended (l x 10^s cells per ml) in RPMI 1640 medium/1% FBS for 2 hours before stimulation with lipid activators and differentiating agents.

3. Membrane Preparation

Cells (3xl0⁷/ml) were homogenized with a tight-fitting Dounce homogenizer at 4°C in buffer (25 mM Hepes, pH 7.4/5 mM EGTA/50mM NaF/leupeptin at 10 μg/ml) according to methods well known to those skilled in the art [17] . The homogenate was centrifuged at 500 x g for 5 minutes, and the postnuclear supernatant was centrifuged at 200,000 x g for 30 minutes. The microsomal membrane pellet was resuspended (7.5 μg of membrane protein per μl for A-431 cells and 1.5 μg/μl for HL-60 cells) in homogenizing buffer. Membranes were prepared fresh daily.

4. Assay of Kinase Activity For most experiments, the reaction mixture contained 25 μl of microsomal membrane or postnuclear supernatant, 50 μl of EGFR peptide (4 mg/ml in 25 mM_^Hepes, pH 7.4) and 125 μl of buffer (50 mM Hepes, pH 7.4/20 mM MgCl₂) [17] . Phosphorylation was initiated at 22°C by addition of 50 μl of [γ-³²P] ATP (150 μM final concentration; 4000 dpm/pmol) . For studies with lipid activators, ceramide and other lipids were dried under N₂ and resuspended in the kinase assay buffer by bath sonication for 2 minutes at 37°C. The reaction was terminated at the indicated times by addition of 50 μl of 0.5 M ATP in 90% formic acid. Unless otherwise indicated, all assays were done under conditions determined as linear for time and enzyme concentration. Enzyme activity was determined from the transfer of ³²P from the γ position of ATP to EGFR peptide and the specific radioactivity of [γ-³²P] ATP.

Phosphorylated peptide was quantified by two separate methods. For initial studies, samples were spotted on phosphocellulose paper, washed in 1 M acetic acid/4 mM pyrophosphate and subjected to liquid scintillation counting, according to methods well known to those skilled in the art [25] . Values obtained from a boiled blank or a sample lacking peptide were subtracted from each determination. Alternatively, HPLC was done according to methods well known to those skilled in the art [17] . For these studies, samples were first applied to a C₁₈ Sep-Pak cartridge and eluted with 99.9% acetonitrile/0.1% trifluoroacetic acid. The eluates were lyophilized, resuspended in 6 M guanidine hydrochloride/200 mM Tris, pH 8.5 and applied to a C₁₈ reverse-phase column (Dynamax, 4.6 mm i.d., Rainin, Woburn, MA) . The peptide was eluted with a linear gradient (1% per minute) of acetonitrile at a flow rate of 1 ml/minute and was detected by measuring the Cerenkov radiation associated with 1-ml fractions.

5. Phosohoamino Acid Analysis To determine which amino acid was phosphorylated, phosphoamino acid analysis of the peptide was done. The phosphopeptide peak obtained by HPLC was subjected to partial acid hydrolysis (1 hr at 110°C in 6 M HCl) . The hydrolysates were dried, resuspended in 250 μl of water, and applied to a Dowex AG1-X8 column (Bio-Rad) . Amino acids were eluted with 0.5 M HCl, dried, and analyzed by thin-layer electrophoresis, according to methods well known to those skilled in the art [26] . Individual phosphoamino acids were identified by ninhydrin staining of carrier phosphoamino acids and by autoradiography.

6. Other Procedures

Protein was measured by the method of Bradford [27] .

7. Statistics Statistical analysis was performed by t test and linear regression analysis by the method of least squares.

D. Results

Davis and coworkers [15, 17] showed that addition of sphingosine to A-431 cells enhanced phosphorylation of the EGFR on Thr-669. Subsequently, the subject experiments show that sphingosine was rapidly converted to ceramide in these cells and that exogenous ceramide induced identical effects [18] . To investigate the kinase that mediated ceramide action, a synthetic peptide corresponding to the sequence around Thr-669 was used as substrate.

Initial studies were done to determine the kinetics of phosphorylation of the EGFR peptide. The conditions for this assay were adapted from Davis and coworkers [17, 25] . Briefly, postnuclear supernatant was used as a source of enzyme activity, and samples were spotted on phosphocellulose paper to measure phosphorylated peptide. The kinetics of phosphorylation appeared biphasic. Initial rapid incorporation of ³²P into peptide for 10 minutes was followed by incorporation at a slightly reduced rate for as long as 30 minutes (Fig. 1) .

Subsequent studies were done to optimize the assay. Kinase activity was found by Lineweaver-Burke analysis to be linearly related (r=0.98) to substrate concentration for ATP (0.3-300μM) and EGFR peptide (0.02-1 mg/ml) at 5 minutes of stimulation. Apparent K-_ values for ATP of 15 μM and for EGFR peptide of 0.25 mg/ml were derived. Apparent V.,^ values ranged from 100-200 pmol»min^"1/mg of protein^"1. All subsequent studies were done with 150 μM ATP and EGFR peptide at 4 mg/ml. Under these conditions, substrate concentration was not rate limiting.

An additional set of studies assessed the pH optimum for the kinase activity. There was no measurable activity at pH values <5. Thereafter, peptide phosphorylation increased to a maximum at pH 7-7.4 and rapidly dropped to undetectable levels at pH 8. Hence, this kinase appears active within the physiologic pH range.

The divalent cation requirement for kinase activity was also investigated. In the presence of EGTA (1 mM) alone, peptide phosphorylation did not occur. Mg²⁺ induced dose- dependent peptide phosphorylation (Fig. 2) . As little as 0.1 mM Mg²* increased kinase activity to 4 pmol^βmin^"1 (mg of protein"¹) , and maximal activity occurred with 10 mM

Mg²⁺; the ED_S0 was «3.5 mM. An increase in Mg²⁺ to 25 mM did not further increase activity. Mn²⁺ (1-10 mM) , Zn²⁺

(1-10 mM) , and Ca²⁺ (0.001-lOmM) did not support kinase activity toward the EGFR peptide. These studies indicate that this kinase activity is Mg²*-dependent.

Cell-fractionation studies were done to compare levels of kinase activity in the postnuclear supernatant, cytosol, and membrane. Activity detected in the postnuclear supernatant was equally divided between membrane and cytosolic fractions. Only membrane activity was enhanced by ceramide (see below) .

A more specific method for detection of phosphorylated peptide used reverse-phase HPLC. Peptide was eluted with a linear gradient of acetonitrile, and fractions were monitored for Cerenkov radiation. A peak of Cerenkov radiation was eluted at 30% acetonitrile in samples containing peptide but was absent when the peptide was omitted from the reaction mixture (Fig. 3, Left) . Phosphoamino acid analysis of the eluate demonstrated the presence of [³²P]phosphothreonine (Fig. 3, Right) .

These studies indicate that of the two potential phosphorylation sites contained within the EGFR peptide, corresponding to Thr-669 and Ser-671, only threonine-669 served as substrate.

To determine whether ceramide and sphingosine enhance EGFR peptide phosphorylation, these lipids were added to a reaction mixture containing peptide and membrane. Ceramide (0.5 μM) stimulation of EGFR peptide phosphorylation was evident by -30 s (Fig. 4) and demonstratable for at least 15 minutes. Ceramide (0.001- 3 μM) enhanced peptide phosphorylation in a concentration-dependent manner at 2 minutes of stimulation (Fig. 5) . As little as 1 nM ceramide was effective, and a maximal effect to 2.1-fold of control occurred with 1 μM ceramide; the ED₅₀ was «=30 nM. Synthetic C₈-cer and natural ceramide (Sigma type III) were similarly effective. As with basal phosphorylation, ceramide-enhanced phosphorylation occurred exclusively on the threonine residue of the EGFR peptide.

Figure 6 shows that sphingosine also stimulated EGFR peptide phosphorylation to a level 1.6 fold of control at 2 minutes of stimulation. The concentration-dependence of this stimulatory effect was similar to that of ceramide. In contrast, palmitic acid, the predominant fatty acid in natural ceramide, failed to increase EGFR peptide phosphorylation.

TNF-Q! has been shown to increase cellular levels of ceramide within minutes of activation of HL-60 cells, and a synthetic ceramide replaced the requirement of TNF-α: in monocytic differentiation of these cells [6] . Hence, studies were done to determine whether TNF-α treatment of HL-60 cells activated a kinase similar to that detected in A-431 cells. For these studies, cells were stimulated with TNF-α, and then membranes were isolated and used to assess kinase activity toward the EGFR peptide. Figure 7 demonstrates kinetics of the effect of 30 nM TNF-αr, a maximally effective concentration for generation of ceramide and monocyte differentiation of these cells [6] . Cellular stimulation for as little as 5 minutes increased membrane-bound kinase activity to 1.5-fold of control, and activity continued to increase for as long as 2 hours to 2.2-fold of control. The effect of TNF-α was concentration-dependent when measured at 60 minutes of stimulation (Fig. 8) . As little as 3 pM TNF-α increased activity to 1.1-fold of control, and a maximal effect of 1.8-fold of control occurred with 30 nM TNF-c.; the ED₅₀ was «200 pM. Additional studies assessed the effect of the cell-permeable synthetic ceramide, C₈-cer, on enzyme activity. In three separate studies, addition of as little as 0.3 μM C₈-cer to the medium of HL-60 cells increased kinase activity in membranes derived from stimulated cells to 1.2-fold of control, and a maximal effect was achieved with 10 μM C₈-cer to 1.5-fold of control. This value was quantitatively similar to that obtained with a maximal concentration of TNF-cx in this set of studies. Sphingosine was similarly as effective as C8-cer. Kinase activity was not stimulated by maximally effective concentrations of other HL-60 differentiating agents, including cholera toxin (10 nM) , retinoic acid (0.5 μM) , and butyrate (0.5 mM) (28, 29) . In sum, these studies demonstrate that HL-60 cells, like A-431 cells, contain a ceramide-activated protein kinase and that TNF-o:, which generates ceramide as an early event in cellular activation, enhances kinase activity. E. Discussion

Davis et al. [15, 17] originally demonstrated that sphingosine stimulated phosphorylation of the EGFR on Thr-669 in A-431 cells. It is shown here that sphingosine was rapidly converted to ceramide in these cells and that ceramide induced identical effects. To investigate the kinase that mediated ceramide action, the present studies used a 19-amino acid synthetic peptide corresponding to the sequence around Thr-669 of the EGFR as a substrate. These studies have demonstrated that A- 431 cells contain a Mg²⁺-dependent kinase activity with a physiologic pH optimum that was stimulated by ceramide in a time- and concentration-dependent manner. This kinase has some distinctive features. It appears exclusively membrane-bound, does not utilize Ca²⁺ as cofactor, and is also activated by the protein kinase C inhibitor sphingosine. These features distinguish this kinase from any other known protein kinase. A similar activity was detected in HL-60 cells and was enhanced rapidly by TNF- a, which elevates ceramide (but not sphingosine) levels, as an early event in cellular activation.

Several studies have demonstrated that TNF-o. stimulates protein phosphorylation as a proximal event in cellular stimulation [19-24] . A variety of substrates have been identified, including a 28-kDa stress protein in bovine aortic endothelial cells [21] , the eukaryotic initiation factor 4E [22, 23] , an uncharacterized 26-kDa cytosolic protein in U937 human monoblastoid cells [20] , and the EGFR [24] . In most of these studies serine/threonine phosphorylation of these proteins was seen, and different investigators have suggested that the cAMP-dependent protein kinase [30] , protein kinase C [31, 32] , or some other protein kinase mediates TNF action [33] . The present studies strongly suggest that another serine/threonine protein kinase, in some systems, mediates TNF action. The amino acid sequence surrounding Thr-669 of the EGFR is unusual, containing three proline residues within a span of 9 amino acids. This unusual structure has no homology to the consensus substrate sequences for any of the major protein kinases [34, 35] . In fact, Gill and coworkers [36] reported that a peptide corresponding to residues 662-673 of the EGFR failed to serve as substrate for a variety of purified protein kinases in vitro, including the cAMP-dependent protein kinase, protein kinase C, calcium-calmodulin-dependent protein kinase, and S6 kinase. Only casein kinase II and glycogen synthase kinase 3 demonstrated significant activity toward this substrate, but the peptide proved to be a poor substrate for both of these kinases, as evidenced by high Km values. Glycogen synthase kinase 3 has a known preference for proline-rich substrates, which may account for the low level of activity detected in these studies [37] .

The region corresponding to Thr-669 of the EGFR is located between the transmembrane domain and the ATP- binding site within the catalytic domain. This region also contains Thr-654, the major protein kinase C phosphorylation site, and the region, in general, is considered to be involved in modulation of receptor function [38] . Mutational removal of Thr-669 has been shown to alter receptor down-regulation and substrate specificity [36] . This region is also highly conserved in the v-erbB and neu oncogene products and may represent a site for phosphorylation of these proteins by ceramide- activated protein kinase.

In sum, these studies characterize a ceramide-activated protein kinase activity in A-431 and HL-60 cells. Evidence has been presented that this kinase is activated by TNF-α, which triggers the generation of ceramide as an early event during cellular stimulation. Hence, this kinase may mediate, in whole or in part, signal transduction by TNF-o. in some systems. In this paradigm, binding of TNFor to its cell-surface receptor stimulates a neutral plasma membrane-bound sphingomyelinase that cleaves sphingomyelin to yield ceramide. Ceramide would then enhance kinase activity, resulting in the phosphorylation of specific substrates.

II - Tumor Necrosis Factor-αt Activates the Sphingomyelin Signal Transduction Pathway in a Cell-Free System

A. Abstract

The mechanism of tumor necrosis factor (TNF) -α signaling is unknown, however, TNF-α signaling most likely involves sphingomyelin hydrolysis to ceramide by a sphingomyelinase and stimulation of a ceramide-activated protein kinase. In a cell-free system, TNF-o. induced a rapid reduction in membrane sphingomyelin content and a quantitative elevation in ceramide concentrations. Ceramide-activated protein kinase activity also increased. Kinase activation was mimicked by addition of sphingomyelinase but not by phospholipases A₂, C, or D. Reconstitution of this cascade in a cell-free system demonstrates tight coupling to the receptor, suggesting that this is a signal transduction pathway for TNF-o;.

B. Experimental Procedure and Discussion Sphingomyelin can be metabolized to generate molecules that have various functions within the cell [1-6] . Ceramide, which is generated by sphingomyelinase action, can be deacylated to sphingoid bases [1, 14] , which are potential inhibitors of protein kinase C [9-il] or phosphorylated to ceramide 1-phosphate [4] by a ceramide kinase [5,^' 13] . Ceramide appears to have bioeffector properties [7, 8, 18] . Cell-permeable ceramide analogs stimulate monocytic differentiation of human leukemia (HL-60) cells [7, 8] and the phosphorylation of the epidermal growth factor receptor (EGFR) at Thr⁶⁶⁹ in A431 human epidermoid carcinoma cells [18] . TNF-o; activates a neutral sphingomyelinase to generate ceramide in HL-60 cells, and it was postulated that this initiated TNF-o; action [6] . A ceramide-activated protein kinase with a synthetic peptide derived from the amino acid sequence surrounding Thr⁶⁶⁹ of the EGFR (residues 663 to 681) was defined [40] . Kinase activity was membrane-associated, Mg²⁺-dependent, and activated by natural or synthetic ceramide in a concentration-and time-dependent manner. This ceramide-activated protein kinase activity was rapidly increased in membranes derived from HL-60 cells treated with TNF-o;. The present studies were undertaken to evaluate coupling of this sphingomyelin pathway to stimulation of the TNF receptor in a cell-free system.

The binding of TNF-o; to its receptor is detectable within 2 minutes and maximal by 5 to 10 minutes at 4°C in membranes derived from HL-60 cells [41] . Therefore, supernates from HL-60 cells, collected after a low-speed centrifugation to remove nuclei, were first incubated with TNF-o; for 5 minutes at 4°C to allow the formation of TNF-receptor complexes. Thereafter, reactions were initiated by warming supernates to 22°C in a reaction mixture containing adenosine triphosphate (ATP) and Mg²⁺ at pH 7.4.

These conditions were adopted to allow for activation of neutral sphingomyelinase [1, 42] . Under these conditions, TNF-o; induced a time- and concentration- dependent reduction in sphingomyelin content (Fig. 9A) . The effect of TNF-o; was evident at 1 minute and maximal by 7.5 minutes. Sphingomyelin concentrations decreased 27% from- a control concentration of 10.4 ± 0.5 (mean + SEM) to 7.6 ± 0.2 nmol per milligram (nmol mg^"1) of supernate protein (P < 0.001) . In contrast, the concentration of sphingomyelin in control incubations did not change. Concentrations of TNF-o; of 300 pM were effective, with a maximal effect at 3 nM TNF-α [effective dose (ED_S0) *-= 500 pM] . Under the same conditions, ceramide increased quantitatively from 1.8 ± 0.3 to 4.0 ± 0.5 nmol mg^"1 (Fig. 9B) . This effect was detectable at 1 minute (P<0.001) and maximal by 7.5 minutes. Thus, 2.8 nmol of sphingomyelin per milligram of supernate protein were lost for each 2.2 nmol of ceramide per milligram of supernate protein that was generated. Similar kinetics of sphingomyelin degradation and ceramide generation were determined in intact HL-60 cells (n=3) , confirming previous studies [6]. Other choline-containing lipids, including phosphatidyl-choline, lysophosphatidylcholine, sphingosylphosphorylcholine [1], and 1,2-diacylglycerol were not affected by TNF-o;. Thus, TNF-α activated a neutral sphingomyelinase in a cell-free system, which resulted in the generation of the potential second messenger ceramide.

The effect of TNF-o; on ceramide-activated protein kinase activity was assessed. Nuclei-free supernates contain ceramide-activated protein kinase activity that can phosphorylate EGFR peptide with a maximum velocity (V.^) of 50 to 100 pmol per minute per milligram (pmol min^mg^"1) of protein and a Michaelis constant (K.. of 15 μM) for ATP and 0.25 mg ml^"1 for peptide [40] . Ceramide (0.001 to 3 μM) enhances kinase activity to a maximum of two-fold of the control [40] . TNF-o;, which increased ceramide concentrations, similarly enhanced kinase activity in intact cells [40] . For studies assessing the effect of TNFo; in broken cell preparations, nuclei-free supernates were incubated under conditions_^ sufficient for stimulation of neutral sphingomyelinase in a reaction mixture that also contained EGFR peptide and γ-³²P-labeled ATP. Phosphorylated peptide was resolved by high- performance liquid chromatograph (HPLC) and quantified by Cerenkov counting [40] . Kinase activity was calculated from the specific activity of [γ-³²P] ATP and incorporation of ³²P into EGFR peptide. Background activity was subtracted from each point. TNF-o; (30 nM) treatment enhanced kinase activity (P<0.001) in a time- dependent manner (Fig. 10A) . TNF-α stimulation of kinase activity was evident by 1 minute and demonstrable for at least 10 minutes. If the initial incubation with TNF-o; at 4°C was omitted and TNF was added directly to the reaction mixture at 22°C, the reaction was delayed. Under these conditions, enhancement of activity by TNF-o; did not occur for 2 minutes, presumably until after TNF- receptor complexes had formed. TNF-α; enhanced kinase activity in a concentration-dependent manner at 5 minutes

(Fig. 10B) . TNF-o; was effective at 10 pM and had a maximal effect at 3 nM; the ED₅₀ was « 300 pM TNF-o;. This is similar to the ED₅₀ of 200 pM for stimulation of ceramide-activated protein kinase by TNF-o; in intact cells [40] . TNF-o; enhanced kinase activity in a total of 20 separate studies. Guanosine triphosphate (GTP) and guanosine-5' -O- (3-thiotriphosphate) (GTP-γ-S) (0.25 to 200 μM) did not affect kinase activity.

To demonstrate that the effect of TNF-o; is mediated by sphingomyelin hydrolysis to ceramide, a sphingomyelinase or a phospholipase (A_2/ C, or D) was added to the kinase reaction mixture and measured EGFR peptide. phosphorylation was measured. For some studies, the reaction mixture contained free Ca²⁺ (1 mM) , which did not affect results. Control activity reflects several TNF-o;- independent protein kinases that are known to phosphorylate EGFR peptide on Thr⁶⁶⁹. Exposure of the nuclei-free supernates to sphingomyelinase (1 x 10^"3 U ml" ¹) from Staphylococcus aureus for 5 minutes induced and increase in kinase activity comparable to that induced by TNF-o; (1 nM) (Fig. 11) . This concentration of sphingomyelinase stimulates a two-fold elevation in ceramide levels in HL-60 cells [3, 4] . Concentrations of phospholipases A₂, C, and D, which were 40- to 400-fold higher than sphingomyelinase and which are effective for phospholipid hydrolysis under conditions used in these assays, did not enhance kinase activity. Hence, the effect of TNF-o; in broken cell preparations was mimicked by a sphingomyelinase but not by other phospholipases.

The mechanism of coupling of the TNF receptor to sphingomyelinase is unknown. Neutral sphingomyelinase appears to be ubiquitous in mammalian cells and is externally oriented in the plasma membrane [44] . Similarly, sphingomyelin is preferentially localized to the outer leaflet of the plasma membrane [45] . This colocalization of receptor, phospholipase, and substrate at the plasma membrane suggests that ceramide is generated at this site. The exact intracellular site of the ceramide-activated protein kinase has not yet been investigated. However, preliminary evidence suggests it is an intrinsic membrane protein [40] . In this regard, ceramide-activated protein kinase would not have to be present in the outer leaflet of the plasma membrane for signaling to occur, as ceramide can redistribute across a membrane bilayer [46] .

Ceramide-activated protein kinase may be a member of an emerging family of serine/threonine protein kinases that includes microtubule-associated protein 2 (MAP2) kinase

[extracellular signal-regulated kinase (ERK1)] [35, 47,

48] , EGFR threonine (ERT) kinase [49] , glycogen synthase kinase-3 [35, 47, 48] and p34^cdc2-containing proline- directed and histone HI kinases [49, 50] . The substrates for these kinases appear to have a minimal recognition sequence, X-Ser/Thr-Pro-X, in which the phosphorylated site is flanked by a COOH-terminal proline residue [50,

51] and X can be any amino acid. Substrates for this class of kinases include EGFR, proto-oncogene products

Jun and Myc, tyrosine hydroxylase, histone HI, glycogen synthase, synapsin I, and protein phosphatase inhibitor II [37, 49-51] . TNF-induced, proline-directed phosphorylation of these proteins has not yet been demonstrated. The X-Ser/Thr-Pro-X sequence is different from consensus substrate sequences for other major serine/threonine kinases, including cyclic adenosine monophosphate (cAMP) - and cyclic guanosine monophosphate

(cGMP) -dependent-protein kinases, Ca²⁺/calmodulin- dependent-protein kinase, and ribosomal S6 protein kinase

[49] . In fact, these kinases have limited activity toward this proline-containing sequence [50] .

It has been proposed that various distinct signaling systems, including protein kinases A and C, phospholipases A₂ and C, the EGFR tyrosine kinase, and a novel serine kinase, may mediate TNF-o; action [19] . It is clear that no single second messenger pathway can account for the entirety of the reported biologic effects of TNF-o;. The role of the sphingomyelin pathway in events other than monocytic differentiation has not been investigated nor has the relation to these other signaling systems. This issue is further complicated by the recent cloning of two distinct TNF receptor forms of 55 kD and 75 kD [52-55] with homologous extracellular domains with dissimilar intracellular portions.

In sum, the rapid kinetics of activation of the sphingomyelin pathway by TNF-o; in intact cells, the ability of cell-permeable ceramide analogs to bypass receptor activation and mimic TNF-α action, and the reconstitution of this cascade in a cell-free system provide strong support for the notion that this pathway serves to couple TNF receptor activation to cellular stimulation. Hence, these studies suggest that TNF-α; activates a plasma membrane-bound neutral sphingomyelinase to generate at the second messenger ceramide, which stimulates the ceramide-activated protein kinase to phosphorylate a distinct set of protein substrates, thereby altering their function.

Il - Interleukin-lβ Signals Through the

Sphingomyelin Pathway in Intact EL-4 Cells and in a Cell-Free System

A. Abstract

The mechanism of interleukin-1 (IL-1) signaling is unknown. Recent investigations demonstrated that tumor necrosis factor-o; utilizes a signal transduction pathway involving sphingomyelin hydrolysis to ceramide and stimulation of a ceramide-activated protein kinase. In intact EL-4 thymoma cells, IL-lβ similarly stimulated rapid reduction in sphingomyelin and elevation in ceramide levels, and enhanced ceramide-activated protein kinase activity. This cascade was also activated by IL- lβ in a cell-free system demonstrating tight coupling to the receptor. Further, exogenous sphingomyelinase but not phospholipases A₂, C or D, replaced IL-lβ to stimulate IL-2 secretion in combination with phorbol ester. These studies demonstrate that IL-lβ signals through the sphingomyelin pathway.

B. Experimental Methods and Discussion

Hydrolysis of sphingomyelin to ceramide at the plasma membrane by a neutral sphingomyelinase may initiate a cascade that functions in signaling [6-8, 18, 40, 58, 59] . Evidence has been provided that ceramide may stimulate a serine/threonine kinase termed ceramide- activated protein kinase to transduce the signal [18, 40, 59] . Ceramide-activated protein kinase is membrane- bound, Mg⁺²-dependent and defined by its capacity to phosphorylate a synthetic peptide derived from the amino acid sequence surrounding Thr⁶⁶⁹ of the epidermal growth factor receptor (EGFR) . Ceramide-activated protein kinase may be a member of an emerging family of proline- directed serine/threonine kinases that includes the extracellular-signal regulated (also referred to as mitogen-activated) and p34^cdo2 kinases [47] . Substrates for these kinases contain the minimal recognition sequence, X-Ser/Thr-Pro-X, in which the phosphorylated site is flanked on its carboxy terminus by a proline residue and X can be any amino acid.

Evidence has been provided that tumor necrosis factor (TNF) -α; may utilize the sphingomyelin pathway for signaling [6, 40, 59] . TNF stimulates this pathway early during HL-60 cell differentiation into monocytes [6, 59] and synthetic ceramide analogs bypass receptor activation and directly induce differentiation [7] . Further, this cascade has been reconstituted in a cell-free system comprised of extracts of HL-60 cells, demonstrating tight coupling of this pathway to the TNF receptor [59] . The present studies were performed because of numerous reports that TNF and IL-1 stimulate a common set of events in diverse biologic systems [60] .

The murine thymoma EL-4 cell line is a well-defined IL-1 responsive cell line that expresses functional IL-1 receptors [61, 62] . Upon stimulation with IL-1, these cells up-regulate the IL-2 receptor and secrete IL-2 [62] . Initial studies were designed to investigate the effects of IL-lβ on cellular sphingomyelin content. Cells, grown in Dulbecco's Modified Eagle's (DME) /Ham's F-12 medium containing 10% horse serum and [³H] choline (1 μCi ml^"1) , were resuspended back into the same medium at 10 x 10⁶ cells ml^"1 and stimulated with IL-lβ. IL-lβ is commercially available. Under these conditions, IL-lβ induced time- and concentration-dependent sphingomyelin hydrolysis (Fig. 12 A, B) . A maximally effective concentration of IL-lβ, 40 ng ml^"1, induced a detectable reduction in sphingomyelin content by 2 minutes from a baseline of 800 ± 14 pmol 10⁶ cells^"1 (mean ± SEM) and the level decreased to 648 ± 16 pmol 10⁶ cells^"1 (p<0.005) at 30 minutes. Concentrations of IL-lβ of 0.01 ng ml"¹ were PCΪ7US95/07405

53 effective, with a maximal effect at 10 ng ml^"1 [effective dose (ED_S0) ^•= 2 ng ml^"1 (Fig. 12B) ] . A similar reduction in sphingomyelin content after IL-1 stimulation was determined by direct measurement of phosphorous content [63] . In contrast, the content of phosphatidylcholine, the other major choline-containing phospholipid, was unchanged.

Under the same conditions, IL-lβ induced a statistically significant increase in ceramide content (Fig. 13) .

Ceramide increased from 360 to 403 pmol 10⁶ cells^"1 at 2 minutes (p<0.005) and to a maximum of 450 pmol 10⁶ cells^"1 at 15 minutes. In separate studies (n=4) , a statistically significant increase in ceramide content was evident by 30 seconds. Maximally effective concentrations of other agents known to stimulate EL-4 cells [65, 66] including 12-0-tetradecanoylphorbol-13- acetate (TPA) , concanavalin A, epinephrine and an anti-

CD3 antibody failed to elicit a ceramide response (n=5) . Hence, sphingomyelinase activation appeared specific for stimulation by IL-lβ.

Subsequent studies assessed whether IL-lβ also enhanced ceramide-activated protein kinase activity. EL-4 cells were found to contain a membrane-bound ceramide-activated protein kinase activity similar to that reported in A431 human epidermoid carcinoma cells and HL-60 cells [40, 59] . Activity was measured by the transfer of ³²P from the γ-position of ATP to EGFR peptide (AA 663-681 of the EGFR) . The effect of IL-lβ on kinase activity was determined using microsomal membranes_^derived from cells stimulated with IL-lβ. IL-lβ enhanced kinase activity in a time- ^• and concentration-dependent manner. In cells treated with 10 ng ml^"1 IL-lβ, a maximally effective concentration, an increase in kinase activity was detectable at 30 seconds and maximal at 2 minutes (Fig. 14, p<0.005) . Activity increased to 2.1-fold of control from 5 to 11 pmol per minute per milligram (pmol min"¹ mg^" ¹) and then gradually declined over 15 minutes. Concentrations of IL-lβ of 0.03 ng ml^"1 were effective, with a maximal effect at 10 ng ml^"1 [effective dose (ED₅₀) «= 2 ng ml^"1] . This is the same range of concentrations found effective for sphingomyelin hydrolysis. Stimulation by IL-lβ was detected in a total of 10 experiments. Cytosolic fractions of EL-4 cells also contained kinase activity toward EGFR peptide of 2.6 ± 0.3 (mean ± range) pmol min"¹ mg"¹. Cytosolic activity, which represents proline-directed protein kinase activities other than ceramide-activated protein [68] , was not enhanced by Il-lβ during these studies. Further, protein kinase C activity as determined by phosphorylation of lysine-rich histone (Sigma Chem. Co., type III-S) [69] was not enhanced in either membrane or cytosolic fractions.

Early kinetics of activation of a potential signaling system provide some support that the pathway might be involved in the signaling process. However, signal transduction pathways are highly regulated and often interrelated [70] . Hence, activation of one system often results in rapid activation of another. To provide additional support for tight coupling of the sphingomyelin pathway to activation of the IL-lβ receptor, studies were performed with subcellular fractions derived from EL-4 cells. For these studies, supernates, collected after a low-speed centrifugation to remove nuclei, were first incubated with IL-lβ for 10 minutes at 4°C to permit formation^ of IL-1 receptor complexes [59] . Thereafter, reactions were initiated by warming supernates to 22°C in a reaction mixture containing Mg²⁺ at pH 7.4. These conditions were adopted to allow for activation of endogenous neutral sphingomyelinase [1, 42] . For studies measuring kinase activity, reaction mixtures also contained [³²P] ATP and EGFR peptide. Under these conditions, IL-lβ stimulated a rapid reduction in sphingomyelin content and a quantitative increase in ceramide content (Fig. 15A) . In separate studies, a statistically significant reduction in sphingomyelin content (n=10) and elevation in ceramide content (n=6) were detected at 1 minute of stimulation

(p<0.005 vs. control) . Ceramide-activated protein kinase activity also increased (Fig. 15B) . These effects were quantitatively similar to those determined in the intact cells. Hence, the effect of IL-lβ to activate the sphingomyelin pathway was also observed in a cell-free system.

To determine whether the sphingomyelin pathway mediated the biologic response to IL-lβ, direct activation of ,the sphingomyelin pathway with exogenous sphingomyelinase

[59] was compared to stimulation by IL-lβ. For these studies, cells were treated with IL-lβ, sphingomyelinase and/or phorbol ester and, after 24 h IL-2 secreted into the media was measured. As previously reported [62, 65,

71, 72] , IL-lβ (1-30 ng ml^"1) alone did not induce detectable IL-2 secretion (Table 1) , nor did TPA (1-20 ng ml^"1) alone.

However, in combination IL-lβ (10 ng ml^"1) and TPA (20 ng- l^"1) stimulated secretion maximally. Sphingomyelinase alone also failed to stimulate IL-2 secretion, but again, in combination with TPA, induced secretion. Concentrations of sphingomyelinase between 5 x 10"^s U ml"¹ and 1 x 10^"1 U ml^"1) were effective. In separate studies (n=2) , sphingomyelinase (1 x 10^"3 U ml^"^) induced secretion at all concentrations of TPA from 0.5 to 20 ng ml"¹. This concentration of sphingomyelinase induced an increase in ceramide content quantitatively similar to that induced by maximally effective concentrations of IL-lβ, and has previously been shown to mimic TNF action in HL-60 cells

[59] . In contrast, phospholipases (PL) A₂, C and D at concentrations 10-50 times higher than maximally effective sphingomyelinase, did not stimulate IL-2 secretion alone or in combination with TPA. Hence, the effect of IL-1 to co-stimulate IL-2 secretion in EL-4 cells was mimicked by activation of the sphingomyelin pathway with sphingomyelinase.

Table I Induction of IL-2 secretion by IL-1 and sphingomyelinase. EL4 cells (1.5 x lb⁶ ml^"1) were treated with IL-lβ (10 ng ml"¹) , sphingomyelinase (SMase, Staphylococcus aureus) ,

PLA₂ ( Vipera ruselli) , PLC (Bacillus cereus) and PLD

(Streptomyces chromofuscus) at the indicated concentrations, in the absence or presence of TPA (20 ng ml^"1) . Boiled sphingomyelinase preparations had no activity. Culture supernates were harvested at 24 h and assayed for secreted IL-2 using an anti-mouse IL-2 ELISA kit (Genzyme Corp.) according to the manufacturer's instructions. The lower limit of sensitivity of this assay was 15 pg IL-2 and the assay was linear up to 960 pg IL-2. These data (mean ± range) represent duplicate determinations from 2 experiments.

Diluent +TPA

Control nd* nd IL-lβ (10 ng ml^"1) nd 261 ± 2

SMase (lxlO"³ u ml"¹) nnrdl - 3-π13--. ± 60 PLA₂ (1-5X10"² u ml"¹) nd nd

PLC (l-5xl0^"2 u ml"¹) nd nd

PLD (l-5xl0^"2 u ml^"1) nd nd

*nd, not detectable

Although signaling for IL-1 has been ascribed to various protein kinases including protein kinases A and C and a novel serine/threonine protein kinase [30, 73-77] , no coherent picture has emerged to account for all of the data. Two distinct IL-1 receptors of 60 kDa and 80 kDa have recently been cloned [78-80] . The receptors are homologous in their extracellular binding domains but have little homology in their cytoplasmic portions. In fact the 60 kDa receptor has only a short intracellular portion. There is no empiric or structural evidence suggesting that these receptors themselves might serve as protein kinases [78-80] . In addition, there is no homology between these receptors and any protein known to be involved in signal transduction. The present studies define a new mechanism by which the IL-1 receptor might activate a protein kinase. Preliminary studies with the human natural killer-like cell line, YT [81] , demonstrate that IL-1 also induces rapid generation of ceramide in this system.

Despite the often reported similarities in action of TNF- a and IL-1 there is limited primary sequence homology between their receptors. Hence, the mechanism by which these two cytokines activate the sphingomyelin signal transduction pathway is not readily apparent.

In sum, these studies provide evidence that the effects of IL-lβ may be mediated by the sphingomyelin signal transduction pathway. In this paradigm, ligand binding to the receptor • activates a neutral sphingomyelinase hydrolysing sphingomyelin to ceramide. Neutral sphingomyelinase appears to be ubiquitous in mammalian cells and like sphingomyelin is externally oriented in the plasma membrane [44] . This co-localization of receptor, phospholipase and substrate at the plasma membrane suggests that ceramide is generated at this site. Ceramide, which can redistribute across a lipid bilayer, then stimulates ceramide-activated protein kinase which phosphorylates a specific subset of cellular proteins hereby altering their function.

IV - Renaturation and TNFo; Stimulation of a 97 kD Ceramide- Activated Protein Kinase

A. Summary

Recent investigations identified a new signal transduction pathway, termed the sphingomyelin pathway, which may mediate the action of tumor necrosis factor (TNF)α; and interleukin- lb. This pathway is initiated by hydrolysis of sphingomyelin to ceramide by a neutral sphingomyelinase and stimulation of a ceramide-activated Ser/Thr protein kinase. Recent investigations demonstrated that kinase activity is proline-directed, recognizing substrates in which the phosphoacceptor site is followed by a proline residue. Until now, the kinase has been defined only as a membrane-bound activity capable of phosphorylating a peptide derived from the sequence surrounding Thr669 of the epidermal growth factor receptor (EGFR) . In these experiments, the kinase was quantitatively extracted from membrane with detergent and separated from protein kinase C by anion exchange chromatography and isoelectric focusing. Ceramide-activated protein kinase was resolved as an exclusively membrane-bound, 97 kD protein with a pi of 7.05. Kinase activity toward the EGFR peptide co-purified with activity toward a generic proline-directed substrate, myelin basic protein (MBP) . Kinase activity was* reconstituted by a denaturation-renaturation procedure and demonstrated activity towards self (autophosphorylation) and exogenous substrate (MBP) . Autophosphorylation occurred exclusively on serine residues. These activities were enhanced to seven-fold of control by ceramide and TNFo;. These data provide additional evidence for a role for ceramide- activated protein kinase in signal transduction for TNFα. B. Introduction

The sphingomyelin pathway is a signal transduction pathway mediating the action of the cytokines tumor necrosis factor (TNF)o; and interleukin-1 (IL-l)b [59] . In this paradigm, receptor stimulation initiates hydrolysis of plasma membrane sphingomyelin to ceramide by the action of a neutral sphingomyelinase. Ceramide then acts as a second messenger, stimulating a membrane-bound serine/threonine kinase, termed ceramide-activated protein kinase, thereby, transducing the cytokine signal [58] . Three lines of evidence support the notion that the sphingomyelin pathway mediates cytokine action. First, the sphingomyelin pathway is activated within seconds to minutes by TNF in human leukemia (HL-60) cells and by IL-1 in human dermal fibroblasts and mouse EL4 T-helper cells. Second, cell-permeable ceramide analogs can bypass receptor activation and directly mimic the effect of these cytokines. In this regard, synthetic ceramide analogs have been shown to stimulate differentiation of HL-60 cells into monocytes [6] , phosphorylation of the epidermal growth factor receptor (EGFR) at a specific site, Thr669, in human A431 epidermoid carcinoma cells [18] , induce cyclooxygenase gene expression in fibroblasts [83] and IL-2 secretion in EL4 cells. Third, the effects of TNF and IL-1 on this cascade of events have been reconstituted in cell-free extracts indicating tight coupling of this pathway to the respective receptors [59] . Recently, Wiegmann et al. [90] transfected the 55 kD TNF receptor into the pre-B cell line 70Z/3 which contains no TNF receptor. Under these conditions, transfection conferred TNF-induced sphingomyelin hydrolysis onto these cells.

A key element in this signal transduction cascade is stimulation of ceramide-activated protein kinase. Until now, this activity has only been defined as a membrane-bound activity capable of phosphorylating a peptide derived from the amino sequence surrounding Thr669 of the EGFR (amino acids 663-681) . This is the same site recognized by the mitogen-activated protein (MAP) kinases (also known as extracellular signal-regulated kinases or ERKs) . Hence ceramide-activated protein kinase may be a member of the emerging family of serine/threonine kinases, termed proline-directed protein kinases, which recognize substrates containing the motif X-Ser/Thr-Pro-X. Ceramide-activated protein kinase activity is Mg²⁺-dependent and has a physiologic pH optimum. Activity has been found in every cell type has so far been examined including HL-60 cells, A431 cells and EL4 cells. The purpose of the present experiments was to characterize this activity more definitively. In the present studies ceramide-activated protein kinase has been solubilized from HL-60 cell membranes, separated from protein kinase C (PKC) by anion exchange chromatography, partially purified by isoelectric focusing (IEF) and defined as a single band of 97 kD using a denaturation-renaturation protocol. Further, renatured kinase activities towards self (autophosphorylation) and exogenous substrate (myelin basic protein) are markedly enhanced by TNFo; and ceramide consistent with the proposition that this kinase is involved in signal transmission.

C. Experimental Procedures 1. Reagents

Buffers, lipids (phosphatidylserine and diolein) , lysine-rich histone (type Ills) , insulin, transferrin, leupeptin, soybean trypsin inhibitor, hexokinase, glucose-6-phosphate dehydrogenase, phosphoamino acid standards and bovine myelin basic protein (MBP) were purchased from Sigma Chemical Co. HPLC grade solvents were from Fisher Scientific. [γ-³²P]ATP (3000 Ci/mmol) was from NEN/Du Pont. Fetal bovine serum was obtained from GIBCO-BRL. P81 phosphocellulose paper and DE52 anion exchange resin were from Whatman. The EGFR peptide was synthesized as described [40] . C8-ceramide may be readily prepared by those skilled in the art. TNFo; was provided by Genentech, and is commercially available.

2. Cells

HL-60 cells were grown in suspension culture in RPMI 1640 medium supplemented with 10% fetal bovine serum, penicillin (10,000 units ml^"1), streptomycin (10,000 mg ml"¹), serine (16 mg ml^"1), L-asparagine (8.4 mg ml"¹) and glutamine (16.7 mg ml^"1) as described [3] . Prior to treatment with ceramide or TNFo;, cells were resuspended (1 x 10^s ml"¹) in serum-free RPMI containing bovine insulin and human transferrin (5 μg ml^"1 each) for 2 h at 37°C.

3. Membrane preparation

All procedures were conducted at 4°C. Cells (3 x 10⁸ ml^"1) were homogenized with a tight fitting dounce homogenizer in 25 mM HEPES (pH 7.4) containing 5 mM EGTA, 50 mM NaF, and protease inhibitors (10 mg ml^"1 each of leupeptin and soybean trypsin inhibitor) . The homogenate was centrifuged at 1000 x g for 10 min and the nuclei-free supernatant was centrifuged at 12,000 x g for 7 min. The resulting supernatant was centrifuged at 250,000 x g for 30 min. To eliminate cytosolic contaminants, the membrane pellet was washed with HEPES buffer and recentrifuged at 250,000 x g for 30 min. For solubilization of ceramide-activated protein kinase activity, the following detergents were tested: Triton X-100 (ι%) , CHAPS (10 mM) , b-octylglucoside (1.5%) and sodium deoxycholate (1 mM) . Membranes were incubated in detergent for 10 min at 4°C. Particulate material was then removed by centrifugation at 15,000 x g for 10 min prior to assay for kinase activity. 4. Assay of protein kinase C activity

The membrane pellet was resuspended in HEPES buffer (25 mM HEPES, pH 7.4, 5 mM EGTA with protease inhibitors) and solubilized with 1% Triton X-100 as described above. DE52 anion exchange chromatography was employed for fractionating proteins by stepwise elution with increasing concentrations of NaCl from 0.05 M to 0.4 M. PKC activity was measured by the transfer of ³²P from [γ-³²P]ATP to lysine-rich histones using a standard reaction mixture containing 20 mM HEPES (pH 7.5), 10 mM

MgCl₂, 2 mM dithiothreitol, 1 mg ml^"1 lysine-rich histones,

100 mM [γ-³²P] ATP (1000 dpm pmol^"1) , phosphatidylserine

(80 mg ml^"1), 1,2-diolein (8 mg ml^"1) and portions of the column eluate as described [2] . The final concentration of free Ca²⁺ was 0.5 mM. Phospholipid- and Ca²⁺-independent activity was measured in the absence of Ca²⁺, phosphatidylserine and diolein. Portions of the reaction mixture were transferred to P81 phosphocellulose paper, washed extensively and ³²P-labeled histones were quantified by Cerenkov counting. All incubations were performed under conditions determined as linear for time and enzyme concentration.

5. Assay of Ceramide-activated protein kinase The kinase reaction mixture contained 15 ml intact membrane, solubilized membrane or DE52 column fractions, 75 μl assay buffer (50 mM HEPES, pH 7.4, 20 mM MgCl₂) and 30 μl ATP (150 μM final concentration for EGFR peptide and 50 mM final concentration for MBP; 4000 dpm pmol^"1) . The substrate used was 30 μl EGFR peptide (4 mg ml^"1 in 25 mM HEPES, pH 7.4) or 30 μl bovine MBP (5 mg ml"¹) . For studies of lipid activation, C8-ceramide was dried under N₂ and added to the assay buffer by sonication for 2 min at 37 °C. All assays were performed under conditions determined as linear for time and enzyme concentration.

Phosphorylation of the EGFR peptide was terminated by addition of 30 μl of 0.5 M ATP in 90% formic acid.

Samples were applied to a C18 Sep-pak cartridge

(Millipore) and phosphorylated peptide was eluted with

99.9% acetonitrile/O.1% trifluoroacetic acid. The eluates were lyophilized, resuspended in buffer (6 M guanidine hydrochloride, 200 mM Tris, pH 8.5) and resolved by HPLC using a C18 reverse phase column

(Dynamax, 4.6 mm i.d., Rainin) . The peptide was eluted with a linear gradient of acetonitrile and quantified by measuring the Cerenkov radiation associated with 1 ml fractions [17, 40] . Enzyme activity was determined from the transfer of ³²P from [γ-³²P]ATP to the EGFR peptide and from the specific radioactivity of [γ-³²] ATP.

Phosphorylation of MBP was terminated by addition of 30 μl of 5X Laemmli sample buffer. Samples were boiled for 2 min and electrophoresed on 15% SDS-polyacrylamide gels, followed by autoradiography and Cerenkov counting of labeled bands.

6. Isoelectric focusing of the kinase activity toward the EGFR peptide Membrane extracts prepared from 1 x 10⁹ cells were fractionated by DE52 anion exchange chromatography as described above. The flow-through fraction containing protein kinase activity toward the EGFR peptide was subjected to focusing by a Rotofor preparative isoelectric focusing cell (Bio-Rad) . The pH gradient was achieved by pre-focusing a 2% ampholyte solution (pH range 3-10) for 1.5 h. To prevent protein aggregation, Triton X-100 and glycerol were added to final concentrations of 1% and 5%, respectively. Focusing was performed at a constant power of 12 W for 4 h and 20 fractions of 2 ml each were collected. The pH of individual fractions was measured and a 5 μl aliquot from each fraction was analyzed for EGFR peptide phosphorylation and MBP phosphorylation as described above. To assess protein purification after isoelectric focusing, an aliquot of each fraction was applied to a 12% SDS gel and stained with Coomassie blue. Measurement of the protein content in individual fractions was not possible by the method of Bradford [27] due to interference of ampholytes with the reagents in these assays.

7. Renaturation of ceramide-activated protein kinase Microsome membranes were prepared as described above from untreated, ceramide-treated, or TNF-treated cells. For some experiments, the membrane was solubilized with 1% Triton X-100 and chromatographed on a DE52 anion exchange column. Intact membrane and the flow-through column fraction were electrophoresed on a 10% SDS-gel polymerized with or without MBP (50 μg ml^"1) or the EGFR peptide (500 μg ml^"1) . Renaturation of the kinase activity was performed according to the method described for calmodulin-dependent protein kinase II [86] . Briefly, the gel was washed with two changes of wash buffer (50 mM Tris, pH 7.4, 5 mM 2-mercaptoethanol) containing 20% isopropanol at room temperature for 1 h, and once in wash buffer without isopropanol for 1 h. Denaturation was accomplished by incubation of the gel in two changes of 6 M guanidine HCl in wash buffer for 1 h each. Renaturation was accomplished by incubation of the gel overnight at 4°C in wash buffer containing 0.04% Tween-20. The gel was then equilibrated for 10 min at room temperature in kinase reaction mixture (25 mM HEPES, pH 7.4, 10 mM MgCl₂, 0.5 mM EGTA and 5 mM NaF) . After equilibration, [γ.³²P]ATP (50 μM final concentration) was added to the reaction mixture for varying lengths of time. Autophosphorylation was terminated by removal of the reaction mixture and the gel was washed with water for 10 min followed by 6 changes of buffer (5% trichloracetic acid, 1% sodium pyrophosphate) for 2 h. The gel was then autoradiographed. In some cases phosphorylated bands were excised and subjected to Cerenkov counting.

8. Measurement of ATP and depletion of ATP in 5 membranes

The endogenous ATP level was measured by an enzymatic assay using hexokinase and glucose-6-phosphate dehydrogenase [88] . For this procedure, an aliquot of membrane (equivalent to 15 x 10⁶ cells) was incubated with

10 reaction mixture containing 25 mM HEPES, 200 mM D-glucose, 8 mM MgCl₂, 1 mM NADP, 3 units of glucose-6-phosphate dehydrogenase, and 0.03 units of hexokinase. Absorbance was read on a spectrophotometer at 340 nm. The limits of detection of ATP by this assay

15 was 7 mM. Additionally, to remove trace amounts of ATP, a membrane fraction equivalent to 15 x 10⁶ cells was pre-treated with 0.03 units of hexokinase for 5 min prior to incubation with ceramide.

20. 9. Phosphorylation of the exogenous substrate,

MBP, by renatured kinase contained within a polyacrylamide gel

Membrane from untreated or TNF-treated cells was electrophoresed and kinase was renatured and

25 autophosphorylated as above. The gel was then washed for 20 min with 4 changes of 50 mM Hepes buffer, pH 7.4. Bands corresponding to 97 kD and to 75-90 kD were excised, crushed and incubated for varying lengths of time with 10 ml MBP (5 mg ml^"1) and 40 ml kinase reaction

30 mixture in the presence of [γ-³²P]ATP (50 mM final concentration) . MBP phosphorylation reactions were terminated as above. After centrifugation to pellet gel particles, the supernates were electrophoresed and phosphorylation was measured as above.

35

10. Phosphoamino acid analysis Phosphoamino acid analysis was carried out by a modification of the method of Boyle et al. [84] . Briefly, autophosphorylated bands were excised from the denaturation-renaturation gel and hydrolyzed in 6 N HCl- at 110°C for 1 h. The hydrolysates were dried by Speed-vac overnight, resuspended in dH₂0 and applied to an AG 1-X8 anion exchange column (Bio-Rad) . Phosphoamino acids were eluted from the column with 0.5 N HCl and analyzed by one-dimensional thin-layer electrophoresis. Individual amino acids were detected by ninhydrin staining and autoradiography.

11. Protein assay Protein content was measured according to the method of Bradford [27] using BSA as a standard.

D. Results

1. Solubilization of membrane-bound protein kinase activity toward the EGFR peptide Previous studies showed that exogenous ceramide enhanced phosphorylation of the EGFR immunoprecipitated from A431 cells on a specific site Thr669 [40, 18] . Subsequently, a ceramide-activated protein kinase was defined as a membrane-bound activity capable of phosphorylating a synthetic peptide derived from the amino acid sequence surrounding Thr669 of the EGFR (residues 663 to 681) . Activity was optimal in a buffer containing Mg²⁺ at neutral pH. Other cations including Ca²⁺, Mn²⁺ and Zn²⁺ were incapable of sustaining kinase activity. To extract this activity from membranes, several detergents were tested. For these studies microsomal membranes were treated with detergent, centrifuged to-remove particulate material, and recovery of kinase activity was determined by measuring phosphorylation of the EGFR peptide as described. Table II shows that kinase activity toward the EGFR peptide was quantitatively recovered after treatment with the nonionic detergent Triton X-100 or the zwitterionic detergent CHAPS. In comparison, treatment with b-octylglucoside or sodium deoxycholate resulted in 27% and 47% loss of the activity, respectively. For all subsequent studies with solubilized membrane, Triton X-100 was used.

Table II Effect of detergents on ceramide-activated protein kinase. Membrane proteins were extracted by incubation of HL-60 cell membranes in the absence (control) or presence of detergents. Extracts were assayed for kinase activity towards the EGFR peptide as described in Experimental Procedures.

Activity Treatment

Concentration Recovery

(pmol-min^"1-mg^"1) (%)

Control 51 100

Triton X-100 1% 51 100

CHAPS 10 mM 52 1 0 1 b-octylglucoside 1.5% 37 73

Deoxycholate 1 mM 27 53

2. Fractionation of protein kinase activity toward the EGFR peptide and PKC by anion exchange chromatography In an attempt to separate the other major membrane-bound serine/threonine protein kinase, protein kinase C (PKC) , from the activity phosphorylating the EGFR peptide, the solubilized membrane preparation was subjected to mild anion exchange chromatography using a DE52 column at neutral pH. Membrane proteins were eluted from the column with a stepwise gradient of NaCl. Each fraction was tested for PKC activity by phosphorylation of histone or for ceramide- activated protein kinase activity by phosphorylation of the EGFR peptide. Fig. 16 shows that the PKC activity was, as previously described [2] , found entirely in the fractions eluting between 0.05 and 0.1 M NaCl. In contrast, 80% of protein kinase activity toward the EGFR peptide was detected in the flow-through fraction of the column and 20% was in the 0.2-0.4 M NaCl fractions. Hence, under the conditions employed for this assay, the fractions containing PKC activity did not phosphorylate the EGFR peptide.

3. Phosphorylation of myelin basic protein by the fractions containing activity toward the EGFR peptide

The ceramide-activated protein kinase apparently recognizes the same minimal substrate motif

(X-Ser/Thr-Pro-X) as MAP kinases (mitogen-activated protein kinases) [18] . Since bovine MBP has often been used as a substrate to assess activity of MAP kinases, it was tested as a substrate for the kinase which displays activity towards the EGFR peptide using the fractions separated by anion exchange chromatography. Fig. 17 shows that the majority of activity toward MBP was observed in the column flow-through and wash, and a smaller amount of the activity was detected in the fractions eluted with 0.2-0.4 M NaCl. In contrast, little activity, if any, could be detected in fractions eluted with 0.05-0.1 M NaCl (Fig. 17) . Hence, the kinase activity toward MBP correlated closely with the activity toward the EGFR peptide.

4. Isolation of the kinase activity toward the EGFR peptide by isoelectric focusing

To further^'resolve the protein kinase activity toward the EGFR peptide, the flow-through fraction of the anion exchange column was mixed with 2% ampholytes and subjected to isoelectric focusing at a constant power of 12 W for 4 h. A linear pH gradient was achieved under these conditions. Protein kinase activity towards MBP was observed in fractions 8-15. The majority of the activity was detected in fractions 11 and 12 (Fig. 18) . The pi of the protein kinase in fraction 12 was 7.05. Recovery of total activity applied to the focusing apparatus in fractions 8-15 was 30%, and 60% of this activity was contained in fractions 11 and 12. An identical phosphorylation profile was observed when the EGFR peptide was used as a substrate. In addition to phosphorylated MBP, a faint band of 97 kD was observed in fraction 12. This phosphorylated band became even more obvious when a large sample volume was applied to an SDS-polyacrylamide gel and its appearance was independent of the presence of exogenous substrate (Fig. 19) .

5. Renaturation of the protein kinase activity isolated by isoelectric focusing To determine whether the 97 kD phosphorylated protein isolated by IEF might represent an autophosphorylating kinase, subcellular fractions were first separated by SDS-PAGE and then renaturation of kinase activity was attempted according to the method described for calmodulin-dependent protein kinase II [86] . It was reasoned that if the 97 kD protein was an auto- phosphorylating kinase, its activity might be reconstituted by this procedure. Alternatively, if the 97 kD protein was a membrane protein that co-focused by IEF with the kinase that recognized MBP and also served as its substrate, it should not be detected, since potential substrates would be physically separated from the kinase by SDS-PAGE prior to renaturation and initiation of phosphorylation. For these studies, samples of intact membrane, flow-through fraction from DE52 anion exchange chromatography of solubilized membrane, and of cytosol were compared. Fig. 20 shows that a single band at 97 kD was observed in intact and fractionated membrane samples. In contrast, 4 bands appeared in the cytosolic sample, with Mr values of 57, 44, 42 and 34 kD. The 44/42 kD bands correspond to the molecular weights reported for MAP kinases in these cells. Polymerization of MBP or of the EGFR peptide into the gel did not affect phosphorylation. These results indicate that the 97 kD protein is an autophosphorylating protein kinase. Phosphoamino acid analysis showed that phosphorylation occurred on serine residues. It should be noted that the small amount of kinase activity toward MBP that eluted from the DE-52 column from 0.2-0.4 M NaCl renatured as a 57 kD protein.

6. Ceramide enhances the kinase activity detected by renaturation To demonstrate activation of the kinase by ceramide,

HL-60 cells were treated at 37°C with C8-ceramide for 15 min prior to preparation of membranes and SDS-PAGE. Fig.

21 shows that treatment of intact cells with ceramide

(0.03 to 3 mM) induced a dose-dependent enhancement of auto-phosphorylation of the 97 kD protein with a maximal effect at 0.3 mM. At this concentration, ceramide enhanced autophosphorylation four-fold. Incorporation of ceramide into the gel or addition of ceramide to the kinase reaction mixture during in situ phosphorylation did not enhance kinase activity.

Alternatively, microsomal membranes were used to assess the effect of ceramide on autophosphorylation of the 97 kD protein kinase. For these studies, membranes were incubated with ceramide, subjected to SDS-PAGE and then denaturation- renaturation was performed. Under these conditions, ceramide enhanced autophosphorylation to the same extent as when added to intact cells. The concentration of ATP in the membranes was below the limit of 7 mM detected by the hexokinase assay. Moreover, the enhancement of autophosphorylation by ceramide was not affected by pre-treatment of the membrane fraction with hexokinase to deplete traces of ATP. These studies provide additional evidence that the membrane-bound 97 kD protein was a ceramide-activated, autophosphorylating protein kinase. Furthermore, these investigations show that activation of the kinase by ceramide was independent of ATP.

Addition of ceramide to the DE52 flow-through fraction resulted in a small (1.5-fold) increase in phosphorylation of the EGFR peptide under Vmax conditions. It should be noted that solubilization exposes membrane protein to the entirety of membrane constituents and that many of these components, including much of the ceramide, co-purify with the kinase even through isoelectric focusing. Alternately, it is possible that a target other than ceramide-activated protein kinase is the direct target for ceramide action.

7. TNFo: enhances the autophosphorylating kinase activity detected by renaturation

To demonstrate stimulation of the 97 kD kinase by TNEα, HL-60 cells were treated with TNFo; for 20 min prior to preparation of membranes, SDS-PAGE and renaturation of kinase activity. Renatured kinase activity exhibited time-dependent autophosphorylation. Enhancement of autophosphorylation in response to TNFo; (1 nM) , was detectable at 2 min and reached four-fold of control at 15 min (Fig. 22) . Treatment of intact cells with TNFo; (0.1 to 10 nM) induced a dose-dependent enhancement of auto-phosphorylation of the 97 kD protein with a maximal effect at 10 nM. This is the same range of TNF concentrations which have been shown to activate the sphingomyelin pathway in human leukemia (HL-60) cells [59] and stimulate differentiation of these cells into monocytes [6] . Phospho-amino acid analysis showed that kinase autophosphorylation occurred exclusively on serine residues in TNF-stimulated cells as in unstimulated cells .

8. TNF enhances renatured kinase activity toward the exogenous substrate, MBP To demonstrate that the enhancement of autophosphorylation by TNFo; represents a general increase in kinase activity, gel slices corresponding to 97 kD were excised and incubated with the exogenous substrate MBP in kinase reaction mixture containing [γ-³²P]ATP [89] . After termination of reactions, samples were electrophoresed and visualized by autoradiography. Fig. 23 shows that stimulation of cells with TNFo; resulted in seven-fold enhancement of MBP phosphorylation by the 97 kD kinase. Similar results were obtained with ceramide-treated cells. Enhancement of MBP phosphorylation was detectable at 10 min and maximal at

1 h. A brief period of autophosphorylation

(approximately 10 min) markedly enhanced the rate of utilization of exogenous substrate. MBP was not phosphorylated by incubation with a gel slice from the region corresponding to 75-90 kD, which would contain PKC.

E. Discussion The present studies identify a 97 kD protein kinase as ceramide-activated protein kinase. Kinase activity was quantitatively extracted from microsomal membrane with Triton X-100 and separated from PKC by anion exchange chromatography. Isoelectric focusing showed that the kinase resolved as a single peak with a pi of 7.05. Co-purification of the kinase activity toward MBP and the EGFR peptide provides further support for the notion that this activity is proline-directed. Like other members of the proline-directed family of protein kinases, ceramide- activated protein kinase undergoes autophosphorylation. Reconstitution of the kinase activity by denaturation- renaturation demonstrated that ceramide-activated protein kinase was a single band of 97 kD and that it was exclusively membrane-bound. Ceramide and TNFo; were observed to enhance the kinase activity whether added to intact cells or to a membrane fraction in the absence of ATP. Furthermore, TNFα and ceramide enhanced both autophospho-rylation and phosphorylation of the exogenous substrate MBP.

The 97 kD ceramide-activated protein kinase appears to be a member of the proline-directed class of protein kinases. All other proline-directed protein kinases are either cytosolic or nuclear with molecular weights between 34-62 kD [85, 87] whereas ceramide-activated protein kinase is membrane-bound with a molecular weight of 97 kD. Hence by at least these two criteria, subcellular localization and size, ceramide-activated protein kinase appears to be a novel member of the family of proline-directed protein kinases. Further, ceramide-activated protein kinase represents a new lipid-activated protein kinase involved in signal transduction and its role in the sphingomyelin pathway appears analogous to that of PKC in the phosphoinositide pathway.

V- Bacterial Lipopolysaccharide has Structural

Similarity to Ceramide and Stimulates Ceramide- activated Protein Kinase in Myeloid Cells

A. Summary Bacterial lipopolysaccharide (LPS) , tumor necrosis factor (TNF) -o; and interleukin-1/S (IL-ljβ) stimulate similar cellular responses. TNF-α; and IL-13 initiate signaling through a pathway involving hydrolysis of sphingomyelin to ceramide. In this system, ceramide acts as a second messenger stimulating a ceramide-activated serine/threonine protein kinase. The present studies demonstrate the LPS, like TNF and IL-1, stimulates ceramide-activated protein kinase activity in human leukemia (HL-60) cells and in freshly isolated human neutrophils. Lipid A, the biologically active core of LPS, enhanced kinase activity in a time- and concentration-dependent matter. As little as 10 nM lipid A was effective and a maximal effect occurred with 500 nM lipid A, increasing kinase activity 5-fold. Native LPS similarly induced kinase activation. This effect of LPS was markedly enhanced by LPS binding protein (LBP) and required the LPS receptor CD14. In contrast to TNF and IL-1, LPS does not cause sphingomyelin hydrolysis and thus stimulates ceramide-activated protein kinase without generating ceramide. Molecular modeling showed strong structural similarity between ceramide and a region of lipid A. Bases on these observations, it is proposed that LPS stimulates cells by mimicking the second messenger function of ceramide.

B. Introduction TNF, IL-1 and LPS initiate a common spectrum of cellular activities associated with the inflammatory response. (The abbreviations uses herein are: LPS, lipopolysaccharide; TNF, tumor necrosis factor; IL-lβ, interleukin-lβ; LBP, LPS binding protein; MAP, microtubule-associated protein; NF, nuclear factor; MBP, myelin basic protein; CAP, ceramide-activated protein) .

These activities include induction of adhesion molecules

(E-selectin, ICAM-1, and VCAM-1) on endothelium [91,

92], integrin-mediated adhesion of neutrophils [93, 94], and cytokine synthesis in mononuclear cells [95] . These three stimuli appear to use a common set of kinases, transcription factors, and promoter elements to provoke these responses. An early event common to all three stimuli is phosphorylation and activation of microtubule- associated protein (MAP) -2 kinases [96-99] . MAP kinases are proline-directed serine/threonine protein kinases that serve as intermediaries in numerous signaling cascades from the cell surface [100] . TNF, IL-1 and LPS also activate NF-/.B, a factor that promotes transcription of a large family of genes. NF-KB exists in the cytoplasm of many cells complexed to an inhibitor, IκB [101-104] . Treatment of cells with TNF, IL-1 and LPS lead to proteolytic degradation of IKB-Q; [105] and the release of NF-/.B. NF-KB then translocates to the nucleus and binds its cognate DNA sequence on responsive genes

[101-104, 106-108] . /.B-like motifs are found in the TNF- 0! promoter and the HIV long terminal repeat and are activated by TNF, IL-1 and LPS [109, 110] . Thus, these agents stimulate a common set of early events in sensitive cells.

These early events stimulated by TNF and IL-1 are likely mediated through generation of ceramide. In this regard, ligation of the TNF and IL-1 receptors results within seconds in ceramide generation, and elevation of cellular ceramide levels with ceramide analogs or exogenous sphingomyelinase mimics cytokine action. Ceramide may utilize a serine/threonine kinase to initiate these evens

[110-112] . Ceramide-activated protein kinase is a membrane-bound, proline-directed protein kinase that recognizes the minimal amino acid motif, Leu-Thr-Pro [114] . Ceramide-activate protein kinase has been solubilized from HL-60 cell membranes, partially purified, and renatured [115] as a band of 97 kDa. Renatured kinase autophosphorylates on serine residues, and autophosphorylated kinase recognizes a generic substrate, for proline-directed kinases, myelin basic protein (MBP) . Both autophosphorylation and phosphorylation of MBP are enhanced 5-10 fold by treatment of intact cells with TNF-α; or ceramide, consistent with the proposition that this kinase is involved in signal transmission. Further, kinase activation appears specific for ceramide as generation of other lipid second messengers such as arachidonic acid, 1, 2, -diacylglycerol or phosphatidic acid failed to enhance kinase activity [111, 112] .

The similarity of actions of TNF, IL-1 and LPS suggests that some effects of LPS may be mediated by ceramide- activate protein kinase. LPS is a membrane-forming phospholipid expressed on the surface of gram-negative bacteria. Purified LPS provokes profound responses including septic shock, an often fatal consequence of bacterial infection. All of the biological activity of LPS resides in a highly conserved portion of the molecule known as lipid A. LPS stimulates cells by binding stoichiometrically to CD14 [116] , a receptor expressed on monocytes and polymorphonuclear leukocytes. Spontaneous diffusion of LPS to CD14 is a slow process, and efficient binding requires a serum factor such as lipopolysaccharide binding protein (LBP) to catalyze this reaction. Binding to CD14 is followed by activation of MAP kinase [99, 117] and NF-KB [118] , but the molecules coupling LPS to these responses have not been described.

C. Experimental Procedures

1. Materials

Buffers, lipids (phosphatidylserine and diolein) , insulin transferrin, leupeptin, soybean trypsin inhibitor, bovine myelin basic protein (MBP) , LPS (Salmonella typhosa) , and lipid A (Escherichia Coli) were purchased from Sigma Chemical Co. HPLC grade solvents were from Fisher Scientific. [γ-³²P]ATP (300 Ci/mmol) was from NEN/Du Pont. MRF34 autoradiographic film was from Cronex, DuPont.

2. Cell culture

HL-60 cells were grown in suspension culture in RPMI 1640 supplemented with 10% fetal bovine serum, penicillin

(10,000 units ml^"1), streptomycin (10,000 units ml.i) , serine (16μg ml^"1) , asparagine (8.4 μg ml^"1) , and glutamine ( 16 . 7 μg ml^"1) .

3. Stimulation of ceramide-activated protein kinase On the day of an experiment, cells were resuspended (1 x 10⁶ cells ml"¹) into serum-free RPMI 1640 containing 5 μg ml^"1 insulin and transferrin. After 2 h, cells were stimulated with lipid A or diluent (DMSO, <0.01%) . For isolation of microsomal membranes [111,112], cells were resuspended into homogenizing buffer (25mM HEPES, pH 7.4, 5 mM EGTA, 30 mM NaF, and 10 μg ml-2 each of leupeptin and soybean trypsin inhibitor) , disrupted with a tight fitting Dounce homogenizer, and the homogenate was centrifuged at 500 x g for 5 min to remove cell debris and nuclei. The postnuclear supernate was centrifuged at 200,000 x g for 30 min and microsomal membranes were resuspended (1.5 μg ml^"1) into homogenizing buffer. Ceramide-activated protein kinase was detected by renaturation and autophosphorylation. Briefly, membrane proteins (200 μg per lane) were separated by SDS-PAGE

(10%) , and the gel was washed with two changes of buffer

(50mM Tris, pH 7.4, 5mM 2-mercaptoethanol) containing 20%

2-propanol at room temperature for 1 h, and once in buffer without 2-propanol for 1 h. Denaturation was accomplished by incubation of the gel in two changes of 6M guanidinium HCl in wash buffer for 1 h each. Renaturation was accomplished by incubation of the gel overnight at 4°C in wash buffer containing 0.04% Tween- 20. The gel was then equilibrated for 10 min at room temperature in kinase reaction mixture (25mM HEPES, pH

7.4, 10 mM MgCl₂, 0.5 mM EGTA and 5mM NaF) and [γ0³²P] ATP

(50 μM final concentration; 1000 dpm pmol^"10.

Autophosphorylation was terminated by removal of the reaction mixture. The gel was washed with 6 changes of buffer (5% trichloracetic acid, 1% sodium pyrophosphate) for 2 h and subjected to autoradiography. For studies involving activation of ceramide-activated protein kinase by LPS, HL-60 cells (13 x lO^l"¹) were handled as above, and LPS (Salmonella typhosa, 50 ng ml" ^x) , recombinant LBP (1.7 μg ml^"1) or both LPS and LBP were added for the times indicated. Isolation of microsomal membranes and autophosphorylation of ceramide-activated protein kinase were performed as• above.

4. Measurement of ceramide-activated protein kinase enzymatic activity toward MBP

Membrane proteins (200μg) from treated and untreated cells were subjected to SDS-PAGE and kinase activity was renatured as above. Autophosphorylation was allowed to proceed for 10 min and the gel was washed for 20 min with four changes of 50 mM HEPES buffer, pH 7.4. Gel slices were then excised from regions corresponding to 97kDa, crushed, and incubated for 1 h with 10 μ MBP (5 mg ml"¹) and 40 μl kinase reaction mixture in the presence of [γ^" ³²P]ATP (50 μM final concentration) . Reactions were terminated by addition of 10 μl Laemmli buffer, boiling for 3 min, and centrifugation of gel particles. The supernates were subjected to electrophoresis and autoradiography.

5. Molecular modeling studies

Molecular modeling of lipid A and ceramide was performed using the SYBYL (version 6.03) molecular modeling program

(Tripos Associates, Inc.) implemented on a Silicon

Graphic Personal Iris 4D/35. The structures are based on energy minimization calculations using the tripos force field, a molecular mechanics method, _^and conformational analysis in search of global minima.

D. Results and Discussion Initial studies were designed to test the effect of lipid A on ceramide-activated protein kinase activity. HL-60 cells were treated at 37°C with lipid A and microsomal 81

Untreated cells contained 100 ± 3 pmol ceramide 10⁶ cells" ². This level was unaffected for up to 15 min by incubation with LBP and LPS. This observation indicates that LPS stimulates ceramide-activated protein kinase independent of the generation of ceramide. Additional assays showed that the preparations of LPS and lipid A used in these studies did not contain detectable ceramide contamination.

Since LPS stimulates a ceramide-activated protein kinase in the absence of the generation of ceramide, the possibility that LPS chemically resembles ceramide was explored. A portion of the reducing end of the lipid A molecule closely resembles a protein of ceramide. Recent studies using synthetic analogs of both LPS and ceramide have shown that this precise region is conserved in all biologically active LPS and ceramide analogs, and that nearly all other portions of the molecules can be removed or altered without destroying the ability to stimulate cells. Carbons 1, 2 and 3 of LPS are normally part of a pyranose ring which is in turn connected to the non- reducing acylated sugar, but neither the pyranose ring nor the nonreducing acylated sugar are needed for biological activity. Acyclic derivatives of lipid A in which the reducing acylated sugar is replaced with a linear, acylated carbon chain [123] retain biological activity, and the nonreducing acylated sugar may be removed and activity is retained if an additional fatty acid is esterified to carbon 4 [124, 125] . In LPS, carbon 1 may bear phosphate, phosphonooxyethyl [126] , phosphonate [127] , CH₂COOH [128] or OOH [123] and retain activity, while carbon 1 of ceramide may bear a hydrogen atom, hydroxyl group [129] or a phosphate group [114] and retain activity. Carbon 3 of ceramide generally bears a 15 carbon chain alkyl tail and a hydroxl group, but the alkyl tail may be replaced with a phenyl group [129] of the hydroxl group replaced with a hydrogen atom [130] 82 without loss of activity. Carbon 3 of LPS bears an esterified fatty acid, but this substituent may be removed [123] with modest reduction of biological activity. Attention was focused on portions of the molecules as a possible "core region" that participates in stimulating cells. Consistent with this view, the amide-linked fatty acid on carbon 2 of ceramide analogs appears essential for activity [129] , and the optimal chain length of the fatty acid is 14 carbon atoms. Analogs with alterations at carbon 2 of LPS have not been prepared, but nearly all active species of LPS bear a 14 carbon fatty acid at this position. No analogs of either ceramide or LPS have been prepared in which carbons 1, 2, or 3 were deleted or altered, thus precluding further comparison.

Certain residues are nearly identical in formal structure. Molecular modeling studies were therefore undertaken to more closely determine their three- dimensional resemblance. The molecular structures of the reducing glucosamine of lipid A (GlcN-1, dephosphorylated form) and ceramide were obtained using molecular mechanics and by global conformational analysis. The results for lipid A are similar to those published by Kastowsky et al. [131] . The relative positions of C-l, C-2 and C-3 are nearly identical for GlcN-1 and ceramide. The molecular modeling was also carried out with the 1- phosphate present on each lipid, and demonstrated the same molecular similarity. In contrast, comparison of a model for 1,2- diacylglycerol generated using molecular mechanics with ceramide or GlcN-1 yielded far less similarity. Thus, overlay of either one or both fatty acyl chains, and carbons 1, 2 and 3, could not be simultaneously achieved in low energy conformations.

The chemical structures of lipid A and ceramide are summarized as follows. Carbon atoms 2 and 3 are 79 membranes were prepared. Ceramide-activated protein kinase was detected by measuring autophosphorylation after SDS-PAGE and renaturation of kinase activity. Fig. 24, panel A shows that enhancement of autophosphorylation in response to lipid A (5 μM) was detected at 5 min and was demonstratable for 60 min. In studies designed to assess very early kinetics, an increase in autophosphorylation of ceramide-activated protein kinase could be detected as early as 30 s after treatment of cells with lipid A. Rapid stimulation of ceramide- activated protein kinase precludes the possibility that synthesis of cytokines in response to LPS mediates kinase activation. As little as 10 nM of lipid A was effective and 500 nM induced a maximal activation of the kinase to 5-fold of control (Fig. 24, panel B) . This effect of lipid A was quantitatively similar to that induced by ceramide and TNF in HL-60 cells.

Since prior studies correlated enhanced autophosphorylation of ceramide-activated protein kinase with increased kinase activity toward exogenous substrate, the effect of kinase activity toward MBP was examined after treatment of HL-60 cells with lipid A. For these studies, gel slices corresponding to ceramide- activated protein kinase were subjected to autophosphorylation for 10 min in buffer containing ATP (50 μM final concentration) , and then were washed and incubated with MBP in a kinase reaction mixture containing [γ-³²P]ATP [119] . After termination of reactions, samples were subjected electrophoresis and visualized by autoradiography. Fig. 25 shows that stimulation of cells with lipid A resulted in 5-fold enhancement of MBP phosphorylation by ceramide-activated protein kinase. Enhancement of MBP phosphorylation was detectable at 0.5 min of lipid A treatment and maximal at 15 min. 80

Additional studies showed that stimulation of ceramide- activated protein kinase by LPS exhibits the same requirements as stimulation of other biological effects of LPS. Responses of cells to low doses of LPS are dramatically enhanced by the addition of LBP [120] , which catalyzes binding of LPS to CD14. Similarly, 50 ng/ml LPS caused negligible activation of ceramide-activated protein in HL-60 cells, but addition of LBP enabled a strong response (Fig. 26, panel A) . Enhanced autophosphorylation of ceramide-activated protein kinase was evident by 2 min of treatment with LPS and LBP and persisted for at least 15 min (Fig. 26, panel B) . Studies performed with freshly isolated human neutrophils showed similar results.

Biological responses to LPS require CD14 and are blocked by anti-CD14 mAb 3C10 [116] . Flow cytometry revealed that the HL-60 cells used in these experiments showed uniform low expression of CD14 (mean fluorescent intensity 21.4 in cells stained with anti-CD14 vs. 6.6 in unstained cells) , consistent with previous findings

[121] . Further, addition of anti-CD14 mAb 3C10 (10 μg/ml) 15 min prior to stimulation with LPS (50 ng/ml) and LBP (1.7μg/ml) resulted in an 82% inhibition of LPS- induced autophosphorylation of ceramide-activated protein kinase. In contrast, addition of anti-CD18 mAb IB4 [122] did not affect LPS-induced autophosphorylation. Thus, stimulation of ceramide-activated protein kinase by LPS is mediated by CD14.

TNF and IL-1 stimulate production of ceramide which then enhances ceramide-activated protein kinase activity. It was thus asked whether LPS also stimulates production of ceramide. For these investigation, HL-60 cells were incubated with 50 ng/ml LPS and 1.7 μg/ml LBP for varying times from 0.5 to 15 min, and ceramide levels were determined by the DG kinase reaction as described [112] . asymmetric in both lipid A and ceramide, with the absolute configuration identical at carbon 2 and opposite at carbon 3. The configurations at carbon 3 are considered opposite because the oxygen at carbon 3 of lipid A is positioned opposite from the oxygen in ceramide. However, the long carbon chains attached to carbon 3 are identically placed on lipid A and ceramide.

In conclusion, LPS and ceramide initiate similar effects in cells and these lipids show similarity of structure. Further, these lipids both originate on the outside of the cell, LPS from extracellular micelles and ceramide from sphingomyelin on the outer leaflet of the plasma membrane, and both rapidly stimulate a common membrane- bound target, ceramide-activated protein kinase. The topography of ceramide-activated protein kinase in the membrane is not currently known, but the inability of large LPS molecules to cross the bilayer suggests an interaction site for lipids at the outer membrane surface. It is suggested that LPS provokes cellular responses by mimicking the second messenger function of ceramide. It is further suggested that stimulation of ceramide-activated protein kinase represents and important early event in cellular responses to LPS, and as such represents a novel target for pharmacologic intervention on the treatment of septicemia.

Despite the close resemblance of LPS and ceramide, these lipids show important distinctions in mode of action. Responses to LPS require CD14, but cells lacking CD14 such as L929 fibrosarcoma cells, and Swiss 3T3 and human dermal fibroblasts respond well to ceramide [114] . Biologically active LPS molecules must contain not only the "core" region of similarity to ceramide but additional structures, usually a second acylated glucosamine. These distinctions may arise from the fact that ceramide is generated in cell membranes by the action of a sphingomyelinase, whereas LPS originates outside the cell and must be transported by proteins that may confer additional specificities.

VI. Ceramide-activated Protein Kinase is a Raf Kinase

A. Abstract

Kinases that phosphorylate and activate Raf presumably exist, although they need identification. Here, we show ceramide-activated protein kinase is a Raf kinase. In vitro, ceramide-activated protein kinase phosphorylated Raf-1 on τhr^269, increasing its activity toward MEK. In intact HL-60 cells, ceramide-activated protein kinase complexes tightly with Raf-1, and in response to TNF and ceramide analogs phosphorylates and activates Raf-1. These investigations identify ceramide-activated protein kinase as a link between the TNF receptor and Raf-1.

B. Background Raf-1 (c-Raf) is a Ser/Thr kinase that is ubiquitously present in mammalian cells [132, 133] . Raf-1 is upstream in a cascade of protein kinases that link some cell surface receptors through to the cellular interior. Raf- 1 directly phosphorylates and activates MEK (MAP or ERK Kinase) , which in turn phosphorylates and activates MAP kinase (also known as extracellular signal-regulated protein kinase or ERK) [134-136] . In resting cells, Raf- 1 is inactive and localizes to the cytoplasm. Upon cellular stimulation, Raf-1 interacts with the GTP-bound form of Ras, translocates to the plasma membrane, and is activated [137, 138] . Evidence suggests that the primary role of Ras in this process is to recruit Raf-1 to the membrane. ^" This is based on studies which show that^" binding of Raf-1 to Ras fails to activate Raf in vi tro [137] and that targeting of Raf-1 to membranes by addition of a membrane-localization signal allows Raf-1 activation independent of Ras [138-139] . Although the mechanism by which membrane-bound Raf-1 becomes active is at present uncertain, evidence suggests that regulation of the kinase activity of Raf-1 may involve its phosphorylation. There are numerous reports showing mitogens induce rapid phosphorylation of Raf-1 and stimulation of its kinase activity [132-133] . Although a low incidence of tyrosine phosphorylation is observed in these instances, the majority of phosphorylation is on serine residues with lesser amounts on threonine residues. Further, when activated Raf-1 from insulin-stimulated cells was treated with a serine- specific phosphatase, the majority of its kinase activity was abolished, confirming that serine phosphorylation mediates kinase activation [140] . Raf-1 was also activated by tyrosine phosphorylation in vitro through the platelet-derived growth factor (PDGF) receptor, and in this instance was inactivated by a tyrosine-specific phosphatase [132, 133, 141] . Raf-1 phosphorylation may also be inhibitory as it has been shown that elevation of the level of cAMP results in phosphorylation of Raf-1 on Ser 43 and prevention of Raf-1 activation [142, 143] .

Morrison and co-workers recently mapped the sites of Raf- 1 phosphorylation in resting and PDGF-stimulated Balb/3T3 cells and human skin fibroblasts, and in Sf9 insect cells co-expressing human Raf-1 and activated PDGF receptors [144] . These studies showed that Ser259 and Ser621 are phosphorylated in vivo and that phosphorylation of these sites regulates the kinase activity of Raf-1. Investigations by Kolch et al. [145] showed that PKCa may phosphorylate Raf-1 on Ser499 and enhance its activity. However, Raf-1 can be activated normally in many cells depleted of PKC [146-147] and it has been suggested that in most instances a protein kinase other than PKC is most likely involved in phosphorylation and activation of Raf-1 at the plasma membrane [149] . Candidate kinases capable of performing this function have yet to be identified. The present study tests the hypothesis that ceramide-activated protein (CAP) kinase may serve as a Raf-1 kinase.

5 CAP kinase is a central kinase in the recently described sphingomyelin signal transduction pathway that mediates the action of cytokines such as TNFa and interleukin-lb [150-153] . This pathway is initiated by hydrolysis of sphingomyelin to ceramide in the plasma membrane by the

10 action of a sphingomyelinase, a sphingomyelin specific form of phospholipase C. Ceramide acts as second messenger stimulating a number of targets including CAP kinase [154] . CAP kinase is a member of an emerging family of proline-directed Ser/Thr protein kinases that

15 recognize Ser/Thr phosphoacceptor sites which are amino- terminal to a proline residue. CAP kinase is distinguished from other proline-directed protein kinases by being exclusively membrane-bound and by its ability to recognize the minimal substrate sequences -L-T-P- and -T-

20. L-P- [155] . CAP kinase activity can be assessed after renaturation in SDS polyacrylamide gels by demonstrating its ability to undergo either autophosphorylation or by phosphorylation of exogenous substrates such as myelin basic protein. Treatment of cells with either TNF,

25 cell-permeable ceramide analogs or with exogenous sphingomyelinase to generate an endogenous ceramide load enhance CAP kinase activity 5-10 fold [154] .

Preliminary evidence suggests that Raf-1 may be involved 30 in signal transduction through the sphingomyelin pathway. Recent investigations from a number of groups showed that TNF induces rapid phosphorylation and activation of MAP kinases [156-157] . Raines et al. [157] provided evidence that TNF-induced p42 MAP kinase activation was mediated 35 by ceramide generation, since these effects were mimicked by treatment of cells with exogenous sphingomyelinase and synthetic ceramide analogs. Additionally, transfection with dominant negative Raf-l abolished TNF-induced activation of nuclear factor kB and HIV replication

[159,160], events ascribed to ceramide generation [161-

163] . Raf-1 contains a number of Ser/Thr residues in the amino-terminal regulatory domain and in the carboxyl- terminus that conform to proline-directed sites that might be recognized by CAP kinase [144] . In the present study, we demonstrate that Raf-1 is a component of the sphingomyelin pathway. Signaling through Raf-1 involves formation of a complex containing Raf-1 and CAP kinase, and the phosphorylation of Raf-1 by CAP kinase.

C. CAP kinase phosphorylates and activates Raf-1 in vi tro To investigate whether Raf-1 can be phosphorylated by CAP kinase in vitro, CAP kinase from HL-60 cells was resolved by SDS-PAGE and renatured as a 97 kDa protein as described [154] . Prior investigations demonstrated that CAP kinase was the only kinase to renature from HL-60 cell membranes and that this activity was exclusively membrane-bound, since no 97 kDa activity could be renatured from cytosolic fractions. To determine whether Raf-1 was a substrate for CAP kinase, Raf-1 protein was immunoprecipitated with anti-Raf-1 antibody-conjugated Sepharose beads from a lysate of insect Sf9 cells that co-expresses human Raf-1, p21ras and activated pp60src proteins as reported previously [144] . Raf-1 was then incubated with gel slices containing renatured CAP kinase in the presence of a reaction buffer containing [g- ³²P]ATP. Conditions for CAP kinase activity were optimized previously using myelin basic protein (MBP) or a peptide derived from the amino acid sequence surrounding Thr669 of the epidermal growth factor receptor as substrates [154] . Minimal autophosphorylation of Raf-1 could be detected in the absence of CAP kinase (Fig. 27A) , but Raf-1 phosphorylation was markedly enhanced by CAP kinase. Phosphorylation of Raf-1 by renatured CAP kinase was linear for 30 min under the conditions employed. A preparation of CAP kinase purified to homogeneity from bovine brain and renatured as above yielded similar results. CAP kinase activity toward Raf-1 was TNF- dependent. If CAP kinase was obtained from TNF- stimulated HL-60 cells, Raf-1 phosphorylation was enhanced 4-5 fold (Fig. 27B) . Similar results were obtained if CAP kinase was derived from ceramide (25 mM) - or sphingomyelinase (10 mU/ml) -treated cells (Figure 27C) or when a FLAG-tagged Raf-1 was used as substrate. These studies demonstrate that Raf-1 can serve as a substrate for CAP kinase in vitro and that CAP kinase activity towards Raf-1 is increased by TNF stimulation.

To investigate whether Raf-1 phosphorylation by CAP kinase leads to Raf-1 activation, the kinase activity of Raf-1 was monitored using MEKl as substrate (Fig. 28) . For these studies, Raf-1 was first phosphorylated by CAP kinase for 30 min using unlabeled ATP, and then MEKl and [g-³²P]ATP were added to the reaction mixture. Raf-1, pretreated by CAP kinase, was 4-fold more active in phosphorylating MEKl than untreated Raf-1, indicating that phosphorylation of Raf-1 by CAP kinase enhanced its kinase activity (Fig. 28A) . Control experiments showed that CAP kinase did not phosphorylate MEKl directly (Fig. 28B) . Raf-1, singly expressed in Sf9 cells, possessed no intrinsic kinase activity, and was neither a substrate for, nor activated by, CAP kinase, consistent with the notion that phosphorylation by CAP kinase enhances Raf-1 activation.

Additional investigations evaluated whether phosphorylation of Raf-1 by CAP kinase represents a physiologic mechanism for activation of the MAP kinase cascade. For these studies, we reconstituted the entire MAP kinase cascade from CAP kinase to MAP kinase in vitro with purified reagents (Figure 28C) . As in prior investigations [149] , Raf-1 and MEKl together increased MAP kinase phosphorylation and enhanced MAP kinase activity 5-fold. Addition of CAP kinase to these incubations induced a marked further effect, increasing MAP kinase phosphorylation and activity to 30-fold of control. The effect of CAP kinase on MAP kinase activation was indirect and required Raf-1, as CAP kinase failed to activate MEKl or MAP kinase directly. Further, dephosphorylation of CAP kinase-treated Raf-1 with potato acid phosphatase resulted in abolition of

MEKl phosphorylation and signaling through to MAP kinase.

These investigations provide substantive evidence that signaling of Raf-1 activation through CAP kinase is physiologic.

D. Mapping of the site on Raf-1 phosphorylated by CAP kinase To determine the site on Raf-1 which is phosphorylated by CAP kinase, FLAG/Raf-1 was phosphorylated by CAP kinase in the presence of a reaction buffer containing [g-³²P]ATP as described above. Phosphorylated FLAG/Raf-1 was subsequently digested with trypsin and the tryptic phosphopeptides were separated using a C₁₈ reverse-phase HPLC column. The profile of the radioactivity released from the C₁₈ column revealed the presence of one major peak detected in fractions 28 and 29 (Fig. 29A) . To determine the exact residue phosphorylated, the ³²P- labeled phosphopeptide isolated in fraction 29 was subjected to Edman degradation and phosphoamino acid analysis (Fig. 29B) . Phosphoamino acid analysis revealed that CAP kinase phosphorylated Raf_-1 exclusively on threonine residues (Fig. 29B right hand panel) . Edman degradation of the peptide showed that the radioactivity was recovered in cycles 12 and 13 on Thr268 and Thr269 (Fig. 29B left hand panel) . Based on the obligate losses that occur during each progressive cycle of Edman degradation, and the ratio of counts in Thr268 and Thr269, it would appear that there is a slight preference for Thr269 as the phosphoacceptor site. This site is contained within a -T-L-P- motif, corroborating prior investigations defining this as the preferred recognition site for CAP kinase [164] .

To further provide evidence that Raf-1 served as a substrate for CAP kinase, CAP kinase was used to phosphorylate peptides derived from the amino acid sequence surrounding Thr268 and Thr269 of Raf-1 (amino acids 254-278) . Fig. 29C shows that bovine brain CAP kinase phosphorylated a peptide containing the wild-type Raf-1 sequence TTLP. Phosphoamino acid analysis of the phosphorylated peptide revealed that phosphorylation occurred exclusively on threonine residues. In contrast, CAP kinase failed to phosphorylate a peptide in which Thr268 and 269 were substituted with alanine residues, generating the site AALP. Additional studies were performed using peptides with alanine substituted for either Thr268 or Thr269, generating the sites ATLP and TALP. respectively. These studies showed that replacement of Thr269 with alanine also abolished phosphorylation on Thr268, whereas replacement of Thr268 did not affect phosphorylation of Thr269. Hence, the availability of Thr269 for phosphorylation appears requisite for Thr268 phosphorylation by CAP kinase. Identical results were obtained using CAP kinase from HL-60 cells.

To further clarify the relevance of phosphorylation of Thr268 and 269 to signaling through βa.f-1 , a mutant of Raf-1 was used in which these sites were substituted with valine residues. This mutant was triply expressed with p21ras and activated src in Sf9 cells. Although this mutant retained activity toward MEK, it was not a substrate for CAP kinase. Further, it did not support CAP kinase-induced activation of the MAP kinase cascade in vi tro (Figure 29D) . This study provides additional support for a physiologic role of Thr268 and 269 phosphorylation in Raf-1 activation by CAP kinase.

E. TNF and ceramide induce phosphorylation and activation of Raf-1 in intact cells

The rapid activation of MAP kinase in HL-60 cells in response to TNF, C6-ceramide or exogenous sphingomyelinase, suggested that Raf-1 might be a component of the TNF signaling pathway in these cells [157,158] . To examine this possibility, intact HL-60 cells were metabolically labeled with ³²P-orthophosphate for 4 hr followed by stimulation with TNF for the indicated times (Fig. 30A) . Post-nuclear lysates were subsequently prepared from TNF-treated and control cells, and Raf-1 protein immunoprecipitated with anti-Raf-1 antibody-conjugated Sepharose beads. Raf-1 proteins were resolved by SDS-PAGE and autoradiographed. A time course of TNF-induced Raf-1 phosphorylation is shown in Fig. 30A. The level of Raf-1 phosphorylation was increased within seconds of TNF treatment and remained elevated for at least 20 min. Ceramide (25 mM) and sphingomyelinase

(10 mU/ml) similarly induced Raf-1 phosphorylation in intact HL-60 cells. It was not possible to accumulate sufficient phosphorylated Raf-1 in these studies to map the phosphorylation site.

To evaluate whether the phosphorylated Raf-1 possessed increased kinase activity, Raf-1 was immunoprecipitated from TNF-stimulated cells and incubated for 15-45 min in vitro with recombinant MEKl in a reaction buffer containing [g-³²P]ATP (Fig. 30B) . These studies show that TNF treatment enhanced the kinase activity of Raf-1 towards MEKl, its natural substrate, 10-20 fold. Raf-1 derived from ceramide- or sphingomyelinase-treated cells possessed similarly enhanced activity toward MEKl (Fig. 30C) . Dephosphorylation of immunoprecipitated Raf-1 with potato acid phosphatase abolished the enhanced activity toward MEKl. These studies indicate that Raf-1 is a component of the sphingomyelin pathway mediating TNF action.

Studies were performed comparing the effect of TNF to other agents known to activate Raf-1. A concentration of the phorbol ester 12-O-tetradecanoylphorbol 13-acetate (TPA) sufficient to induce macrophage differentiation of HL-60 cells (100 ng/ml) , which did not activate CAP kinase, increased Raf-1 activity to a maximum of -fold of control after 10 min. Similarly, insulin (100 nM) increased Raf-1 activity 4-fold in cells pre-incubated in serum-free medium as described [164] . Granulocyte- macrophage colony stimulating factor (500 pM) , which induces monocytic differentiation of HL-60 cells [165] , only enhanced Raf-1 activity 2.5-fold, whereas PDGF (5 nM) treatment of HL-60 cells, after induction of PDGF receptors with TPA [166] , resulted in a 4-fold increase in Raf-1 activation. Hence, the effect of TNF on Raf-1 activity in HL-60 cells is larger than that of other agonists known to stimulate Raf-1 activity.

F. CAP kinase and Raf-1 exist in complex Since prior studies suggested that Raf-1 might participate in a multi-protein complex [167] , Raf-1 was immunoprecipitated with anti-Raf-1 antibody-conjugated

Sepharose beads from TNF-treated and untreated HL-60 cells. An immune-complex kinase assay was then performed by addition of reaction buffer containing [g-³²P]ATP to

Raf-1 while bound to the beads. Fig. 31 shows a spectrum of proteins immunoprecipitated with Raf-1 that become phosphorylated in vi tro under these conditions (Fig.

31A) . Other than Raf-1, these proteins were not directly recognized by the primary anti-Raf-1 antibody by western blotting (Fig. 3IB) . When immunoprecipitation was performed with non-specific antibody, no phosphorylated bands were observed. TNF treatment of cells resulted in enhanced phosphorylation of numerous proteins within this complex. A band at 97 kDa that was barely detectable in control incubations was readily observed in Raf-1 immunoprecipitates from TNF-stimulated cells.

Studies were also performed to establish whether the 97 kDa protein that immunoprecipitated with Raf-1 was CAP kinase (Fig. 32) . When MBP was added as a substrate in the immunecomplex kinase assay, the complex of proteins precipitated from TNF-treated cells expressed higher activity toward MBP than the proteins derived from unstimulated cells (Fig. 32A) . To determine whether the activity toward MBP resulted from CAP kinase within the immunecomplex or another protein kinase, the proteins contained within the complex were separated by SDS-PAGE and renatured. Subsequently, gel slices corresponding to regions of different molecular weight were assayed for activity toward MBP. Prior investigations [166] demonstrated that multiple kinases, including MAP kinases, could be renatured from cytosol of HL-60 cells under the conditions employed, whereas only CAP kinase renatured from membrane. In this regard, the gel slice containing the 97 kDa CAP kinase contributed the large majority of MBP phosphorylating activity (Fig. 32B) .

G. Experimental Methods.

1. CAP kinase phosphorylates recombinant human Raf-1 in vitro and the level of phosphorylation is enhanced by TNF and ceramide._^

Figure 27A - HL-60 cells were incubated in serum-free

RPMI [RPMI containing 16 mg/ml serine, 8.4 mg/ml asparagine, 16.7 mg/ml glutamine, 25 mM HEPES, pH 7.4 and 0.5 mg/ml each of insulin and transferrin] at 1 x 106 ml"¹ for 2 hr, followed by stimulation with TNF (1 nM) for 20 min. The cells were collected in Homogenizing Buffer [25 mM HEPES, pH 7.4, 5 mM EGTA, 50 mM NaF containing 10 mg/ml of the protease inhibitors, soy bean trypsin inhibitor (SBTI) and leupeptin; 500 x 10⁶ ml^"1] and homogenized using a Bellco drive unit (catalog # 1981- 01900) on setting 6 for 4 min. Centrifugation for 5 min at 700 x g yielded a post-nuclear supernate from which microsomal membranes were prepared by centrifugation at 250,000 x g for 30 min. Microsomal membranes were resuspended into homogenizing ^"buffer and proteins (30 x 10⁶ cell equivalents/lane) were resolved on a 7.5 % SDS- polyacrylamide gel. CAP kinase was renatured as described by Liu et al. [153] . Briefly, the acrylamide gel harboring CAP kinase was incubated for 2 hr in buffer A [50 mM Tris, pH 7.4, 5 mM b-mercaptoethanol] containing 20% isopropanol and washed once in buffer A for 1 hr. Subsequently, the gel was denatured in buffer A containing 6 M guanidine HCl for 2 hr and renatured in buffer A containing 0.04% Tween-20 overnight. The entire procedure was performed at 4°C. The gel slice (1.5 x 5 x 8 mm³) containing the 97 kDa CAP kinase was excised and used for Raf-1 phosphorylation. To immunoprecipitate recombinant human Raf-1, lysates from Sf9 cells coexpressing Raf-1, p21ras, and activated pp60src proteins, which were prepared in RIPA lysis buffer [137 mM NaCl, 20 mM Tris, pH 8.0, 10% glycerol, 1% NP-40, 0.1%. SDS, 0.1% sodium deoxycholate and 10 mg/ml each of SBTI and leupeptin] , were incubated with anti-Raf-1 antibody- conjugated Sepharose beads [144] . Anti-Raf-1 antibody- conjugated Sepharose beads were prepared by incubating 1 ml of rabbit anti-Raf-1 antibody (generated against the last 12 amino acid residues of wild type Raf-1) with protein A Sepharose CL-4B beads (Pharmacia) overnight at 4°C in NP-40 lysis buffer [137 mM NaCl, 20 mM Tris, pH 8.0, 10% glycerol, 1% NP-40] . Antibody-conjugated beads were washed 3 times with NP-40 lysis buffer, and incubated with 200 ml Sf9 cell lysate and 600 ml RIPA lysis buffer to immunoprecipitate Raf-1 protein. The Raf- bound beads were washed 3 times with NP-40 lysis buffer containing 1 mM NaV0₄. To phosphorylate Raf-1, blank or CAP kinase-containing gel slices (equivalent to microsomes from 30 x 10⁶ HL-60 cells) were cut into small pieces (1.5 x 1.5 x 2 mm³) and mixed with Raf-bound beads in a 40 ml reaction mixture containing 30 mM HEPES, pH 7.4, 5 mM MgCl₂, 10 mM MnCl₂, 1 mM dithiothreitol, 5 mM ATP and 20 mCi [g-³²P]ATP. The reaction was terminated after 30 min by the addition of laemmli buffer and boiled for 5 min. Phosphorylated Raf-1 was resolved by 7.5% SDS-PAGE and autoradiographed. Identical results were obtained using CAP kinase renatured from cells stimulated for 5 min with TNF. Figure 27B-27C - Experiments were performed as in Fig. 27A except cells received C8-ceramide (25 mM) or S. aureus sphingomyelinase (Boehringer; 10 mU/ml) .

2. Phosphorylation of recombinant human Raf-1 by CAP kinase in vitro enhances the kinase activity of Raf-1 towards recombinant human MEKl.

Figure 28A - CAP kinase was prepared from TNF-stimulated HL-60 cells (30 x 10⁶/incubation) as described in Fig. 27. Raf-1, immunoprecipitated with anti-Raf-1 antibody- conjugated Sepharose beads, was phosphorylated for 30 min with a gel slice containing renatured CAP kinase by incubation in Raf-1 reaction buffer without radiolabeled ATP. Control reactions (Raf) received blank gel pieces. The kinase activity of Raf-1 was then measured by phosphorylation of purified recombinant human MEKl (0.1 mg per reaction) in 50 ml MEKl reaction buffer [30 mM NaCl, 10 mM MgCl₂, 100 mM ATP and 50 mCi [g-³²P]ATP] . The reaction was terminated at the indicated times by the addition of laemmli buffer and boiled for 5 min. Phosphorylated MEKl was resolved by 10% SDS-PAGE and autoradiographed. MEKl autophosphorylation (MEK auto) was performed for 20 min in the absence of Raf-1 or CAP kinase. This figure represents one of five similar experiments.

Figure 28B- Autophosphorylation of MEK, and phosphorylation by Raf-1 or CAP kinase from TNF-treated cells, were performed for 1 hr as described in Fig. 28A. Figure 28C - CAP kinase was purified to homogeneity from bovine brain using the following procedure: Bovine brain (800 g) was homogenized and a post-nuclear supernate prepared as described in Fig. 27. Thereafter, a "heavy" microsomal membrane fraction enriched in plasma membrane was generated according to the method of Morre et al.

[177] by centrifugation at 43,000 x g for 0.5 hr. This fraction is enriched 10-fold in the plasma membrane marker alkaline phosphodiesterase I (EC 3.1.4.1) and contains virtually all of the cellular CAP kinase. This plasma membrane-enriched fraction was further sub- fractionated over a discontinuous sucrose density gradient. The CAP kinase-enriched fraction was extracted with 1 M KC1, precipitated with ammonium sulfate, eluted from a FPLC hydroxyapatite column with a continuous gradient of phosphate buffer (0.1-0.4 M) , and the fractions containing CAP kinase activity were resolved completely with the use of a Prep Cell (BIO-RAD) .

For reconstitution of the MAP kinase cascade, purified renatured bovine brain CAP kinase or blank gel pieces were incubated with or without recombinant Flag/Raf-1 and MEKl (0.1 μg per reaction) in a buffer containing 40 mM Tris, pH 7.5, 5 mM MgCl₂, 10 mM MnCl₂, 1 mM DTT and 5 μM ATP at 22ΦC. After 30 min, CAP kinase was removed by centrifugation at 10,000g x 5 min. For studies measuring phosphorylation of MAP kinase, agarose-conjugated human GST-MAP kinase (6.25 μg per reaction, UBI, Lake Placid, NY) was added to the supernate in 40 mM Tris, pH 7.5, 10 mM MgCl₂, 10 mM MnCl₂, 30 mM NaCl₂, 50 μM ATP and 50 μCi [g-³²P]ATP and after 20 min, the agarose-conjugated GST- MAP kinase was spun down at 10,000g x 5 min, washed three times in the same buffer without ATP, and resuspended into Laemmli sample buffer. For measurement of MAP kinase activity, experiments were performed as above except MAP kinase was phosphorylated in cold ATP and then incubated with 40 mM Tris, pH 7.5, 10 mM MgCl₂, 30 mM NaCl, 50 μg MBP, 50 μM ATP and 50 μCi [g-³²P] ATP for 20 min. ³²P-labeled MAP kinase and MBP were resolved by 12% SDS-PAGE. Qualitatively similar results were obtained with CAP kinase from HL-60 cells.

FLAG/Raf-1 was provided by Dr. Debbie Morrison (NCI) and synthesized as described [43] . Briefly, to generate the FLAG/Raf-1 construct, sequences encoding the FLAG epitope tag (amino acids DYKDDDDK) were inserted proximal to the amino terminal methionine of Raf-1 by site-directed mutagenesis. The cDNA fragments encoding the FLAG/Raf-1 protein was inserted into the pVL941 baculoviral transfer vector, expressed in Sf9 cells along with p21ras and activated pp60src, and purified from Sf9 lysates using an anti-FLAG affinity resin.

3. Mapping of the site of Raf-1 phosphorylation by CAP kinase. Figure 29A - For separation of tryptic Raf-1 fragments, aliquots of tryptic digests were lowered to pH 2 with 20 % trifluoroacetic acid (TFA) and loaded onto a Waters 3.9 X 300 mm C₁₈ reverse-phase HPLC column. The column was developed with an increasing gradient of acetonitrile in 0.05% TFA. The stepwise gradient at a flow rate of 1 ml/min was 0-40% CH₃CN for 10 min, 40-60% CH₃CN over 10 min, and 60% CH₃CN for 10 min. 1-Min fractions were collected and Cerenkov-counted for ³²P content in a Beckman LS 5801 scintillation counter.

Figure 29B- Semi-automated amino-terminal sequence analysis was performed in a Beckman 890C spinning cup sequencer. 2.5 mg of polybrene (Aldrich Chemical Co.) was applied to the spinning cup along with 120 nmol of the dipeptide Tyr-Glu and subjected to four cycles of Edman degradation. ³²P containing peptide was added in CH₃CN/water along with an equine apomyoglobin carrier (9 nmol) to the spinning cup, dried, and subjected to 20 cycles with no prewashes. Aliquots of each fraction were dried and quantified by liquid scintillation counting. Phosphoamino acid analysis is performed according to the methods described previously [144] . Figure 29C - Purified bovine brain CAP kinase was renatured as described in Fig. 27 and used to phosphorylate Raf-1 peptides. A peptide derived from the amino acid sequence surrounding Thr269 of Raf-1 (amino acids 254-278) containing the wild type sequence TTLP was synthesized using an Applied Biosystems model 431A synthesizer and used as a substrate in the CAP kinase assay. An identical peptide was sequenced with the two threonine residues replaced by alanine to generate the site AALP. These peptides are slightly longer than the natural tryptic peptide from intact Raf-1 corresponding to amino acids 257-275. The reason for extending the peptide was so that the potential CAP kinase phosphorylation site was situated in the middle rather than the carboxyl-terminus. Each of the synthetic peptides (40 mg) were phosphorylated for 30 min by CAP kinase under the conditions described in Fig. 27B. The reactions were terminated by adding 0.5 M ATP in 90% formic acid and the supernates were brought to a final TFA concentration of 20% (v/v) . The phosphorylated peptides were resolved by reverse-phase HPLC as described in 29B using a linear gradient of acetonitrile from 2-60% in 0.1% TFA at a rate of 1%/min with a flow rate of 1 ml/min. Fractions (1 ml) were collected for Cerenkov counting. The data represent one of five identical experiments. Figure 29D - Reconstitution of the MAP kinase cascade was performed as described in Fig. 28D except for the use of mutant FLAG/Raf-1 which was co-expressed in Sf9 cells with p21ras and activated pp60src and contains substitutions of valine for threonine at residues 268 and 269.

4. TNF stimulates Raf-1 phosphorylation and its kinase activity in vivo.

Figure 30A- 300 x 10⁶ cells were resuspended at 37°C in 15 ml of serum-free phosphate-free RPMI medium (1 x 10^s ml^"1) containing 6 mCi ³²P-orthophosphate. After 2 hr, cells were resuspended into the same buffer without radiolabel

(5xl0⁷/point) and stimulated with TNF (1 nM) . At the indicated times, ice-cold serum-free phosphate-free RPMI was added and cells were homogenized as described in Fig.

27 in RIPA lysis buffer. Cell debris was removed by centrifugation at 700 x g for 5 min. Raf-1 protein was immunoprecipitated from the supernate with anti-Raf-1 antibody-conjugated Sepharose beads as described above, washed 4-5 times with 1.5 ml of NP-40 lysis buffer, boiled in laemmli buffer and resolved by 7.5 % SDS-PAGE. After SDS-PAGE, Raf-1 was transferred to an Immobilon PVDF (Millipore) membrane according to the vendor's instructions. An autoradiogram was obtained and a western blot (described below) with anti-Raf-1 antibody was employed to monitor recovery of Raf-1 protein. The data (CPM/Raf protein) are presented as fold of control and represent one of three studies performed in triplicate. Figure ^' 30B - HL-60 cells (30 x 10⁶/incubation) were stimulated with TNF (1 nM) for 20 min, lysed in RIPA buffer, and Raf-1 was immunoprecipitated as in Fig. 30A. MEKl phosphorylation by immunoprecipitated Raf-1 was performed as described in Fig. 28. MEKl autophosphorylation was for 45 min. An autoradiogram of MEKl phosphorylation (top panel) and recovery of MEKl protein by western blot (bottom panel) are shown. For western blot, proteins separated by SDS-PAGE were electrotransferred to an Immobilon PVDF membrane at 12 volts overnight at 4°C. Membranes were then blocked with 2 % BSA in TBS [20 mM Tris, pH 7.6, 137 mM NaCl] for 1 hr and washed with TBST (TBS containing 0.2 % Tween-20) . Membranes were incubated for 1 hr with rabbit anti-MEKl antibody (1:2000 dilution in TBST; Anti-MEKl antibody, generated against MEKl peptide CPKKKPTPIQLNPNPEG-NH2 and washed 3 times for 5 min in TBST, followed by a 1 hr incubation with anti-rabbit IgG antibody (1:20,000 dilution in TBST) . An ECL detection system (Amersham Life Science) was used following the vendor's instructions to develop the western blot. The procedure used for the Raf-1 western blot was identical except for the use of the rabbit anti-Raf-1 antibody described in Fig. 27. The data represent one of three similar experiments. Figure 30C - These studies were performed as in Fig. 30B except cells were stimulated for 20 min with C8-ceramide

(25 mM) or S. aureus sphingomyelinase (10 mU/ml) , and

MEKl was phosphorylated for 30 min in vi tro by immunoprecipitated Raf-1. These data represent one of two similar studies.

5. Raf-1 complexes with 97kDa kinase Figure 31A - HL-60 cells (70 x 10⁶/incubation) were stimulated with TNF as described in Fig. 1. Cell lysates were prepared and Raf-1 protein was immunoprecipitated with anti-Raf-1 antibody-conjugated Sepharose beads as described in Fig. 27. The beads were incubated in a reaction buffer containing 30 mM HEPES, pH 7.4, 5 mM MgCl₂, 10 mM MnCl₂, 1 mM PTT, 5 mM ATP .and 20 mCi [g- 32P]ATP. After 30 min, laemmli buffer was added and phosphorylated proteins were separated by 7.5% SPS-PAGE, transferred to an Immobilon PVDF membrane, and autoradiographed. These results represent one of three similar experiments. Figure 31B- Western blot analysis using anti-Raf-1 antibody was performed as described in Fig. 30. 6. Characterization of the 97. kDa protein as CAP kinase. Figure 32A - Cell lysates were prepared in RIPA lysis buffer from control and TNF-stimulated HL-60 cells (30 x 10⁶/incubation) and Raf-1 was immunoprecipitated using an anti-Raf-1 antibody as described in Fig. 31. The immune¬ complex was assayed for kinase activity toward MBP by incubation in the presence of 30 mM HEPES, pH 7.4, 10 mM MgCl₂, 5 mM NaF, 50 mM ATP, 15"mCi [g-³²P]ATP and 50 mM MBP. After 20 min, the beads containing immunoprecipitated Raf-1 were removed by centrifugation at 700 x g, the reaction supernate containing phosphorylated MBP was mixed with laemmli buffer, and proteins were separated on a 13% SDS polyacrylamide gel and autoradiographed. The data represent one of two similar experiments.

Figure 32B - Proteins contained within the immune-complex from Figure 32A were separated on a 7.5% SDS polyacrylamide gel and renatured for CAP kinase activity as described in Fig. 27. Gel slices (1.5 x 5 x 8 mm³) were cut according to the chromatogram defined by the molecular weight markers as indicated and renatured as described in Fig. 27. The gel slices were cut into smaller pieces (1.5 x 1.5 x 2 mm³) and MBP phosphorylating activity was determined by incubation of gel pieces for 60 min in the reaction buffer described in Fig. 32A and separation of phosphorylated MBP as above. The data represent one of two similar experiments.

H. Conclusions

Three lines of evidence demonstrate that CAP kinase is a Raf-1 kinase. Firstly, CAP kinase, renatured from bovine brain or from TNF- or ceramide-stimulated HL-60 cells, phosphorylates recombinant human Raf-1 in vitro, increasing Raf-1 activity toward MEK. Secondly, in intact HL-60 cells, TNF and ceramide analogs induce hyperphosphorylation of Raf-1, increasing its activity toward MEK 10-20 fold. Thirdly, CAP kinase, activated by TNF and ceramide treatment of HL-60 cells, associates tightly with Raf-1 in a multi-protein complex. Further, the major phosphorylation site on Raf-1 Thr²⁶⁹ exists within an -L-T-P- motif that conforms to the recognition site previously recognized as preferred by CAP kinase [155] . These investigations suggest that TNF-receptor interaction, through ceramide generation, stimulates CAP kinase to complex with, phosphorylate and activate Raf-1, linking the sphingomyelin pathway at the cell surface through to MAP kinase.

These observations have a number of implications with regard to mechanisms of signal transmission across the plasma membrane. An obvious question is whether there is a role for Ras in this process. In this regard, recent investigations by Green and co-workers [168] suggest that Ras is involved in ceramide-mediated apoptosis through Fas. In these studies, activation of Fas resulted in rapid ceramide generation, Ras activation, and apoptosis in Jurkat cells and a mastocytoma cell line transfected with Fas cDNA. Stimulation of cells with ceramide analogs directly induced Ras activation and apoptosis, and inactivation of Ras by transfection of dominant negative ras^"-¹¹ or microinjection of inactivating anti- Ras antibodies blocked apoptosis. These studies suggest a role for Ras in signal transmission through the sphingomyelin pathway as an element downstream of ceramide generation.

The present studies also have__^ implications for understanding signaling through the MAP kinase cascade. Numerous reports have documented the MAP kinase cascade to be evolutionary conserved [169] . As in the human system, in most instances upstream kinases capable initiating this cascade have not been identified. Ceramide is a major lipid in all eucaryotes and recent evidence suggests that it activates a protein phosphatase in Sacchar.omyces cerevisiae [170] . Hence, a search for a ceramide responsive kinase in lower eucaryotes would appear warranted.

A number of groups have reported that Raf-1 exists in mammalian cells in large multi-protein complexes ranging from 300-500 kD [138, 167] . Davis and co-workers [167] showed that Raf-1 existed in the cytoplasm of CHO cells in a pre-formed complex consisting of Raf-1 and the heat shock proteins hsp90 and hsp50. Under some conditions, MEK was also found. Recently the 14-3-3 proteins, which may be involved in Raf-1 activation, were also detected in Raf-1 immune complexes [171-173] . Further, Raf-1 appears to complex with the EGF receptor [174] and Bcl-2 [175] in some cells. These studies suggest that definition of the panoply of Raf-1 associated proteins is still incomplete. Any potential role for proteins other than CAP kinase that complex with Raf-1 in Raf-1 activation in myeloid cells will necessarily require identification of these proteins.

Last, it should be noted that this system may not be operative in all TNF- and ceramide-responsive cells. TNF has at least two major functions in mammalian cells, to induce either apoptosis or inflammation. Kronke and co¬ workers [164] have provided evidence that this functional dichotomy may reflect activation of two separate sphingomyelinases, an acidic and neutral isoform. These isoforms are activated by different domains of the 55 kDa TNF receptor, triggering distinct downstream signaling pathways. These investigators postulated that the neutral sphingomyelinase, which is membrane-bound, initiates the MAP kinase cascade, whereas the acidic isoform, which is endosomal, might initiate apoptosis. The present studies performed in HL-60 cells support this supposition. In these cells, TNF- activates only the neutral sphingomyelinase and results in stimulation of the Raf/MAP kinase cascade and arachidonic acid release, without apoptosis. This dichotomy may explain the recent observation that TNF does not activate Raf-1 in some cell systems [176] .

In sum, the present studies provide evidence that CAP kinase is a Raf-1 kinase linking activation of the sphingomyelin pathway at the cell surface through to MAP kinase in the cellular interior.

References

1. Kolesnick, R.N., J. Biol. Chem. 262:16759-16762 (1987) . 2. Kolesnick, R.N. and Clegg, S., J. Biol. Chem. 263:6534-6537 (1988).

3. Kolesnick, R.N. , J. Biol. Chem. 264:7617-7623 (1989) .

4. Dressier, K.A. and Kolesnick, R.N. , J. Biol. Chem. 265:14917-14921 (1990) .

5. Kolesnick, R.N. and Hemer, M.R., J. Biol. Chem. 265:18803-18808 (1990) .

6. Kim, M.-Y., Linardic, C. , Obeid, L. and Hannun, Y., J. Biol. Chem. 266:484-489 (1991) . 7. Okazaki, T. , Bielawska, A., Bell, R.M. and Hannun, Y.A., J. Biol. Chem. 265:15823-15831 (1990) . 8. Okazi, T., Bell, R.M. and Hannun, Y.A. , J. Biol. Chem. 264:19076-19080 (1989) . 9. Hannun, Y.A. , Loomis, C.R., Merill, A.H. , Jr., and Bell, R.M. , j. Biol. Chem. 261:12604-12609 (1986) .

10. Merrill, A.H., Jr., Sereni, A.M., Stevens, V.L., Hannun, Y.A., Bell, R.M. and Kinkade, J.M., Jr., J. Biol. Chem. 261:12610-12615 (1986) .

11. Wilson, E., Olcott, M.C., Bell, R.M. , Merril, A.H., Jr., and Lambeth, J.D., J. Biol. Chem. 261:12616-12623 (1986).

12. Hannun, Y. and Bell, R.M. , Science 235:670-674 (1987) .

13. Bajjalieh, S.M., Martin, T.F.J. and Floor, E., J. Biol. Chem. 264:14354-14360 (1989) .

14. Slife, C.W., Wang, E., Hunter, R. , Burgess, C. , Liotta, D. and Merrill, A.H. , Jr., J. Biol. Chem 264:1-7 (1989) .

15. Faucher, M. , Girones, N. , Hannun, Y.A. , Bell, R.M. and Davis, R.J., J. Biol. Chem. 263:5319- 5327 ( 1988 ) .

16. Davis, R.J., Girones, N. and Faucher, M. , J. Biol. Chem. 263:5373-5379 (1988) .

17. Countaway, J.L., Northwood, I.C. and Davis, R.J., J. Biol. Chem. 264:10828-10835 (1989) .

18. Goldkorn, T. , Dressier, K.A. , Muindi, J. , Radin, N.S., Mendelsohn, J. , Menaldino, D., Liotta, D. and Kolesnick, R.N. , J. Biol. Chem. 266:16092- 16097 (1991) . 19. Vilcek, J. , and Lee, T.H. , J. Biol. Chem. 266:7313-7316 (1991) .

20. Schutze, S., Scheurich, P., Pfizenmaier, K. and Kronke, M. , J. Biol. Chem. 264:3562-3567 (1989) .

21. Robaye, B., Hepburn, A., Lecocq, R. , Fiers, W. , Boeynaems, J.-M. and Dumont, J.E., Biochem.

Biophys. Res. Commun. 163:301-308 (1989) .

22. Marino, M.W. , Feld, L.J., Jaffe, E.A., Pfeffer, L.M., Hanm H.-M. and Donner, D.B., J. Biol. Chem. 266:2685-2688 (1991) . 23. Marino, M. W. , Pfeffer, L.M. , Guidon, P.T., Jr. and Donner, D.B., Proc. Natl. Acad. Sci. USA 86:8417-8421 (1989) .

24. Donato, N.J., Gallick, G.E., Steck, P.A. and Rosenblum, M.G., J. Biol. Chem. 264:20474-20481 (1989) .

25. Bowen, S., Stanley, K. , Selva, E. and Davis, R.J., J. Biol. Chem. 266:1162-1196 (1991) .

26. Cooper, J.A. , Septon, B.B. and Hunter, T., Methods Enzymol. 99:387-402 (1983) . 27. Bradford, M.M. , Anal. Biochem. 72:248-254 (1976) .

28. Chaplinski, T.J. and Niedel, J.E., J. Clin. Invest. 70:953-964 (1982) .

29. Breitman, T.R., Selonick, S.E. and Collins, S.J., Proc. Natl. Acad. Sci. USA 77:2936-2940 (1980) . 30. Zhang, Y. , Lin, J.-X., Yip, Y.K. and Vilek, J. , Proc. Natl. Acad. Sci. USA 85:6802-6805 (1988) . 31. Kronke, M. , Schutze, S., Scheurich, R. and Pfizenmaier, K. in "Tumor Necrosis Factor: Structure, Function and Mechanism of Action, " eds. Aggarwal, B.B. and Vilcek, J. (Dekker, New York), pp.189-216 (1991) . 32. Brenner, D.A., O'Hara, M. , Angel, P., Chojkier, M. and Karin, M., Nature (London) 337:661-663 (1989) . 33. Shiroo, M. and Matsushima, K. , Cytokine 2:13-20 (1990) . 34. Edelman, A.M., Blumenthal, D.K. and Krebs, E.G., Annu. Rev. Biochem. 56:567-613 (1987) .

35. Heisermann, G.J. and Gill, G.N. , J. Biol. Chem. 263:13152-13158 (1988) .

36. Heisermann, G.J., Wiley, H.S., Walsh, B.J., Ingraham, H.A. , Fiol. C.J. and Gill, G.N., J.

Biol. Chem. 265:12820-12827 (1990) .

37. Aitken, A., Holmes, C.F.B., Campbell, D.G., Resink, T.J., Cohen, P., Leung, C.T.W. and Williams, D.H., Biochim. Biophys. Acta. 790:288- 291 (1984) .

38. Carpenter, G. and Cohen, S., J. Biol. Chem. 265:7709-7712 (1990).

39. R.N. Kolesnick, Prog. Lipid Res. 30, 1 (1991) .

40. S. Mathias, K.A. Dressier, R.N. Kolesnick, Proc. Natl. Acad. Sci USA 88, 10009 (1991) .

41. K. Imamura, M.L. Sherman, D. Spriggs, D. Kufe, J. Biol. Chem. 263, 10247 (1988) .

42. B.G. Rao and M.W. Spence, J. Lipid Res. 17, 506

(1976) . 43. Michaud, N.R. , Fabian, J.R., Mathes, K.D. & Morrison, D.K. (1995) Mol_._ Cell. Biol. (in press) . 44. D.V.M. Das, H.W. Cook, M.W. Spence, Biochim. Biophys. Acta 777, 339 (1984) . 45. Y. Barenholz and T.E. Thompson, Biochim. Biophys. Acta 604, 129 (1980) . 46. N.G. Lipsky and R.E. Pagano, Cell Biol. 80, 2608 ( 1983 ) .

47. R. Seger et al., Proc. Natl. Acad. Sci USA 88, 6142 (1991) .

48. T.G. Boulton et al., Science 249, 64 (1990) . 49. I.C. Northwood, F.A. Gonzalez, M. Wartmann, D.L. Raden, R.J. Davis, J. Biol. Chem. 266, 15266 (1991) . 50. F.L. Hall et al., J. Biol. Chem. 266, 17430 (1991) . 51. E. Alvarez et al., J. Biol. Chem. 266, 15277 (1991) .

52. H. Loetscher et al., Cell 61, 351 (1990) .

53. T. Schall et al., J. Biol. Chem.351, 361 (1990) .

54. CA. Smith et al., Science 248, 1019 (1990) . 55. Z. Dembic et al., Cytokine 2, 231 (1990) .

56. K.A. Dressier, C.-C. Kan, R.N. Kolesnick, J. Biol. Chem. 266, 11522 (1991) .

57. J. Preiss et al., J. Biol. Chem. 261, 8597 (1986) . 58. R.N. Kolesnick, Trends in Cell Biol. (in press) .

59. K.A. Dressier, S., Mathias, R.N. , Kolesnick, Science 255, 1715 (1992).

60. R. Neta, T.J. Sayers, J.J. Oppenheim, in "Tumor Necrosis Factors: Structure, Function and Mechanism of Action" pp. 499-566 (eds. B.B.

Aggarwal, J. Vilcek) Marcel Dekker, Inc., NY, NY ^' (1992) .

61. S.K. Dower et al., J. Exp. Med. 162, 501 (1985) .

62. J.W. Lowenthal, J-C Cerottini, H.R. MacDonald, J. Immunol. 137, 1226 (1986) .

63. P.S. Chen, Jr., T.Y. Toribora, H. Warner, Anal. Chem. 28, 1756 (1956) .

64. M.J. Rebecchi, R.N. Kolesnick, M.C. Gershengorn, J. Biol. Chem. 258, 227 (1983) . 65. J.J. Farrar et al., J. Immunol. 125, 2555 (1980) . 66. L.A.J. O'Neill, T.A. Bird, A.J.H. Gearing, J. Saklatvala, J. Biol. Chem. 265, 3146 (1990) . 67. J. Dornand, C. Sekkat, J.-C. Mani, M. Gerber, Immunol. Lett. 16, 101 (1987) .

68. T.A. Bird, P.R. Sleath, P.C. deRoos, S.K. Dower, G.D. Virca, J. Biol. Chem. 266, 22661 (1991) . 69. R. Ballester, O.H. Rosen, J. Biol. Chem. 260, 15194 (1985) .

70. Y. Nishizuka, J. Am. Med. Assoc. 262, 1826

(1989) .

71. I. von Hoegen, W. Falk, G. Kojouharoff, P.H. Krammer, Eur. J. Immunol. 19, 329 (1989) .

72. J. Dornand et al. , J. Cell Physiol. 151, 71 (1992) .

73. F. Shirakawa, U. Yamashita, M. Chedid, S.B. Mizel, Proc. Natl. Acad. Sci. 85, 8201 (1988) . 74. M. Chedid, F. Shirakawa, 0. Naylor, S. Mizel, J. Immunol 142, 4301 (1989).

75. E. Munoz, U. Beutner, A. Zubiaga, T. Huber, J. Immunol. 144, 964 (1990) .

76. T.A. Bird, J. Saklavata, J. Immunol. 142, 126 (1989) .

77. L.A.J. O'Neill, ^' T.A. Bird, J. Saklatvala, Immunol. Today 11, 392 (1990) .

78. J.E. Sims et al., Science 241, 585 (1988) .

79. J.E. Sims et al., Proc. Natl. Acad. Sci. USA 86, 8946 (1989) .

80. C.J. Mcmahan et al., EMBO J. 10, 2821 (1991) .

81. J. Yodoi et al., J. Immunol. 134, 1623 (1985) .

82. R.N. Kolesnick, A. Paley, J. Biol. Chem. 262, 9204 (1987) . 83. Ballou, L.R. , Chao, P.C, Holness, M.A. , Barker, S.C., and Raghow, R. , J. _^Biol. Chem. 267, 20044-20050 (1992) .

84. Boyle, W.J., Van Der Geer, P., and Hunter, T., Methods in Enzymology, T. Hunter and B.M. Sefton, eds. (Academic Press, Inc., San Diego, CA) , vol.,

201, pp. 110-148 (1991) .

85. Cobb, M. H., Boulton, T. G. , and Robbins, D. J. , Cell Regulation 2, 965-978 (1991) .

86. Kameshita, I., and Fujisawa, H. , Anal. Biochem. 183, 139-143 (1989) .

87. Kennelly, P. J., and Krebs, E. G. , J. Biol. Chem. 266, 15555-15558 (1991) .

88. Lamprecht, W. , and Trautschold, I., Methods of Enzymatic Analysis, H.U. Bergmeyer, ed. (Academic Press, Inc., Weinheim) , pp. 543-551 (1963) .

89. Wang, H. R. , and Erikson, R. L. , Mol. Biol. of the Cell 3, 1329-1337 (1992) .

90. . Wiegmann, K. , Schutze, S., Kampen, E., Himmler,

A., Machleidt, T., and Kronke, M. , J. Biol. Chem. 267, 17997-18001 (1992) .

91. Bevilacqua, M.P. (1993) Anno. Rev. Immunol. 11, 767-804.

92. Frey, E.A., Miller, D.S., Jahr, T.G., Sundan, A., Bazil, V., Espevik, T., Finlay, B.B., and Wright, S.D. (1992) J. Exp. Med. 176, 1665-1671.

93. Loike, J.D., Sodeik, B., Cao, L., Leucona, S., Weitz, J.I., Detmers, P.A., Wright, S.D., and

Silverstein, S.C. (1991) Proc. Natl. Acad. Sci. (USA) 88, 1044-1048.

94. Wright, S.D., Ramos, R.A. , Hermanowski-Vosatka, A., Rockwell, P., and Detmers, P.A. (1991) J. Exp. Med. 173, 1281-1286.

95. Tracey, K.J., and Lowry, S.F. (1990) Adv. Surg. 23, 21-56.

96. Bird, T.A., Sleath, P.R. deRoos, P.R., Dower, S.K., and Duke Virca, G. (1991) J. Biol. CHem. 266, 22661-22670.

97. Han, J., Lee, J-D, Tobias, P.S., and Ulevitch, R.J. (1993) J. Biol. Chem. 268, 25009-25014.

98. Raines, M.A. , Kolesnick, R.N. , and Golde, D.W.

(1993) J. Biol. Chem. 268, 14572-14575. 99. Weinstein, S.L., Sanghera, J.S., Lemke, K. , DeFranco, A.L., and Pelech, S.L. (1992) J. Biol. Chem. 267, 14955-14962. 100. Davis, R.J. (1993) J. Biol. Chem. 268, 14553- 14556.

101. Baeuerle, P.A., and Baltimore, D. (1988) Cell 53, 211-217. 102. Lenardo, M.J., and Baltimore, D. (1990) Cell 58, 227-229.

103. Ghosh, S., and Baltimore, D. (1990) Nature 344, 678-682.

104. Zabel, U., and Baeuerle, P. . (1990) Cell 61, 255-265.

105. Henkel, T. , Machleidt, T., Alkalay, I., Kronke, M., Ben-Neriah, Y., and Baeuerle, P.A. (1993) Nature 365, 182-185.

106. Bohnlein, E., Ballard, D.W., Bogerd, H. , Peffer, N.J., Lowenthal, J.W., and Greene, W.C. (1989) J.

Biol. Chem. 264, 8475-8478.

107. Bomsztyk, K. , Rooney, J.W., Iwasaki, T., Rachie, N.A., Dower, S.K., and Sibley, CH. (1991) Cell. Regul. 2, 329-335. 108. Yang, Z., Costanzo, M., Golde, D.W., and Kolesnick, R.N. (1993) J. Biol. Chem. 268, 20520- 20523. 109. Osborn, L., Kunkel, S., and Nabel, G.J. (1989) Proc. Natl. Acad. Sci. (USA) 86, 2336-2340. 110. Collart, M.A. , Baeuerle, P., and Vassalli, P.-

(1990) Mol. Cell. Biol. 10, 1498-1506.

111. Mathias, S., Dressier, K.A. , and Kolesnick, R.N.

(1991) Proc. Natl. Acad. Sci. (USA) 88, 10009- 10013. 112. Dressier, K.A. , Mathias, S., and Kolesnick, R.N.

(1992) Science 255, 1715-1718_^.

113. Mathias, S., Younes, A., Kan, C-C , Orlow, I., Joseph, C, and Kolesnick, R.N. (1993) Science 259, 519-522. 114. Joseph, C.K., Byun, H-S., Bittman, R. , and Kolesnick, R.N. (1993) J. Biol. Chem. 268, 20002- 20006. 115. Kameshita, I., and Fujisawa, H. (1989) Anal. Biochem. 183, 139-143.

116. Wright, S.D., Ramos, R.A. , Tobias, P.S., Ulevitch, R.J., and Mathison, J.C (1990) Science 249, 1431-1433.

117. Weinstein, S.L., June, CH. , and DeFranco, A.L.

(1993) J. Immunol. 151, 3829-3838.

118. Bagasra, O., Wright, S.D., Thikkavarapu, S., Oakes, J.W., and Pomerantz, R.J. (1992) Proc. Natl. Acad. Sci. (USA) 89, 6285-6289.

119. Wang, R.H. , and Erikson, R.L. (1992) Mol. Cell. Biol. 3, 1329-1337.

120. Shumann, R.R. , Leong, S.R., Flaggs, G.W. , Gray, P.W., Wright, S.D., Mathison, J.C, Tobias, P.S., and Ulevitch, R.J. (1990) Science 249, 1429-1431.

121. Ball, E.D., Guyre, P.M., Shen, L., Glynn, J.M. , Maliszewski, C.R. , Baker, P.E., and Fanger, M.W.

(1984) J. Clin. Invest. 73, 1072-1077.

122. Wright, S.D., Rao, P.E., Van Voorhis, W.C, Craigmyle, L.S., Iida, K. , Talle, M.A., Westburg,

E.F., Goldstein, C, and Silverstein, S.C. (1983) Proc. Natl. Acad. Sci. (USA) 80, 5699-5703.

123. Bulusu, M.A.R.C, Waldstatten, P., Hildebrandt, J., Schutze, E., and Schultz, G. (1992) J. Med. Chem. 35, 3463-3469.

124. Perera, P.Y., Manthey, C.L., Stutz, P.L., Hildebrandt, J. , and Vogel, S.F. (1993) Infect. Immun. 61, 2015-2023.

125. Lam, C.E., Schutze, E., Hildebrandt, J. Aschauer, H. Liehl, E. Macher, I., and Stutz, P. (1991)

Antimicrob. Agents Chemothera. 35, 500-505.

126. Kusama, T. , Soga, T., Shioya, E., Nakayama, K. , ^•Nakajima, H., Osada, Y. , Ono, Y. , Kusumoto, S., and Shiba, T. (1990) Chem. Pharm. Bull. 38, 3366- 3372.

127. Scholz, D., Bednarik, K. , Ehn, G., Neruda, W. , Janzek, E., Loibner, H., Briner, K. , and Vasella, A. (1992) J. Med. Chem. 35, 2070, 2074. 128. Vyplel, H., Scholz, D. , Macher, I., Schindlmaier, K., and Schutze, E. (1991) J. Med. CHem. 34, 2759-2767. 129. Bielawska, A., Linardic, CM., and Hannun, Y.A. (1992) J. Biol. Chem. 267, 18493-18497. 130. Bielawska, A., Crane, H.M. , Liotta, D. , Obeid, L.M., and Hannun, Y.A. (1993) J. Biol. Chem. 268, 26226-26232. 131. Kastowsky, M. , Sabesch, A., Gutberlet, T. , and Bradaczek (1991) Eur. J. Biochem. 197, 707-716. 132. Daum, G. , Eisenmann-Tappe, I., Fries, H.-W., Troppmair, J. & Rapp, U.R. (1994) TIBS 19, 474- 480. 133. Williams, N.G. & Roberts, T.M. Cancer and Metastasis (1994) Reviews 13, 105-116. 134. Kyriakis, J.M., Force, T.L., Rapp, U.R. , Bonventre, J.V. & Avruch, J. (1993) J. Biol. Chem. 268, 16009-16019. 135. Kyriakis, J.M. , et al. (1992) Nature 358, 417- 421. 136. Lange-Carter, CA. , Pleiman, CM., Gardner, A.M., Blumer, K.J. & Johnson, G.L. (1993) Science 260, 315-319. 137. Zhang, X.-F., et al. (1993) Nature 364, 308-313.

138. Stokoe, D., MacDonald, S.G., Cadwallader, K., Symons, M. & Hancock, J.F. (1994) Science 264, 1463-1467.

139. Leevers, S.J., Paterson, H.F. & Marshall, C.J. (1994) Nature 369, 411-414.

140. Kovacina, K.S., et al. (1990) T. Biol. Chem. 265, 12115-12118.

141. Morrison, D.K., et al. (1989) Cell 58, 649-657.

142. Wu, J., et al. (1993) Science 262, 1065-1068. 143. Cook, S.J., and McCormick, F. (1993) Science 262,

1069-1072. 144. Morrison, D.K., Heidecker, G. , Rapp, U.R. & Copeland, T.D. (1993) J. Biol. Chem. 268, 17309- 17316.

145. Kolch, W., et al. (1993) Nature 364, 249-252.

146. Lee, R.-m., Rapp, U.R. & Blackshear, P.J. (1991) J. Biol. Chem. 266, 10351-10357.

147. Lai, W.S., Stumpo, D.J. & Blackshear, P.J. (1990) J. Biol. Chem. 265, 16556-16563.

148. Blackshear, P.J., Haupt, CM., App, H. & Rapp, U.R. (1990) J. Biol. Chem. 265, 12131-12134. 149. Macdonald, S.G., et al. (1993) Mol. Cell. Biol. 13, 6615-6620.

150. Kolesnick, R. & Golde, D.W. (1994) Cell 77, 325- 328.

151. Hannun, Y.A. (1994) J. Biol. Chem. 269, 3125- 3128.

152. Heller, R.A. & Kronke, (1994) M. Cell Biol. 126, 5-9.

153. Michell, R.H. & Wakelam (1994) M.J.O. Curr. Biol. 4, 370-373. 154. Liu, J., Mathias, S., Yang, Z. & Kolesnick, R.N. (1994) J. Biol. Chem. 269, 3047-3052.

155. Joseph, C.K., Byun, H.-S., Bittman, R. & Kolesnick, R.N. (1993) J. Biol. Chem.

156. Van Lint, J. , et al. (1992) J. Biol. Chem. 267, 25916-25921.

157. Raines, M.A. , Kolesnick, R.N. & Golde, D.W.

(1993) J. Biol. Chem. 268, 14572-14575.

158. Vietor, I., Schwenger, P., Li, W. , Schlessinger, J., and Vilcek, J. (1993) J. Biol. Chem. 268, 18994-18999.

159. Finco, T.S. & Baldwin, A.S., Jr. (1993) J. Biol. Chem. 268, 17676-17679.

160. Bruder, J.T., Heidecker, G. & Rapp, U.R. (1993) Genes & Dev. 6, 545-556. 161. Rivas, C I., Golde, D. W. , Vera, J. C & Kolesnick, R.N. (1994) Blood 83, 2191-2197: 162. Wiegmann, K. , Schutze, S., Machleidt, T. , Witte, D. & Kronke, M. (1994) Cell 78, 1005-1015.

163. Yang, Z., Costanzo, M. , Golde, D.W. & Kolesnick, R.N. (1993) J. Biol. Chem. 268, 20520-20523.

164. Izumi, T., Tamemoto, H. , Nagao, M. , Kadowaki, T., Takaku, F. & Kasuga, M. (1991) J. Biol. Chem.

266, 7933-7939.

165. Tσmonaga, M. , Golde, D.W. and Gasson, J.C. (1986) Blood 67, 31-36.

166. Pantazis, P, Kharbanda, S, Goustin, AS, Galanopoulos, T and Kufe, D (1991) Proc. Natl.

Acad. Sci. USA 88,2481-2485.

167. Wartmann, M. , R. & Davis, R. J. (1994) Biol. Chem. 269, 6695-6701.

168. Guilbins, E., Bissonnette, R., Mahboubi, A., Martin, S., Nishioka, W. , Brunner, T., Baier, G. ,

Bitterlich-Baier, G. , Byrd, C, Lang, F., Kolesnick, R., Altman, A. & Green, D. Immunity. (In Press) .

169. Errede, B. & Levin, D.E. (1993) Curr. Opin. Cell Biol. 5, 254-260.

170. Fishbein, J.D., Dobrowsky, R.T., Bielawska, A., Garrett, S. & Hannun, Y.A. (1993) J. Biol. Chem. 268, 9255-9261.

171. Fantl, W.J. et al. (1994) Nature 371, 612-614. 172. Freed, E., Symons, M. , Macdonald, S.G.,

McMormick, F. & Ruggieri, R. (1994) Science 265, 1713-1716.

173. Fu, H., et al. (1994) Science 266, 126-129.

174. App, H., et al. (1991) Mol. Cell. Biol. 11, 913- 919.

175. Wang, H.-G., et al. (1994) Onc gene 9, 2751-2756.

176. Minden, A., et al. (1994) Science 266, 1719-1723.

177. Morre et al. (1994) Methods in Enzymology 228, 448-450.

Claims

What is claimed is:

1. A purified membrane-bound ceramide-activated protein kinase having an apparent molecular weight of about 97 kD as determined by SDS polyacrylamide gel electrophoresis, which protein kinase is capable of specifically phosphorylating the threonine residue in a Pro-Leu-Thr-Pro-containing polypeptide.

2. A method of determining whether an agent is capable of specifically inhibiting the phosphorylation activity of the ceramide- activated protein kinase of claim 1 which comprises:

(a) contacting the protein kinase with a predetermined amount of a polypeptide containing the amino acid sequence Pro-Le -Thr-Pro, and the agent, under conditions i) which would permit activity of the protein kinase to be linear with respect to time and protein kinase concentration in the absence of the agent, and ii) which would permit the specific phosphorylation by the protein kinase of a predetermined number of the threonine residues in such predetermined amount of the Pro-Leu-Thr-Pro-containing polypeptide in the absence of the agent;

(b) quantitatively determining the percentage of such predetermined number of threonine residues which are specifically phosphorylated in the presence of the agent, thereby determining whether the agent is capable of inhibiting the activity of the ceramide-activated protein kinase; and

(c) determining whether the agent inhibits the activity of a non-ceramide-activated kinase, so as to determine whether the agent is capable of specifically inhibiting the activity of the ceramide-activated protein kinase.

3. The method of claim 2, wherein the polypeptide containing the amino acid sequence Pro-Leu-Thr-Pro is human epidermal growth factor receptor.

4. A method of determining whether an agent is capable of specifically stimulating the phosphorylation activity of the ceramide- activated protein kinase of claim 1 which comprises:

(a) contacting the protein kinase with a predetermined amount of a polypeptide containing the amino acid sequence Pro-Leu-Thr-Pro, and the agent, under conditions i) which would permit activity of the protein kinase to be linear with respect to time and protein kinase concentration in the absence of the agent, and ii) which would permit the specific phosphorylation by the protein kinase of a predetermined number of the threonine residues in such predetermined amount of the Pro-Leu-Thr-Pro-containing polypeptide in the absence of the agent;

(b) quantitatively determining the percentage of such predetermined number of threonine residues which are specifically phosphorylated in the presence of the agent, thereby determining whether the agent is capable of stimulating the activity of the ceramide-activated protein kinase; and (c) determining whether the agent stimulates the activity of a non-ceramide-activated kinase, so as to determine whether the agent is capable of specifically stimulating the activity of the ceramide-activated protein kinase.

5. The method of claim 4, wherein the polypeptide containing the amino acid sequence Pro-Leu-Thr-Pro is human epidermal growth factor receptor.

6. A method of treating a subject having an inflammatory disorder which comprises administering to the subject an agent capable of inhibiting the phosphorylation activity of a ceramide-activated protein kinase of T helper cells and macrophage cells of the subject in an amount effective to inhibit said phosphorylation activity, thereby reducing the inflammation associated with the disorder.

7. The method of claim 6, wherein the inflammatory disorder is rheumatoid arthritis.

8. The method of claim 6, wherein the inflammatory disorder is ulcerative colitis.

9. The method of claim 6, wherein the inflammatory disorder is graft versus host disease.

10. The method of claim 6, wherein the inflammatory disorder is lupus erythematosus.

11. The method of claim 6, wherein the inflammatory disorder is septic shock.

12. A method of treating a human subject infected with HIV so as to reduce the proliferation of HIV in the human subject which comprises administering to the human_^ subject an agent capable of inhibiting the phosphorylation activity of a ceramide-activated protein kinase of HIV-infected cells of the human subject in an amount effective to inhibit said activity, thereby reducing the proliferation of HIV in the human subject.

13. A method of treating a subject having a disorder associated with poor stem cell growth, which comprises administering to the subject an agent capable of stimulating the phosphorylation activity of a ceramide-activated protein kinase of the stem cells of the subject in an amount effective to stimulate said phosphorylation activity, thereby stimulating stem cell growth.

14. The method of claim 13, wherein the disorder associated with poor stem cell growth is aplastic anemia.

15. The method of claim 13, wherein the agent is interleukin-I.

16. A method of determining whether an agent is capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide- activated protein kinase of claim l which comprises:

(a) contacting the protein kinase with a predetermined amount of a polypeptide containing the amino acid sequence Pro-Leu-Thr-Pro, a predetermined amount of lipopolysaccharide, and the agent, under conditions (i) which .would permit activity of the protein kinase to be linear with respect to time, lipopolysaccharide concentration and protein kinase concentration in the absence of the agent, and (ii) which would permit the specific phosphorylation by the

.protein kinase of a predetermined number of the threonine residues in such predetermined amount of the Pro-Leu-Thr-Pro-containing polypeptide in the absence of the agent;

(b) quantitatively determining the percentage of such predetermined number of threonine residues which are specifically phosphorylated in the presence of the agent, thereby determining whether the agent is capable of inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide- activated protein kinase; and (c) determining whether the agent inhibits the ability of a non-lipopolysaccharide agent to stimulate the phosphorylation activity of the ceramide-activated protein kinase, ' said non- lipopolysaccharide agent being known to stimulate said activity in the absence of the agent, so as to determine whether the agent is capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide- activated protein kinase.

17. A method of treating a subject suffering from a lipopolysaccharide-related disorder which comprises administering to the subject an agent capable of specifically inhibiting the ability of lipopolysaccharide to stimulate the phosphorylation activity of the ceramide activated protein kinase of CD14-positive cells of the subject in an amount effective to specifically inhibit said stimulatory ability, so as to thereby treat the subject.