WO1998020347A1

WO1998020347A1 - Assay for lipopolysaccharide antagonists

Info

Publication number: WO1998020347A1
Application number: PCT/US1997/020438
Authority: WO
Inventors: Fabian Gusovsky; Melissa Ferrin
Original assignee: Eisai Co., Ltd.
Priority date: 1996-11-01
Filing date: 1997-10-31
Publication date: 1998-05-14
Also published as: AU5106898A

Abstract

An assay is disclosed for the identification and screening of antagonists that inhibit lipopolysaccharide (LPS) mediated stimulation of cytokine production in tissue cells. LPS is a bacterial cell wall component which, when released into the circulation of a host animal, can induce a cascade of biological events that lead to the physical symptoms of gram-negative sepsis. The assay described is targeted to identify compounds that inhibit LPS mediated activation of tissue cells, such as U373 human astrocytoma cells, which are negative for an LPS receptor protein, such as CD14. The assay is conducted in the absence of serum. This way, the LPS antagonist properties of test compounds are evaluated and quantified without concern of interference from serum factors.

Description

ASSAY FOR LIPOPOLYSACCHARIDE ANTAGONISTS

An assay is disclosed for the identification and screening of lipopolysaccharide (LPS) antagonists. LPS is a bacterial cell wall component associated with the symptoms and the etiology of gram-negative sepsis, a debilitating and often fatal syndrome which can result from bacterial infection. LPS antagonists may be useful for treating and studying sepsis and related conditions. The invention provides a serum-free assay for the identification and screening of compounds that inhibit the lipopolysaccharide-mediated stimulation of cytokine production by suitable target cells. In a preferred embodiment, production of the cytokine interleukin-6 (IL-6) by target cells exposed to LPS and a candidate antagonist is evaluated. The ability of the candidate antagonist to inhibit LPS, as shown by a decrease in IL-6 production of the target cells, is an indication of antagonist efficacy. Potency can be further quantified by testing for a dose-dependent IL-6 response.

Target cells suitable for use in the assay are preferably deficient in a cell membrane receptor which is necessary for production of the cytokine marker (e.g. IL- 6), but which can be functionally supplied from an exogenous source in a known quantity. In a preferred embodiment, U373 human astrocytoma cells may be used. U373 cells lack the LPS cell membrane receptor CD14. When exposed to LPS, U373 cells do not produce IL-6 unless soluble CD14 is also present. Thus, in a preferred embodiment, U373 cells are exposed to known quantities of LPS, soluble CD14, and a candidate antagonist. IL-6 production is then measured. The assay is particularly advantageous as a rapid and highly sensitive screen for LPS antagonist activity, without interference from serum components or other contaminants. Background of the Invention

Sepsis, due to gram-negative bacteria, is one of the most serious infectious conditions today and has no reliable cure. Risk factors for sepsis include immunodeficient and immunosuppressed conditions including leukemia, lymphoma, advanced solid tumors, pneumonia, meningitis, AIDS, biliary tract infections, uremia, chronic liver disease, and diabetes mellitus. Other risk factors include infections secondary to intravenous drug use, severe burns, and open wounds. Symptoms of sepsis include fever, inflammation, hypotension, acute renal failure, acute respiratory distress, hepatocellular destruction, cardiac failure and death. The pathogenesis of gram-negative sepsis is complex and is still not completely understood. The condition occurs from infection by gram-negative bacteria, with release of endotoxin into the circulation of a host animal. Lipopolysaccharide (LPS) is an outer cell membrane component of bacteria, and has been identified as an endotoxin associated with sepsis. When tissue cells of a host animal, such as monocytes and macrophages, are exposed to LPS from invading bacteria, adverse effects are triggered which can result in endotoxemia, or a condition known as sepsis, systemic inflammatory response syndrome (SIRS) or septic shock. This process is believed to be mediated by one or more host cell membrane receptors, such as CD 14, which respond to LPS by signaling certain host cells to produce cytokines, such as interleukin-6 (IL-6). The cytokines in turn are thought to induce the immunological imbalance which results in sepsis. Thus, pathological exposure to the LPS endotoxin can be serious and often fatal, with as many as 300,000 cases annually in the United States each year with a fatality rate of 20 to 60 per cent. (Am. J. Hosp. Pharm. 47, Supp.3:S3 (1990)). Symptoms of sepsis include fever, inflammation, hypotension, acute renal failure, acute respiratory distress, hepatocellular destruction, cardiac failure and death.

The structure of lipopolysaccharide has been described in studies of gram-negative bacteria such as E. coli and Salmonella typhimurium. LPS possesses three component regions. Imbedded in the cell membrane and extending out of the cell is the lipid A component, which is responsible for the toxic properties of the molecule and has a highly conserved structure ._ The middle region of the LPS molecule is also highly conserved and is called the core oligosaccharide. The third and most distal region of the LPS molecule is the somatic region comprised of highly variable polysaccharide components.

Current therapies for sepsis are directed towards eliminating the bacterial infection, as by treatment with antibiotics, and controlling the symptoms of sepsis. These conventional treatments do not address the underlying syndrome and have met with limited success. For example, antibiotics are necessary to combat the bacteria, but may result in the increased release of LPS and may exacerbate rather than alleviate the sepsis syndrome. One form of treatment for sepsis, described in U.S. Patent Nos. 4,185,090 and 4,057,685, relates to methods of reducing the toxicity of LPS. Other attempted therapies have been to find antigens or antibodies which eliminate the LPS molecule, as described in U.S. Pat No. 4,337,243 and Ziegler et al. , New Eng. J. Med. 307: 1225, (1982). Treatments have also been directed at controlling symptoms such as hypotension, using naloxone (Sheagren et al., Shock Syndromes related to Sepsis. In: Wyngaarden and Smith, eds., Cecil Textbook of Medicine, 18th ed. Philadelphia, 1988, pp. 1538-41), or controlling inflammation using corticosteroid treatment (Bone, N. Eng.J.Med. 317:653, 1987; Braunwald et al. , Harrison's Principles of Internal Medicine, 11th ed. , McGraw-Hill Book Co. , New York, 1987). Other methods of controlling inflammation have concentrated on nonsteroidal anti-inflammatory drugs (Am. J. Hosp. Pharm. 47, Supp.3:s6, 1990). All of these approaches have drawbacks and have experienced only limited success, and none have achieved widespread adoption and acceptance as viable treatments for sepsis. In a recently published article, "Clinical Trials In Sepsis and Septic Shock In 1994 and 1995" (Freeman et al., Current Opinion in Critical Care 1995 1:349 - 357 [1995]), results from six clinical trials were published. The article described research results on attempted therapies for sepsis, including the use of dobutamine to augment oxygen delivery to cells, the use of core lipid A-directed antibodies to bind endotoxin, and the use of anti-cytokine drugs to block inflammation. All were considered failures as a possible therapy for sepsis. Thus, there continues to be a pressing need for new, reliable and effective therapies.

As disclosed in Christ et al. , U.S. Patent No. 5,530,113, the use of LPS antagonists which inhibit harmful LPS activity is a particularly promising treatment option. The present invention provides an efficient and reliable method for screening and identifying LPS antagonists.

An LPS antagonist, as used herein, is a compound which interferes with or inhibits the ability of LPS to stimulate cytokine production by tissue cells. In theory, a clinically effective LPS antagonist should block the endotoxin in vivo and prevent the adverse responses which result in sepsis. Compounds of this kind have been identified, and act as LPS antagomsts, as shown in Christ et al. , U.S. Patent No. 5,530,113.

The identification, design and synthesis of LPS antagonists is difficult and unpredictable. For example, analogues of LPS, where part of the native molecule has been modified, may or may not act as LPS antagonists, e.g. by competitive inhibition of LPS itself. LPS analogues, if found to be LPS antagonists, may themselves be toxic, or otherwise unsuitable for therapeutic use. Thus, there is a continuing need to identify LPS antagonists and to determine the structure-activity relationships which contribute to LPS antagonist activity.

Previous methods for evaluating LPS antagonist activity are known but they have serious flaws. Principle among these is the presence of serum or serum components in the assay system, which may interfere with the sensitivity of the assay and indeed can result in false negatives: compounds which actually do have LPS antagonist activity are not recognized because they are inactivated or otherwise inhibited by serum components in the assay mixture.

One such assay for screening lipid A analogues of LPS is described in U.S. Patent No. 5,530,113 (the '113 Patent). In the '113 Patent, monocytes were isolated from blood and were cultured with 10% human serum. The lipid A analog and LPS were added to the cell culture, which was allowed to grow for three hours. The culture supernatant was then collected and analyzed for the presence of tumor necrosis factor (TNF) and interleukin-lB using an ELISA assay. The '113 patent also discloses assays for LPS inhibition using animal models. These assays measure the effects of lipid A analogues on LPS-induced TNF production and LPS-induced lethality in five week old mice. LPS was injected into the mice simultaneously with a candidate lipid A analog. Plasma was obtained from each mouse one hour after the LPS injection. The plasma was then analyzed by an ELISA assay for TNF levels. Several analogues were selected as potential antagonists based upon their ability to inhibit TNF levels.

These assays are conducted in the presence of serum. Serum factors can interfere with assay results in a number of ways. For example, serum factors can create false positives inhibiting the activity of LPS, giving the impression of a more potent antagonist than actually exists. Also, serum factors such as lipopolysaccharide binding protein (LBP) can create false negatives by enhancing LPS activity and artificially increasing cytokine levels, giving the false impression of poor antagonist activity when the cytokine marker is measured. Thus, interaction of serum factors with the assay reagents is undesirable in that it may produce imprecise results.

Summary of the Invention

The assay of the invention is serum free and overcomes a problem in the art by providing a highly sensitive means to evaluate the ability of a candidate antagonist to directly suppress LPS from inducing cytokine production in suitable target cells. The assay uses cells which are negative for the expression of a cell receptor needed for production of the cytokine marker, i.e. CD 14. Instead, a soluble CD14 is used. Soluble CD14 induces a dose-dependent production of IL-6 in the presence of LPS in U373 cells and therefore it is important to know how much CD 14 is present in the assay to accurately evaluate the potency of a candidate antagonist. Conventional assays are less precise and less sensitive, in that they utilize monocytes which have CD 14 or other receptors on their cell membranes in unknown quantities. The assay of the present invention utilizes precise amounts of soluble CD 14 which gives more accurate data on antagonist potency.

It has been found that LPS binds to a receptor called CD 14, a Glycosyl Phosphatidylinosito (GPI) linked membrane protein which is present in specific cell lines such as macrophages and monocytes. CD 14 is also present in a soluble form in plasma. CD 14 can bind to LPS and activate tissue cells, such as endothelial cells, which do not express CD 14 on their cell membranes. When LPS binds to CD 14, this triggers a cytokine release from the targeted cells which in turn causes an inflammatory response in other cells. Juan et al., Journal of Biological Chemistry, Volume 270, pg. 1372 (1995).

The assay measures the release of a marker cytokine (IL-6) by target cells in response to stimulation by LPS. An active or potent LPS antagonist will block or reduce that response, which the assay shows as a decrease in the amount of cytokine produced. More particularly, target cells which do not produce CD 14 are exposed to soluble CD14, LPS, and a candidate antagonist, to determine the ability of the candidate to block production of the cytokine IL-6.

Because LPS activation of IL-6 is dose dependent on CD 14, it is preferable to know the quantity of CD 14 present in the assay. Suitable target cells for this purpose are U373 human astrocytoma cells which do not have any measurable CD 14 on their cell membranes. U373 human astrocytoma cells do not respond to LPS by generating and releasing IL-6. Thus, when the U373 cells are exposed to LPS, no appreciable cytokine release is detected. However, when the U373 cells are exposed to LPS in the presence of soluble CD 14, measurable cytokine production occurs. In the assay of the present invention, soluble CD 14 is added to the assay in measured amounts, which provides a baseline or control for the assay and facilitates a more accurate analysis of the potency of a particular antagonist.

A particular advantage of this assay is that cytokine production is observed despite the absence of any other protein in the incubation media. Therefore, other factors affecting cell response to LPS, such as lipopolysaccharide binding protein (LBP), are not required in order to elicit IL-6 production by U373 cells. In addition, previously identified LPS antagonists, such as those described in U.S. Patent No. 5,530,113, are able to inhibit LPS/CD14-induced IL-6 generation in U373 cells. The assay can therefore be utilized to determine the potency of compounds that can block LPS activation without interference from serum factors. This provides an accurate and reliable measure of cytokine production which occurs as a direct result of LPS activation.

The assay can also be provided in the form of a kit, which ideally would contain most of the essential equipment and reagents to run the assay. The kit may contain a vial of purified LPS, a vial of medium necessary to run the assay as described, well plates and an ELISA kit for measuring IL-6.

Brief Description of the Drawings FIG. 1 is a silver stain of MY-4 affinity column fractions; FIG. 2 is a Western Blot analysis of the MY-4 Affinity Column fractions; FIG. 3 is a graph illustrating the stimulation of IL-6 production in U373 cells for each of the affinity column fractions;

FIG. 4 is a graph demonstrating the dose-dependent stimulation of IL-6 generation in U373 cells with soluble CD 14;

RECTIFIED SHEET (RULE 91) FIG. 5 is a graph demonstrating the dose-dependent stimulation of IL-6 generation in U373 cells with LPS;

FIG. 6 is a graph demonstrating dose-dependent inhibition of rhCD14/LPS induced IL-6 production in U373 cells by two exemplary antagonists (A and B).

Detailed Description of the Invention

This invention discloses an assay for the identification of LPS antagonists including the ability to determine the potency of the individual antagonist. Generally, the assay requires (a) purified CD 14, (b) cells which are negative for the expression of CD 14 cultured in serum free medium, (c) LPS, and (d) a proposed LPS antagonist. The CD 14, LPS and proposed antagonist are added to the cell culture. The cells are allowed to incubate, and thereafter the medium is removed and analyzed for cytokine levels.

Through experimentation, it has been determined that LPS binds to a plasma protein called lipopolysaccharide binding protein (LBP) to form an LPS-LBP complex. The LPS-LBP complex transfers to a GPI-linked macrophage membrane protein CD 14, which in turn activates the release of chemical activators called cytokines. Interleukin-6 (IL-6), interleukin-1 (IL-1) and tumor necrosis factor alpha (TNF-α) are cytokines which act as chemical messengers between cells. Cytokines also act as inflammatory mediators in cells. For example, it is believed that the production of inflammatory agents such as prostaglandins, leukotrienes, platelet activating factor and interleukin-1 is signaled by the release of cytokine α-TNF from macrophages and monocytes in reaction to LPS. One of the most abundant cytokines released in response to LPS is IL-6.

RECTIFIED SHEET (RULE 91) The assay measures production of a cytokine marker, such as IL-6, by cultured tissue cells in response to stimulation by LPS and in the presence of a candidate LPS inhibitor or antagonist. Decreased cytokine production in comparison with a antagonist-free control indicates LPS antagonist activity. In one embodiment of the assay, cells are cultured which are negative for the expression of membrane receptor CD 14. These cells do not respond to LPS, unless they are exposed to soluble CD 14, which enables them to respond to LPS stimulation without LBP or other serum factors. As a control model, some of the cells are incubated in minimum essential medium (MEM) with LPS and soluble CD 14 in the absence of an antagonist. The medium is then analyzed for cytokine levels, in particular IL-6, by conventional means.

For the test system, LPS, soluble CD14 and a candidate antagonist are incubated with the cells. After a sufficient time, the medium is removed and analyzed for cytokine levels, particularly IL-6. If the cytokine levels are similar to the control, the particular antagonist has a low activity. However, if the cytokine level is low in comparison to the control, the antagonist is an active LPS inhibitor. Based on a comparison of IL-6 levels from a series of control and test groups, the activity of a proposed antagonist to block cytokine production can be determined on a dose- dependent basis.

LPS REAGENT The assay of the invention tests the ability of a candidate LPS antagonist to inhibit LPS-induced cytokine production in a model cell system that is advantageously serum-free. LPS suitable for use in the invention can be obtained, for example, from Sigma (Re595, Salmonella minnesota).

GENERATION AND PURIFICATION OF RECOMBINANT HUMAN CD14 (rhCD14) FROM TRANSFECTED CHINESE HAMSTER OVARY CELLS

LPS-induced cytokine production can be mediated by interaction with certain proteins. The invention exploits this interaction in a suitable cell system to provide an accurate and reproducible way to evaluate whether a candidate protein inhibits LPS in a serum-free environment.

Purified soluble CD 14 is used as a mediating protein in a preferred embodiment of the assay, and can be obtained according to the method of Golenbock, et al. , J. Biol. Chem. , vol. 265:29, 22055-59 (1995). To obtain CD14 by this method, cells are genetically altered to express CD 14 in large quantities. For this purpose, Chinese Hamster Ovary-Kl (CHO-K1) fibroblasts are suitable and can be obtained from the American Type Culture Collection (ATCC HTB 17) of Rockville, Maryland for purposes of making a cell culture.

Preparation of CHO-K1 Cell Cultures

The CHO-K1 cells are plated at 10⁵ cells per 100-mm diameter tissue culture dish overnight. The cells are cultured at 37°C utilizing Ham's F12 medium (Whittaker Bioproducts) in a 5 % CO₂ environment. The medium is supplemented with 10% human serum, and the antibiotic ciprofloxacin (Miles Pharmaceuticals) is added at a dose of lOug per milliliter of medium. The cells are then washed free of medium with PBS. The CHO cells are then genetically altered to express CD 14 by the following method.

Complementary DNA (cDNA) for hCD14 is obtained through conventional techniques utilizing reverse transcriptase and the appropriate primer. The cDNA is cloned into the expression vector pcDNAI, utilizing conventional cloning techniques, and is ligated through the use of restriction endonuclease EcoRI into the EcoRI site present in the multiple cloning region of pcDNAI. The pcDNAI can be obtained from Invitrogen in San Diego, California. Large quantities of the vector can be generated through transformation and expansion of the MC1061/P3 strain of E. coli in medium containing ampicillin and tetracycline. The pure plasmid DNA is purified from detergent lysates of transformed bacteria, for example, by using a commercial plasmid preparation kit available from Qiagen in Chatsworth, CA.

CHO-K1 cells are co-transfected with the recombinant plasmids for CD 14 and pKoNeo (Stratagene in La Jolla, CA) using 3 ug of pcDNAI-CD14 and 1 ug of pKoNeo. Plasmid pKoNeo encodes for the gene neomycin phosphotransferase and confers resistance to the cells. Transfection is accomplished using the calcium phosphate precipitation method as described by Chen and Okayama in Molecular Cell Biology 7(8):2745-52 (1987). This method involves mixing the

RECTIFIED SHEET (RULE 91) plasmid DNA and .25 M CaCl₂ and 2x BBS and adding the mixture dropwise to a plate of the CHO-K1 cells. The mixture is swirled and allowed to incubate for 15 to 24 hours at 35 °C under 2 - 4% CO2. The medium is then removed and the cells are rinsed twice with growth medium and incubated for an additional 24 hours at 37 degrees C under 5% CO₂.

The resulting transfected cells are split 1 : 10 in Ham's F12 medium containing 400 micrograms per ml of active G418, a soluble analogue of neomycin obtained from Gibco at Gaithersburg, Maryland. The G418 acts to eliminate any cells for which transfection was unsuccessful. The G418-resistant cells are then cultured in Ham's F12 medium for 10 days at 37°C. After 10 days in medium, the G418- resistant cells are cloned by limiting dilution in 96 well culture dishes at a density of approximately 1 cell per three wells. Cloned cells can then be analyzed for CD14 expression by flow microfluorocytometry using an mAB leuM3 antibody available from Coulter Cytometry in Miami, Florida. Subclones are also analyzed for rhCD14 expression by flow microfluorocytometry using the mAB leuM3 antibody. This ensures that the most productive cells for producing CD 14 are isolated and cultured for use in the assay.

Extraction and Purification of rhCD14 from Transfected CHO Cells The cells selected for CD 14 expression are cultured in suspension in protein-free media at 37 °C (CHO Excell Pf media from JRH Biosciences). The media is supplemented with L-glutamine. The culture supernatant is then collected and concentrated five-fold in a protein concentrator made by Amicon Diaflo (PM30) with a 30Kd filter under pressure (50 psi) at 4°C. The supernatant is then filtered through a .22 u nitrocellulose filter and loaded onto an anti-CD 14 affinity column.

Preparation of MY-4 affinity column and Isolation of CD 14

An affinity column is prepared by crosslinking 1.5 mg of MY-4 antibody obtained from Coulter to 2 ml of Immobilized protein-A 64 using dimethylpimelimidate (Immuno Protein A IgG Orientation Kit, Pierce, Rockford,

Illinois). The column is then washed with 0.2 M triethanolamine at a pH of 8.2. The remaining active sites are blocked with O.1M ethanolamine at a pH of 8.2. The column is then washed with 0.1M glycine at pH 2.8 to remove any MY-4 antibody that is bound but not crosslinked to protein A. The column is equilibrated with 50 mM sodium borate wash buffer at pH 8.2.

Two milliliters of CD14/CHO supernatant collected after the filtration step are then loaded onto the MY-4 affinity column and passed through twice. The column acts to both separate rhCD14 protein from the supernatant and to purify the rhCD14 for later use in the assay. The column is washed twice with .2M triethanolamine buffer (pH 8.2) before eluting the column with five sequential 1 ml aliquots of 0.1M glycine at a pH of 2.8. The five sequential 1 ml aliquots act to release the rhCD14 from the column free of any impurities of the supernatant. Samples are collected from each fraction for testing to determine which fraction contains the rhCD14. The pH of the fractions are then immediately raised to neutral with 50 ul of 1M Tris-HCl at a pH 9.5. The column is regenerated by extensive washing with .1M glycine (pH 2.8) followed by a wash buffer of .2M triethanolamine (pH 8.2).

The column fractions obtained by following this procedure were identified as FT (Flow Through), Wl and W2 (primary and secondary washes), and E1-E5 (Eluates 1-5).

Silver Stain and Western Blot Analysis of Column Fractions

The purpose of this analysis is to determine if the fractions retained from the affinity column step contain pure rhCD14 which can be utilized for the LPS assay. Each of eight fractions (20ul) are mixed with 2x Sodium Dodecyl Sulfate (SDS) Loading Buffer and heated for three minutes at 95 °C. The samples are then loaded in duplicate on two 10% Tris-Glycine No vex gels. The gels are run for 60 minutes at 175 volts. One of the gels is selected for silver staining as described by Sambrook, Fritsch and Maniatis in Molecular Cloning, pgs. 18.56 - 18.57, (1989). Figure 1 demonstrates the silver stain of two protein bands of 53 and 58 Kd that were eluted in fractions E3 and E4 of the affinity column chromatography step.

The other gel is transferred to nitrocellulose and is run at 30 volts for 1.5 hours. The resulting blot is then rocked at room temperature for 30 minutes in 3% dry milk/TTBS (Composition: 20mM Tris HCL, pH 7.4; 0.9% NaCl; 0.05% Tween 20). The blot is then incubated with MEM- 18 anti-CD 14 antibody (Accurate Chemical & Scientific Co., Westbury NY) at a ratio of 1:1000 dilution in TTBS for one hour at room

RECTIFIED SHEET (RULE 91) temperature. The blot is washed several times in TTBS and is then incubated for 30 minutes with rabbit anti-mouse Horse Radish Peroxidase-conjugated secondary antibody (Pierce, Rockford, IL) at a ratio of 1:10,000 in TTBS. The incubation is performed at room temperature with extensive washing in TTBS. Immunoreactive proteins are visualized using ECL detection reagents (Amersham, Arlington Heights, II).

As shown in Figure 2, the Western Blot analysis of fractions E3 and E4 were strongly immunoreactive when probed with anti-CD 14 antibody MEM-18. This demonstrates that rhcdl4 was successfully obtained in fraction E3 and E4 eluted from the column. These proteins (53 and 58 Kd) were detected in the cell lysate, but not in either the flow through or the wash fractions. This indicates that the rhCD14 was efficiently retained by the affimty column containing the MY4 antibody. Other proteins were eluted, as shown in Figure 1, in fraction 5, at 69 kilodaltons. However, this protein was not immunoreactive to MEM 18 as shown in Figure 2 and thus is not rhCD14. The observed bands (-53 and 58 kd) represent CD14 in different glycosylation states.

Soluble CD 14 obtained by affinity chromatography, or by any other suitable method, can be used in the assay.

PREPARATION OF U373 CELLS U373 human astrocytoma cells were obtained from ATCC (American

Tissue Culture Collection, Rockville, MD, Catalog # HTB17). They were cultured in complete medium (composition: MEM (Minimum Essential Medium, Life Technologies, Catalog # 11095-080), 10% fetal calf serum, non-essential amino acids (Life Technologies, Catalog #11140-050) and sodium pyruvate (Life Technologies, Catalog # 11360-070)).

IL-6 DETERMINATIONS IN U373 CELLS The following protocol was used to determine IL-6 levels in U373 cells. U373 cells were cultured in 90 mm plates in complete medium (see composition above). After reaching confluency, cells were trypsinized (1 ml

Trypsin/EDTA, Life Technologies, Catalog #25300-054 for 10 min at room temperature) and were collected in 5ml of fresh complete medium. After centrifugation

RECTIFIED SHEET (RULE 91) (10 min, 4°C 300xg), the cell pellet was resuspended in 10 ml of fresh complete medium. Cells were plated in 48 well plates at a density of 3-4 X 10* cells/well in a total volume of 250μl/well. After overnight incubation at 37°C, cells were washed 3 times with 0.5 ml Hank's Balance Salt Solution (HBSS, Life Technologies, Catalog #14025-084). Five hundred microliters of serum-free medium (MEM, Life

Technologies, Catalog #11140-050) and sodium pyruvate (Life Technologies, Catalog #11360-070) were added.

LPS and CD14, obtained as described above, were added to the U373 cells in serum-free medium, together with a test compound, i.e. a known or potential LPS antagonist to be evaluated. After an incubation of 18-24h at 37 °C, aliquots of the culture media were analyzed for IL-6 levels using a commercial ELISA kit (Endogen, Cambridge, MA or Genzyme, Cambridge, MA).

EXPERIMENTAL RESULTS

Eaxmple 1 Response of U373 Cells to LPS in the Presence of CD 14

In order to characterize the responsiveness of U373 cells to rhCD14 and LPS, experiments were carried as shown in Figures 3, 4 and 5.

A. Baseline Production of IL-6. Using the protocol for IL-6 determination described above, but without adding a test compound, aliquots of the column fractions obtained in the purification of rhCD14 were tested. In these experiments, U373 cells were incubated with medium, 100 ng/ml LPS and each of the column fractions (identified along the horizontal axis of the graph in Figure 3). As Figure 3 shows, only fractions E3 and E4 were able to induce IL-6 generation, confirming that they contain purified CD 14. This stimulation was only observed in the presence of 100 ng/ml LPS. Neither CD 14 alone nor LPS alone induced any IL-6 generation.

B. IL-6 Dose Response to CD 14. Samples corresponding to fractions E3 and E4 were pooled and protein concentrations were determined using a commercial kit (Bio-

Rad Protein Assay Kit, catalog #500-0002). In these experiments, U373 cells were incubated with medium, 100 ng/ml LPS and increasing concentrations of sCD14

RECTIFIED SHEET (RULE 91) obtained from the pooled E3 and E4 fractions. The results of this experiment are shown on Figure 4. IL-6 production rises sharply, from about 100 pg/ml to a maximum of more than 1000 as CD 14 is increased from 0.05 nM to 0.5 nM, where it rises slightly at about 1 nM CD 14 and thereafter levels off.

C. IL-6 Dose Response to LPS. The dose dependency for LPS to induce IL- 6 generation in U373 was examined. In these experiments, U373 cells were incubated with 1 nM rhCD14 and increasing concentrations of LPS. IL-6 levels were measured as described above. The results (Figure 5) indicate that IL-6 production rises sharply with increasing amounts of LPS, to a maximum IL-6 level in the presence of about 30 ng/ml LPS in the presence of 1 nM rhCD14.

Example 2 Antagonist-Mediated Inhibition of LPS/CD14-induced IL-6 in U373 Cells Under the described assay conditions, LPS/CD14-induced IL-6 production in U373 cells can be inhibited by endotoxin antagonists. This inhibition is dose dependent as shown in Figure 6. Compound A (Figure 6) is a potent LPS antagonist according to the invention, and has been shown in other studies to inhibit LPS-induced cytokine generation in whole human blood and prevent LPS-induced effects in animal models (See, e.g., US Patent No. 5,530,113). Compound B is also a potent LPS antagonist according to the invention, but has not previously shown meaningful activity in other, serum-based, assays.

Compound A

The assay was conducted according to the protocol described above. U373 cells, cultured as described, were centrifuged (10 min, 4°C 300xg), and the cell pellet was resuspended in 10 ml of fresh complete medium. Cells were plated in 48 well plates at a density of 3-4 X 10⁴ cells/well in a total volume of 250μl/well. After overnight incubation at 37 °C, cells were washed and 500 microliters of serum-free MEM containing sodium pyruvate were added. LPS (100 ng/ml) and CD 14 (1 nM), obtained as described above, were added to the U373 cells in serum-free medium, together with various concentrations of Compound A as a test compound, ranging from 0.1 to 1000 nM.

RECTIFIED SHEET (RULE 91) The test antagonist, Compound A, is an LPS antagonist described in U.S. Patent No. 5,530,113. See, e.g. column 146 (Compound 21) and columns 166-167 (Compound B531-35). See also, Christ et al., Science, 268: 80-83 (1995) (Compound

E5531).

COMPOUND A

As shown in Figure 6, concentrations of less than 1 nM of Compound A have little or no effect on LPS-induced production of IL-6. A marked drop in cytokine production, showing LPS inhibition, occurs at concentrations greater than 1 nM. The sharpest reduction in IL-6 production is at concentrations of Compound A in the range of 10 to 100 nM. Higher concentrations further inhibit IL-6, but at a reduced rate. This dose response curve demonstrates the utility of the assay for identifying and evaluating LPS antagonists in serum-free conditions.

Compound B

An assay composition was prepared according to the protocol for Compound A, above. U373 cells were plated in 48 well plates at a density of 3-4 X 10⁴

RECTIFIED SHEET (RULE 91) cells/well in a total volume of 250μl/well. After overnight incubation at 37°C, cells were washed and 500 microliters of serum-free MEM and sodium pyruvate were added. LPS (100 ng/ml) and CD 14 (1 nM), obtained as described above, were added to the U373 cells in serum-free medium, together with various concentrations of Compound B as a test compound, ranging from 0.01 to 100 nM.

The test antagonist, Compound B, is an LPS antagonist as described in U.S. Patent No. 5,530,113.

COMPOUND B

As shown in Figure 6, IL-6 production is immediately and strongly inhibited by Compound B at concentrations between 0.01 and 1 nM. LPS-induced production of IL-6 by the CD14/U373 system is further inhibited at concentrations between 1 and 10 nM, and is virtually zero at concentrations above 10 nM. This dose response curve demonstrates the utility of the assay for identifying and evaluating LPS antagomsts in serum-free conditions.

Discussion of Results

RECTIFIED SHEET (RULE 91 Compounds A and B both blocked LPS/CD14-induced IL-6 generation in U373 cells, showing that their antagonistic activity is not dependent on any serum factors. This is of interest because the serum protein LBP has a critical role in LPS stimulatory activity. The results presented in the foregoing Examples strongly suggest that LPS antagonists do not require LBP in order to block LPS/CD14 actions.

Furthermore, the results present a simple, direct and easily applicable procedure to test the activity of endotoxin antagonists. In addition to the absence of LBP, the absence of serum factors involved in the deactivation of LPS is advantageous. Recently, it was shown that serum factors are involved in neutralizing LPS activity (Hailman et al., J. Biol. Chem., 271, 12172-12178, 1996). Therefore, assays containing serum may interfere with the results by neutralizing LPS and possibly antagonists which resemble LPS structurally.

In summary, a serum free assay for LPS antagonists has the advantages of: 1) absence of LBP, which enhances LPS activity; 2) absence of deactivators of LPS, which may reduce LPS activity; and 3) consistency, in that cultured cells can be handled in such a way to eliminate response variability, whereas in whole blood assays, donor to donor variability is unavoidable.

Example 3 The ability of LPS antagonists to inhibit LPS activity in a whole blood

(serum-containing) assay was compared with their ability to inhibit LPS activity in the serum-free assay of the invention. The whole blood assay was performed as described in U.S. Patent No. 5,530,113. Briefly, human fresh blood samples were incubated at 37 C for 3 hours in the presence of 10 ng/ml LPS and various concentrations of antagonist. At the end of this incubation, samples were centrifuged and levels of TNF- α (tumor necrosis factor alpha) were determined in the supernatant, using a commercially available ELISA kit. IC₅₀ values were determined for each antagonist as the concentration that inhibits 50% of the response elicited by LPS. The serum-free assay of the invention, using U373 cells, was performed as described above. Four compounds were tested in both assays. Compounds A and B are described above. Compounds C and D are described in U.S. Patent No. 5,530,113 and are also identified as B-1232 and B-1287 respectively. The results are shown in Table 1, below. TABLE 1

LPS Inhibition in Whole Blood & Serum-Free (U373) Assays Compound Whole Blood Assay Serum-Free Assay

(IC₅₀) in nM (IC₅₀) in nM

A 10 17.4

B no activity 3.2 C 38 27

D 2.0 3.3

These results show that the assay of the invention is a rapid, efficient, accurate and useful screening tool for identifying potent LPS antagonists. Compounds which show an ability to inhibit LPS in a conventional whole blood assay also show that activity in the serum-free assay, and in comparable IC₅₀ concentrations. Further, the assay also identifies compounds which are potent LPS antagonists, but which do not show activity in the conventional whole blood test. The serum-free assay of the invention is therefore more sensitive than the conventional assay, is not compromised by the presence of serum factors, and is easier to perform than conventional assays. The assay is particularly useful as a tool for rapid screening of LPS antagonists, and for determining structure activity relationships (SARs) for LPS antagonists without interference from serum. Thus, structures which show activity in this assay, but not in serum, can be compared to those which show activity in both assays. These and other advantages and embodiments of the invention will be apparent to the practitioner from the foregoing specification and examples, which are illustrative, and do not limit the scope of the invention as claimed.

Claims

We claim:

1. An assay for the detection and screening of LPS antagonists comprising the following steps: providing lipopolysaccharide and a lipopolysaccharide receptor; providing cells negative for the expression of the lipopolysaccharide receptor; adding medium to the cells; providing a proposed lipopolysaccharide antagonist; adding the lipopolysaccharide, the lipopolysaccharide receptor and the proposed lipopolysaccharide antagonist into the medium with the cells; allowing the cells to incubate in the medium; and analyzing the medium for cytokine levels.

2. An assay according to claim 1 wherein the lipopolysaccharide receptor is CD14.

3. An assay according to claim 1 wherein the lipopolysaccharide receptor is soluble recombinant human CD 14.

4. An assay according to claim 1 wherein the medium is a serum-free medium.

5. An assay according to claim 1 wherein the cells are U373 human astrocytoma cells.

6. An assay according to claim 1 wherein the CD 14 is purified soluble recombinant human CD 14.

7. An assay according.to claim 1 wherein the medium is MEM.

8. An assay according to claim 7 wherein the MEM is comprised of at least 10% fetal calf serum, selected non-essential amino acids and sodium pyruvate.

9. An assay according to claim 1 wherein the cytokine levels are analyzed using an Enzyme-linked immunosorbent assay.

10. An assay according to claim 1 wherein the medium is analyzed for levels of interleukin-6.

11. An assay comprising: providing purified soluble recombinant human CD 14 protein and lipopolysaccharide; providing U373 human astrocytoma cells; culturing cells in serum-free medium; providing a candidate lipopolysaccharide antagonist; adding the CD 14 protein, lipopolysaccharide and the candidate lipopolysaccharide antagonist into the medium; allowing the cells to incubate; and analyzing the medium for interleukin-6 levels.

12. The method of claim 11 wherein the interleukin-6 levels are analyzed using an Enzyme-linked immunosorbent assay.

13. An assay for the detection and screening of LPS antagonists comprising the following steps: providing purified soluble recombinant human CD 14 protein and purified lipopolysaccharide; providing U373 human astrocytoma cells; culturing the cells in medium contaimng fetal calf serum, selected non- essential amino acids and sodium pyruvate; culturing the cells in serum-free medium; providing a lipopolysaccharide antagonist; adding the CD 14 protein, lipopolysaccharide and the candidate lipopolysaccharide antagonist into the medium; allowing the cells to incubate; and analyzing the medium for interleukin-6 levels.

14. A kit for conducting an assay for the detection and screening of lipopolysaccharide antagonists comprising: one or more vials, each contaimng one or both of a lipopolysaccharide and a lipopolysaccharide receptor; cells capable of producing a cytokine response only in the presence of liposolysaccharide and the lipopolysaccharide receptor; serum-free medium in a container suitable for culturing the cells; means for combining the liposolysaccharide and the lipopolysaccharide receptor with the cells and serum-free medium; and means for analyzing cytokine levels in the medium.

15. A kit according to claim 14 wherein the lipopolysaccharide receptor is CD 14 and the cells are U373 cells.

16. A kit according to claim 14 wherein the means for analyzing cytokines in the medium includes an enzyme linked immunosorbent assay.

17. A kit according to claim 15 wherein the means for combining includes a well plate and the means for analyzing cytokines in the medium includes an enzyme linked immunosorbent assay.