AU2005244514B2 - Antagonists of HMG1 for treating inflammatory conditions - Google Patents

Description

0 SANTAGONISTS OF HMG1 FOR TREATING INFLAMMATORY

CONDITIONS

Technical Field of the Invention The present invention provides a pharmaceutical composition and method for treating diseases characterized by activation of an inflammatory cytokine cascade, particularly sepsis, including septic shock and ARDS (acute respiratory distress syndrome), comprising administering an effective amount of an antagonist to the high mobility group 1 protein (HMG1). The present 0 S 10 invention further provides a diagnostic method for monitoring the severity of sepsis and related conditions, comprising measuring the serum concentration of HMG1 in a patient exhibiting symptoms of a disease characterized by activation of inflammatory cytokine cascade. Lastly, the present invention provides a pharmaceutical composition and method for effecting weight loss or treating obesity, comprising administering an effective amount of an HMG1 protein or a therapeutically active fragment of the gene product of an HMG1 gene.

The reference to any prior art in this specification is not, and should not be taken as an acknowledgement or any form of suggestion that the referenced prior art forms part of the common general knowledge in Australia.

Background of the Invention Sepsis is an often fatal clinical syndrome that develops after infection or injury. Sepsis is the most frequent cause of mortality in hospitalized patients.

Experimental models of gram negative sepsis based on administration of bacterial endotoxin (lipopolysaccharide, LPS) have led to an improved understanding of the pathogenic mechanisms of lethal sepsis and conditions related to sepsis by virtue of the activation of a common underlying inflammatory cytokine cascade. This cascade of host-response mediators includes TNF, IL-1, PAF and other macrophage-derived factors that have been widely studied as acute, early mediators of eventual lethality in severe endotoxemia (Zhang and Tracey, In The Cytokine Handbook, 3rd ed. Ed.

Thompson (Academic Press Limited, USA), 515-547,1998).

SUnfortunately, therapeutic approaches based on inhibiting these 0 individual "early" mediators of endotoxemia have met with only limited success in large prospective clinical trials against sepsis in human patients. It is possible to infer from these disappointing results that later-appearing factors in the host response might critically determine pathogenesis and/or lethality in Ssepsis and related disorders. Accordingly, there is a need to discover such putative "late" mediators necessary and/or sufficient for part or all of the Sextensive multisystem pathogenesis, or for the lethality, of severe endotoxemia, particularly as endotoxemia is representative of clinical sepsis and related clinical disorders.

HMG1 is a 30 kDa chromosomal nucleoprotein belonging to the burgeoning high mobility group (HMG) of non-histone chromatin-associated proteins. As a group, the HMG proteins recognize unique DNA structures and have been implicated in diverse cellular functions, including determination of nucleosome structure and stability, as well as in transcription and/or replication. The HMG proteins were first characterized by Johns and Goodwin as chromatin components with a high electrophoretic mobility in polyacrylamide gels (see in The HMG Chromosomal Proteins, E. W. Johns, Academic Press, London, 1982).

Higher eukaryotes exhibit three families ofHMG proteins: the HMG-1/-2 family, the HMG- 14/-17 family and the HMG-I/-Y family. Although the families are distinguishable by size and o DNA-binding properties, they are similar in their physical properties. HMG proteins are S highly conserved across species, ubiquitously distributed and highly abundant, and are extractable from chromatin in 0.35 M NaCI and are soluble in 5% perchloric or trichloroacetic acid. Generally, HMG proteins are thought to bend DNA and facilitate binding of various transcription factors to their cognate sequences, including for instance, progesterone receptor, S estrogen receptor, HOX proteins, and Octl, Oct2 and Oct6. Recently, it has become apparent t that a large, highly diverse group of proteins including several transcription factors and other DNA-interacting proteins, contain one or more regions similar to HMGI, and this feature has C, come to be known as the HMGI box or HMGI domain, cDNAs coding for HMGI have been S cloned from human, rat, trout, hamster, pig and calf cells, and HMGI is believed to be abundant in all vertebrate cell nuclei. The protein is highly conserved with interspecies sequence identities in the 80% range. In chromatin, HMGI binds to linker DNA between nucleosomes and to a variety of non-1-DNA structures such as palindromes, cruciforms and stem-loop structures, as well as cisplatin-modified DNA. DNA binding by HMGI is generally believed to be sequence insensitive. HMGI is most frequently prepared from washed nuclei or chromatin, but the protein has also been detected in the cytoplasm. (Reviewed in Landsman and Bustin, BioEssays 15:539-546, 1993; Baxevanis and Landsman, Nucleic Acids Research 23:514-523, 1995). To date, no link has been established between the HMG proteins and any clinical condition or disease.

HMGI has been alternatively identified as a heparin-binding protein abundantly expressed in developing brain and dubbed "amphoterin" for its highly dipolar sequence, comprising two internal repeats of a positively charged domain of about 80 amino acids (the HMG1 box) and an acidic C-terminal domain containing a stretch of approximately continuous glutamic or aspartic acid residues. Amphoterin/HMGI has been localized to the outer surface of the plasma membranes of epithelial, and especially neuronal cells, where it has been specifically localized to the filipodia of neural cells. Inhibition studies have suggested that amphoterin/HMGI is required for process (neurite) extension and amphoterin/HMG also may be involved in neuron-glia interactions (Merenmies et al., J. Biol. Chew. 266:16722- 16729,1991; Merenmies et al., J. Biol. Chem. 266:16722-16729, 1991; Milev et al., J. Biol.

Chem. 273:6998-7005, 1998; and Salmivirta et al., Exp. Cell Res. 200:444-451, 1992).

Amphoterin/HMGl can be released from murine erythroleukemia cells after stimulation with the chemical inducer hexamethylenebisacetamide (Melloni et al., Biochem. Biophys. Res.

Commun. 210:82-89, 1995). Previous study suggested that the gene product of the HMG 1 gene functions as a differentiation enhancing factor by stimulating a-PKC (Melloni et al., Biochem. Biophvs. Res. Commun. 210:82-89, 1995; and Melloni et al., FEBSLett. 368:466- 470, 1995).

The HMGI gene product has been shown to interact with plasminogen and tissue-type O plasminogen activator (t-PA) and effectively enhance plasmin generation at the cell surface, a system that is known to play a role in extracellular proteolysis during cell invasion and tissue d remodeling. Amphoterin/HMGl has also been shown to interact with the receptor of advanced glycosylation end products (RAGE) (Mohan et al., Biochem. Biophys. Res. Commun. 182:689- 696, 1992; Yamawaki et al., J. Neurosci. Res. 44:586-593, 1996; Salmivirta et al., Exp. Cell Res. 200:444-451, 1992; and Vassalli et al., J. Clin. Invest. 88:1067-1072, 1991), (Redlitz and S Plow, Baillieres Clin. Haematol. 8:313-327, 1995; and Parkkinen et al., J Biol. Chem.

n 266:16730-16735, 1991).

There is a longstanding need in the art to discover improved agents that can prevent the I cytokine-mediated inflammatory cascade and have therapeutic activity in a large variety of O cytokine-mediated inflammatory diseases. The present invention was made during the course of investigative research to identify agents that mediate toxicity, pathogenesis and/or lethality in sepsis and other disorders related by a common activation of the inflammatory cytokine cascade.

Diseases and conditions mediated by the inflammatory cytokine cascade are numerous.

Such conditions include the following grouped in disease categories: Systemic Inflammatory Response Syndrome, which includes: Sepsis syndrome Gram positive sepsis Gram negative sepsis Culture negative sepsis Fungal sepsis Neutropenic fever Urosepsis Meningococcemia Trauma hemorrhage Hums Ionizing radiation exposure Acute pancreatitis Adult respiratory distress syndrome (ARDS) Reperfusion Injury, which includes Post-pump syndrome Ischemia-reperfusion injury Cardiovascular Disease, which includes Cardiac stun syndrome Myocardial infarction Congestive heart failure Infectious Disease, which includes o HIV infection/HIV neuropathy O Meningitis 0 Hepatitis Septic arthritis Peritonitis Pneumonia Epiglottitis E. coli 0157:H7 S Hemolytic uremic syndromc/thrombolytic thrombocytopcnic purpura Malaria 10 Dengue hemorrhagic fever SLeishmaniasis 0 Leprosy C Toxic shock syndrome Streptococcal myositis Gas gangrene Mycobacterium tuberculosis Mycobaclerium aviun intracellulare Pneumocystis carinii pneumonia Pelvic inflammatory disease Orchitis/epidydimitis Legionella Lyme disease Influenza A Epstein-Barr Virus Viral associated hemiaphagocytic syndrome Viral encephalitis/aseptic meningitis Obstetrics/Gynecology, including: Premature labor Miscarriage Infertility Inflammatory Disease/Autoimmunity, which includes: Rheumatoid arthritis/seronegative arthropathies Osteoarthritis Inflammatory bowel disease Systemic lupus erythematosis Iridoeyelitis/uveitistoptic neuritis Idiopathic pulmonary fibrosis Systemic vasculitis/Wegener's gramilornatosis Sarcoidosis o Orchitis/vasectomy reversal procedures C Atlergic/Atopic Diseases, which includes: o Asthma 0 Allergic rhinitis Eczema Allergic contact dermatitis Allergic conjunctivitis Hypersensitivity pneumonitis t/n Malignancy, which includes:

ALL

AML

0 CML ci CLL Hodgkin's disease, non-Hodgkin's lymphoma Kaposi's sarcoma Colorectal carcinoma Nasopharyngeal carcinoma Malignant histiocytosis Paraneoplastic syndrome/hypercalcemia of malignancy I Transplants, including: Organ transplant rejection Graft-versus-host disease Cachexia Congenital, which includes: Cystic fibrosis Familial hematophagocytic lymphohistiocytosis Sickle cell anemia Dermatologic, which includes: Psoriasis Alopecia Neurologic, which includes: Multiple sclerosis Migraine headache Renal, which includes: Nephrotic syndrome Hemodialysis Uremia Toxicity, which includes: OKT3 therapy Anti-CD3 therapy o Cytokine therapy o Chemotherapy Radiation therapy Chronic salicylate intoxication Metabolic/ldiopathic, which includes: Wilson's disease Hemachromatosis tln Alpha-I antitrypsin deficiency Diabetes SHashimoto's thyroiditis 0 Osteoporosis C< Hypothalamic-pituitary-adrenal axis evaluation Primary biliary cirrhosis Summary of the Invention The present invention provides a pharmaceutical composition for treating conditions (diseases) mediated by the inflammatory cytokine cascade, comprising an effective amount of an antagonist or inhibitor of HMGI. Preferably, the HMGI antagonist is selected from the group consisting of antibodies that bind to an HMGI protein, HMG I gene antisense sequences and HMG1 receptor antagonists. The present invention provides a method for treating a condition mediated by the inflammatory cytokine cascade, comprising administering an effective amount of an HMG1 antagonist. In another embodiment, the inventive method further comprises administering a second agent in combination with the HMG 1 antagonist, wherein the second agent is an antagonist of an early sepsis mediator, such as TNF, IL-la, IL- 1 P, MIF or IL-6. Most preferably, the second agent is an antibody to TNF or an IL-I receptor antagonist (IL-Ira).

The present invention further provides a diagnostic and prognostic method for monitoring the severity and predicting the likely clinical course of sepsis and related conditions for a patient exhibiting shock-like symptoms or at risk to exhibit symptoms associated with conditions mediated by the inflammatory cascade. The inventive diagnostic and prognostic method comprises measuring the concentration ofHMG1 in a sample, preferably a serum sample, and comparing that concentration to a standard for HMG1 representative of a normal concentration range of HMG1 in a like sample, whereby higher levels of HMG1 are indicative of poor prognosis or the likelihood of toxic reactions. The diagnostic method may also be applied to other tissue or fluid compartments such as cerebrospinal fluid or urine. Lastly, the present invention provides a pharmaceutical composition and method for effecting weight loss or treating obesity, comprising administering an effective amount of HMGI or a therapeutically active fragment thereof.

Brief Description of the Drawings 0 Figure 1 shows two graphs that profile the induction ofHMGI release by LPS in vitro 0 (Figure 1A) and in vivo (Figure 1B). Specifically, Figure 1A shows the accumulation of 0 HMGI in culture supematants ofmacrophage RAW 264.7 cells after stimulation with LPS (100 ng/ml). The inset is a Western blot (using antibodies raised against recombinant HMG1) showing induction of HMGI release from RAW 264.7 cells after induction with TNF. Figure IB shows accumulation of HMGl in serum of LPS-treated mice. Serum from Balb/C mice was collected at various time points after LPS administration, and assayed for HMG1 by Western blotting using antibodies raised against recombinant HMG1.

Figure 2 illustrates that HMGI is a mediator of pathogenesis and lethality in endotoxemia. Figure 2A shows the protective effect ofanti-HMGI antibodies against LPS ,I lethality, tested in mice. Administration ofanti-HMGI antiserum in the indicated amounts at O 0.5 (if one dose), -0.5 and 12 (if two doses), or 12 and 36 (if three doses) hours relative to C1 LPS challenge (at lime 0) was protective against LPS-induced lethality, and repeated dosing schedules provided better protection. Figure 2B illustrates that rHMG caused dose-dependent lethality in endotoxic mice. Male Balb/C mice (20-23 grams) were randomized in groups of ten to receive LPS (3.15 mg/kg; a non-lethal dose) alone or in combination with purified recombinant HMGI protein. Administration of HMGl at the indicated doses 2, 16, 28 and hours after LPS challenge significantly increased the lethality of the underlying endotoxemia.

Figure 2C illustrates independent lethal toxicity of HMGI as a function of dose. Purified rHMGI was administered to male Balb/C mice (five mice per treatment group) as a single i.p.

bolus at the indicated dosage. Mice were observed for at least 48 hours, and 60% of mice treated with rHMGI at a dose of 500 pg/mouse died within 24 hours ofrHMG challenge, indicating a single dose LDso of less than 500 pg/mouse.

Figure 3 shows that HMGI induced TNF release both in vitro (Figure 3A) and in vivo (Figure 3B). Specifically, Figure 3A shows that HMGI induces TNF release from huPPBMCs in dose-dependent fashion. Freshly isolated huPBMC cultures were stimulated with purified recombinant HMG I protein at the indicated doses, and culture media were sampled four hours later to be assayed for TNF according to known immunologic methods (ELISA). Figure 3A shows the mean S.E.M. of the induced TNF response in two experiments (in triplicate).

Figure 3B shows that administration of HMG1 induced accumulation of TNF in serum of treated mice. Balb/C mice (20-23 g) were treated intraperitoneally with purified recombinant HMGI at the indicated doses and blood samples were taken two hours later for assay of TNF by an L929 bioassay and (TNF levels expressed as mean N=3).

Figure 4 shows that HMG1 caused body weight loss in mice. Purified HMGI was administered intraperitoneally to mice at 100 pg/mnouse/day for three days, and body weight was monitored. Figure 4 shows the mean S.E.M. of net body weight change of three mice per group.

Figure 5 shows the tissue distribution of HMGl mRNA. Human RNA master blots containing poly(A)' RNA of various tissues (Clontech, Palo Alto, CA. USA) were hybridized S with a 0.6 kb digoxigenin-I l-dUTP-labeled HMGI cDNA probe synthesized by PCR using recombinant plasmid containing the HMG I cDNA insert, all in accordance with methods well- O known in the art. Briefly, hybridization was performed in a hybridization buffer (5X SSC/2% S blocking reagent/0.1% SDS/50% formamide, Boehringcr Mannheim, Indianapolis, IN) with a probe concentration of 10 ng/ml for 16 hours at 65 After hybridization, the filter was subjected to two washes of 0.5X SSC/0.1% SDS for 5 minutes, and two washes of 0.2X SSC/0.1% SDS for 10 minutes at room temperature. Signal was detected using antidigoxigenin antibodies conjugated to phosphotase and detection reagents 4- S nitrobluetetrazolium chloride (NBT) and 5-cromo-4-chloro-3-indolyl-phosphate

(BCIP)

0 (Boehringer-Mannheim) according to standard methods. The blots were scanned with a silver S image scanner (Silverscanner II, Lacie Limited, Beaverton, OR), and relative optical density (in arbitrary units, AU) was quantified using NIH 1.59 image software. Note that highest S levels were observed in macrophage-rich tissues.

Figure 6 shows, in comparison to a group of normal control subjects, increased human serum HMG I levels as detected in hospitalized human subjects with sepsis, wherein the septic patients have been further categorized as to whether the patient died or survived.

Detailed Description of the Invention The present invention is based upon the discovery and isolation of a highly inducible kDa protein that is released by, and accumulates in media conditioned by, cultured murine macrophage-like cells (RAW 264.7) following stimulation with LPS, TNF, or IL-1. A partial amino acid sequence of this isolated polypeptide was identical to the sequence of the HMG protein, also known as amphoterin, a protein not before linked to the pathogenesis of any disease. This information was used to clone a cDNA encoding HMGl, which sequence was expressed to provide recombinant protein, which protein was used to generate specific antii HMGl antibodies.

Therapeutic and diagnostic efficacy was determined in a series of predictive in vitro and in vivo experiments. The experiments are detailed in the Examples section. For example, following administration ofendotoxin (LDtoo) to mice, serum HMGI levels increased later (at 16 h) than well-known "early" mediators of sepsis (such as TNF and IL-1) and plateau levels of HMGI were maintained for 16 to 32 hours. Patients with lethal sepsis had high serum HMGI levels, which were not detected in normal healthy volunteers. Moreover, acute experimental administration of rHMGI to test animals, whether alone or in combination with sub-lethal amounts of LPS, caused marked pathological responses and even death. More distributed dosing schedules of lower amounts of rHMG led to significant weight loss in treated animals. These results give evidence that HMG1 is a mediator of endotoxemia and particularly a late mediator, as opposed to known "early" mediators such as TNF and IL-I.

These data further show the importance of serum HMG as a marker for the severity or potential lethality of sepsis and related conditions.

o In addition, treatment with anti-HMGl antibodies provided full protection From LDoo O doses of LPS in mice. HMGI is inducible by TNF and IL-IP, and dose-dependently stimulates TNF release from huPBMCs. TNF is a marker ofmacrophage activation, so it is likely 0 (without limitation as to implied mechanisms or being bound by theory) that HMG i promotes downstream re-activation ofcytokine cascades which, in turn, mediates late pathogenesis and lethality in sepsis and related conditions involving activation of pro-inflammatory cytokine responses. Thus, HMGI likely occupies a central role in mediating the inflammatory response to infection and injury, and antagonists of HMGI will be of therapeutic benefit in sepsis and lt' related conditions of inflammatory cascade activation. The appearance of HMGI in the S0 inflammatory cytokine cascade is suitable to propagate later phases of the host response and C, contribute to toxicity and lethality. The predictive data provided herein support the therapeutic S efficacy of HMGI antagonists and provide evidence in support of the aforementioned theory Cl regarding mechanism of action. The in vivo treatment data showed the efficacy of HMG 1 antagonists in general, and anti-HMGl antibodies in particular, for treating conditions mediated by the inflammatory cytokine cascade in general and particularly sepsis conditions, including, for example, septic shock, sepsis syndrome or other "sepsis-like" conditions mediated by inflammatory cytokines. Further, the independent pathogenicity and toxicity/lethality of HMG1 shows that HMG1 antagonists are particularly effective when coadministered with antagonists of"early" inflammatory mediators such as TNF, MIF, IL- and .0 IL-6.

In summary, HMG I is a cytokine mediator of inflammatory reactions because: 1) HMG1 is released from macrophages and pituicytes following stimulation with bacterial toxins or with pro-inflammatory cytokines (TNF or IL- 2) HMGI accumulates in serum of animals exposed to LPS and in patients with sepsis; and 3) HMG I-specific antibodies protect against mortality in a predictive lethal endotoxemia animal model of clinical sepsis and related conditions.

Pharmaceutical Composition and Method of Administration The inventive pharmaceutical composition or inventive pharmaceutical combination can be administered to a patient either by itself (complex or combination) or in pharmaceutical compositions where it is mixed with suitable carriers and excipients. The inventive pharmaceutical composition or inventive pharmaceutical combination can be administered parenterally, such as by intravenous injection or infusion, intraperitoneal injection, subcutaneous injection, or intramuscular injection. The inventive pharmaceutical composition or inventive pharmaceutical combination can be administered orally or rectally through appropriate formulation with carriers and excipients to form tablets, pills, capsules, liquids, gels, syrups, slurries, suspensions and the like. The inventive pharmaceutical composition or inventive pharmaceutical combination can be administered topically, such as by skin patch, to achieve consistent systemic levels of active agent. The inventive pharmaceutical composition or inventive pharmaceutical combination can be formulated into topical creams, skin or 1Tt mucosal patches, liquids or gels suitable for topical application to skin or mucosal membrane Ssurfaces. The inventive pharmaceutical composition or inventive pharmaceutical combination can be administered by inhaler to the respiratory tract for local or systemic treatment.

d The dosage of the inventive pharmaceutical composition or inventive pharmaceutical combination of the present invention can be determined by those skilled in the art from this disclosure. The pharmaceutical composition or inventive pharmaceutical combination will contain an effective dosage (depending upon the route of administration and pharmacokinctics of the active agent) of the inventive pharmaceutical composition or inventive pharmaceutical t .combination and suitable pharmaceutical carriers and excipients, which are suitable for the 10 particular route of administration of the formulation oral, parenteral, topical or by C inhalation). The active agent is mixed into the pharmaceutical formulation by means of S mixing, dissolving, granulating, dragee-making, emulsifying, encapsulating, entrapping or c lyophilizing processes. The pharmaceutical formulations for parenteral administration include aqueous solutions of the active agent or combination in water-soluble form. Additionally, suspensions of the active agent may be prepared as oily injection suspensions. Suitable lipophilic solvents or vehicles include fatty oils such as sesame oil, or synthetic fatty acid esters, such as ethyl oleate or triglycerides, or liposomes. Aqueous injection suspensions may contain substances that increase the viscosity of the suspension, such as sodium carboxymethyl cellulose, sorbitol, or dextran. The suspension may optionally contain stabilizers or agents to increase the solubility of the active agent or combination to allow for more concentrated solutions.

Pharmaceutical formulations for oral administration can be obtained by combining the active agent with solid excipients, such as sugars lactose, sucrose, mannitol or sorbitol), cellulose preparations starch, methyl cellulose, hydroxypropylmethyl cellulose, and sodium carboxymethyl cellulose), gelaten, gums, or polyvinylpyrrolidone. In addition, a disintegrating agent may be added, and a stabilizer may be added.

Antisense Oligomers The present invention provides antisense oligomers having a sequence effective to inhibit or block the expression of the HMG1 gene or mRNA sequence. Antisense technology, which uses specific- oligonucleotides to inhibit expression of target gene products, is developing as a therapeutic modality for human disease. Several selection criteria are available to contribute to the optimization of antisense oligonucleotide antagonists. For example, it is advisable to choose sequences with 50% or more GC content. Preferred sequences span the AUG initiation codon of the target protein, but sites in the coding region and 5' UTR may perform equally well. Such sequences are generally about 18-30 nucleotides long and chosen to overlap the ATG initiation codon from the HMGI cDNA sequence to inhibit protein expression. Longer oligomers are often found to inhibit the target to a greater extent, indicating that a preferred length is about 25 mer for the first oligonucleotides chosen as antisense reagents. Typically, three oligonucleotide sequences are chosen with regard to these criteria, and compared for antagonist activity to control oligonucleotide sequences, such as O "reverse" oligonucleotides or those in which about every fourth base of the antiscnse sequence Sis randomized. Therefore, a preferred sequence for making antisense oligomer sequences to HMGI is a 25 mer sequence chosen to overlap the ATG initiation codon (underlined) from the HMGI cDNA sequence: GAGGAAAAATAACTAAACATGGGCAAAGGAGATCCTAAGAAG [SEQ ID NO. and such preferred antisense sequences are used to construct antisense oligonucleotide agents (and suitable controls) for an in vitro comparison as antagonists of HMG1. These in vitro data are predictive of human clinical utility using antisense agents of comparable design.

HMG1-directed Antibodies SThe antibodies disclosed herein may be polyclonal or monoclonal; may be from any of O a number of human, non-human eukaryotic, cellular, fungal or bacterial sources. The c antibodies may be encoded by genomic or vector-borne coding sequences, and may be elicited against native or recombinant HMGI or fragments thereof with or without the use of adjuvants. The methods for making antibodies are methods and procedures well-known in the art for generating and producing antibodies. Generally, neutralizing antibodies against HMGI those that inhibit biological activities of HMGI particularly with regard to its proinflammatory cytokine-like role) are preferred for therapeutic applications while nonneutralizing antibodies may be as suitable for diagnostic applications. Examples of such useful antibodies include but are not limited to polyclonal, monoclonal, chimeric, single-chain, and various human or humanized types of antibodies, as well as various fragments thereof such as Fab fragments and fragments produced from specialized expression systems.

Diagnostic Assay The diagnostic assay provided here uses anti-HMG antibodies that can be either polycolonal or monoclonal or both. The diagnostic procedure can utilize standard antibodybased techniques for measuring concentrations of the gene product of HMGI genes in a biological fluid. Preferred standard diagnostic procedures are ELISA assays and Western techniques.

Example 1: Identification of HMG as a "late" mediator of endotoxemia This example provides the results of an experiment to identify and isolate later released macrophage-derived factors that play a role in sepsis and in related conditions typified by inflammatory cytokine activity. The experiment reported in this example examined murine macrophage RAW 264.7 cell-conditioned media after stimulation of the cultures with TNF.

Murine macrophage RAW 264.7 cells were obtained from American Type Culture Collections (ATCC, Rockville, MD, USA), and proliferated in culture under DMEM supplemented with fetal bovine serum and 1% glutamine. When confluency reached 70-80%, the medium was replaced by serum-free OPTI-MEM I medium and cultures were stimulated with proinflammatory cytokines TNFa or IL-1) or bacterial endotoxin (LPS).

0 The proteins released from the above stimulated macrophage cultures were surveyed.

S Specifically, at different time points, cells and cell-conditioned media were separately collected by centrifugation (3000 rpm, 10 minutes). Proteins in the conditioned medium were S concentrated by ultrafiltration over Amicon membranes with Mr cutoff of 10 kDa (Amicon Inc., Beverly, MA, USA), subsequently fractionated by SDS-PAGE, and stained with Coomassie blue (1.25% Coomassie Blue R250 in 30% methanol/10% acetic acid). After destaining with 30% methanol/7% acetic acid, protein(s) of interest those that preferentially accumulated in conditioned media of stimulated cultures) was isolated by S excision from the SDS-PAGE gel, and subjected to N-terminal sequencing analysis (Commonwealth Biotechnologies, Inc., Richmond, VA, USA).

Comparison of SDS-PAGE gel analysis of profiles of proteins accumulated in control (without TNFa stimulation) versus TNF-stimulated RAW 264.7 cells revealed a strongly 0 inducible 30 kDa protein whose concentration in the cell-conditioned medium was significantly increased after stimulation for 16 hours. Amino acid sequence analysis of this isolated protein revealed its N-terminal sequence as Gly-Lys-Gly-Asp-Pro-Lys-Lys-Pro-Arg- Gly-Lys-Met-Ser-Ser [SEQ ID NO. A review of relevant gene databases found a 100% identity to the N-terminal amino acid sequence of HMG1.

These data identified HMG as a "late-appearing" product of LPS-stimulated macrophage cultures, and therefore as a candidate pro-inflammatory mediator. This activity !0 was confirmed by administration of recombinantly produced HMGI and/or of anti-HMGI antibodies in cellular and animal model systems that are predictive of human clinical conditions.

Example 2: Cellular sources of HMG1 This example shows which cell sources are capable of releasing HMGI in response to TNF, IL-I and/or LPS. Cells studied include GH3 pituicytes, murine macrophage RAW 264.7 cells, human primary peripheral blood mononuclear cells (huPBMCs), human primary T cells, rat adrenal PC-12 cells, and rat primary kidney cells (Table The rat pituitary GH3 cell line was obtained from American Type Culture Collection (ATCC, Rockville, MD, USA), and cultured in DEME supplemented with 10% fetal bovine serum and 1% glutamine. Human PBMCs and T cells were freshly isolated from whole blood of healthy donors and cultured in RPMI 1640 supplemented with 10% human serum as previously described (Zhang et al., J Exp. Med. 185:1759-1768, 1997). When confluency reached 70-80%, the medium was replaced by serum-free OPTI-MEM I medium and cultures stimulated with proinflammatory cytokines TNFa or IL-1) or bacterial endotoxin (LPS).

Although human T cell, rat adrenal (PC-12) cells, and rat primary kidney cells contained cell-associated HMG1 as demonstrated by Western blotting analysis of whole cell lysates using HMG1-specific antibodies (see example 4 below), HMGI did not significantly accumulate in the medium of these cultures after stimulation with either TNF, IL-l 3, or LPS (Table 1).

STable 1. Induced release of HMGI from various types of cells.

O Cell Type Stimulus STNF IL-t3 LPS 01 Murine RAW 264.7 cells Yes Yes Yes Human PBMCs Yes Yes Yes Human primary T cells No No No Rat adrenal PC-12 cells No No No SRat pituitary GH3 cells Yes Yes No t Rat primary kidney cells No No No Note: PBMCs, peripheral blood mononuclear cells.

t TNF, IL-I p (minimal effective concentration 5 ng/ml for each) and bacterial endotoxin S 5 (LPS, minimal effective concentration 10 ng/ml) induced the release of HMGI from human PBMCs in a time- and dose-dependent manner (Table IFN-y alone (0-200 U/mi) did not induce HMG1 release from any of the above cells, but when added in combination either with TNF or IL-I IFN-y dose-dependently enhanced HMG1 release from macrophages, with a maximal 3-fold enhancement by IFN-y at a concentration of 100 U/ml. The release of HMGI was not due to cell death, because cell viability was unaffected by TNF, IL-l1, or LPS, as judged by trypan blue exclusion (90-92 5% viable for control vs. 88-95 4% in the presence of 100 ng/ml TNF, IL-lp or LPS). The amount of HMGI released by pituicytes and macrophages inversely correlated with the intracellular concentration of HMGI, as determined by Western blotting analysis, indicating that the released material is, in part, derived from preformed cell-associated HMGI protein.

Potential sources of circulating HMGI in vivo were assessed by hybridization of an HMG 1-specific probe to mRNA prepared from various normal human tissues (blot substrate available from commercial sources), with the results summarized in Figure 5. Several macrophage-rich tissues (lung, liver, kidney, pancreas and spleen) exhibited the most abundant HMG 1 mRNA expression; less was observed in pituitary, bone marrow, thymus, lymph node and adrenal gland. In addition to providing information as to the relative tissue distribution of- HMG1 expression, this study shows the practicality and utility of assaying for HMGi-specific nucleic acid sequences in tissue samples.

Example 3: Recombinant HMGI Administration, in vitro and in vivo This example details procedures to produce HMG1 by well-known recombinant DNA technologies. The HMGI open reading frame was amplified by PCR and subcloned into an expression vector (pCAL-n). Briefly, the 648-bp open reading frame of HMGI cDNA was PCR amplified (94 "C 56 °C 72 °C 45", 30 cycles) from 5 ng Rat Brain Quick-Clone cDNA (Catalog 7150-1, Clontech, Palo Alto, CA, USA) using primers containing the following sequences, CCC GCG GA CCA TCG AGG GAA GGA TGG GCA AAG GAG ATC CTA [SEQ ID NO. and CCC GCA AGC TTA TTC ATC ATC ATC ATC TTC o T [SEQ ID NO. The 680 bp PCR product (4 ug) was digested with Bar HI and Hind O II, and cloned into the Barn HI/Hind III cloning sites of the pCAL-n vector (Stratagene, La o Jolla, CA, USA). The recombinant plasmid was transformed into E. coli BL2I(DE3)pLysS S (Novagen, Madison, WI, USA), and positive clones were screened and confirmed by DNA sequencing on both strands using a Tag DyeDeoxy terminator cycle sequencing kit on the ABI 373A automated fluorescent sequencer (Applied Biosystems, Foster City, CA, USA).

To express recombinant HMGI, positive clones were cultured at 37 °C with vigorous S shaking (250 rpm) until OD600 reached 0.6, when IPTG (1 mM) was added. Twelve hours S after IPTG induction, bacterial cells were harvested by centrifugation (6500 rpm, 15 minutes), and lysed by freeze-thaw cycles. The water-soluble fraction was collected after centrifugation minutes, 12,000 rpm), and recombinant HMGI was purified on a calmodulin-binding resin S column as instructed by the manufacturer (Stratagene). Bacterial endotoxin was removed from C(N the recombinant HMGI by using Detoxi-Gel endotoxin-removing gel (Pierce, Rockford, IL USA, Cat. #20344), and residual LPS content was determined by the Limulus Amebocyte Lysate Test (LAL test, Cat. 50-648U, QCL-1000 Chromogenic LAL, Bio-Whittaker, Inc., Walkersville, MD, USA). Purified recombinant HMG1 was added to cultures of human peripheral blood mononuclear cells (HuPBMCs), and supernatants assayed for TNF by ELISA four hours after stimulation. The LPS-neutralizing agent polymyxin B (10 pg/ml) was added concurrently with recombinant HMG 1 to eliminate the effect of any contaminating LPS on TNF release. Additionally, recombinantly derived HMG 1 was administered to test animals, with or without the additional endotoxemic challenge of exogenous LPS, to study the pathogenic potential of high levels of HMG in vivo (see Figures 2B and 2C). In some experiments, serum samples were secured from HMGl-treated animals to be assayed for TNF as detailed herein (see Figure 1B).

The above procedure provides recombinant HMGI as a fusion peptide comprising a kDa calmodulin-binding domain and a thrombin cleavage site as an amino terminal extension in register with the HMGI peptide sequence. In some experiments, the fusion tag was removed from an aliquot of the recombinant protein and the bioactivity of the full fusion protein was compared to the cleaved HMGI peptide; no significant difference in bioactivity was noted and additional experiments (especially those requiring administration of recombinantly produced HMGl to animals) typically were conducted with the (uncleaved) fusion protein.

As demonstrated in Figures 3A and 3B, in vitro or in vivo administration of recombinantly derived HMGI induced a brisk TNF response, confirming the identification of HMG1 as a late-appearing LPS-induced macrophage-derived endogenous mediator with proinflammatory activity.

Example 4: Anti-HMGI antibodies and immunodetection This example provides the results of experiments to generate and use polyclonal antibodies against HMG1. Briefly, polyclonal antibodies against an oligopeptide o corresponding to the N-terminal amino acid sequence of HMGI, or against purified O recombinant HMGI, were generated in rabbits according to standard procedures well known in the an. Briefly, eight copies of an oligopeptide with the sequence GKGDPKKPRGKMSSC 0 [SEQ ID NO. 4] were anchored to radially branching lysine dendrites (small immunogenically inert core). These large macromolecules were injected three times both subcutaneously and intradermally (0.5 1.0 mg per injection) into rabbits at week 1, 2, and 4 after pre-bleed at Day 0. Two weeks after the last immunization, rabbits were bled and boosted intramuscularly with mg of antigen, followed by a second bleeding two weeks later. Alternatively, to produce polyclonal antibodies against recombinant HMGI, rabbits were immunized with recombinant HMGI fusion peptide (100 jig per injection) following a similar protocol. Monoclonal .I antibodies reactive against HMG1 that bind, and in some cases, neutralize or antagonize O the biological activity of HMGI) are conveniently prepared according to methods well known in the art using the HMG] antigens described herein or other HMGI peptide fragments as immunogens. Such monoclonal antibodies, and/or the hybridomas that produce them, are useful to produce various "humanized" antibodies reactive against HMG1 (all according to methods known in the art), which humanized antibodies are useful as taught herein.

HMG I-specific antibodies were used to measure by Western blotting analysis the inducible release of HMG1 from RAW 264.7 cells after treatment with TNF or LPS (Figure 1).

Briefly, proteins were fractionated by SDS-PAGE on a 4-20% gradient gel, transferred to a PVDF membrane, and blotted with rabbit antiserum raised against either the N-terminal synthetic HMGI antigen or against recombinant HMGI. The signal was detected using a ECL kit as instructed by the manufacturer (Amersham Life Science Inc., Arlington Heights, IL, USA), and levels of HMG1 were determined by measuring optical intensity of bands on Western blots digitized for analysis using NIH 1.59 image software, with reference to a standard curve of purified recombinant HMGI.

No HMGI protein was detected in RAW 264.7 cells-conditioned medium in the absence of TNF or LPS treatment, but HMGI accumulated in conditioned medium to high levels after such stimulation, reaching a plateau at 8-28 hours after stimulation (Figure IA). In summary, the data presented in Examples 1, 3 and in Figure 1 A show that the release of HMGI from macrophages is stimulus-specific and time- and dose-dependent, with maximal accumulation observed within 8 hours after stimulation with TNF at concentrations as low as ng/ml. It is well appreciated that sepsis, septic shock and related conditions may occur in humans in response to stimuli that differ qualitatively or quantitatively from the single large, lethal LPS bolus used in this predictive model. Nevertheless, experimental endotoxemia has been a valuable and predictive model system by which to identify critical components of the inflammatory cytokine cascade and by which to identify specific antagonists with predicted clinical utility. In this regard, HMGI antagonists are perhaps more therapeutically attractive than TNF antagonists in view of the later appearance of HMG1 versus TNF in the response to endotoxin.

SExample 5: Detection of HMGI in in vivo animal models SThis example illustrates an in vivo experiment in rodents measuring scrum HMG1 levels after administration of a sublethal dose ofLPS (LDso). Mice or rats were treated with LPS, and sera were collected at different time points, and assayed for levels of HMG1 by Western blotting analysis. The serum concentrations of HMGI were estimated by measuring the optical band intensity with reference to a standard curve of purified HMGI. Serum levels increased significantly by 16 hours after LPS, and remained high for at least 32 hours (Figure 1B), and were not detectable in vehicle-treated control animals. These data show that HMGI t represents a particularly attractive target for diagnosis of, and pharmaceutical intervention against, sepsis and related disorders of cytokine toxicity because HMG is a late-appearing C mediator in the inflammatory cytokine cascade.

Example 6: Benefits of protection against HMGI C, This example provides the results of a predictive in vivo assay to measure therapeutic activity of antagonists of HMG in relation to treatment of sepsis and related conditions of cytokine-mediated toxicity. In this example, the HMGI antagonist was an anti-HMGI antibody preparation. Controls treated with pre-immune serum developed lethargy, piloerection, diarrhea, and succumbed to death within 48 hours. These clinical signs of endotoxemia were significantly prevented by administration of anti-HMGI antibodies. Male Balb/C mice (6-7 weeks, 20-23 grams) were randomly grouped (10 animals per group) and pre-treated either with control (pre-immune) or anti-HMG1 serum (as made in Example 4) minutes before administration (intraperitoneally) of a lethal dose of LPS (50 mg/kg in 1 x PBS). Other experimental groups received additional doses of anti-HMGl serum at +12 or, +12, and +36 hours after LPS administration. Animals were observed for appearance and survival for at least two weeks.

Polyclonal antibodies against recombinant HMG 1 were generated in rabbits, and antiserum was assayed for specificity and titer by ELISA and Western blotting procedures.

The polyclonal antiserum immunospecifically recognized (bound to) recombinant HMGI in Western blot analysis, for instance, and discriminated rHMGI from other proteins in both crude bacterial lysates and as a purified protein that had been diluted into mouse serum. Using chemiluminescence-amplified detection methods in Western blotting analysis, polyclonal anti- HMGI antiserum at dilutions up to 1:1000 was useful to detect as little as 50 pg rHMGl protein. Administration of anti-HMGI antiserum in the indicated (Figure 2A) amounts at (if one dose), -0.5 and 12 (if two doses), or 12 and 36 (if three doses) hours relative to LPS challenge (at time 0) was protective against LPS-induced lethality, and repeated dosing schedules provided better protection.

Figure 2B illustrates that rHMG causes dose-dependent lethality in endotoxic mice.

Male Balb/C mice (20-23 grams) were randomized in groups of ten to receive LPS (3.15 mg/kg; a non-lethal dose) alone or in combination with purified recombinant HMG I protein.

O Administration ofHMG1 at the indicated doses 2, 16, 28 and 40 hours after LPS challenge S significantly increased the lethality of the underlying endotoxemia.

SFigure 2C illustrates the independent lethal toxicity of HMGl as a function of dose.

C) Purified rHMGI was administered to male Balb/C mice (five mice per treatment group) as a single i.p. bolus at the indicated dosage. Mice were observed for at least 48 hours, and 60% of mice treated with rHMG 1 at a dose of 500 gg/mouse died within 24 hours of rHMG 1 challenge, indicating a single dose LD 5 0 of less than 500 pg/mouse.

The protection conferred by anti-HMGI antibodies was specific, because administration of pre-immune serum, which showed no immunospecific reactivity to HMGI on Western blots, did not spare subjects from LPS-mediated mortality (Figure 2A). Moreover, C HMGl-specific antibodies did not cross-react with other macrophage-derived cytokines (e.g.

O IL-1 and TNF), eliminating the possibility that antibodies conferred protection by binding and C thereby neutralizing these mediators. Protection against sepsis, sepsis associated pathogenesis and sepsis-related diseases involving activation of pro-inflammatory cytokine cascades may be improved by combination therapy targeted against more than one component of the cytokine cascade. Antagonists ofHMGI in this regard can be combined with specific antagonists of TNF, IL-1, MIF and other inflammatory mediators, or with more broadly active antagonists of inflammatory responses that inhibit multiple components of the inflammatory cascade aspirin, NSAIDS, anti-inflammatory steroids, etc.), to provide even more effective therapeutic modalities. Protection against LPS toxicity was antibody dose-related, and more frequent dosing with higher amounts of antibody reduced mortality by up to 70% (Figure 2A). Mice were observed for at least 2 weeks in all experiments, and no late mortality occurred, indicating that anti-HMGI antibody treatment confers lasting protection against LPS lethality, and does not merely delay the time of death.

Example 7: HMGI in human disease This example provides data that establish an association between HMGI and human sepsis, and thereby support an indication for using HMGI antagonists generally and anti- HMGI antibodies in particular in human sepsis and related conditions of cytokine toxicity.

Serum HMGI levels in normal healthy individuals and critically ill patients were measured using the polyclonal antibodies generated as in Example 4 in a Western blot format with reference to a standard curve of rHMGl. HMGI was not detectable in normal controls, but accumulated to high levels in critically ill patients with sepsis (Table 2).

Table 2. Serum appearance of HMG in sepsis patients.

Patient Age HMGI Diagnosis Outcome (year) (ng/ml) 1 i l. Normal Healthy Normal Healthy Normal Healthy <dl. Normal Healthy 3 5 61 Normal Healthy O 6 31 Normal Healthy 7 55 10 Sepsis, anastomotic leak Recovered S8 70 7-20 Sepsis, colonic perforation Recovered S9 44 10-60 Sepsis, MOF, spinal reconstruction Died 60 >120 Sepsis, MOF, perforated gastric Died ulcer 1 1 47 >120 Sepsis, MOF, pneumonia Died Note: below detection limit; MOF Multiple Organ Failure.

I These data show that elevated serum HMGI levels are observed in patients with sepsis, and the highest levels of serum HMG1 are observed in lethal cases (Table These data further indicate the therapeutic importance of HMGI antagonists in sepsis and also provide evidence for the diagnostic utility of an assay for sepsis and severity potential lethality) of CN sepsis by measuring serum concentrations of HMG1. This diagnostic assay is also useful for diagnosing the severity of allied conditions involving activation of the inflammatory cytokine cascade.

Additional subjects were screened for serum HMG1 levels in association with lethal versus non-lethal sepsis, with results (cumulative with Table 2) as described in Figure 6. The data summarized in Figure 6 represent serum samples obtained from eight healthy subjects and twenty-five septic patients infected with Gram positive [Bacillus fragilis (1 patient), Enterococcusfacecalis (1 patient), Streptococcus pneumonia (4 patients), Listeria monocytogenes (1 patient), or Staphylococcus aureus (2 patients)], Gram negative [Escherichia coli (7 patients), Klebsiella pneumonia (1 patient), Acinetobacter calcoaceticus (1 patient), Pseudomonas aeruginosa (I patient), Fusobacterium nucleatum (1 patient), Citrobacter freundii (1 patient)], or unidentified pathogens (5 patients). Serum was fractionated by SDS- PAGE gel electrophoresis, and HMG1 levels were determined by Western blotting analysis with reference to standard curves of purified rHMGI diluted in normal human serum. The detection limit by Western blotting analysis is 50 pg. Note that HMGI is not detectable in normal controls, but significantly increased in septic patients. The average level of HMG1 in serum of non-surviving septic patients (N 13 patients, mean HMGI level 83.7 22.3 ng/ml) is significantly higher than in survivors (N 12, mean HMGI level 25.2 15.1 ng/ml, P 0.05). These data provide direct evidence of the utility of screening tissue (including, without limitation blood or serum) samples for HMG I sequences (protein or nucleic acid) as a diagnostic and prognostic indicator of the presence of sepsis and related disorders of cytokine activation and of the severity and likely clinical course of such diseases and conditions.

Example 8: HMG1 induces pro-inflammatory mediators and weight loss The present results provide evidence that HMG1 is a late released mediator element of the inflammatory cytokine cascade. Addition of recombinant HMG1 to primary human peripheral blood mononuclear cells led to the dose-dependent induction of TNF within four S19

O

O

hours after stimulation (Figure 3A). This stimulation by recombinant HMG1 of 0 TNF release by HuPBMCs was not due to LPS contamination because:(i) purified recombinant HMGI was not contaminated by LPS as judged by an LAL endotoxin assay; ii) addition of the LPS-neutralizing agent polymyxin B did not affect HMG1-induced TNF release; and iii) proteolytic cleavage of Srecombinant HMG1 preparations with trypsin completely abolished the TNF release activity for the PBMC cultures. HMG1 stimulation also induced f macrophages to release nitric oxide (NO).

o To confirm that HMG1 induced serum TNF release in vivo, purified recombinant HMG1 was administered intraperitoneally to Balb/C mice, and blood samples were collected to be assayed for TNF by the L929 assay. As shown in Figure 3B, TNF was not detectable in serum of control animals, but was significantly increased two hours after administration of recombinant HMG1 protein.

Repetitive administration of recombinant gene product of the HMG1 gene (100pg/mouse/day) caused significant body weight loss (Figure 4) in mice. Without limitation as to mechanism and without being bound by theory, these data are consistent with the hypothesis that HMG1 acts as a feedforward stimulator of the pro-inflammatory cascade under both in vitro and in vivo conditions. These in vivo data in a predictive model of weight loss also provide predictive evidence that a pharmaceutical formulation comprising HMG1, or a therapeutically active fragment thereof, is an effective weight loss therapy.

Example 9: In vivo sources of HMG1 Serum HMG1 levels in hypophysectomized versus control rats also were measured by quantitation of Western blot intensities as described above.

There were significantly higher HMG1 levels within 12 hours after endotoxic challenge (LPS at 1.0 mg/kg) in hypophysectomized rats (approximately ng/ml) as compared to controls (approximately 25 ng/ml). These results indicate that pituicytes are not the major source of serum HMG1 levels and that macrophages may play a quantitatively more important role.

O

O

In the specification the term "comprising" shall be understood to have a broad meaning similar to the term "including" and will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps. This definition also applies to variations on the term "comprising" such as "comprise" and "comprises".

Claims (19)

1. A method for diagnosing a condition characterized by activation of the inflammatory cytokine cascade in a mammalian subject, comprising measuring the concentration of HMG1 in a test sample from the subject, and comparing that concentration to a standard for HMG1 representative of a Snormal range of HMG1 in a like sample, wherein a level of HMG1 in the test sample higher than the standard indicates a diagnosis of a condition characterized by activation of the inflammatory cytokine cascade.

2. The method of Claim 1, wherein the condition is selected from the group consisting of sepsis, arthritis, systemic lupus erythematosus, trauma hemorrhage and malaria.

3. The method of Claim 1, wherein the condition is arthritis.

4. The method of Claim 1, wherein the condition is systemic lupus erythematosus.

5. The method of Claim 1, wherein the condition is trauma hemorrhage.

6. The method of Claim 1, wherein the condition is malaria.

7. A prognostic method for monitoring the severity and predicting the likely clinical course of a mammalian subject having a condition characterized by activation of the inflammatory cytokine cascade, comprising measuring the concentration of HMG1 in a test sample from the subject, and comparing that concentration to a standard for HMG1 representative of a normal range of HMG1 in a like sample, wherein the magnitude of the difference between the level of HMG1 in the test sample and the standard positively correlates with the severity of the condition characterized by activation of the inflammatory cytokine cascade and with a poor prognosis.

8. The method of Claim 7, wherein the condition is selected from the group consisting of sepsis, arthritis, systemic lupus erythematosus, trauma hemorrhage and malaria.

9. The method of Claim 7, wherein the condition is arthritis.

The method of Claim 7, wherein the condition is systemic lupus erythematosus.

11. The method of Claim 7, wherein the condition is trauma hemorrhage.

12. The method of Claim 7, wherein the condition is malaria.

13. A method for diagnosing sepsis in a mammalian subject, comprising measuring the concentration of HMG1 in a test sample from the subject, and comparing that concentration to a standard for HMG1 representative of a normal range of HMG1 in a like sample, wherein a level of HMG1 in the test sample higher than the standard indicates a diagnosis of sepsis.

14. A prognostic method for monitoring the severity and predicting the likely clinical course of a mammalian subject having sepsis comprising measuring the concentration of HMG1 in a test sample from the subject, and comparing that concentration to a standard for HMG1 representative of a normal range of HMG1 in a like sample, wherein the magnitude of the difference between the level of HMG1 in the test sample and the standard positively correlates with the severity of sepsis and with a poor prognosis.

A method for diagnosing rheumatoid arthritis in a mammalian subject, comprising measuring the concentration of HMG1 in a test sample from the subject, and comparing that concentration to a standard for HMG1 representative of a normal range of HMG1 in a like sample, wherein a level of HMG1 in the test sample higher than the standard indicates a diagnosis of rheumatoid arthritis.

16. A prognostic method for monitoring the severity and predicting the likely clinical course of a mammalian subject having rheumatoid arthritis comprising measuring the concentration of HMG1 in a test sample from the subject, and n comparing that concentration to a standard for HMG1 representative of a 0normal range of HMG1 in a like sample, wherein the magnitude of the difference between the level of HMG1 in the test sample and the standard positively correlates with the severity of rheumatoid arthritis and with a poor prognosis.

17. The method of any one of Claims 1 to 16, wherein the test sample is a blood or serum sample.

18. The method of any one of Claims 1 to 17, wherein the mammalian subject is a human.

19. The method of any one of Claims 1 to 18, wherein the concentration of HMG1 is measured using an antibody that binds to an HMG1 protein. DATED THIS ELEVENTH DAY OF JULY 2006 NORTHSHORE LONG ISLAND JEWISH RESEARCH INSTITUTE BY PIZZEYS PATENT AND TRADE MARK ATTORNEYS