WO1993020851A1 - Methods for identifying agents useful in treating septicemia - Google Patents

Abstract

This invention relates to a method for identifying compounds useful in the treatment of septicemia comprising: a) forming an infective fibrin clot; b) implanting said clot in the peritoneal cavity of an experimental animal; c) administering to a population of said implanted animals a compound to be tested in the presence of a known anti-infective agent; and d) identifying the compounds which increase the percentage survival in the experimental animal population when compared to a control population. The invention also relates to the treatment of septicemia employing alpha-2-macroglobulin soluble complement receptor-1 and combinations thereof.

Description

METHODS FOR IDENTIFYING AGENTS

USEFUL IN TREATING SEPTICEMIA This invention relates to methods for identifying compounds useful in the treatment of septicemia and to methods for treating septicemia employing the compounds so identified.

Septicemia (or sepsis), as used herein, is broadly defined to mean situations when the invasion of a host by a microbial agent is associated with the clinical manifestations of infection.

Sepsis can occur in hospitalized patients having underlying diseases that render them susceptible to infection . In many cases of sepsis, the predominant pathogen is Escherichia coli, followed by other Gram-negative bacteria such as the Klebsiella-Enterobacter-Serratia group and then Pseudomnnas. Although

comprising a relatively smaller percentage of infection, Gram-positive microbes such as Staphylococcus and systemic viral and fungal infections are comprehended by the term sepsis as used herein. The genitourinary tract is the most common site of infection, the

gastrointestinal tract and respiratory tract being the next most frequent sources of sepsis. Other common foci are wound, burn, and pelvic infections and infected intravenous catheters.

A serious consequence of bacteremia often is septic shock. Septic shock is characterized by inadequate tissue perfusion, leading to insufficient oxygen supply to tissues, hypotension and oliguria. Septic shock occurs because bacterial products, e.g. endotoxins from Gram-negative bacteria, react with cells and components of the coagulation, complement, fibrinolytic and

bradykinin systems to release proteases which injure cells and alter blood flow, especially in the

capillaries. Endotoxins, which are of great structural diversity and unique to Gram-negative bacteria, are associated with the lipopolysaccharide ("LPS") component of the outer membrane of the bacterial cells. The LPS from different Gram negative bacterial species are structurally similar.

Microorganisms frequently activate the classical complement pathway, and endotoxin activates the

alternative pathway. Complement activation, leukotriene generation and the direct effects of endotoxin on neutrophils lead to accumulation of these inflammatory cells in the lungs, release of their proteolytic enzymes and toxic oxygen radicals which damage the pulmonary endothelium and initiate the adult respiratory distress syndrome ("ARDS"). ARDS is a major cause of death in patients with septic shock and is characterized by pulmonary congestion, granulocyte aggregation,

haemorrhage and capillary thrombi.

Septic shock is a major cause of death in intensive care units. There are an estimated 200,000 cases per year of septic shock in the United States, and despite advances in technology (i.e., respiratory support) and antibiotic therapy, the mortality rate for septic shock remains in excess of 40%. In fact, mortality for established septic shock has decreased very little since the comprehensive description by Waisbren (Arch. Intern. Med. 88:467-488 (1951)). Although effective antibiotics are available, and there is an increased awareness of the septic shock syndrome, the incidence of septic shock over the last several decades has actually increased. With the appreciation that antimicrobial agents have failed to completely abrogate septic mortality, it is clear that interventions must be developed to be used in conjunction with antimicrobials in order to rectify the deficiencies of current established therapy. To

facilitate the process of identifying agents for the treatment of septic shock, animal models of sepsis that more closely resemble the human septic syndrome must be developed. Despite the development of a variety of animal models of endotoxemia and septic shock, it is clear that most animal models of septic shock are not

representative of the clinical septic syndrome. For example, many models utilize a bolus intravenous

injection of bacterial endotoxin or live organisms to initiate the sequelae of septic shock. Others have employed neutrophil depleting agents, such as nitrogen mustard or vinblastine (See for example Braude, A.I. etal., J. Bacteriol 98:979-991 (1969)), or hepatotoxins such as galactosamine (Galanos, C. et al. , Proc. Nat'l. Acad. Sci . USA 76:5939-5943 (1979)) to induce

sensitization to the lethal effects of bacterial

endotoxin or live organisms. However, each of these paradigms produce pathophysiologic responses that are frequently dissimilar to the clinical septic syndrome.

Ahrenholz and Simmons (Surgery 88:41-47 (1980)), and Natanson et al., (J. Clin. Invest. 83:243-251

(1989) ) have described models of sepsis produced by implanting infected fibrin clots into the peritoneal cavity of the rat or dog. In contrast to other

experimental models of sepsis, this particular approach has the advantage of initiating a slowly evolving septic process. The canine model is reported to display similar cardiovascular characteristics to the human septic syndrome, the model described does not

incorporate antibiotic therapy, and furthermore, has been optimized to study the effects of pharmacologic interventions on cardiovascular responses as opposed to the identification of agents for improving survival following septic insult.

This invention relates to a method for identifying compounds useful in the treatment of septicemia

comprising forming an infective fibrin clot; implanting said clot in the peritoneal cavity of an experimental animal; administering to a population of said implanted animals a compound to be tested in the presence of a known anti-infective agent; and, identifying the

compounds which increase the percentage survival in the experimental animal population when compared to a control population.

This invention further relates to a method of the identifying compounds for use in the prophylactic treatment of septicemia comprising forming an infective fibrin clot; administering to a population experimental animals said compound to be tested; implanting said clot in the peritoneal cavities of experimental animals in said population; administering to said implanted animals a known anti-infective agent; and identifying those compounds which increased the percentage survival in the test population as compared to a control group that had not been exposed to said compound to be tested.

This invention further relates to a method of treating sepsis comprising administering to an animal in need thereof an effective amount of alpha-2-macroglobulin.

This invention further relates to a method of treating sepsis comprising administering to an animal in need thereof an effective amount of soluble complement receptor-1.

This invention also relates to a pharmaceutical composition for the treatment of sepsis comprising alpha-2-macroglobulin.

This invention also relates to a pharmaceutical composition for the treatment of sepsis comprising soluble complement receptor-1.

This invention also relates to a pharmaceutical composition for the treatment of sepsis comprising an effective amount of soluble complement receptor-1 and alpha-2-macroglobulin.

Figure 1 illustrates the effect of gentamicin therapy on survival in animals implanted with a fibrin clot containing increasing doses of E. coli. Animals were inoculated with 3 × 10⁷ CFU (panel A), 1 × 10⁸ CFU (panel B), 3 × 10⁸ CFU (panel C), or 1 × 10⁹ CFU E. coli

(panel D), with or without follow-up treatment of gentamicin (5 mg/kg, b.i.d., s.c). Data are depicted for 7 d survival, although there was no further change in survival in any of the groups between 7 and 14 d. Refer to Table 1 for numbers of animals per group, and level of statistical significance between groups.

Figure 2 illustrates the effect of antibiotic and steroid therapy on survival in animals implanted with a fibrin clot containing 10⁹ CFU E. coli. Gentamicin (5 mg/kg, b.i.d., s.c.) and dexamethasone (1 mg/kg, u.i.d., i.v.) therapy (n = 10) were initiated 2 hr after

initiation of the septic insult.

Figure 3 illustrates the comparative effects of a bolus of E. coli endotoxin vs. initiation of a septic insult by implanting a fibrin clot containing 10⁹ CFU live E. coli on cardiac index. A bolus intravenous injection of 30 mg/kg endotoxin (LPS: n = 10) produced a prolonged cardiodepression, whereas implanting an infected fibrin clot (n = 6) or a sterile clot (n = 5) into the peritoneal cavity did not result in any

significant change in cardiac index.

It is an object of this invention to develop and characterize a model of peritoneal sepsis in the rat that resembles closely the clinical syndrome of septic shock in humans.

Although the term sepsis has been defined herein above, the following terms are also useful for

understanding the invention and have the following meanings:

Bacteremia means a form of septicemia in which the infective microbe is a bacterium.

Endotoxemia is a form of bacteremia in which an infective gram-negative organism produces an endotoxin (LPS) mediator.

Experimental animals as that phrase is used herein refers to those well-known animal species useful for laboratory testing. Preferred are mammalian species susceptible to infection by Gram-positive and Gram-negative microbes. More preferred are rodents (mice, rats, guinea pigs, rabbits, and the like), cats, dogs, pigs, sheep, cows, and the like. Most preferred are rats.

Known anti-infective agents include, without limitation, anti-microbial agents routinely used for the treatment of septicemia such as amino-glycosides (such as amikacin, tobramycin, netilmicin, and gentamicin), cephalosporin, related beta-lactam agents such as maxalactam, carbopenems such as imipenem, monobactam agents such as aztreonam, ampicillin and broad-spectrum penicillins (e.g., penicillinase-resistant penicillins, ureidopenicillins, or antipseudomonal penicillins) that are active against P . aeruσinosa, Enterobacter species, indole-positive Proteus species, and Serratia. Also included within the definition of anti-infective agents are antifungal agents, nystatin, and the like as well as anti-viral agents.

In developing models for use in the practice of this invention, the following general protocol may be employed. The appropriate laboratory animal is

anesthetized with sodium pentobarbital (30 mg/kg, i.p.). When appropriate, the abdominal surface is shaved, and a midline laporatomy performed. Septic insult such as peritonitis is induced by implanting a fibrin clot containing viable microbes, e.g., bacteria (gram-positive or gram-negative) or fungi, into the abdominal cavity. A sterile fibrin clot is implanted into the abdomen of control animals. After implantation of an infected or sterile clot, the muscle layers are closed with 4-0 silk suture, and the wound is closed with surgical staples. The animals are then returned to their cages where they recover consciousness. The animals are closely observed, and animals obviously moribund were euthanized. Animals were permitted free access to food and water both prior to and following the surgical procedure.

Although predicated on the work of Ahrenholz and Simmons (supra, 1980) and Natanson et al., (supra,

1989), a number of additional criteria were considered which lead to the development of the claimed invention. First, it was desired to develop a model that could be applied to a rodent species since adequate numbers of animals could be studied to more precisely characterize differences in the survival between treatment groups. Another important criteria was that the model utilized live microbes and thereby avoids the use of a bolus injection of bacterial endotoxin or live organisms.

Finally, it was considered important that the models employ antibiotic therapy, since patients with suspected or documented sepsis are administered appropriate antimicrobial agents. In this regard, it was desired that antibiotic therapy in this model would provide a partial improvement in survival, to approximate the clinical septic shock mortality rates of 40-60%. In this respect, development of a model with these

characteristics provides a novel, useful and unobvious means to identify pharmacologic agents for use singly or in combination with antimicrobials for the therapy of prophylaxis of sepsis.

As mentioned above, a variety of infectious agents can be employed. For example, a clinical isolate of E. coli can be isolated from sputum and utilized. The E. coli infected fibrin clots are formed by adding

sequentially 160 μl 100 U/ml human thrombin, 1 ml bacterial suspension and 2 ml 1% (wt/vol) fibrinogen solution to 24 well plastic plates. Bacterial numbers of 3 × 10⁷, 1 × 10⁸, 3 × 10⁸ or 1 × 10⁹ CFU/ml are used to inoculate the fibrin clots. The resulting mixture is then incubated at room temperature for 30 minutes before surgical implantation into the abdominal cavity.

Although fibrin is a convenient matrix, other matricies can be used such as matrigel or other material that will fulfill the entrapment and release profile for live infectious agents as disclosed herein for fibrin.

When the model is used to identify therapeutically useful compounds, two hours after clot implantation animals are treated with antibiotic therapy continuing for 7 d. Concurrently, with the antibiotic, the

compound to be tested is given to a group of test animals and survival is evaluated against antibiotic treated controls.

In order to more fully characterize the model of peritoneal sepsis, the change in hemodynamic parameters of infected animals is determined in a separate group of animals prepared for determination of cardiac output by the thermal-dilution technique. Correspondingly, mean arterial blood pressure and heart rate are also

continuously measured. Additionally, the production of a cytokine implicated as one of the mediators of the sequelae of septic shock, TNFα, can be measured in the serum of animals in which an infected fibrin clot was implanted into the abdominal cavity.

Other applications of the model are contemplated by this invention. For example, the animals can be

pretreated with the compound tested prior to onset of the septic insult to ascertain which test compound has application in a prophylactic modality.

In yet another embodiment, the model can be- used without the known anti-infective present. In this form, the model can be used as part of an in vivo confirmatory screen for the discovery of new anti-infective molecular entities.

This model of sepsis can also be used to further understand the pathophysiologic sequelae and mediator involvement in septic shock, as well as having utility in the identification of novel pharmacologic

interventions for the treatment of septic shock leading to decreased mortality. Although a variety of agents may be employed in this model, several specific applications are disclosed in the Examples which follow.

One of the agents employed was alpha-2-macroglobulin (α₂M), a major proteinase inhibitor in blood plasma. It accounts for 10% of the total trypsin inhibiting capacity of serum and makes up to about 3-4% of the total protein content. The molecule has been purified to homogeneity and its amino acid sequence is known (Sottrup-Jensen, L., et al., J. Biol. Chem.

259(13) : 8293-8331 (1984)). The following review

provides helpful background information concerning (α₂M ("Chemistry and Biology of α₂-Macroglobulin", Feinman, R.D. ed., "Annals of New York Acad. of Sci." Vol. 421 (1983)).

A second class of inhibitor was also tested, specifically, soluble complement receptor-1 (sCRl), a complement inhibitor, described in detail inter alia in PCT Application PCT/US89/01358, filed March 31, 1989, and published as WO89/09220 October 5, 1989, and PCT Application PCT/US90/05454, filed September 25, 1990, and published as WO91/05047, April 18, 1991, the

contents of each are incorporated herein by reference for purposes of background information, was also shown to be effective.

It is contemplated that both α₂M and sCR1 can be used therapeutically singly or in combination with currently accepted clinical treatments, as well as in combination with one another. The compounds are also useful prophylactically. For example, treatment prior to contemplated bowel surgery would be useful in

reducing the likelihood of post-operative septicemia.

Pharmaceutical compositions of the invention are particularly useful for parenteral administration, i.e., subcutaneously, intramuscularly or intravenously. The compositions for parenteral administration will commonly comprise a solution of the compounds of the invention or a cocktail thereof dissolved in an acceptable carrier, preferably an aqueous carrier. A variety of aqueous carriers may be employed, e.g., water, buffered water, 0.4% saline, 0.3% glycine, and the like. These

solutions are sterile and generally free of particulate matter. These solutions may be sterilized by

conventional, well known sterilization techniques. The compositions may contain pharmaceutically acceptable auxiliary substances as required to approximate

physiological conditions such as pH adjusting and buffering agents, etc. The concentration of the

compound of the invention in such pharmaceutical

formulation can vary widely, i.e., from less than about 0.5%, usually at or at least about 1% to as much as 15 or 20% by weight and will be selected primarily based on fluid volumes, viscosities, etc., according to the particular mode of administration selected.

Thus, a pharmaceutical composition of the invention for intramuscular injection could be prepared to contain 1 ml sterile buffered water, and 50 mg of a compound of the invention. Similarly, a pharmaceutical composition of the invention for intravenous infusion could be made up to contain 250 ml of sterile Ringer's solution, and 150 mg of a compound of the invention. Actual methods for preparing parenterally administrable compositions are well known or will be apparent to those skilled in the art and are described in more detail in, for

example, Remington's Pharmaceutical Science. 15th ed., Mack Publishing Company, Easton, Pennsylvania.

To effectively treat septic shock in a human or other animal, a single dose of approximately 1 mg/kg to approximately 500 mg/kg of a molecule of this invention should be administered parenterally, e.g., s.c.

(subcutaneously) i.p. (intraperitonealy), preferably i.v. (intravenously) or i.m. (intramuscularly) .

The preferred daily dosage amount to be employed of a compound of the invention to prophylactically or therapeutically treat septic shock in a human in need thereof is from about 10 mg/kg to about 100 mg/kg per day.

The compounds described herein can be lyophilized for storage and reconstituted in a suitable carrier prior to use. This technique has been shown to be effective with conventional proteins and art-known lyophilization and reconstitution techniques can be employed.

Depending on the patient condition, the

pharmaceutical composition of the invention can be administered for prophylactic and/or therapeutic

treatments. In therapeutic application, compositions are administered to a patient already suffering from a disease in an amount sufficient to cure or at least partially arrest the disease and its complications. In prophylactic applications, compositions containing the present compounds or a cocktail thereof are administered to a patient not already in a disease state to enhance the patient's resistance.

Single or multiple administrations of the

pharmaceutical compositions can be carried out with dose levels and pattern being selected by the treating physician. In any event, the pharmaceutical composition of the invention should provide a quantity of the compounds of the invention sufficient to effectively treat the patient.

The pharmaceutical formulation of the compounds of this invention, preferably also includes conventional anti-infectives as disclosed herein above, such as gentamicin. The particular anti-infective chosen should be one to which the infective organism is susceptible and is selected or modified during therapy as the infecting microorganism is more particularly identified.

Additionally, various adjunctive agents in the treatment of septic shock also are useful in combination with the components of this invention. These include sympathomimetic amines (vasopressors) such as

norepinephrine, epinephrine, isoproterenol, dopamine, and dobutamine; anti-inflammatory agents such as

methylprednisolone anti-inflammatory agents such as indomethacin and phenylbutazone; and corticosteroids such as betamethasone, hydrocortisone,

methylprednisolone, or dexamethasone; anti-coagulants such as heparin anti-thrombin III or coumarin type drugs for certain conditions and schedules; diuretics such as furosemide or ethacrynic acid; an antagonist of opiates and beta-endorphins such as naloxone; an antagonist of tumor necrosis factor or of interleukin-1;

phenothiazines; anti-histamines; glucagon; α-adrenergic blocking agents, vasodilators; plasma expanders, packed red blood cells; platelets; cryoprecipitates; fresh frozen plasma; bacterial permeability protein;

clindamycin; and antibodies to the J5 mutant of E . coli or to endotoxin core glycolipids (lipid A). Methods for preparing such antibodies are described widely in the literature.

As employed herein, the term alpha-2-macroglobulin (α₂M) comprehends not only the native tetramer form, but also those biologically functionally forms such as dimers which arise from treatment of the tetramer with denaturants such as urea. Also included within the definition are α₂M-like proteins which inhibit proteases by the "trapping" mechanism found in α₂M and are

homologous to α₂M, such as human pregnancy zone protein.

Because the above-mentioned "trapping" mechanism is independent of the class of protease or species from which the (α₂M is obtained; therapy is contemplated across widely divergent species lines. Accordingly, any species having (α₂M or (α₂M-like activity in its serum can be treated with α₂M or α₂M-like molecules from its own species or from a second donor species. As exemplified hereinbefore, bovine (α₂M is used to treat sepsis in rats. However, α₂M from human or other species could also be employed. Because of the common mechanism of action therapeutic intervention is also contemplated for exotic, rare and agriculturally important species, zoo animals, and the like. Furthermore, although α₂M and α₂M-like molecules can be purified using conventional techniques, this invention also contemplates the use of certain blood fractions which are merely enriched in α2M. It is appreciated because these compositions are less pure, higher doses (by weight) may be required. A useful human blood fraction which may be employed is one of the so-called Cohn fractions.

Furthermore, by the term soluble complement

receptor (sCR1) protein is meant any soluble form of human CR1 including soluble forms that contain all 30 extra-cellular sCR domains, including alleles,

truncates, chemically modified and fusion proteins derived from the CR-1 allotype.

The following examples are provided merely to illustrate certain specific embodiments and should not be construed as limiting the invention.

EXAMPLE I

Development of Fibrin Clot Model of Septic Peritonitis Surgical Procedure

Fischer 344 rats, obtained from Taconic Farms and weighing approximately 300 g, were anesthetized with sodium pentobarbital (30 mg/kg, i.p.). The abdominal surface was shaved, and a midline laporatomy performed. Bacterial peritonitis was induced by implanting a fibrin-thrombin clot containing viable E. coli into the abdominal cavity. A sterile fibrin clot was implanted into the abdomen of control animals. After implantation of the infected or sterile clot, the muscle layers were closed with 4-0 silk suture, and the wound closed with surgical staples. The animals were then returned to their cages where they recovered consciousness. The animals were closely observed, and animals obviously moribund were euthanized. Animals were permitted free access to food and water both prior to and following the surgical procedure.

A separate group of animals was used to determine the effects of sepsis on cardiac output, mean arterial blood pressure and heart rate. Conscious, chronically instrumented animals were prepared according to

procedures described previously (Smith, E.F., III, et al. Circ. Shock (37:33) (1992)) and (Smith, E.F., III, et al., Circ. Shock 25:21-31 (1988)). Briefly,

following induction of anesthesia with sodium

pentobarbital, a polyethylene catheter was placed into the femoral artery through an inguinal incision, and a catheter inserted into the jugular vein was advanced down to the level of the right atrium. A thermistor-tipped catheter (Columbus Instruments, Columbus, OH) was also placed into the ascending aorta via the carotid artery. All catheters were exteriorized through a ventral tail incision, and the animals were tethered via a syringe barrel fastened to the tail, as described previously (Smith, E.F., III, et al., Circ. Shock 25:21-31 (1988)). The syringe barrel also served to protect the catheters. The animals were housed in individual cages , where they regained consciousness. The animals were allowed water and rat chow ad libitum,, and allowed to recover for 1 - 2 d before use.

The arterial catheter was used to measure arterial blood pressure and heart rate, which was measured continuously (Statham P23 DC pressure transducer) on a Columbus Instruments Cardimax-II computer. Cardiac output was determined in triplicate, by injection of 0.2 ml normal saline at room temperature through the venous catheter. Cardiac index was calculated as the quotient of cardiac output and animal body weight.

On the day of study, baseline cardiac index, mean arterial blood pressure and heart rate were determined. Then, the animals were anesthetized with sodium pentobarbital intravenously, and implanted with an E. coli infected clot or a sterile clot following a midline laporatomy (vide supra). Another group of animals was injected with 30 mg/kg E. coli LPS, representing a LD90 dose of bacterial endotoxin. Hemodynamic parameters were recorded at 1 hr intervals for 5 hr, and again at 20 hr.

Bacterial Preparation

A clinical isolate of E. coli. (a gift of Dr. Paul Actor, Magainin, Inc., Plymouth Meeting, PA) isolated from sputum was utilized to produce septic peritonitis in rats. Suitable strains may also be obtained from the American Type Culture Collection or the E. coli Stock Center. The organisms were animal passed in mice and subsequently recovered and plated onto MacConkey's agar plates. The reisolated organisms were grown overnight in brain heart infusion broth, and then stored frozen at -70°C in 1 ml stocks. To inoculate the fibrin clot, organisms from thawed stocks were inoculated into brain-heart infusion broth and incubated overnight on a rotary shaker (120 rpm) at 37°C. The E. coli were harvested by centrifugation, washed 3 X and finally resuspended in normal saline. The number of organisms were quantified by turbidimetry, and the concentration adjusted with normal saline. All inoculum sizes were based on viable counts determined by scoring colony forming units on MacConkey's agar.

The E. coli infected fibrin clots were made from a 1% solution (wt/vol) of bovine fibrinogen (Sigma

Chemical Co., St. Louis, MO) in sterile saline. The fibrinogen solution was sterilized by passage through 0.2 mm Nalgene filters. The clot was formed by adding sequentially 160 μl 100 U/ml human thrombin

(Thrombostat, Parke-Davis, Morris Plains, NJ), 1 ml bacterial suspension, and 2 ml 1% fibrinogen solution to each well in 24-well plastic plates (Nunc, Inc.,

Naperville, IL). Bacterial numbers of 3 × 10⁷, 1 × 10⁸, 3 × 10⁸ or 1 × 10⁹ CFU/ml were used to inoculate the fibrin clots. The resulting mixture was then incubated at room temperature for 30 min before surgical

implantation (Natanson, C. et al., J. Clin. Invest.

78:259-270 (1986)).

TNF Evaluation

A separate group of animals were used to study the effects of E. coli sepsis on the changes in circulating TNFα levels. Fischer 344 rats were anesthetized with sodium pentobarbital (30 mg/kg, i.p.), and prepared with a catheter placed into the femoral artery through an inguinal incision. The animals were placed on a circulating water warming pad, and allowed to stabilize for approximately 0.5 hr. Then, following a midline laporatomy, E. coli infected clots or sterile clots (prepared as described above) were placed into the abdominal cavity, and the incision was closed with surgical staples. Blood samples (approximately 0.5 ml) were removed at specified timepoints, and placed immediately into sterile plastic tubes on ice. After 0.5 hr, the blood samples were centrifuged at 3000 × g for 0.5 hr at 4°C, and the supernatant transferred to plastic tubes. The samples were stored at - 70°C until assayed for TNFα by ELISA.

TNF ELISA Assay

For the determination of TNFα levels in rat serum samples, immunoassay plates (96-well, Immunolon 4) were coated for 2 hr at room temperature with hamster anti-murine TNFα antibody (4 mg/ml in 50 mM phosphate buffered saline, pH 7.5). The plates were blocked with phosphate buffered saline containing 0.5% casein, 0.01% thimersol, 0.001% phenol red and 0.25% Tween-20 (i.e., Block buffer) for 1 hr at 37°C. After three washings with wash buffer (consisting of phosphate buffered saline and 0.05% Tween-20), test samples or TNFα standards were added to the plate and incubated

overnight at 4°C. A standard titration curve was obtained by making serial dilutions of a known sample of murine TNFα in block buffer. The plates were then washed 5 times with wash buffer and incubated with rabbit anti-mouse TNFα antibody (1:3000 dilution) in block buffer for 2 hr at 37°C. Plates were washed 5 times with wash buffer and incubated with peroxidase conjugated goat anti-rabbit antibody (1:5000 dilution in block buffer) for 2 hr at 37°C. Following 5 more washes with wash buffer, substrate (1 mg/ml M-phenylenediamine in 0.1 M citrate buffer containing 0.1% urea peroxide, pH 4.5) was added to the plates for 20 min, and the color reaction was stopped by addition of 0.1 M sodium fluoride. Spectroscopy (462 nm) was performed using a micro-ELISA autoreader (Titertek Multiscan MC). The ELISA had a lower detection limit of 25 pg/ml, and was linear up to 1000 pg/ml TNFα.

Clot Heights

A separate group of animals were used to study the regression of the clots with time. Animals were

implanted with fibrin clots, and treated with gentamicin according to procedures described above. On selected days, the animals were euthanized with an overdose of anesthetic. The clots were removed from the abdominal cavity, weighed in preweighed pans (i.e., wet weight), and then dried to constant weight to determine dry weight.

Substances

Animals were treated with gentamicin sulfate

(Elkins-Sinn, Cherry Hill, NJ; 5 mg/kg, b.i.d., s.c.) and dexamethasone (Sigma Chemical Co., St. Louis, MO;

0.1, 1, or 3 mg/kg, u.i.d., i.v.) for 7 d beginning 2 hr after implantation of the clot. Dexamethasone was dissolved fresh daily in 0.9% NaCl. Bacterial

lipopolysaccharide (LPS; E. coli 055 :B5) and bovine fibrinogen were obtained from Sigma Chemical Co., and dissolved in 0.9% NaCl. Hamster anti-murine TNFα monoclonal antibody (specific activity: 4 x 106 u/mg protein) and rabbit anti-murine TNFα were obtained from Genzyme Corp. (Cambridge, MA). Peroxidase conjugated goat anti-rabbit antibody was purchased from Pierce (Rockford, IL). Human thrombin (Thrombostat) was obtained from Parke-Davis (Morris Plains, NJ).

Ethics

All animals were housed in accordance with the

"Guide for the Care and Use of Laboratory Animals", NIH publication No. 85-23. Procedures involving the use of laboratory animals are in accordance with NIH guidelines for the use of experimental animals.

Statistics

All continuously variable data in the text, figures and tables are expressed as the mean ± SEM of (n) observations. The Wilcoxon rank test for censored data was used to compare the survival rates between different groups over the 14 d observation period (Cox, D.R., et al., "Analysis of Survival Data," Chapman and Hall,

London (pg. 123-24 (1984)), and Fisher's Exact test was used to determine the statistical significance of

differences between the survival rates at 14 d (2). All tests of significance were two-tailed. The differences between groups were considered statistically significant at p < 0.05, after correction for multiple comparisons according to the procedure of Bonferroni ("Practical

Nonparametric Statistics," Conover W., Ed. John Wiley & Sons, New York (1980)).

RESULTS

The effects of increasing doses of E. coli on survival and the attenuation of septic mortality with gentamicin therapy is summarized in Figure 1 and Table I. Implanting a sterile clot had no effect on 14 d survival (n = 10, 100% survival; data not shown). Following implantation of a fibrin clot with 3 × 10⁷ - 10⁹ CFU E. coli, there was a dose-dependent reduction in survival. At a dose of 3 × 10⁷ CFU E. coli, survival at 14 d was 47%, decreasing to 10% with 1 × 10⁸ CFU E. coli. At 3 × 10⁸ CFU E. coli, most animals had died by 24 hr, with the remaining animals succumbing within 48 hr. At 1 × 10⁹ CFU E. coli, mortality was 100% within 24 hr. In all groups implanted with an infected clot, the mortality rates were significantly different from the group of animals implanted with a sterile clot (p < 0.01). The differences in the mortality between the 3 × 10⁷ CFU E.coli and the 1 × 10⁸ CFU E. coli groups was not

statistically significant (p > 0.05). However, the mortality rate of the 3 × 10⁸ and the 1 × 10⁹ CFU E. coli groups were significantly different from both the 3 × 10⁷ CFU E. coli group and the 1 × 10⁸ CFU E. coli group (p < 0.001).

Therapeutic administration of gentamicin improved the survival in all groups of septic animals (Figure 1 and Table I). In animals inoculated with 3 × 10⁷ CFU E. coli, gentamicin improved survival from 47% to 90% (p < 0.05). Gentamicin similarly improved septic survival of animals infected with larger numbers of viable E. coli. These data indicate that septic mortality in this model can be partially attenuated with antibiotic therapy, and that the magnitude of the improvement was clearly dependent upon the number of live organisms incorporated initially into the fibrin clot. It was particularly noteworthy that in the group of animals inoculated with 1 × 10⁹ CFU E coli and treated with gentamicin, there was both an early benefit on survival (i.e., a

progressive decrease in mortality over 4 d), in addition to a benefit in long term survival (i.e., 30% vs. 0% survival for untreated animals). The dose regimen of gentamicin used (i.e., 5 mg/kg, b.i.d., s.c.) gave a maximal effect since increasing the dose to 15 mg/kg (b.i.d.) did not provide a further attenuation of septic mortality.

Table I. Therapeutic Administration of Gentamicin Improves Septic Mortality Produced by Abdominal Implantation of an E. coli Infected Fibrin

Clot.

E. coli n Control % n Gentamicin Wilxocon Fisher

[CFU] survival % survival Rank Test Exact Test

3 × 10⁷ 15 47% 10 90% p < 0.05 p < 0.05

1 × 10⁸ 20 10% 10 60% ρ < 0.01 p < 0.01

3 × 10⁸ 30 0% 20 70% p < 0.001 p < 0.001

1 × 109 30 0% 30 33% p < 0.001 p < 0.001

Percent survival data are indicated for 14 d. Gentamicin therapy (5 mg/kg, b.i.d., s.c.) was administered beginning 2 hr after fibrin clot implantation, and continued for 7 d. The Wilcoxon test examined the differences in the mortality rate between the untreated and gentamicin treated groups over the 14 d period, and the Fisher Exact test analyzed the differences between the two groups only at 14 d.

The effects of an intraperitoneal bolus injection of E. coli, and the efficacy of antibiotic therapy are summarized in Table II. Intraperitoneal injection of a bolus of 1 × 10⁹ CFU E. coli (i.e., without using the fibrin clot as a vehicle) resulted in an all-or-none mortality response within 24 hr. Although

intraperitoneal injection of 1 × 10⁹ E. coli was not 100% lethal, gentamicin failed to significantly improve survival. Thus, the paradigm of a bolus intraperitoneal injection of live organisms produced a septic model refractory to antibiotic therapy. These data indicate that placing bacteria into a fibrin clot confers a substantial advantage in providing a model of sepsis which is responsive to antimicrobial therapy, and thus one which exhibits a gradual decline in the number of surviving animals over a 4 d period.

Table II. Effect of Antibiotic Therapy on Peritoneal Sepsis Produced by a Single Intraperitoneal Bolus Injection of Live E. coli.

Survival [%]

Group n 1 2 3 4 7 14 [d]

Control 20 20% 20% 15% 15% 15% 15%

Gentamicin 10 30% 30% 30% 30% 30% 30%

All animals received a single intraperitoneal bolus injection of 1 × 10⁹ CFU E. coli. Gentamicin therapy (i.e., 5 mg/kg, b.i.d., s.c.) or vehicle (i.e., control group) was started 2 hr after injection of bacteria, and continued for 7 d. The differences in the survival rates between the two groups, as assessed by the Wilcoxon rank test or Fisher Exact test were not statistically significant (p > 0.05).

One of the most important aspects in the treatment of the clinical septic shock syndrome is its apparently intractable resistance to the effects of a variety of highly potent antimicrobial agents. Despite the

development of newer antimicrobial agents, the overall incidence of clinical sepsis has increased, and

mortality remains unacceptably high, often approaching 60% of diagnosed patients. The partial attenuation of septic mortality with antibiotic therapy in this rat model of sepsis was thus a desirable response since it provides a model which incorporates antibiotic therapy, and one in which antibiotic therapy provides a

significant, but only partial, improvement in survival. Utilizing this paradigm it could be possible to

investigate the effects of other pharmacologic

interventions used in conjunction with an antibiotic regime in the attempt to achieve improved survival above that provided by an antibiotic alone.

Hinshaw et al., (Surg. Gynecol. Obstet. 149:545-553 (1979)) have reported the combination of gentamicin and steroid therapy with methylprednisolone significantly improved survival of dogs infused with live E. coli., above that achieved with either antibiotic or steroid therapy alone. Subsequently, Hinshaw and colleagues (Circ. Shock 8:291-300 (1981)) demonstrated that

administration of the combination of gentamicin and methylprednisolone effectively improved survival of baboons infused with live E. coli, even when therapy was delayed until the onset of systemic hypotension. The synergistic effectiveness of steroid and antibiotic therapy has also been demonstrated in rodent models of sepsis (Ottosson, J., et al. Circ. Shock 35:14-24

(1991)). Thus, there is convincing evidence in several experimental animal models of sepsis that the

combination of antibiotic and steroid therapy

effectively improves survival to a greater extent than that achieved with either agent alone. Therefore, it seemed reasonable to consider this pharmacologic

approach to reduce septic mortality in the model of fibrin clot-induced septic peritonitis.

The effect of antibiotic and steroid combination therapy on septic mortality induced with 1 × 10⁹ CFU E. coli is depicted in Figure 2. As described above, antibiotic therapy with gentamicin significantly

improved survival as compared to the group of animals in which antibiotic therapy was withheld. Administration of 1 mg/kg dexamethasone alone did not improve septic mortality. However, the combination of 1 mg/kg

dexamethasone with antibiotic therapy (n = 10) beginning 2 hr. after initiation of the septic insult

significantly improved the survival rate above that achieved with either therapy alone (p < 0.05, compared to the gentamicin group), although on day 14 there was no significant difference in the number of permanent survivors between the two groups (p > 0.05, Fishers Exact test). The combination of gentamicin and 3 mg/kg dexamethasone (n = 10) provided no further improvement in survival (i.e., 60% survival on day 6 and 50%

survival on day 14), suggesting that 1 mg/kg

dexamethasone provided the maximal improvement in survival. A lower dose of dexamethasone (i.e., 0.1 mg/kg) in conjunction with gentamicin therapy had no beneficial effect on either the survival rate or the number of permanent survivors on day 14 (i.e., 40% survival at d 14; p > 0.05). These results indicate that in this model steroid therapy with dexamethasone

provided a short-term improvement in the duration of survival, but did not significantly improve long-term survival. It is important to note that these findings are more consistent with the clinical experience of steroid therapy for septic shock than with previously reported observations in experimental animal models.

The effect of fibrin-clot peritoneal sepsis on hemodynamic parameters is illustrated in Figure 3 and summarized in Table III. Baseline (i.e, prior to implanting a clot or injection of LPS) values for cardiac index, mean arterial blood pressure and heart rate are in accord with values reported in the

literature, and at baseline there were no significant differences between the three experimental groups. Mean arterial blood pressure or heart rate were not altered in animals implanted with a sterile clot. In animals implanted with an infected clot, mean arterial blood pressure was not significantly changed from baseline at any time point, whereas heart rate was elevated from the baseline value only at 5 hr. There were no significant differences in cardiac index at any timepoint between the group of animals implanted with a sterile clot vs. animals implanted with infected clots (p > 0.05). In comparison, the bolus injection of 30 mg/kg E. coli LPS produced a 50% decrease in cardiac index by 1 hr., which persisted for 5 hr (p < 0.01, compared to the baseline value or to the group of animals implanted with infected clots). Most animal models of septic shock utilizing a bolus intravenous injection of bacterial endotoxin

(Young, J.S., et al. Circ. Shock 32-243-55 (1990)) or the bolus intraperitoneal injection of live organisms (Ottosson, (1991) (supra)) are characterized by a relatively rapid and persistent decrease in cardiac output. In contrast, early clinical sepsis is generally associated with a hyperdynamic phase. Although in the present study it did not appear that cardiac index was significantly increased in septic animals, it is clear that acute, profound cardiodepression did not occur following implantation of an infected clot. Therefore, it may be appropriate to consider this as a high cardiac output sepsis model as opposed to a hyperdynamic model. Table III. Effect of Fibrin Clot Peritoneal Sepsis on

Mean Arterial Blood Pressure and Heart Rate.

ParamBaseline 1 2 3 4 5 20[hr] eter

Sterile Clot (n=5)

MABP 134±3 118±14 121111 113112 113113 114113 103121

HR 443±13 422±23 423122 42215 437111 452118 415128

1 10⁹ CFU E. coli Clot (n =6)

MABP 130±5 129±2 135117 13413 13213 12813 92110

HR 433±6 428±12 42217 42817 45015 46217* 410134

30 mg/kg E. coli LPS (n=10)

MABP 12214 108±6 119±4 11812 11214 11418 112112

HR 414±8 418118 401110 407116 393118 381118 372123

Animals were implanted with a sterile clot (n = 5), a clot containing 1 × 10⁹ CFU E. coli (n - 6), or subjected to an intravenous injection of E. coli LPS (n = 10). Mean arterial blood pressure (MABP) in mm Hg, and heart rate (HR) in beats per minute (bpm). *, p < 0.05, compared to the baseline value.

The changes in blood levels of TNFα following implantation of an E. coli infected clot are depicted in Table IV. In naive animals (i.e., animals not subjected to laporatomy; n = 5) or animals implanted with a sterile clot (n = 5), the serum TNFα levels were

typically less than 50 pg/ml. Similarly, in animals prepared with a sterile clot and receiving gentamicin therapy there were no further elevations in serum TNFα levels. In comparison, TNFα levels were significantly elevated in animals prepared with a fibrin clot infected with 1 × 10⁹ CFU E. coli (p < 0.05, compared to the sterile clot group, in both untreated and gentamicin-treated animals). These results indicate that the peritoneal implantation of an infected fibrin clot was associated with significant increase in the circulating levels of a cytokine considered important in initiating pathogenic responses in septic shock. Moreover, the magnitude of the increase in TNFα in these studies (i.e., 200 - 400 pg/ml) is considerably less than those typically observed in experimental animal models

following a bolus injection of endotoxin or a rapid infusion of live organisms, but instead more closely approximate the levels of TNFα detected in clinical studies of septic shock. Further studies will be required to describe the changes in circulating TNFα levels beyond 5 hr.

Table IV Effect of an Infected Peritoneal Fibrin Clot on Serum TNFa levels.

TNFa [pg/ml]

Base0.5 1 1.5 2 3 4 5[hr] line

Naive (n=6)

34±9 48±30 28±2 26±10 106±9 26±3 25±1 ND

Sterile Clot (n=5)

45±10 37±28 27±4 50±7 104±21 35±10 31±11 25±1

Sterile Clot & Gentamicin (n=6)

33±8 ND 25±2 ND 35±10 26±1 42±10 27±2

1 × 10⁹ CFU E. coli Clot (n=8)

41±16 111±28 189±103 206±75 86±28 152±67* 198±65* 172±87*

1 × 10⁹ CFU E. coli Clot & Gentamicin (n=8)

44±8 83±21 159±69* 212167 199±104 159±56* 238±73* 292±112*

TNFa levels are expressed in pg/ml. N.D.: not determined. *, p < 0.05, compared to the mean value of the corresponding sterile clot group. There were no differerences between the group of animals implanted with an infected clot and the group of animals implanted with an infected clot and treated with gentamicin (p > 0.05).

In a separate group of animals, the time-dependent resorption of sterile or infected fibrin clots was assessed (Table V). For this study, different groups of animals were euthanized on different days, the residual clot was removed from the peritoneal cavity, and the wet and dry clot weights were determined. The mean weight of fibrin clots aged for 30 min at room temperature was 2,955 ± 28 mg (n = 10). After implantation of sterile clots into the peritoneal cavity for 24 hr (n = 10) or 48 hr (n = 10), the clot wet weights decreased to 362 ± 12 mg and 661 ± 37 mg, respectively. However, from day 1 to day 4 following fibrin clot implantation both the wet and dry weights of the clots increased. From days 4 - 10, the clots appeared to resolve. Similar to

previous observations by Ahrenholz and Simmons (supra, 1980), there was no evidence of bowel obstruction in animals implanted with sterile clots. Although the number of animals implanted with an infected clot was more variable due to the inherent lethality of animals implanted with an infected clot, similar trends in resorption of the clot seemed to exist in both groups. One noted difference was the formation of peritoneal abscesses with the infected clots. Interestingly, the clot dry/wet ratios increased from approximately 1% at baseline to greater than 11% for days 2 - 10.

Table V. Time-dependent Resorption of Fibrin Clots Implanted into the Peritoneal Cavity of the Rat

Sterile Clot Infected Clot

Day n Wet Dry n Wet Dry

Weight Weight Weight Weight [mg] [mg] [mg] [mg]

Baselin 10 2,955128 42±1 10 2,999±10 4511 e

1 10 362±12 31±3 5 331±36 27±5

2 10 661±37 77±3 5 663±26 95±5

3 10 998±56 111±6 ND

4 5 1,197±187 136±19 5 590±99 60±6

7 5 788±68 91±9 5 579±70 74±10

10 5 360±60 48±7 5 193±44 25±6

Baseline values indicate the weight of fibrin clots aged for 30 min at room temperature, but not implanted into the peritoneal cavity. All animals were treated with gentamicin (5 mg/kg, b.i.d., s.c). Infected clots contained 1 x 10⁹ CFU E.coli. N.D.: not determined.

Analysis of the clots by light microscopy or myeloperoxidase activity indicated that a considerable influx of polymorphonuclear leukocytes had occurred within 24 - 48 hr after clot implantation. These observations are consistent with those of Ahrenholz and Simmons (supra, 1980) who reported that microscopic analysis revealed that neutrophil penetration of the fibrin clots within the first 48 hr was limited to the outer 2 mm of the clot, and that bacteria deep within the clot continued to divide within this protected environment. This observation suggests that fibrin clot peritoneal sepsis creates a continuous septic focus with a on-going release of bacteria and their end products.

The foregoing results demonstrate that a suitable model of sepsis in the rat that mimics the clinical observations of sepsis can be provided. The paradigm of implanting a live bacteria infected fibrin clot into the peritoneal cavity produced a highly reproducible model with consistent lethality. Utilizing the fibrin clot to induce septic peritonitis has the advantage of trapping the bacteria within a protective fibrin capsule, thereby presenting an in situ barrier to phagocytic cells.

Within this protected field the bacteria can continue to proliferate and hence release toxins for an extended period of time. Unlike many other animal models of endotoxemia or sepsis, this invention incorporates the routine administration of antibiotic therapy, does not require prior sensitization of the animals with

neutrophil depleting agents or hepatotoxins, and the model exhibits cardiovascular and cytokine responses that are similar to the clinical septic syndrome.

Moreover, most animal models employing a bolus of endotoxin generally require that pharmacologic

interventions be administered as a pretreatment. In the present model therapeutic administration of a steroid in conjunction with an antibiotic significantly reduced early mortality, indicating clearly that the model can Table Via.

Effect of α₂-MACROGLOBULIN ON SURVIVAL

IN THE FIBRIN CLOT MODEL

150mg/kg × 1 (@ 2 hour post clot)

α2-M Gentamicin

TRIAL No. N% SURVIVAL N%SURVTVAL P value

1. 5/100 10/40 Significant

(0.0440)

2. 10/100 10/10 Significant

(0.0001)

3. 5/80 5/60 Not Significant

(1.0000)

4. 5/100 10/30 Significant

(0.0256)

5. 6/50 8/50 Not Significant

(1.4079)

6. 10/20 13/23 Not Significant

(1.0000)

7. 10/10 26/0 Not Significant

(0.2778)

8. 6/67 29/14 Significant

(0.0161)

Total 57/60 111/20 Significant

(<0.0001)

N=Number of rats in study. Table Vlb.

Effect of α₂-MACROGLOBULIN ON SURVIVAL

IN THE FIBRIN CLOT MODEL

15mg/kg × 1 (@ 2 hour post clot)

α2-M Gentamicin

TRTAL No. N%SURVTVAL N%SU RVIVAL P value

1. 10/50 10/10 Not Significant

(0.1409)

2. 7/57 10/30 Not Significant

(0.3500)

Total 17/53 20/20 Marginally Significant

(0.0808)

N=Number of rats in study.

Table Vic.

Effect of α₂-MACROGLOBULIN ON SURVIVAL

IN THE FIBRIN CLOT MODEL

150mg/kg × 2 (@ 2 & 24 hours post clot)

α2-M Gentamicin

TRIAL No. N%SURVIVAL N%SURVIVAL P value

1. 5/80 30/40 Not Significant

(0.1558)

2. 4/75 29/14 Significant

(0.0231)

3. 10/10 10/30 Not Significant

(0.5820)

Total 19/42 69/28 Not Significant

(0.3480) N=Number of rats in study.

be useful for the evaluation of therapeutically

administered interventions. This invention contemplates the use of the model of sepsis to further understand the pathophysiologic sequelae and mediator involvement in septic shock, as well as to identify novel pharmacologic interventions for the treatment of septic shock.

EXAMPLE II

Identification of Compounds Useful for Treatment of Sepsis

Alpha-2-macroglobulin was tested in the model described in Example I. All animals, both treatment and control, were administered 5 mg/kg b.i.d., s.c. gentamicin beginning 2 hours after initiation of sepsis. Alpha-2-macroglobulin at the concentration indicated was also given 2 hours post insult. The results of a variety of trials are shown in Table Vla-c.

Also tested was a preparation of Alpha-2-macroglobulin that had been treated with methylamine. Such treatment is known to inactivate the protease inhibitory activity of Alpha-2-macroglobulin. In that study no significant

improvement in survival (20%), n=5) over control (10%, n=10) was observed. Therefore, protease inhibitory activity is important for the therapeutic efficacy of Alpha-2-macroglobulin.

Employing the same protocol, two complement inhibitory substances were tested: Cobra venom factor (CVF), an agent that depletes complement generally; and sCRl a specific inactivator of C3 and C5 convertase. Since complement is involved in a host's defense against bacterial infection, CVF demonstrated the expected effect (i.e., depleting complement further compromises the animals so that the septic mortality is exacerbated). Completely unexpected in view of the CVF response was the result that sCR1 can actually have a beneficial effect in prolonging survival when given as five separate doses over a 3-day period beginning 2 hours after insult. The results are given in Table VII. Table VII. sCR1 Improves Survival in Fibrin Clot Septicemia Model in Rat

Group Dose [mg/kg] Survival [%]

Pre Post n 1 2 3 7[d]

Control 70 69 49 43 33

CVF 500 U 10 30 20 20 20*

500 U 5 40 20 0 0* sCR1 10 mg/kg × 5 14 100 93 93 79**

10 mg/kg × 1 10 70 60 60 30

All animals in all groups were administered gentamicin (5 mg/kg b.i.d., s.c.) beginning 2 hrs. after initiation of sepsis; other agents used in conjunction with gentamicin were administered either 2 hrs. after initiation of the septic insult (i.e., POST), or as a pretreatment prior to initiation of sepsis (i.e., PRE). *, p < 0.05, **, p < 0.01, compared to the control group (Wilcoxon rank-sum test).

Claims

WHAT IS CLAIMED IS:

1. A method for identifying compounds useful in the treatment of septicemia comprising:

a) forming an infective fibrin clot;

b) implanting said clot in the peritoneal cavity of an experimental animal;

c) administering to a population of said implanted animals a compound to be tested in the presence of a known anti-infective agent; and

d) identifying the compounds which increase the percentage survival in the experimental animal population when compared to a control population.

2. The method according to Claim 1 wherein said test compound is administered from about 2 to about 24 hours after the implanting of said clot.

3. The method according to Claims 1 or 2 wherein the administration of said compound is about 2 to about 4 hours after the implanting of said clot.

4. The method according to any of the Claims 1-3 wherein the fibrin clot is formed with live infective organisms.

5. The method according to any of Claims 1-4 wherein the organism is a bacterium.

6. The method according to Claim 5 wherein said bacterium is Gram negative.

7. The method according to Claim 6 wherein said Gram negative bacterium is selected from the group consisting of Neisseria, Shiσella, Escherichia, Edwardsiella, Salmonella Klebsiella, Serratιa, Proteus, Yersinia, Hemophilus,

Rordetella, and Pasteurella.

8. The method according to any of the Claims 4-7 wherein said organism is E. coli.

9. The method according to Claim 5 wherein said bacterium is Gram positive.

10. The method according to Claim 4 wherein said organism is a fungus.

11. The method according to Claim 4 wherein said organism is a virus.

12. A method for identifying compounds for the

prophylactic treatment of septicemia comprising:

a) forming an infective fibrin clot;

b) administering to a population experimental animals said compound to be tested;

c) implanting said clot in the peritoneal cavities of experimental animals in said population;

d) administering to said implanted animals a known anti-infective agent; and

e) identifying those compounds which increased the percentage survival in the test population as compared to a control group that had not been exposed to said compound to be tested.

13. A method according to any of the Claims 1-12 wherein the known anti-infective is omitted.

14. A method of treating sepsis comprising

administering to an animal in need thereof an effective amount of (a) alpha-2-macroglobulin, (b) soluble complement receptor-1 or combinations of (a) and (b).

15. The method according to Claim 14 wherein said animal is a mammal.

16. The method according to Claims 14 and 15 wherein said mammal is selected from the group consisting of humans, cattle, horses, sheep, goats, camels, whales and zoo

mammals .

17. The method according to any of the Claims 14-16 wherein said mammal is a human patient.

18. The method according to any of the Claims 14-17 wherein said effective amount of alpha-2- macroglobulin and/or soluble complement receptor is from about 1 to about 500 mg/kg/day.

19. The method according to any of the Claims 14-18 wherein said amount is from about 10 to about 100 mg/kg/day.

20. The method according to any of the Claims 14-19 wherein the alpha-2-macroglobulin and/or soluble complement receptor is administered intravenously, subcutaneously, intramuscularly, or intraperitoneally.

21. The method according to any of the Claims 14-20 wherein said administration is given therapeutically.

22. The method according to any of the Claims 14-20 wherein said administration is given prophylactically.

23. A pharmaceutical composition for the treatment of sepsis comprising an effective amount of alpha-2-macroblobulin and a pharmaceutically acceptable carrier.

24. The pharmaceutical composition according to Claim 23 including an effective amount of an anti-invective agent.

25. A pharmaceutical composition for the treatment of sepsis comprising an effective amount of soluble complement receptor-1 and a pharmaceutically acceptable carrier.

26. The pharmaceutical composition of Claim 25

including an effective amount of an antimicrobial agent.

27. A pharmaceutical composition for treatment of sepsis comprising an effective amount of alpha-2-macroglobulin, soluble complement receptor and a

pharmaceutically acceptable carrier and optionally including an anti-infective agent.