CA1338658C - Method and compositions for treatment of trauma-associated sepsis with gamma interferon - Google Patents

Abstract

Trauma-associated sepsis is effectively treated using gamma interferon alone or in combination with an antibiotic.

Description

` 1 338658 METHOD AND COMPOSITIONS FOR TREATMENT OF
TRAUMA-ASSOCIATED SEPSIS WITH GAMMA INTERFERON

Hiram C. Polk, Jr. and Gerald Sonnenfeld Background The present invention relates in general to methods and compositions for treatment of trauma-associated sepsis, and in particular to methods and compositions including gamma interferon ("IFN-y"), especially human IFN-y, for treatment of trauma-associated sepsis.
Infection remains the major cause of late death in trauma patients. Antibiotics, sterile techniques, mechanical barriers, and conventional treatment have met with limited success in preventing or controlling these infections.
A number of immunological parameters have been shown to be depressed after trauma. Expression of HLA-DR antigen (which corresponds to the Ia antigen in mouse models) is decreased on monocytes of human trauma patients.
Viral infections and tumors may be treated using -interferon or IFN-y as may intracellular bacterial infections. In a suggested mechanism for the effect of IFN-y on intracellular bacterial infections, it is proposed that the IFN-y prevents uptake of the bacteria by cells. This mechanism does not relate to infections which are not intracellular in nature.

Summary of the Invention A method according to the present invention for treatment of trauma-associated sepsis, particularly of a bacterial blood infection caused by an enteric bacterium, includes the step of administering a therapeutically effective dose of IFN-y to a victim of trauma, and in particular, to a trauma patient not known to have a viral infection, a tumor or an intracellular bacterial infection. The method involves introducing IFN-y into a bodily fluid of a patient. This introduction may be effected by injecting IFN-y subcutaneously, intramuscularly or intravenously.
Intravenous administration may be by infusion or bolus injection. The IFN-y used according to the present invention is preferably desCysTyrCys human IFN-y.
A method according to the present invention preferably also involves co-administering an antibiotic with the IFN-y. Co-administration may be effected by administration of the antibiotic and IFN-y during the course of the same infection, or may be effected by inclusion of the antibiotic and IFN-y in the same composition (particularly where administration is by intravenous infusion).
The antibiotic may be selected from the group consisting of cefazolin, nafcillin, vancomycin, cefoxitin, neomycin plus erythromycin, penicillin G, trimethoprim plus sulfamethoxazole, and clindamycin or clindamycin plus gentamycin or tobramycin.
A composition for treatment of trauma-associated sepsis according to the present invention includes IFN-y, an antibiotic and a diluent or carrier. The antibiotic may be selected from the group consisting of cefazolin, nafcillin, vancomycin, cefoxitin, neomycin plus erythromycin, penicillin G, trimethoprim plus sulfamethoxazole, and clindamycin or ` ` 1 338658 clindamycin plus gentamycin or tobramycin.

Brief Description of Drawings FIG. 1 iS a graphic depiction of the improvement in survival after IFN-y treatment according to the present invention in a suture-challenge model;
FIG. 2 is a graphic depiction of blood borne bacterial counts for cultures from suture-challenged mice in a control group of suture challenged mice and in a group of mice treated with IFN-y according to the present invention;
FIG. 3 is a graphic depiction of the effect of IFN-y treatment according to the present invention on survival of mice with a Klebsiella burn wound infection;
FIG. 4 iS a graphic depiction of the effect of burn and an IFN-y treatment according to the present invention on the expression of Ia antigen on peripheral mononuclear cells of mice;
FIG. 5 iS a graphic depiction of the survival of mice following peritoneal inoculation of different concentrations of E. coli prior to laparotomy;
FIG. 6 is a graphic depiction of the effect of administration of IFN-y according to the present invention on survival of mice after laparotomy when a second bacterial challenge was administered five days after a first challenge;
FIG. 7 iS a graphic depiction of the effect of administration of IFN-y according to the present invention on survival of mice after laparotomy when a second bacterial challenge was administered three days after the first challenge;
FIG. 8 iS a graphic depiction of the effect of laparotomy IFN-y treatment on the expression of Ia antigen on peripheral mononuclear cells;

FIG. 9 is a graphic depiction of the effect of IFN-y treatment of peripheral blood mononuclear cell cultures with respect to the density of monocyte HLA-DR
expression of trauma patients;
FIG. 10 iS a graphic depiction of the effect of IFN-y treatment on peripheral blood mononuclear cell cultures with respect to the percentage of monocytes expressing HLA-DR antigen;
FIG. 11 iS a graphic depiction of the effect of IFN-y treatment of purified monocyte cultures on the density of monocyte HLA-DR antigen expression of trauma patients;
FIG. 12 iS a graphic depiction of the effect of IFN-y in purified monocyte cultures on the percentage of monocytes expressing HLA-DR antigen;
FIG. 13 iS a graphic depiction of the effect of IFN-y in peripheral blood mononuclear cell culture on the percentage of monocytes expressing HLA-DR antigen in a patient who died of burn wound sepsis; and FIG. 14 iS a graphic depiction of the synergistic effect on abcess diameter of coadministration of IFN-y and an antibiotic according to the present invention.

Detailed Description Trauma-associated sepsis or infection remains a serious problem despite advances in antibacterial chemotherapy. According to the present invention, augmenting host immune defenses with IFN-y increases the frequency of survival of patients suffering from such sepsis or infection.
Trauma-associated sepsis is defined herein as the presence of extracellular bacteria in the blood or tissues ("sepsis") associated with a wound or injury.
Bacteria characteristic of trauma-associated sepsis include enteric bacteria, and in particular include coliform bacteria, bacilli of the genus Klebsiella and bacilli of the genus Pseudomonas. Trauma includes physical injury such as surgical incision.
Bacteria which are recognized as being associated with trauma include: Gram-positive bacteria such as Staphylococcus aureus, Streptococcus faecalis, Pneumococci, anhaemolytic Enterococci, Sarcina species and haemolytic Streptococci; and Gram-negative bacteria such as Escherichia coli, Pseudomonas species, Klebsiella species, Proteus species, Enterobacter cloacae, coliform bacteria, Serratia species, Citrobacter species and Providencia species [Allgower et al., Surg. Clin. N. Am., 60, 133-144 (1980)].
Thus, trauma-associated sepsis may be distinguished from other forms of bacterial infection ~s being characterized by the presence of trauma (i.e., injury, such as a burn or a wound) and the relative predominance of bacteria which are not intracellular in location. Sepsis after trauma usually occurs within two weeks after injury in patients who are immunosuppressed as a result of the trauma. Sepsis may be due to microbial contamination at the time of trauma, or to spillover of organisms from the gut.

` ` t 3386~8 Human IFN-y is defined herein as a polypeptide having the sequence of native gamma interferon as set forth in European Publication1,No. 77,670 and all amino acid sequence or other variants thereof which are capable of activation of monocytes from trauma patients. Examples of such variants are alleles or the products of site directed mutagenesis in which residues are deleted, inserted or substituted. For example, see European Publication No. 146,354.
IFN-y according to the present invention may be derived from any source, including IFN-y isolated from natural sources, and chemically synthesized or recombinantly-produced IFN-y.
IFN-y should be used which is homologous to the animal species to be treated since it may not be active across species lines. In human therapy, the desCysTyrCys variant of the sequence shown in EP 77,670 is preferably employed, and optionally the C-terminal variant in which the last 4 residues are deleted in post-translational processing. IFN-y having native sequences is obtained by purification from natural sources using known methods. The same molecule or its variants are obtained from recombinant sources, also by known methods.
A typical formulation may contain IFN-y (20 X
106 U) at 1.0 or 0.2 mg/ml, succinic acid at 0.27 mg/ml, disodium succinate hexahydrate 0.73 ml/injection at pH
5Ø This aqueous formulation is administered at therapeutic doses, which are less than the maximum tolerated dose in humans as determined by the clinician. IFN-y may also suitably administered from a reconstituted lyophilized preparation.
IFNy may be administered by any conventional route that directs a therapeutic dose to the site of the trauma, for example, by intravenous or intrapulmonary (European Publication No. 257,956) delivery routes.

` 1 338658 Administration may be by continuous infusion or bolus dosing in sufficient amounts to maintain therapeutic levels. IFN-y may be administered prior to the trauma, in the case of surgical trauma, or as soon as possible following accidental or non-surgical trauma, and then continued for a time sufficient to permit proper healing of the trauma, typically about from 3 to 14 days or such other period as is determined by the clinician.
IFN-y iS preferably administered according to the present invention subcutaneously at doses of from 0.01 to 0.1 mg/m2/day for 7-10 days. However, other routes of administration, such as intramuscular or intravenous administration, may be employed. Moreover, immunologically effective dosage levels may generally be determined for a particular application according to the procedure of Maluish et al., J. Clin. Oncol., 6, 434-435 (1988).
According to the present invention, IFN-y may be administered therapeutically (i.e. after appearance of sepsis) or prophylactically (i.e., prior to appearance of sepsis).
IFN-y and antibiotics may be co-administered with synergistic effect according to the present invention. Thus, trauma-associated sepsis which is not effectively treated by antibiotics alone may be treated according to the present invention. Antibiotics which may be useful according to the present invention include penicillin, ampicillin, amoxacillin, methicillin, oxacillin, cloxacillin, diclosacillin, nafcillin, carbenicillin, cephalothin, cefazolin, cephradine, cephalexin, erythromycin, lincomycin, clindamycin, chloramphenicol, tetracycline, doxycycline, minocycline, gentamicin, kanamycin, tobramycin, vancomycin, trimethoprim and sulfamethoxazole. [Cushing, Surg.
Clin. N. Am., 57, 165-177 (1977)]

The following Examples more particularly illustrate the invention.
In Example 1, the ability of IFN-y treatment to alter the course of infection is demonstrated in a model of surgical wound infection. This model simulates a classical surgical wound infection involving pathogenic bacteria, a foreign body, tissue injury and an open wound. Simulation is obtained by placing a suture impregnated with Klebsiella pneumoniae ("K.
pneumoniae") into the thigh of a mouse. Mortality rates are high for untreated animals in this model.
In Example 2, the simulated wound infection-treatment model involves subcutaneous injection of IFN-y. Bacterial challenge in this model consists of intramuscular injection of K. pneumoniae.
In Example 3, an infected burn wound model was used to study the efficacy of IFN-y therapy. Two different Gram-negative wound infections, specifically those involving K. pneumoniae and Pseudomonas aeruginosa ("P. aeruginosa"), were studied.
Example 4 includes a description of a laparotomy model of clinical conditions following trauma. Bacterial challenge involved intraperitoneal injections of Escherichia coli ("E. coli") or intramuscular injection of K. pneumoniae.
Example 6 is a description of a determination of values for MO2DR (i.e. for monocyte HLA-DR antigen expression) in cultures of blood samples from normal subjects and trauma patients before and after incubation with IFN-y.
Example 5 involves a description of a study of the effect of IFN-y on monocyte HLA-DR antigen expression in trauma patients.
HLA-DR antigen expression by monocytes and macrophages is necessary for effective antigen presentation to lymphocytes. Traumatic injury and g sepsis reduce HLA-DR antigen expression of monocytes.
In Example 7, a study is described in which patients with severe injury received recombinant IFN-y therapy along with prophylactic antibiotic therapy.
Example 8 illustrates the increased effectiveness against bacterial infection observed with the use of an antibiotic plus IFN-y, as opposed to the use of either alone.

Example 1 Adult male CBA/J mice, weighing 20-25 g each, (Jackson Laboratories, Bar Harbor, Maine) were deprived of food but allowed free access to water for 24 hours prior to bacterial challenge and refed 4 hours following that challenge. This model has been shown to parallel infections of surgical wounds. Polk et al., Surgery, 90, 376 (1981).
Lengths of 3-0 twisted cotton suture were incubated overnight in trypticase soy broth (BBL
Microbiological Systems, Cockeysville, Maryland) and inoculated with K. pneumoniae (Capsular Type 2).
Lengths of suture attached to a French eye needle were inserted aseptically into the right thigh of each mouse and the suture was cut flush with the skin at either end and buried subcutaneously.
Murine IFN-y, produced by recombinant DNA
technology, was a gift of Genentech, Inc., South San Francisco, California. The IFN-y had specific activity of approximately 2.3 X 107 U/mg protein and was diluted with RPMI-1640 medium (Gibco Laboratories, Grand Island, New York). The interferon was titered prior to use by means of a microplaque reduction assay using the Indiana strain of vesicular stomatitis virus on L-929 cells.
Campbell et al., Can. J. Microbial., 21, 147 (1975). In this assay, one interferon unit was equivalent to 0.88 ` 1 338658 NIH G-002-904-511 reference standard units.
IFN-y was administered subcutaneously into the left or opposite hind limb in daily 0.1 ml injections of 7500 units each. Treatment commenced 5 days prior to suture challenge and the last dose was administered 1 hour prior to this challenge. In all experiments, control animals received 0.1 ml of RPMI-1640 medium subcutaneously at the same time that IFN-y was administered.
A first experiment assessed the effect of IFN-y on survival following suture challenge. Survival of an IFN-y group (n=16) was compared to a control group (n=16). Surviving mice were observed for a minimum of 3 weeks after bacterial challenge to assess late mortality.
A second experiment determined the effect of IFN-y on local and systemic bacterial recovery following suture challenge. An IFN-y group (n=26) was compared to a control group (n=35). Six to eight animals from each group were sacrificed by cervical dislocation at 1, 2 and 3 days after suture challenge. Blood was withdrawn immediately from each sacrificed animal by cardiac puncture. The medial compartment musculature, including the inserted suture, was excised "en bloc" and the suture separated and its length measured with a micrometer. Blood and muscle then underwent quantitative bacteriology. The number of mice that died during the experiment was recorded.
To estimate the initial bacterial inoculum, the number of bacteria per millimeter of suture was determined by homogenizing a 1 cm length of the suture with 10 ml of sterile phosphate buffer saline (PBS), using a glass mortar and electrically-driven Teflon~
pestle. After agitation for 5 minutes, the homogenate was serially diluted in PBS, plated on nutrient agar (BBL Microbiology Systems, Cockeysville, Maryland), and 1 33~

incubated overnight at 37C., after which the bacterial colonies were counted. The number of bacteria initially introduced into an animal was equal to the suture length in millimeters multiplied by the number of bacteria per millimeter in the suture.
Quantitative determination of wound bacteriology was performed in a similar manner. Each excised specimen of muscle, together with the inoculating suture, was homogenized and plated as above. Values for bacterial quantity were determined as log10 of blood bacterial counts or log10% local bacterial recovery from muscle. Of the blood removed by cardiac puncture, 0.1 ml was serially diluted with PBS
and plated on nutrient agar. After overnight incubation, the number of colony-forming units was counted.
The Student's t Test for independent means was used to analyze muscle and blood culture results, and the Chi Square test with Yates' correction for continuity was used in the analysis of survival data.
In all cases, differences were considered significant at p < 0.05.
The intramuscular dose of Klebsiella introduced with the suture ranged between 1.5 and 2.2 X 106 organisms. The IFN-y treated group had significantly greater survival than the control group from day 3 onwards as is illustrated in FIG. 1. In FIG.
1 the survival of mice treated for 5 days prior to suture challenge is plotted using squares to indicate treatment with RPMI-1640 medium (controls) or using circles to indicate treatment with IFN-y. There were no further deaths after day 6, and 10 mice in the IFN-y treated group survived as compared to only 2 mice in the control group.

In the second experiment, significantly more mice in the control group died (15 of 35) than in the IFN-y group (4 of 26) as set forth in Table 1.

Number of Animals Sacrificed for Quantitative Cultures and Mortality At Each Time Period Mortality atControl Group IFN-y Treated Each Time Period (n = 35) Group (n = 26) Day 1 Sacrificed 7 7 Died 6 Day 2 Sacrificed 7 7 Died 4 Day 3 Sacrificed 6 8 Died 5 2 The number of animals sacrificed for quantitative bacteriology is also shown in Table 1. As illustrated in FIG. 2, the IFN-y treated group had significantly less systemic blood-borne bacterial counts than the control group for each time period. In FIG. 2, the effect of administration of IFN-y on blood cultures of mice treated for 5 days prior to suture challenge is plotted using squares to indicate treatment with RPMI-1 33~

1640 medium (controls) or using circles to indicate treatment with IFN-y.
There was no difference in local bacterial recovery at the site of infection between the IFN-y treated group and control group for any time period.
Pretreatment of mice with IFN-y significantly reduced the mortality observed using the model. This indicates that IFN-y treatment alters the course of surgical wound infection.
In the second experiment, recovery of bacteria at the local site of infection was not altered in IFN-y treated mice; however, recovery of bacteria from the circulation of IFN-y treated mice was reduced significantly compared to controls. This suggests that the IFN-y treatment enhances the ability of the host to limit the spread of bacterial infections. The limitation on the spread of bacterial infection may contribute to the decreased mortality observed in mice treated with IFN-y.
Example 2 The efficacy of IFN-y in prophylaxis and therapy was studied in a simulated wound infection model using a typical surgical pathogen. Groups of 12 CBA/J
mice were subcutaneously injected with either murine IFN-y or RPMI-1640 medium (controls). The murine IFN-y was produced by recombinant DNA technology (specific activity 2.3 X 107 units/mg protein) and was a gift of Genentech, Inc. Bacterial challenge consisted of intramuscular injections of K. Pneumoniae (103 organisms in 0.1 ml of physiological saline).
Mice pretreated with IFN-y at a dose of 7,500 or 750 units per day for 3 days, infected, and then treated therapeutically for 2 days survived significantly longer than controls or mice treated with i 1 338658 150 units of IFN-y per day. Significantly (p < 0.05) greater survival than controls was observed with pretreatment for 5 or 3 days with IFN-y, but not with 1 day of pretreatment.
Administration of IFN-y to the opposite hind limb from the one receiving bacterial challenge was as effective as treatment in the same limb. When IFN-y treatment was commenced 1 hour after bacterial challenge and continued for 7 days, 13 of 60 mice survived, a significantly (p < 0.05) greater number than the 4 of 60 surviving controls.
These data confirm that IFN-y is effective in pretreatment of Klebsiella pneumoniae infection of mice. Furthermore, these results support the therapeutic efficacy of IFN-y in treatment of an extracellular bacterial infection associated with trauma.

Example 3 Adult male CBA/J mice, each weighing 20-25 g (Jackson Laboratories, Bar Harbor, Maine), were anesthetized and were burned. Immediately after administration of a burn wound, bacteria were topically applied to the burn wound. The mice were anesthetized using 3.75 mg Ketamine~ hydrochloride (Bristol Laboratories, Syracuse, New York) and 0.5 mg Xylazine~
(Miles Laboratories, Shaunee, Kansas) in 0.1 ml of physiological saline injected intramuscularly into the right hind limb. The dorsum was clipped with an animal clipper.
The animals were then placed in a mold which was specially designed to expose a predetermined amount of the dorsum while protecting the remainder of the mouse from thermal injury. The mold consisted of a plastic cylinder lined with styrofoam with a large opening on one side through which the mouse was inserted and a smaller opening on the opposite side which exposed an area of the clipped dorsum equal to 30~ of the total body surface of the mouse. Full thickness burns were consistently produced by immersion of the exposed dorsum in an 85 C water bath for 8 seconds. The mice were resuscitated with 2 cc of Ringers lactate administered intraperitoneally.
Two different bacterial challenges were used:
K. pneumoniae (capsular type 2) and P. aeruginosa.
Immediately after resuscitation, bacteria (lo8 organisms) were administered topically to the burn wound in 1 ml of 10~ gelatin using a sterile swab.
Murine IFN-y was administered subcutaneously into the left hind limb in daily doses of 7500 units in 0.1 ml injections.
The murine IFN-y, as described in Example 1, was diluted with RPMI-1640 medium. Treatment commenced 5 days prior to burn and bacterial challenge, and the last dose was administered prior to this challenge. In all experiments, control animals received 0.1 ml of RPMI-1640 medium subcutaneously at the same time that the IFN-y was administered to the experimental animals.
Chi Square tests with Yates correction were used to determine the significance of differences in values obtained in the survival experiments set forth below. The Student's t Test was used to compare groups in experiments involving the expression of Ia antigens on mononuclear cells. In all cases, differences were considered significant at p < 0.05.
Four experiments were performed. The purpose of the first experiment was to assess the effect of thermal injury alone on survival. Twelve untreated mice were burned and given no bacterial challenge.
Burning without bacterial challenge produced no mortality. All twelve mice survived for three weeks following burning.

The second experiment assessed the effect of IFN-y on survival of mice with K. pneumoniae burn wound infection. Survival of an IFN-y group (n=47) was compared to a control group (n=48).
When mice had K. pneumoniae burn wound infection, there was significantly greater survival in the IFN-y treated group than the control group from day 3 onwards, as illustrated in FIG. 3 in which a square denotes a value for the control group, a circle denotes a value for the experimental group and an asterisk denotes a value significantly different from the value for the control group on the indicated day. There were no deaths after day 8 of the experiment. Fifteen of 47 mice survived in the group treated with IFN-y survived compared to 7 of 48 surviving mice in the control group.
Burn alone significantly decreased mononuclear cell Ia antigen expression compared to controls on days 1 and 3, as illustrated in FIG. 4. In FIG. 4, square denotes a value for the control group, a triangle denotes a value for the burned but untreated group, a circle denotes a value for the experimental (burn plus IFN-y) group, an asterisk denotes a value significantly different from the value for the control group on the indicated day, and the error bars indicate the standard error of the mean.
The Ia antigen expression returned to normal by day 5 post-burn. Treatment of burned mice with IFN-y prevented the drop in mononuclear cell Ia antigen expression on day 3 post-burn. In animals which were not burned, there was no difference between the control group and the group receiving IFN-y alone (not shown) at any time.
The third experiment assessed the effect of IFN-y on survival of mice with P. aeruginosa burn wound infection. Survival in an IFN-y group (n=27) was compared to survival in a control group (n=28).

When mice had P. aeruginosa burn wound infection there was no difference observed in survival between the IFN-r treated group and the control group.
In the foregoing three experiments, surviving mice were followed for a minimum of 3 weeks after bacterial challenge to assess late mortality.
The fourth experiment assessed the effect of burn and IFN-y treatment on peripheral mononuclear cell Ia antigen expression. One hundred and forty one mice were divided into four groups. One group received only a burn (n=39), one group received a burn and IFN-y (n=39), and a control group (n=39) received no treatment. The final group received only IFN-y (n=24). Thirteen animals in each of the first three groups and 8 from the IFN-r alone group were sacrificed on days 1, 3 and 5 post burn. Blood was obtained by cardiac puncture and mononuclear cells were assayed for levels of Ia antigen expression.
Aliquots (100 ~1) of heparinized blood from each mouse were stained for 20 minutes with a fluorescein isothiocyanate-conjugated monoclonal anti-Ia antibody (MA3 043, clone OX6, Accurate Chemical and Scientific Corp., Westbury, New York). Erythrocytes were lysed with 1 ml of lysing reagent (Ortho Diagnostics, Raritan, New Jersey). The specimens were then washed in buffer and fixed in 1% paraformaldehyde in phosphate buffered saline (pH 7.4) prior to flow cytometric analyses. Stained samples were analyzed on an Ortho Cytofluorograph IIs flow cytometer (Ortho Diagnostics, Westwood, Massachusetts). Two thousand mononuclear cells were selected for fluorescent analysis by forward versus right angle light scattering properties.
The percentage of cells staining for Ia antigen was recorded for each specimen. Complete white blood cell counts (obtained using a Coulter Counter, ~.

Coulter Electronics, Inc., Edison, New Jersey) and Wright-stained microscope differential counts were made, and the absolute number of mononuclear cells was determined. The total Ia antigen expression per sample was calculated by multiplying the number of mononuclear cells by the value for the percent of cells staining positively for Ia antigen.
When mice were treated for five days with IFN-y prior to burn wound and bacterial challenge, survival was enhanced significantly in K. pneumoniae-infected mice compared to controls. No significant changes in survival were noted in mice which were pretreated with IFN-y, prior to a burn wound, for infection with P. aeruginosa. Different bacteria may respond to varying degrees to this treatment protocol in this burn wound infection model.
Different protocols of IFN-y treatment, perhaps combined with antibiotic chemotherapy, may be useful for successful therapy of P. aeruginosa infections. For example, in preliminary experiments, the addition of 2-5 days of daily subcutaneous injection of 7500 units of IFN-y into the left hind limb after burn and bacterial challenge appeared to increase resistance of burned mice to some strains of P.
aeruginosa. Optimization of treatment protocols or of dosage of IFN-y, or the combination of IFN-y with other materials may contribute to further improved results.
Because of its immunoregulatory properties, IFN-y, may play a crucial role in modulating host defenses against infection of burns. In the above experiments, decreased Ia antigen expression on mononuclear cells was observed in burned mice.
Maintenance of Ia antigen expression in burned mice treated with IFN-y may contribute to the enhanced resistance of mice to infection with K. pneumoniae.

Example 4 Adult male CBA/J mice (H-2k) [Jackson Laboratories, Bar Harbor, Maine] weighing 20 to 25 g each were lightly anesthetized using ether and placed in a supine position. An approximately 2.5 cm midline abdominal incision was made, and care was taken to avoid injury to the abdominal contents. The abdominal cavity was lavaged with 3 ml of lactated Ringer's solution (Abbott Laboratories, Chicago, Illinois) and then closed using 4-0 nylon sutures. Next, the skin was closed with silk sutures, and normal saline (0.5 ml) was administered subcutaneously. Animals were then returned to their cages and were allowed food and water ad libitum. Mice that had undergone laparotomy ate and drank normally and did not show any signs of illness.
Laparotomy without bacterial challenge produced no mortality.
E. coli (105 organisms) were administered in 1 ml intraperitoneal injections one-half hour prior to laparotomy. K. pneumoniae (capsular type 2) was administered in 0.1 ml intramuscular injections into the right hind limb on day 5 following laparotomy, unless otherwise indicated.
Murine IFN-y, as described in Example 1, was diluted with RPMI-1640 medium (GIBCO Laboratories, Grand Island, New York). IFN-y was administered in daily, subcutaneous 0.1 ml injections of 7,500 units each.
Therapy commenced 1 hour after laparotomy and the last dose was administered 1 hour prior to the second bacterial challenge. In all experiments, control animals received 0.1 ml of RPMI-1640 medium subcutaneously.
A first experiment was directed at determination of an inoculum of E. coli that would cause an LD20 when injected intraperitoneally prior to - 20 ~ 1 3 3 8 6 5 8 laparotomy. Twenty-seven mice were divided into three equal groups. The groups received inocula of 103, 105, or 108 organisms/ml and were observed for survival.
Two further experiments assessed the effect of IFN-r on survival in a dual bacterial challenge model.
The first of the two experiments compared an IFN-y-treated group with a control group (n=39 for both groups). Bacterial challenges were 105 E. coli followed by 103 K. pneumoniae five days later. The second of the two experiments assessed the effect of reduction of IFN-y therapy to 3 days (n=32 in both groups). Bacterial challenge was the same as in the previous experiment, except K. pneumoniae was administered 3 days after E.
coli.
An additional six experiments were carried out to determine whether IFN-y used in combination with clinically relevant antibiotics in the laparotomy model caused any toxic effects. The antibiotic regimen and bacterial challenges used in each experiment are summarized in Table 2. In each experiment, animals~ were divided equally into an IFN-r treated group and a control group.

1 3386~8 .

o ~ N

..1 ,.
U ~
I ~7 ~r ~ ~ 1~ u-E
a a~ r u~ u ._ O O o o o o Y
o O 11~ D ~
U ._ o o O O O O

~O
m ~: c~
o o e ~r o a~
~) N t`~ t-l 1`1 _ _ _ _ E E E E

._ ._ ._ ._ + +
a~ _ _ .r + + + + E ~ E Y
G _ _ _ O E u~ E

o --~ ~ a _ ~ _ ._~

._ ._ ._ ._ . . :
.~ .
~ . . ~ _ I _ ~
_ In all of the above experiments, surviving mice were observed for a minimum of 3 weeks to assess any possible late mortality.
Another experiment assessed the effect of laparotomy and IFN-y on peripheral mononuclear cell Ia antigen expression. Two hundred and forty mice were divided into 4 equal groups. One group received only a laparotomy, one received a laparotomy and IFN-r, one received only IFN-y, and one received no treatment.
Fifteen animals in each group were sacrificed on days 1, 3, 5 and 7 post-laparotomy (or following IFN-y treatment in the control group). Blood was obtained by cardiac puncture and mononuclear cells assayed for levels of Ia antigen expression.
Two different antibiotic regimens were used.
In the first, cefazolin and tobramycin ~Eli Lilly and Co., Indianapolis, Indiana) were used. Both drugs were administered in 0.1 ml intramuscular injections in the left hind limb at dosages of 1 mg/kg body weight.
Cefazolin was administered 15 minutes after the first bacterial challenge, but prior to laparotomy. Cefazolin and tobramycin were then administered together daily commencing 1 hour after laparotomy and continuing for 5 days until the second bacterial challenge. Tobramycin was then administered alone daily for 3 days commencing 1 hour after the second bacterial challenge.
In the second antibiotic treatment regimen, Bactrim~ (Roche Laboratories, Nutley, New Jersey), a combination product of trimethoprim (16 mg/ml) and sulfamethoxazole (80 mg/ml), was administered, after appropriate dilution, in 0.1 ml intramuscular injections in the left hind limb. The first injection was administered 15 minutes after first bacterial challenge. Bactrim~ was next administered 1 hour after laparotomy and daily thereafter until 3 days after the second bacterial challenge. Two experiments were performed using this regimen, a high dose experiment (10 mg/kg body weight trimethoprim) and a low dose experiment (5 mg/kg body weight trimethoprim).
One hundred microliter aliquots of heparinized blood from each mouse were stained for 20 minutes with fluorescein-isothiocyanate-conjugated monoclonal anti-Ia (MAS 043p, clone 0X6, Accurate Chemical and Scientific Corp., Westbury, New York). Erythrocytes were lysed with 1 ml of lysing reagent (Ortho Diagnostics, Raritan, New Jersey). The specimens were then washed in buffer and fixed in 1% paraformaldehyde in phosphate buffered saline (pH 7.4) prior to flow cytometric analyses.
Stained samples were analyzed on an Ortho Cytofluorograph IIs flow cytometer (Ortho Diagnostics, Westwood, Massachusetts). Two thousand mononuclear cells were selected for fluorescent analysis by forward versus right angle light scattering properties. A
percent of cells staining for Ia antigen was recorded for each specimen. Complete white blood cell counts (Coulter Counter, Coulter Electronics, Inc., Edison, New Jersey) and Wright-stained microscope differential counts were made, and an absolute number of mononuclear cells was determined. The total Ia antigen expression per sample was calculated by multiplying the number of mononuclear cells by the value for the percent of cells staining positively for Ia antigen.
Chi Square tests were used to determine differences in the survival experiments. The Student's t Test was used to compare groups in the experiments involving the expression of Ia antigens on mononuclear cells. In all cases, differences were considered significant at p < 0.05.
Mice infected with E. coli were assessed for survival 2 days after initial infection. Results are shown in FIG. 5. In FIG. 5, a square denotes values for the group which received 103 organisms/ml, a triangle `` I 338658 denotes values for the group inoculated with 105 organisms/ml, and a circle denotes values for the group which received 108 organisms/ml. No additional deaths occurred after that time. The dosage of 105 organisms/ml of E. coli was chosen as the first bacterial challenge for all additional experiments because this dose yielded the closest dose to an LD20.
When mice received 5 days of therapy with IFN-y, there was greater survival in the IFN-y treated group than in the control group, as illustrated in FIG.
6. In FIG. 6, a square denotes values for the control group, a circle denotes values for the IFN-y-treated group, and an asterisk indicates a statistically significant difference between the values for the IFN-y treated group and the control group on the date shown.
When mice received only 3 days of therapy with IFN-y, there was also greater survival in the group treated with IFN-y than in the control group, as illustrated in FIG. 7. In FIG. 7, a square denotes values for the control group, a circle denotes values for the IFN-y-treated group, and an asterisk indicates a significant difference between the values for the IFN-y-treated group and the control group on the indicated day.
The number of mice surviving after day 15 is shown in Table 2. There were no further deaths after that time in any group. There were no significant differences in survival between IFN-y treated and IFN-y antibiotic-treated groups, as set forth in Table 2.
This indicated no toxic effect of the dual modality therapy. However, this model was not designed to demonstrate a synergistic effect between IFN-y and antibiotics.
Laparotomy alone significantly decreased mononuclear cell Ia antigen expression compared to controls on days 1 and 3 post-surgery, as illustrated in FIG. 8. In FIG. 8, values for the control group are -denoted by a square, values for the laparotomy alone group by a triangle, values for the laparotomy and IFN-y treatment group by a circle, while an asterisk indicates a statistically significant difference. The Ia antigen expression returned to normal by day 5 post-surgery.
Treatment of laparotomized mice with IFN-y prevented the drop in mononuclear cell Ia antigen expression on days 1 and 3 post-surgery.

Example 5 Eleven consecutive patients who sustained major trauma resulting in an Injury Severity Score ("ISS") [Baker et al., J. Trauma, 14, 187-196 (1984)] of at least 20 were studied. These patients sustained the following injuries: burns (5 patients), gunshot wounds (3 patients), motor vehicle accidents (2 patients), and crush injury (1 patient). Blood samples were obtained within 5 days of injury. Ten healthy normal subjects were also studied.
Three series of experiments were performed.
The first assessed the effect of trauma on monocyte HLA-DR expression. An experiment assessed the effect of IFN-y on monocyte HLA-DR antigen expression in mononuclear cell culture in trauma patients (n=ll) and normal subjects (n=10). Four patients had serial tests performed throughout their hospital stay. The final experiment assessed the effect of IFN-y on monocyte HLA-DR antigen expression in pure monocyte culture for trauma patients (n=5) and normal subjects (n=4).
Human affinity-purified IFN-y (Interferon Sciences, Inc., New Brunswick, New Jersey) was used.
The IFN-y had a specific activity of approximately 1.0 X
106 units/mg protein and was diluted with sterile phosphate-buffered saline. IFN-y (500 units) were added - `` 1 338658 to each culture. In all experiments, controls were set up with culture media alone added.
Mononuclear cells were isolated and cultured as follows. Twenty milliliters of venous blood were collected in acid citrate dextrose anticoagulate tubes (Becton-Dickinson, Rutherford, New Jersey). Mononuclear cells were separated from whole blood diluted 1:4 with RPMI-1640 medium (Gibco Laboratories, Grand Island, New York) by centrifugation for 30 minutes at 1500 rpm after layering on Ficoll hypaque-1077~ gradients. (Sigma Chemical Co., St. Louis, Missouri) at a 4:3 ratio. The mononuclear layer was harvested, washed twice with RPMI-1640 medium and resuspended in RPMI-1640 medium supplemented with 5% human AB serum (Hazelton Laboratories, Hazelton, Pennsylvania), 2 X 10 5 M 2-mercaptoethanol, 300 units of penicillin, 300 mg of streptomycin and 75 pg of amphotericin B.
Mononuclear cell cultures were established with 2 X 106 cells in 1.5 ml volumes per siliconized glass tube and incubated for 48 hour at 37 C in 5% CO2.
Monocytes were purified and cultured as follows. Twenty milliliters of venous blood were collected in EDTA anticoagulate tubes (Becton-Dickinson, Rutherford, New Jersey). Whole blood was diluted 10:1 with dextran 500, and allowed to stand for 1 hour in room temperature to permit neutrophils to settle.
Plasma was extracted, passed over Nycodenz~ (Accurate Chemicals, Westbury, New York), and then centrifuged for 20 minutes at 2000 rpm. Purified monocytes were collected and washed with RPMI-1640 medium. These purified monocytes were then resuspended and cultures were established as previously described.
The expression of HLA-DR antigen on monocytes was determined using dual monoclonal antibody staining and flow cytometry. One million cells of each preparation were stained with anti-human fluorescein _ - 27 coupled MO2 monoclonal antibody (Coulter Immunology, Hialeah, Florida), which specifically stains monocytes, and with isothiocyanate-coupled anti-human HLA-DR
antigen (Becton-Dickinson, Sunnyvale, California).
Samples were analyzed on an Orthocytofluorograph IIs flow cytometer (Ortho Diagnostics, Westwood, Massachusetts) configured for simultaneous 2-color (red and green) fluorescent analysis. Gain settings were set on stained calf thymus standards (Fluortrol~, Ortho Instruments, Westwood, Massachusetts). In this analysis, HLA-DR antigen expression was measured only on cells which co-stained positively for MO2 antigen, thereby excluding all cell types other than monocytes from the analysis.
In all experiments, two parameters were measured, the percentage of monocytes which expressed HLA-DR antigen and the density of the HLA-DR antigen expression on monocytes. The density of HLA-DR antigen expression on monocytes was measured as the mean fluorescent channel on gated positive cells (1000 channel resolution).
The Student's t Test was used to determine differences. Differences were considered significant at p < 0.05.
The mean HLA-DR antigen density on monocytes of trauma patients was 85.8 + 15.5 for normal subjects. The mean value for percentage of monocytes expressing HLA-DR antigen of trauma patients was 31.6% +
6.5 and was significantly less than the mean of 91.1~ +
2,5 for normal subjects.
Mixed lymphocyte and monocyte culture alone, without IFN-y addition, increased the density of HLA-DR
antigen expression on monocytes in both normal subjects and trauma patients. IFN-~ treatment significantly increased the density of monocyte HLA-DR antigen expression above culture alone values in both normal ` 1 338658 subjects and trauma patients, although the levels for trauma patients treated with IFN-y were less than the normal level, as illustrated in FIG. 9. In FIG. 9, an empty bar represents a baseline value, a hatched bar represents a value for 48 hour culture, a solid bar represents a value for 48 hour cultures with IFN-y, and an error bar indicates the standard error of the mean for each value.
IFN-y treatment significantly increased the percentage of monocytes expressing HLA-DR antigen above culture alone and baseline values, as illustrated in FIG. 10. In FIG. 10, an empty bar represents a baseline value, a hatched bar represents a value for 48 hour culture, a solid bar represents a value for 48 hour culture with IFN-y, and an error bar indicates the standard error of the mean for each value. This value was not different from the baseline value for normal subjects and approached the IFN-y value for normal subjects.
In purified monocyte culture, culture alone significantly increased both density of monocyte HLA-DR
antigen expression, as depicted in FIG. 11, and increased the percentage of monocytes expressing HLA-DR
antigen above baseline values in the trauma group but not in the normal subjects group, as depicted in FIG.
12. In FIGS. 11 and 12 an empty bar represents a baseline values, a hatched bar represents values for 48 hour culture, a solid bar represents a value for 48 hour culture with IFN-y, and an error bars indicates the standard error of the mean for each value. IFN-y treatment significantly increased the density of HLA-DR
antigen expression in the trauma and normal groups.
When IFN-y treated cultures were considered, there was no difference between the trauma or normal groups in either density of HLA-DR antigen expression on monocytes, as shown in FIG. 11, or in the percentage of monocytes expressing HLA-DR antigen, as shown in FIG.
12.
Of the four serially-followed patients, two made uneventful recoveries, and one developed major sepsis. IFN-y treatment markedly increased monocyte HLA-DR antigen expression at each time. The other patient died of burn wound infection, and IFN-y treatment increased monocyte HLA-DR antigen expression at each time, even for the final sample, which was obtained immediately prior to death.
FIG. 13 is a graphic depiction of monocyte HLA-DR antigen values for the cultures obtained from the burn patient who died. In FIG. 13, baseline values are denoted by squares, 48 hour culture values by triangles, and 48 hour culture with IFN-y values by circles.
The results confirm that severely injured patients have depressed levels of HLA-DR antigen expression as compared to controls. When IFN-y was added to peripheral blood cell cultures obtained from these patients, the IFN-y treatment resulted in an increase in HLA-DR antigen density of expression on monocytes. This increase in density of HLA-DR antigen expression did not result in a return to the normal range of HLA-DR antigen expression of controls, but there was a positive response to the IFN-y treatment.
When the percentage of monocytes expressing HLA-DR antigen was considered, a return to a near normal percentage of cells expressing HLA-DR antigen was observed in IFN-y-treated peripheral blood cell cultures of trauma patients. Similar patterns were observed when cultures of purified monocytes were used, indicating that interaction with lymphocytes in culture may not play a role in this result.
The kinetics of the response to IFN-y were observed in the culture of the peripheral blood cells of a burn patient who succumbed to a septic infection.

From these data, it is clear that HLA-DR antigen expression was depressed on the first day following injury, but that IFN-y was capable of enhancing the percentage of monocytes expressing HLA-DR antigen until the last day of life for this patient.
The results of this study indicate that HLA-DR
antigen expression is depressed on monocytes of trauma patients, but that IFN-y treatment may at least partially restore HLA-DR antigen expression. The resistance to infection of many trauma patients may be compromised as a result of impaired antigen presentation due to depressed HLA-DR antigen expression. In trauma patients, who are often unresponsive to conventional therapy, IFN-y may be used to help restore immune defenses.

Example 6 In an experiment generally following the procedures of Example 4, blood was collected from 10 normal subjects (control group) and 11 patients who sustained severe trauma (trauma group). A baseline MO2DR value for each group was determined using dual monoclonal antibody staining and flow cytometry. Polk et al., Ann. Surg., 204, 282-299 (1986). The mononuclear layer of each blood sample was then harvested and incubated with either 500 I.U. IFN-y (IFN-r culture) or control media (culture alone). The MO2DR
value for each sample was determined after 48 hours of incubation.
The mean baseline MO2DR value for the trauma group was 32%, which was significantly less than the mean MO2DR value of 91% for the control group (p < 0.001). In the trauma group, IFN-y cultured MO2DR
value was 81% which was significantly greater than both the mean baseline MO2DR value of 32% (p < 0.001) and the ` - 1 338658 49% mean value of MO2DR value when cultured without IFN-y (p < 0.01). In the control group, there were no significant differences among the IFN-y culture MO2DR
value (94%), the culture aIone MO2DR value (86%) or the baseline MO2DR value (91%). These data confirm that HLA-DR antigen expression on monocytes of severely injured patients is markedly reduced and show that it may be increased significantly by IFN-y. IFN-y may be an important augmentor of host defense mechanisms in these patients.

Example 7 Patients with severe injury [ISS > 20 and 1+
or 2+ bacterial contamination of their wounds, scored according to the procedure of Hershman et al., Injury, 19, 263-266 (1988)] received recombinant human IFN-y (Genentech, Inc., South San Francisco, California) at doses of 0.01 to 0.1 mg/m2/day for 7-10 days immediately following hospitalization. Standardized prophylactic antibiotic therapy with cefoxitin was administered during the first 72 hours. Patients were monitored for infection and monocyte DR antigen expression, the làtter being an immunologic endpoint used as an indicator of clinical outcome.
Treatment at all dose levels was extremely well tolerated. Fever was the only drug-related toxicity observed, and fever was only observed at the two higher doses. Among the first 13 patients treated, there were only two episodes of major sepsis and no deaths related to sepsis. Monocyte DR antigen expression, which is depressed in severe trauma victims, was restored to normal or greater than normal levels.
Individual case summaries for five of the patients treated on this study are as follows.
Patient LD was a 20 year old male who had suffered full thickness esophageal burns due to the ingestion of a caustic cleaning substance. His ISS was 25 and he had 1+ bacterial contamination in his wound.
He received IFN-y therapy at a dose of 0.01 mg/m2 for seven days and cefoxitin prophylaxis for 72 hours. He recovered fully from this injury without major sepsis.
Patient TR was a 24 year old female who suffered a gunshot wound to the abdomen with injury to the pancreas and inferior vena cava. Her ISS was 25 and she had 1+ bacterial contamination of her wound. She was treated with IFN-y at a dose of 0.01 mg/m2 for seven days and cefoxitin prophylaxis. Nafcillin was also administered. Other than a culture of Staphylococcus aureus from her central venous catheter tip, the patient had no evidence of infection and recovered fully from her injury.
Patient MD was a 28 year old male who suffered multiple stab wounds to the head and chest. His ISS was 26 with 1+ bacterial contamination of the wound. The patient received recombinant IFN-y therapy at a dose of 0.01 mg/m2 for seven days as well as cefoxitin antibiotic prophylaxis. The patient recovered fully and had no evidence of major sepsis.
Patient BD was a 27 year old female who suffered severe pelvic injury due to a motor vehicle accident. The ISS was 25 with 1+ bacterial contamination of the wound. The patient was treated with IFN-y at a dose of 0.01 mg/m2 for seven days as well as cefoxitin antibiotic prophylaxis. In addition, flagyl and amikacin were administered. The patient recovered fully from her injury without evidence of major sepsis.
Patient BD was a 27 year old male who suffered a gunshot wound to the abdomen. His ISS was 25 with 1+
bacterial contamination. He received recombinant IFN-y therapy at a dose of 0.1 mg/m2 for seven days as well as cefoxitin antibiotic prophylaxis. The patient developed an intra-abdominal abscess which required drainage and further antibiotic therapy, but then recovered fully from this injury.

Example 8 Adult female Sprague-Dawley rats (Harlan Sprague-Dawley, Indianapolis, Indiana) weighing 190-230 g were subjected to hemorrhagic shock. The ratswere anesthesized with intraperitoneal Ketamine (50 mg/kg) and Xylazine~ (15 mg/kg) and bled to a mean arterial pressure of 45 mm Hg which was maintained for 45 minutes. Animals were resuscitated with their shed blood and approximately 1.5 volumes normal saline. The mortality of this hemorrhagic shock protocol was approximately 15%; all deaths occurred within 24 hours following shock and these animals were excluded from analysis.
Staphylococcus aureus 502A (ATCC 27172, American Type Culture Collection, Rockville, Maryland) was maintained on trypticase soy agar at 4 C. Prior to use, the bacteria were grown overnight in trypticase soy broth at 37C., washed and resuspended in normal saline at a concentration of 4 X 108 colony forming units (CFU) per ml. The bacterial concentration was determined by measuring the optical density at 620 nm and confirmed by serial dilution and backplating on trypticase soy agar.
Recombinant rat IFN-y was purchased from Amgen Biologicals (Torrence, California) and had a specific activity of 4.8 X 106 U/mg. IFN-y was diluted with phosphate buffered saline containing 2% rat serum and was stored at -70 C prior to use.
One hour after resuscitation from hemorrhagic shock, animals were injected with 5 separate inocula of 1 X 108 CFU (0.25 ml) Staphylococcus aureus ("S.

aureus") subcutaneously on the dorsum. Animals were divided into four treatment groups: a first group of control animals were resuscitated following shock but received no further treatment; a second group of antibiotic-treated animals received cefazolin ("CEF") (Smith, Kline and French, Philadelphia, Pennsylvania) intraperitoneally at 30 mg/kg, 30 minutes prior to inoculation and then again 4 hours later; a third group of IFN treatment-animals received IFN-y 7500 U
subcutaneously, one hour following inoculation and then daily for three days; and a fourth group received both CEF and IFN-y as in groups 2 and 3. Tissue antibiotic levels were measured in the subcutaneous tissue 30 minutes after the initial dose of CEF by a standard bioassay.
Animals were sacrificed on day 7 by cervical dislocation under anesthesia. A dorsal incision was made and the skin reflected. The diameter of each abcess was measured in situ with a micrometer, and abcesses were dissected free from the musculature of the back. Abscesses were then excised from the skin and weighed.
Statistical analysis of abscess number was done using Fisher's Exact Test. Comparisons of abscess diameter and weight were made using a prior analysis of variance followed by a Tukey - Kramer HSD Test.
Significance was set a priori at p < 0.05.
The results for abscess number, diameter and weight are summarized in Table 3. In Table 3, for values indicated by the "" symbol, a significant difference from the control at the p < 0.05 vs. level was determined, and for values identified by an asterisk, a significant difference from the value for CEF alone was determined at the p < 0.05 level.

` ` ` 1 338658 EXPERIMENTAL ABSCESS ABSCESS ABSCESS
GROUPS #(n=20) DIAMETER (mm) WT (mg) Control 20 13.4+1.9 504+125 CEF 20 10.1+1.1 191+78 IFN 20 14.6_3.8 769_212 CEF + IFN 14 4.8+1.6 67+59 The administration of standard antibiotic prophylaxis (CEF) following hemorrhagic shock decreased abscess diameter and weight compared to control but did not decrease abscess number. Animals treated with IFN-y alone had abcesses which were larger than those of control animals. This correlated with a subjective impression of an increased inflammatory reaction around these lesions. The addition of IFN-y to CEF
significantly decreased abscess number, diameter and weight as compared to the control or to the use of CEF
alone. Analysis of the 99~ confidence intervals of the mean abscess diameter for each group illustrates the synergy between these immune adjuvants and CEF in decreasing abscess diameter as compared to~controls or, as illustrated in FIG. 14, as compared to either modality alone.
Tissue CEF levels were 14.9 + 5.2 ~g/gm 30 minutes after the initial dose, exceeding the minimal inhibitory concentration employed for the S. aureus (0.15 ~g/ml).

Although the present invention has been described in terms of particular embodiments, it is understood that modifications and variations will occur to those skilled in the art. For example, when IFN-y is administered via the intravenous route, antibiotics and IFN-y may be co-administered. Therapy according to the present invention could be used for all types of trauma, and may be used for emergency or selective surgery.
Classes of antibiotics useful according to the present invention include, but are not limited to, cephalosporins, penicillins (including semi-synthetic penicillins), bacteriostatic antibiotics and aminoglycosides.
Accordingly, it is intended that the present invention include all variations and modification which come within the scope of the claims.

Claims (15)

1. A composition for the prophylaxis or treatment of a trauma-associated sepsis comprising a therapeutically effective dose of IFN-.gamma.;
a therapeutically effective dose of an antibiotic;
and a diluent or carrier.

2. The composition as recited in claim 1, wherein the antibiotic is selected from the group consisting of cefazolin, nafcillin, vancomycin, cefoxitin, neomycin plus erythromycin, penicillin G, trimethoprim plus sulfamethoxazole, and clindamycin or clindamycin plus gentamycin or tobramycin.

3. The composition as recited in claim 1, wherein he antibiotic is an antibacterial agent.

4. The use of a composition as recited in claim 1 for the prophylaxis or treatment of trauma-associated sepsis.

5. The use as recited in claim 4, wherein the trauma-associated sepsis is not known to be accompanied by viral infection, a tumor, or an intra-cellular infection.

6. The use as recited in claim 4, wherein the prophylaxis or treatment of trauma-associated sepsis comprises subcutaneous injection of a therapeutically effective dose of IFN-.gamma..

7. The use as recited in claim 4, wherein the sepsis is a bacterial blood infection.

8. The use as recited in claim 4, wherein the prophylaxis or treatment of trauma-associated sepsis comprises intramuscular injection of a therapeutically effective dose of IFN-.gamma..

9. The use as recited in claim 4, wherein the trauma is an injury or a surgical incision.

10. The use as recited in claim 4, wherein the prophylaxis or treatment of trauma-associated sepsis comprises intravenous injection of a therapeutically effective dose of IFN-.gamma..

11. The use as recited in claim 7, wherein the sepsis is caused by an enteric bacterium.

12. The use as recited in claim 4, wherein the prophylaxis or treatment of trauma-associated sepsis comprises bodily fluid uptake of an effective dose of IFN-.gamma..

13. The use as recited in claim 4, wherein the IFN-.gamma. is desCysTyrCys human IFN-.gamma..

14. The use as recited in claim 4, wherein the antibiotic is an antibacterial agent.

15. The use as recited in claim 4, wherein the antibiotic is selected from the group consisting of cefazolin, nafcillin, vancomycin, cefoxitin, neomycin plus erythromycin, penicillin G, trimethoprim plus sulfamethoxazole, and clindamycin or clindamycin plus gentamycin or tobramycin.