JOURNAL

OF SURGICAL

RESEARCH

50,421-424

(1991)

Human Recombinant Interleukin-I (X Protection against the Lethality of Endotoxin and Experimental Sepsis in Mice H. RICHARDALEXANDER, M.D., GERARD M.DoHERTY,M.D., DOUGLASL. FRAKER, M.D., MARK I. BLOCK, M.D., JON E. SWEDENBORG,M.D., AND JEFFREYA. NORTON, M.D. Surgical

Metabolism Presented

Section, at the Annual

Surgery Meeting

Branch,

National

of the Association

Cancer

Institute,

for Academic

Human recombinant interleukin-lcu (IL-l) has a diverse range of physiological activities which may be beneficial or deleterious to the host. Pretreatment with doses of IL-1 has been shown to protect mice against a subsequent lethal bacterial injection; however, the protective effects of a single intravenous (iv) dose of IL-l have not been well characterized. The current experiments were performed to determine the best dose, timing, and duration of action of a single iv dose of IL-1 against a subsequent lethal challenge with intraperitoneal endotoxin (LPS) or experimental sepsis induced by cecal ligation and puncture (CLP). Female C5’7B1/6 mice treated with iv IL- 1 24 hr prior to 30 mg/kg LPS ip had improved survival compared to saline-treated controls (P < 0.01). IL-1 was also protective when given 6 to 72 hr, but not 2 or 96 hr, prior to LPS. IL-l protection against LPS lethality was similar to protection seen with an iv dose of tumor necrosis factor (TNF). After CLP, survival was improved with IL-1 versus saline pretreatment (P = 0.02). Unlike previous work with TNF, no toxicity or lethality was observed at any dose of IL-1 administered. A single iv dose of IL-l protects against the lethality of LPS and CLP in mice. IL-l may be a useful treatment strategy in patients at risk for the development of life-threatening sepsis. 0 1991 Academic Press, Iac.

INTRODUCTION Interleukin-1 (IL-l), produced primarily by monocytes and macrophages in response to endotoxin and other stimuli, can provide protection against lethal bacterial infection [l]. The optimal dose, timing, route of administration, and frequency of IL-l treatment are variable and dependent on the animal model as well as the virulence and size of the bacterial innoculum [Z]. In previous work, the maximum beneficial effects of IL-I have required more than one dose [2, 31, were superior when the infectious agent and IL-l were administered via the same route [4], and were optimal when the IL-l was administered prior to the bacterial challenge [5]. 421

National

Institutes

of Health,

Surgery,

Houston,

Texas,

Bethesda, November

Maryland 14-17,

20892

1990

Our laboratory has recently shown that a single intravenous dose of tumor necrosis factor (TNF) protects against endotoxin-induced shock and lethality in the rat [6]. For an agent to be clinically useful against endotoxin-induced shock or sepsis, it should ideally have a rapid onset of action, be of long duration, and have minimal toxicity. We hypothesized that since IL-l and TNF share many physiological actions [7] and since IL-l has less toxicity when administered in large doses, it may be a superior agent in conferring protection against the lethality of endotoxin or experimental sepsis. In mice, a polymicrobial sepsis model has been characterized which develops after cecal ligation and puncture (CLP) [8,9]. Animals develop a polymicrobial bacteremia within several hours after CLP and death occurs from 24 to 48 hr later. This model was used because it develops slowly and appears to reproduce the physiological alterations observed in clinically severe bacterial infections in humans. The present experiments were performed to determine the protective effects of a single iv dose of IL-l against the lethality of intraperitoneal endotoxin (LPS) and a clinically relevant model of sepsis. MATERIALS

AND METHODS

Female C57Bl/6 mice (19-21 g) were group housed in a controlled environment. All experiments were conducted in compliance with the Animal Care and Use Committee of the National Institutes of Health. Human recombinant IL-la (generously supplied by Hoffman-LaRoche, Inc., Nutley, NJ) was reconstituted in endotoxin-free water (Sigma, St. Louis) and brought to final concentration in saline with 0.5% fatty acid-poor endotoxin-free (~6 Eu/mg) bovine serum albumin (BSA, Calbiochem, La Jolla, CA). Control animals received an identical volume of saline with 0.5% BSA (NS). Endotoxin from Escherichia coli serotype 0127:B8 (LPS, Sigma, St. Louis) was reconstituted in saline to a final concentration of 4 mg/cc and administered intraperitoneally (ip). All experiments were conducted in the same general format and initial survival experiments were performed 0022.4804/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

422

JOURNAL

TABLE Protective Effects tered iv 24 hr prior Dose of IL-1 k&d 0” 0.3 1.5 5 27 133 666 3300

SURGICAL

RESEARCH:

(100) (100) (100) (100) (100) (100) (100) (100)

VOL.

50, NO.

5, MAY

1991

1

RESULTS

of Various Doses of IL- 1 Administo LPS ip in Female C57B1/6 Mice

Number surviving IL-l (%) 12 12 12 12 12 12 12 12

OF

Dose of LPS (mdkg) 30 30 30 30 30 30 30 30

72-hr

2 6 10 10 12 11 11 9

survival (%) (17) (50) (83)” (83)* (loo)** (92)** (92)** (75)**

* P < 0.01 versus control. ** P = 0.0001 versus control. a Received equal volume of vehicle.

blinded. In the first experiment mice (n = 12 each group) were injected with increasing doses of IL-l or an equal volume of saline by tail vein 24 hr prior to ip LPS, 30 mg/kg. In the second experiment, the best protective dose of IL-l or NS was administered iv at various intervals prior to ip LPS, 35 mg/kg. To compare the protective effects of IL-1 versus human recombinant tumor necrosis factor (TNF; Cetus Corp., Emeryville, CA) [lo] against LPS, animals were treated iv with either TNF, 50 pg/kg, or IL-l, 27 pg/kg, and challenged 24 hr later with LPS, 40 mg/kg ip. In the final experiment, IL-l or an equal volume of NS was injected by tail vein and 24 hr later mice (n = 12 each group) were anesthetized with pentobarbital, 25 mg/kg ip. A 1.5 to 2-cm midline laparotomy incision was made and the cecum was delivered onto the abdominal wall. The base of the cecum was ligated with a 4-O silk tie and a 16-gauge puncture was made on the antimesenteric surface of the cecum at a point midway between the tip and the ligated base. A small amount of stool was expressed through the puncture wound and the intestines were returned to the abdominal cavity. The wound was then closed with g-mm Autoclip staples (Clay-Adams, Parsippany, NY). All animals then received 0.4 cc of saline (20 cc/kg) subcutaneously on the flank. After CLP all mice recovered from the anesthetic within 1 to 2 hr. Survival was assesseddaily for 3 days at which time animals appeared stable or recovering from LPS or CLP. Animals were observed for 6 days and deaths after 72 hr did not occur. Survival differences after LPS were determined by Fisher’s exact test. Other survival curves were constructed by the method of Kaplan and Meier and analyzed by the Breslow modification of the Kruskall-Wallis test. Significance was defined as P < 0.05.

The protective effects of IL-1 against a lethal challenge of LPS given 24 hr later are summarized in Table 1. Survival was significantly improved by IL-l at doses greater than 1.5 pg/kg (P < 0.01). Animals treated with IL-l, 27 pg/kg, had 100% survival compared to 17% of controls (P = 0.0001). No lethality was observed from any dose of IL-1 administered. The protective effects of IL-l were significant with a pretreatment interval from 6 to 72 hr (P < 0.002) (Table 2). However, there was no improvement in survival when the cytokine was administered either 2 or 96 hr prior to LPS (Table 2). Our laboratory has previously shown that a single nonlethal intravenous dose of TNF will significantly improve survival against a subsequent lethal dose of TNF or LPS [6,11]. To compare the protective effects of IL-l and TNF against the lethality of LPS in this animal model, the best protective dose of each cytokine was administered iv 24 hr prior to a lethal LPS challenge. There was a significant and identical survival benefit observed in both IL-l- and TNF-pretreated animals compared to controls as seen in Fig. 1 (P -c 0.02). In multiple experiments the time course and lethality of CLP were very reproducible with 90 to 100% of untreated animals dying within 48 to 72 h. At 6 days after CLP surviving animals were normally active and appeared to have recovered. Pretreatment with IL-l 24 hr prior to CLP significantly improved survival over concomitant NS-treated control animals (Fig. 2, P < 0.02). DISCUSSION

These experiments demonstrate that a single intravenous dose of IL-l confers significant protection against a subsequent lethal challenge with LPS. The protective effects appear to require a 6-hr pretreatment interval, last for 72 to 96 hr, and were conferred by a wide range of IL-l doses which were well tolerated in this animal

TABLE

2

Onset and Duration of Protection of IL-l, 27 &g/kg iv, against LPS, 35 mg/kg ip, in Female C57B1/6 Mice Time

prior

to LPS

P=

0.002 versus

72-hr

survival

l/12 (8)

NS (-24) -2 -6 -12 -24 -48 -72 -96 *

(hr)

o/12 (0) 11/12(92)* lo/12 (83)* u/12 (92)* u/12 (92)* 9/12 (75)*

O/6 (0) control.

(%)

ALEXANDER

ET AL.: IL-1 PROTECTION

AGAINST LETHAL ENDOTOXIN

model. In addition, a single dose of IL-1 improved survival in a murine model of experimental sepsis, CLP. Previous reports have described the protective effects of IL-l against a subsequent lethal bacterial innoculum [2,5]. The maximum beneficial dose and timing of IL-l, however, are quite variable and appear to be dependent on the animal model, the route of administration of IL-l, or the virulence of the bacterial challenge [2, 41. While some investigators have shown that a single dose of IL-1 improves survival against bacterial infection in mice [ 1, 121 or in a murine model of burn wound sepsis [12], others have shown that multiple doses are most efficacious in improving survival against a lethal bacterial injection [2, 31. The pretreatment interval necessary to confer an optimal survival benefit varies from 4 days to simultaneous administration of IL-l with the bacterial innoculum [3]. Furthermore, IL-1 pretreatment is not uniformly protective against different species of bacteria [2]. For an agent to be clinically useful against the morbidity and mortality of infection, it should be of rapid onset, of long duration, and have minimal toxicity. Our laboratory has previously shown that a sublethal intravenous dose of TNF can improve survival when administered 24 hr before a lethal challenge with LPS [6] and up to 6 hr after CLP in rats [ 141. However, doses two or three times higher than the optimal protective dose of TNF are toxic and occasionally lethal. In the current study, IL-l was well tolerated across a large range of doses and the protective effects were largely preserved. The protective effects of IL-l and TNF against a lethal LPS challenge appear to be quite comparable and suggest that IL-l may

1.0.

(2 5 5

0.9 -

I

0.8 -

I

L

8 l TNF 0 IL-1 l CONTROL

l

0.7 -

I 0.6 -

2

II 0.5 I

$ 0.4

-

0.3

-

0.2

-

0.1

-

I

E %

I

B

0.0

0

L l

1 I 20

I 40

I 60

I 80

I 100

I 120

HOURS

FIG. 1. Comparison of IL-l and TNF against the lethal LPS. Female C57B1/6 mice (n = lo-ll/group) were given fig/kg iv (open circles); TNF, 50 pg/kg iv (solid circles); or volume of saline (squares), 24 hr prior to LPS, 40 mg/kg ip. was significantly improved by IL-1 and TNF (P < 0.001).

effect of IL-l, 27 an equal Survival

423

AND SEPSIS

0 CONTROL 0 IL-1 0.8 z 5

07

f iT 3

06

U-J 0.5 5

0.1 t

0.00I

c, 1

2

3

4

5

6

DAYS

FIG. 2. Protective effects of IL-1 against the lethality of experimental sepsis. Female C57B1/6 mice (n = 11-12/group) were treated with either IL-l, 27 pg/kg iv (open circles), or an equal volume of saline (solid circles) and 24 hr later underwent cecal ligation and puncture as described under Materials and Methods. Mice treated with IL-1 had significantly improved survival (P < 0.02).

be the preferable cytokine due to its less toxic nature and synergistic activity when used in combination with antimicrobial agents [ 151. Cecal ligation and puncture produce a polymicrobial but predominantly gram-negative sepsis which has been previously characterized in rats and mice [8, 9, 161. It appears to be a clinically relevant model in that it develops slowly as bacteria invade devitalized tissue and it produces the physiological alterations which might be anticipated during intraabdominal polymicrobial sepsis in humans. In this study, IL-l significantly improved survival after CLP compared to saline-treated controls. However, the protection was not as complete as that observed against an equally lethal challenge with LPS. It is possible that the best protective dose against CLP is different than that for LPS. It seems more likely, however, that IL-l cannot protect completely against the more complex physiological alterations that occur after CLP. The precise mechanism of IL-l protection is not known but, in light of the pleotropic physiological effects of the cytokine [17, 181, may be secondary to several protective actions. For example, IL-l enhances phagocytic function and induces the production of acute phase proteins in the liver [3, 181. In mice, IL-1 pretreatment also causes an accelerated clearance of E. coli bacteremia [19] and a reduced bacterial burden in the spleen after sublethal L. monocytogenes infection [4]. In a murine model of burn wound sepsis, IL-1 treatment results in significantly fewer positive blood cultures than in control animals [12]. In addition, recruitment of host de-

424

JOURNAL

OF

SURGICAL

RESEARCH:

fenses to the area of infection may play a role in IL-l-induced resistance to infection as ip administration of IL-1 is more protective against a subsequent lethal ip bacterial challenge than is intravenous IL-l [4]. A single iv pretreatment with IL-l has recently been shown to decrease cellular injury after ischemia/reperfusion [20]. This protection is associated with increased glucose-6phosphate dehydrogenase activity, suggesting that IL-l may confer its protective effects by enhancing cellular resistance to a free radical stress. In summary, IL-l administered as a single intravenous injection protects mice from the lethal effects of LPS and experimental sepsis. The cytokine is well tolerated through a large range of doses which confer a significant protective benefit. IL-1 may provide a new strategy to improve the outcome in patients at risk for the development of life-threatening sepsis. REFERENCES 1.

2.

3.

4.

Van der Meer, J. W. M., Barza, M., Wolff, S. M., and Dinarello, C. A. A low dose of recombinant interleukin-1 protects granulocytopenic mice from lethal gram-negative infection. Proc. Natl. Acad. Sci. USA 86: 1682, 1988. Ozaki, Y., Ohashi, T., Minami, A., and Nakamura, S. Enhanced resistance of mice to bacterial infection induced by recombinant human interleukin-1. Znfect. Zmmun. 55: 1436, 1987. Morikage, T., Mizushima, Y., Sakamoto, K., and Yano, S. Prevention of fatal infections by recombinant human interleukin 1 in normal and anticancer drug-treated mice. Cancer Res. 50: 2099, 1990. Czuprynski, C. J., and Brown, J. F. Recombinant murine interleukin-1 enhancement of nonspecific antibacterial resistance. Infect. Immun. 55: 2061, 1987.

5.

Minami, A., Fujimoto, K., Ozaki, Y., and Nakamura, S. Augmentation of host resistance to microbial infections by recombinant human interleukin-1. Infect. Zmmun. 56: 3116, 1988.

6.

Alexander, H. R., Langstein, H. N., Doherty, G. M., Jensen, J. C., and Norton, J. A. Human recombinant tumor necrosis factor protection against endotoxin-induced shock and lethality in the rat. Surg. Forum 41: 103, 1990.

VOL.

50, NO.

5, MAY

1991

7.

Le, J., and Vilcek, J. Tumor necrosis Cytokines with multiple overlapping Invest. 56: 234, 1987.

8,

Wichterman, K. A., Baue, A. E., and Chaudry, I. H. Sepsis and septic shock-A review of laboratory models and a proposal. J. Surg. Res. 29: 189, 1980.

g

factor and interleukin-1: biological activities. Lab.

Urbaschek, B., Ditter, B., Becker, K.-P., and Urbaschek, R. Protective effects and role of endotoxin in experimental septicemia. Circ. Shock 14: 209,1984.

10.

Wang, A., Creasey, A. A., Ladner, M. B., Lin, L. S., Strickler, J., Van Arsdell, J. N., Yamamoto, R., and Mark, D. F. Molecular cloning and the complimentary DNA for human tumor necrosis factor. Science 228: 149, 1985.

ll.

Sheppard, B. C., and Norton, J. A. Tumor interleukin-1 protection against the lethal crosis factor. Surgery, in press, 1991.

r2

necrosis factor effects of tumor

and ne-

Cross, A. S., Sadoff, J. C., Bernton, K. E., and Genski, P. Pretreatment with recombinant murine tumor necrosis factor cachectin and murine interleukin-1 protects mice from lethal bacterial infection. J. Exp. Med. 169: 2021, 1989.

13.

Silver, G. M., Gamelli, R. L., O’Reilly, M., and Hebert, J. C. The effect of interleukin 1 on survival in a murine model of burn wound sepsis. Arch. Surg. 125: 922, 1990.

14.

Sheppard, B. C., Fraker, D. L., and Norton, J. A. Prevention and treatment of endotoxin and sepsis lethality with recombinant human tumor necrosis factor. Surgery 106: 156,1989.

15.

Nakamura, S., Minami, A., Fujimoto, K., and Kojima, nation of effect of recombinant human interleukin-1 microbial agents. Infect. Zmmun. 33: 1804, 1989.

16.

Fink, M. P., and Heard, S. 0. Laboratory models of sepsis and septic shock. J. Surg. Res. 49: 186, 1990. Dinarello, C. A. Biology of interleukin 1. FASEB J. 2: 108,1988.

17.

M. R., Falkenburg, of interleukin-1.

T. Combiwith anti-

18.

Fibbe, W. E., Schaafsma, R. The biological activities

F., and Willemze, Blut 59: 147, 1989.

19.

Pelkonen, S., and Pluschke, G. Recombinant ulates clearance of escherichia coli Kl Pathog. 6: 415, 1989.

20.

Brown, J. M., White, C. W., Terada, L. S., Grosso, M. A., Shanley, P. F., Mulvin, D. W., Banerjee, A., Whitman, G. J. R., Harken, A. H., and Repine, J. E. Interleukin-1 pretreatment decreases ischemia/reperfusion injury. Proc. Natl. Acad. Sci. USA 8’7: 5026, 1990.

interleukin-1 bacteremia.

stimMicrob.