Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
IMMUNOPHARMACOLOGY AND IMMUNOTOXICOLOGY, 1 3 ( 4 ) , 4 8 5 - 4 9 8 ( 1 9 9 1 )
INTERLEUKIN-la ENHANCES HEPATOTOXICITY OF TUMOR NECROSIS FACTOR-a IN GALACTOSAMINE-SENSITIZED MICE J. Nagakawa, I. Hishinuma, K. Hirota, K. Miyamoto, M. Yasuda, T. Yamanaka and K. Katayama
Tsukuba Research Tsukuba-shi,
Laboratories, Eisai Co., Ibaraki 300-26, Japan
Ltd.
ABSTRACT The
possible
involvement
of
interleukin-la
( I L - l a ) in
the
pathogenesis of murine hepatitis model induced with galactosamine and lipopolysaccharide (LPS) was investigated. The injection of 10 ng/mouse of LPS in combination with 10 mg/mouse of galactosamine into mice induced hepatic damage at 24 hours. Treatment with antimouse IL- 1a antiserum 30 min before galactosamine/LPS injection showed a tendency to reduce the liver injury, while pretreatment
Abbreviations: ALT, L-alanine aminotransferase; AST, Laspartate aminotransferase; a n t i - m I L - l a , anti-mouse interleukin- l a antiserum; anti-mTNF, anti-mouse tumor necrosis € a c t o r - a antiserum; C. parvum, Corynebacterium p a r v u m ; GalN, galactosamine; rmIL-la, recombinant murine interleukin- 1a ; rmTNF, recombinant murine tumor necrosis fact or- a Correspondence: Jun-ichi Nagakawa; Tsukuba Research 5-1 -3 Tokodai, Tsukuba-shi, Laboratories, Eisai Co., Ltd., Ibaraki 300-26, Japan 485 Copyright 0 1991 by Marcel Dekker, Inc.
486
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
with
NAGAKAWA ET A L .
anti-mouse
tumor
necrosis
factor-a
(TNF)
antiserum
significantly protected mice from liver injury. The use of recombinant murine TNF, instead of LPS, in combination with galactosamine could elicit hepatic damage, whereas recombinant murine I L - l a could not substitute for LPS. However, recombinant murine IL- la enhanced the hepatotoxic effect of recombinant murine TNF in galactosamine-sensitized mice. These results suggest that TNF plays a major role in the pathogenesis of galactosamine/LPS hepatitis in mice and that IL-la acts synergistically with TNF in this hepatitis model.
INTRODUCTION It is well-known that gram-negative bacterial endotoxin, or lipopolysaccharide (LPS), plays a pathogenic role in liver diseases. Most patients with acute and chronic liver diseases are endotoxemic to an extent which correlates well with the severity of liver injury [l-31. Studies in rats show the injection of LPS causes liver injury [4]. Furthermore, the injection of a smaller amount of LPS causes massive hepatic cell necrosis in mice and rats sensitized with galactosamine (GalN) [5, 61. However, LPS is considered to injure host tissues not directly, but through the action of an endogenous mediator(s) [7]. Tumor necrosis factor-a (TNF)/cachectin and interleukin- 1
(IL-1) are major cytokines secreted by mononuclear cells or macrophages stimulated with LPS [8, 91. Although TNF and IL-1 are biochemically and immunologically distinct proteins, they share a number of biological activities including a role in the synthesis of acute-phase reactants in the liver [lo]. Recently, it has been reported that TNF and IL-1 are involved in the pathogenesis of human liver injury such as fulminant hepatic failure, alcoholic We have recently hepatitis, and chronic liver disease [ 11-14]. reported that endogenous TNF plays an important role in the pathogenesis of LPS-induced hepatitis in mice sensitized with GalN [15]. Here we present several lines of evidence indicating that IL-1
TUMOR NECROSIS FACTOR
48 7
may act synergistically with the hepatotoxicity of TNF in the development of hepatic injury induced with GalN and LPS in mice.
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
MATERIALS AND METHODS Animals Six-week-old male Balblc mice for induction of hepatitis and cytokine generation, and male 4-week-old C3H/HeJ mice for IL- 1 assay were obtained from Charles River Japan (Kanagawa, Japan). One week before the experiment, animals were housed at a constant temperature (23 f 1OC) and humidity (55 f 5 % ) with a fixed lightdark cycle (12-12 h), and provided an ad libitum diet of standard chow pellets and water. Induction of Hepatitis D-(+)-Galactosamine hydrochloride (GalN; Sigma, St. Louis, MO, USA) was dissolved in sterile, nonpyrogenic physiological saline, adjusted to pH 7.0 with NaOH. Bacterial LPS from S a l m o n e l l a enteritidis (Difco, Detroit, MI, USA) was dissolved in nonpyrogenic physiological saline. Hepatitis was induced by i.v. injecting Balb/c mice with the mixture of 10 mg/mouse of GalN and 10 ng/mouse of LPS. The animals were put under ether anesthesia, and blood samples for determinations of TNF, IL- 1 and aminotransferase levels were collected with heparinized syringes from the abdominal aorta. Induction of in vivo Cytokine Generation Corynebacterium parvum (C. parvum; ATCC 6918) was heatkilled (at 6OoC for 30 min), lyophilized and dissolved in nonpyrogenic physiological saline. TNF and IL-1 generation was elicited by priming Balb/c mice with 1 mg/mouse of C. parvum; then after an interval of 7 days, injecting the mice with 1 pg/mouse of LPS. Treatmen ts with Antiserum Polyvalent rabbit anti-mouse TNF-a antiserum (anti-mTNF) and polyvalent rabbit anti-mouse I L - l a antiserum (anti-mIL- la)
488
NAGAKAWA
ET AL.
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
were purchased from Genzyme, Boston, MA, USA. Normal rabbit serum served as control. These materials were diluted i n nonpyrogenic physiological saline, and i.v, injected at 0.1 ml/mouse 30 min before LPS injection.
Substituting TNF and/or I L - l a for LPS Recombinant murine TNF-a (rmTNF) and recombinant murine I L - l a ( r m l L - l a ) were purchased from Genzyme, Boston, MA, USA. These materials were dissolved in nonpyrogenic physiological saline, and i.v. injected into mice with/without GalN. rmTNF and rmIL-1-a solutions at the highest concentration contained less than 50 pg/ml of LPS as determined by the Limulus amebocyte assay using commercial test reagents (Seikagaku Kogyo, Tokyo, Japan). Aminotransferase Assays L-Alanine aminotransferase (ALT; E.C.2.6.2.1) and L-aspartate aminotransferase (AST; E.C.2.6.1.1) activities in plasma were assayed by standard spectrophotometric methods [ 161 using commercial test reagents (Wako, Osaka, Japan) and an autoanalyser (Olympus, Tokyo, Japan).
TNF Assay TNF titer in mouse plasma was determined by the assay of cytotoxicity to highly TNF-sensitive L-P3 cells, a subline of L-929 cells [17], in the presence of 1 pg/ml of actinomycin D as previously described [18]. In brief, plasma samples or standard was prediluted in RPMI 1640 medium containing 10% fetal calf serum, 100 units/ml of penicillin, and 100 pg/ml of streptomycin. Then 50 pl of the prediluted samples or standard rmTNF was added to 96-well multiplates. Next, 50 pl of 7 x 105 cells/ml cell suspension of L-P3 cells, containing 2 pg/ml of actinomycin D, was added, and the plate was incubated for 12-15 h at 37OC. The dilution ratio that had 50% cytotoxicity was evaluated by the dimethylthiazol-2-yldiphenyltetrazolium bromide colorimetric method [ 191. One mg of rmTNF corresponded to 3 x 109 laboratory units.
489
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
TUMOR NECROSIS FACTOR
IL-1 Assay IL-1 titer in mouse plasma were determined by the assay of thymocyte comitogenic response [20]. In brief, thymocytes from C3HkIeJ mice were suspended at a concentration of 1 x 107 cells/ml in RPMI 1640 medium containing 5% fetal calf serum, 5 x 10-5 M of 2-mercaptoethanol, 100 units/ml of penicillin, and 100 pg/ml of streptomycin. 100 p1 of the cell suspension was seeded into 96-well multiplates, and mixed with 50 pl of the prediluted plasma or standard rmIL-la. Then, 50 p1 of phytohemagglutinin-P (final 1 pl/ml) was added and the cells were cultured for 48 h at 37°C. Six hours before harvest, the cells were pulsed with 0.5 pCi of [3H]thymidine. The cells were harvested and radioactivities were counted by a liquid scintillation system (Aloka, Tokyo, Japan). The amount of IL-1 in the samples was calculated by using a curve prepared with the standard rmIL-la (specific activity; 108 units/mg) and expressed as U/ml. Statistical Analysis The data expressed as mean values f SE were statistically analyzed by means of Student’s t test; p < 0.05 was the criterion of significance.
RESULTS Plasma TNF and IL-1 Levels in GalN/LPS-Treated Mice Neither TNF nor IL-I activity could be detected in plasma 1, 3, 6, 12 and 24 h after the injection of GalN (10 mg/mouse) and LPS (10 ng/mouse). Neutralization of Anti-mTNF and Anti-mIL-la A 15000 dilution of anti-mTNF neutralized the cell toxicity induced by lo4 units of rmTNF in the TNF bioassay. A 1:lOOO dilution of anti-mIL-1 a neutralized the thymocyte proliferation induced by 1 unit of rmIL-la in the IL-1 bioassays.
490
NAGAKAWA ET A L . 4000
3000 L L r
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
fr
2000
1000
0 0 1
3
6
12
Tlme after LPS injection (h)
FIG 1. Time-courses of TNF and IL-1 levels in the plasma of Corynebacterium parvum (1 mg/mouse)-LPS (1 pg/mouse)treated mice. The levels of TNF (a), IL-1 ( 0 ) were determined 1, 3, 6, 12 and 24 h after the injection of LPS (n=5 at each time).
To examine the in vivo neutralizing effects of anti-mTNF and anti-mIL- 1a , cytokine generation was elicited by i.v. injection of LPS (1 pg/mouse) in mice primed with C. parvum (1 mglmouse). A large amount of TNF and IL-1 were observed in plasma within 6 h of LPS injection. The increase in plasma TNF activity preceded that in plasma IL-1 activity: the peak activities of TNF and IL-1 were reached at 1 h and 3 h, respectively (FIG 1). TABLE 1 shows the in vivo neutralizing effects of anti-mTNF and anti-mIL-la against both plasma TNF and IL-1 activities in C parvum-LPS-treated mice. The levels of TNF and IL-1 were determined at 1 h and 3 h after LPS injection. Treatment with 1:3 dilution of anti-mTNF 30 min before LPS injection significantly reduced the plasma TNF activity at 1 h and showed a tendency to reduce IL-1 activity at 1 h and 3 h. Pretreatment with 1:3 dilution of anti-mIL-la significantly reduced IL-1 activity at 3 h (p < 0.05, versus serum control at each point) but showed no effect on TNF activity.
491
TUMOR NECROSIS FACTOR
TABLE 1 Neutralizing effects of anti-mTNF and anti-mIL- 1a against plasma TNF and IL-1 activities in C. parvurn-LPS-treated mice Treatment
Dose
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
(dilution) 1 h after LPS iniection Serum control 1:3 Anti-mTNF 1:3 Anti-mIL-la 1:3
.. .
No. of animals 8
8 8
TNF activity (x lo3 U/ml)
2888 f 278 1323 f 118 a ) 2775 f 360
3 h after JPS i n g u m Serum control Anti-mTNF Anti-mIL-la
1:3 1:3 1:3
8
8 8
1.5 f 0.5 0.6 f 0.1 4.8 f 2.2
IL-1 activity (U/ml)
694 f 290 192 f 15 4 0 6 f 123
1135 f 306 514f 95 337 f 600)
0) Significantly different from serum control at each point (p < 0.05, Student’s t test).
Effects of Anti-mTNF a n d Anti-IL-la on GalN/LPS Hepatitis TABLE 2 shows the effects of anti-mTNF and anti-mIL-la on GalN/LPS hepatitis in mice. The injection of LPS (10 ng/mouse) in combination with GalN (10 mg/mouse), neither of which had a n y effect individually, caused hepatic damage at 24 h. Treatment with anti-mTNF 30 min before GalN/LPS injection significantly protected mice from hepatic damage. Pretreatment with anti-mIL-1 a s h o w e d an apparent tendency to reduce the hepatic damage but the effect was not statistically significant. Hepatotoxicity of T N F and/or IL-la in GalN-Sensitized Mice As shown in TABLE 3, the injection GalN (10 mg/mouse) plus rmTNF (100 ng/mouse) elicited hepatic damage in mice. In contrast, the injection of GalN plus rmIL-la (1 ng/mouse) did not elicit hepatic damage. However, treatment with rmIL-1 a (1 ng/mouse) enhanced the hepatotoxic effect of rmTNF (100 ng/mouse) in GalN-sensitized mice (FIG 2).
TABLE 2 Effects of passive immunization against mouse TNF and. IL-la on galactosamineLPS hepatitis in mice.
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
Treatment
Dose Number (dilution) of animals
4 Saline alone Galactosamine alone 4 LPS alone 4 ine/J .PS - t r e u Saline control 6 Serum control 1:3 6 Anti-mTNF 1:30 6
1:lO 1:3 Anti-mIL-la
1:30 1:lO 1 :3
6 6 6 6 6
Plasma ALT (Ub-1
Plasma AST (U/L)
50 f 3 52 f 1 59 f 3
28 f 0 22 f 1 24 f 1
853 f 886 f 43 f 68 f 22f 1099 f 872 f 431 f
131 211 3 b) 24 b) 1b) 314 595 193
451 487 72 73
58 445 629 217
f f f f f f f f
94 134 50)
10 a ) 4b) 112 435 80
~~
b , Significantly different from saline control: p c 0.05, and p c 0.01, respectively (Student's f test).
TABLE 3 Hepatotoxicity of LPS, rmTNF or rmIL-la in combination with galactosamine in mice. Treatment
Dose (/mouse)
Saline alone Galactosamine alone 10 mg LPS alone 10 ng rmTNF alone 100 ng (3x105 U) r m I L - l a alone 1 ng ( 100 U) Galactosamone (10 mdmousebtreated 0.1 n g LPS 1 ng 10 n g rmTNF 1 ng (3x103 U) 10 ng (3x104 U) 100 ng (3x105 U) rmIL- 1 a 0.01 ng (1 U) 0.1 ng (10 U) 1 ng (100 U)
n
Plasma ALT (U/L)
3 3 3 3 3
22f 20 f 33 f 21 f 33 f
4 4 4 4 4 4 4 4 4
21 f 498 f 2585 f 22 f 25 f 1852 f 25 f 20 f 20f
1 1 4 1
15
Plasma AST (U/L)
56f 52f 120 f 62f 127f
8 2 38 18 80
1 58f 3 327 280f 173 355 1351 f 201 2 53f 3 3 60f 6 409 881 f 150 1 76f 16 2 49f 1 1 54f 5
A mixture of galactosamine (10 mg/mouse) and various recombinant murine TNF-a (rmTNF) or recombinant ( r m I L - l a ) was intravenously injected into Balb/c mice. plasma aminotransferase activities are expressed as mean
doses of LPS, murine IL-la The values of f SE.
-
TUMOR NECROSIS FACTOR ALTrctlvlty
2
z a 22
z
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
U
t
-1
!! 3000 0
P 4 0.05
ASTrctlvtty
+
T
I ;
-
i
E
P < 0.05
m
+P
IMMUNOPHARMACOLOGY AND IMMUNOTOXICOLOGY, 1 3 ( 4 ) , 4 8 5 - 4 9 8 ( 1 9 9 1 )
INTERLEUKIN-la ENHANCES HEPATOTOXICITY OF TUMOR NECROSIS FACTOR-a IN GALACTOSAMINE-SENSITIZED MICE J. Nagakawa, I. Hishinuma, K. Hirota, K. Miyamoto, M. Yasuda, T. Yamanaka and K. Katayama
Tsukuba Research Tsukuba-shi,
Laboratories, Eisai Co., Ibaraki 300-26, Japan
Ltd.
ABSTRACT The
possible
involvement
of
interleukin-la
( I L - l a ) in
the
pathogenesis of murine hepatitis model induced with galactosamine and lipopolysaccharide (LPS) was investigated. The injection of 10 ng/mouse of LPS in combination with 10 mg/mouse of galactosamine into mice induced hepatic damage at 24 hours. Treatment with antimouse IL- 1a antiserum 30 min before galactosamine/LPS injection showed a tendency to reduce the liver injury, while pretreatment
Abbreviations: ALT, L-alanine aminotransferase; AST, Laspartate aminotransferase; a n t i - m I L - l a , anti-mouse interleukin- l a antiserum; anti-mTNF, anti-mouse tumor necrosis € a c t o r - a antiserum; C. parvum, Corynebacterium p a r v u m ; GalN, galactosamine; rmIL-la, recombinant murine interleukin- 1a ; rmTNF, recombinant murine tumor necrosis fact or- a Correspondence: Jun-ichi Nagakawa; Tsukuba Research 5-1 -3 Tokodai, Tsukuba-shi, Laboratories, Eisai Co., Ltd., Ibaraki 300-26, Japan 485 Copyright 0 1991 by Marcel Dekker, Inc.
486
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
with
NAGAKAWA ET A L .
anti-mouse
tumor
necrosis
factor-a
(TNF)
antiserum
significantly protected mice from liver injury. The use of recombinant murine TNF, instead of LPS, in combination with galactosamine could elicit hepatic damage, whereas recombinant murine I L - l a could not substitute for LPS. However, recombinant murine IL- la enhanced the hepatotoxic effect of recombinant murine TNF in galactosamine-sensitized mice. These results suggest that TNF plays a major role in the pathogenesis of galactosamine/LPS hepatitis in mice and that IL-la acts synergistically with TNF in this hepatitis model.
INTRODUCTION It is well-known that gram-negative bacterial endotoxin, or lipopolysaccharide (LPS), plays a pathogenic role in liver diseases. Most patients with acute and chronic liver diseases are endotoxemic to an extent which correlates well with the severity of liver injury [l-31. Studies in rats show the injection of LPS causes liver injury [4]. Furthermore, the injection of a smaller amount of LPS causes massive hepatic cell necrosis in mice and rats sensitized with galactosamine (GalN) [5, 61. However, LPS is considered to injure host tissues not directly, but through the action of an endogenous mediator(s) [7]. Tumor necrosis factor-a (TNF)/cachectin and interleukin- 1
(IL-1) are major cytokines secreted by mononuclear cells or macrophages stimulated with LPS [8, 91. Although TNF and IL-1 are biochemically and immunologically distinct proteins, they share a number of biological activities including a role in the synthesis of acute-phase reactants in the liver [lo]. Recently, it has been reported that TNF and IL-1 are involved in the pathogenesis of human liver injury such as fulminant hepatic failure, alcoholic We have recently hepatitis, and chronic liver disease [ 11-14]. reported that endogenous TNF plays an important role in the pathogenesis of LPS-induced hepatitis in mice sensitized with GalN [15]. Here we present several lines of evidence indicating that IL-1
TUMOR NECROSIS FACTOR
48 7
may act synergistically with the hepatotoxicity of TNF in the development of hepatic injury induced with GalN and LPS in mice.
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
MATERIALS AND METHODS Animals Six-week-old male Balblc mice for induction of hepatitis and cytokine generation, and male 4-week-old C3H/HeJ mice for IL- 1 assay were obtained from Charles River Japan (Kanagawa, Japan). One week before the experiment, animals were housed at a constant temperature (23 f 1OC) and humidity (55 f 5 % ) with a fixed lightdark cycle (12-12 h), and provided an ad libitum diet of standard chow pellets and water. Induction of Hepatitis D-(+)-Galactosamine hydrochloride (GalN; Sigma, St. Louis, MO, USA) was dissolved in sterile, nonpyrogenic physiological saline, adjusted to pH 7.0 with NaOH. Bacterial LPS from S a l m o n e l l a enteritidis (Difco, Detroit, MI, USA) was dissolved in nonpyrogenic physiological saline. Hepatitis was induced by i.v. injecting Balb/c mice with the mixture of 10 mg/mouse of GalN and 10 ng/mouse of LPS. The animals were put under ether anesthesia, and blood samples for determinations of TNF, IL- 1 and aminotransferase levels were collected with heparinized syringes from the abdominal aorta. Induction of in vivo Cytokine Generation Corynebacterium parvum (C. parvum; ATCC 6918) was heatkilled (at 6OoC for 30 min), lyophilized and dissolved in nonpyrogenic physiological saline. TNF and IL-1 generation was elicited by priming Balb/c mice with 1 mg/mouse of C. parvum; then after an interval of 7 days, injecting the mice with 1 pg/mouse of LPS. Treatmen ts with Antiserum Polyvalent rabbit anti-mouse TNF-a antiserum (anti-mTNF) and polyvalent rabbit anti-mouse I L - l a antiserum (anti-mIL- la)
488
NAGAKAWA
ET AL.
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
were purchased from Genzyme, Boston, MA, USA. Normal rabbit serum served as control. These materials were diluted i n nonpyrogenic physiological saline, and i.v, injected at 0.1 ml/mouse 30 min before LPS injection.
Substituting TNF and/or I L - l a for LPS Recombinant murine TNF-a (rmTNF) and recombinant murine I L - l a ( r m l L - l a ) were purchased from Genzyme, Boston, MA, USA. These materials were dissolved in nonpyrogenic physiological saline, and i.v. injected into mice with/without GalN. rmTNF and rmIL-1-a solutions at the highest concentration contained less than 50 pg/ml of LPS as determined by the Limulus amebocyte assay using commercial test reagents (Seikagaku Kogyo, Tokyo, Japan). Aminotransferase Assays L-Alanine aminotransferase (ALT; E.C.2.6.2.1) and L-aspartate aminotransferase (AST; E.C.2.6.1.1) activities in plasma were assayed by standard spectrophotometric methods [ 161 using commercial test reagents (Wako, Osaka, Japan) and an autoanalyser (Olympus, Tokyo, Japan).
TNF Assay TNF titer in mouse plasma was determined by the assay of cytotoxicity to highly TNF-sensitive L-P3 cells, a subline of L-929 cells [17], in the presence of 1 pg/ml of actinomycin D as previously described [18]. In brief, plasma samples or standard was prediluted in RPMI 1640 medium containing 10% fetal calf serum, 100 units/ml of penicillin, and 100 pg/ml of streptomycin. Then 50 pl of the prediluted samples or standard rmTNF was added to 96-well multiplates. Next, 50 pl of 7 x 105 cells/ml cell suspension of L-P3 cells, containing 2 pg/ml of actinomycin D, was added, and the plate was incubated for 12-15 h at 37OC. The dilution ratio that had 50% cytotoxicity was evaluated by the dimethylthiazol-2-yldiphenyltetrazolium bromide colorimetric method [ 191. One mg of rmTNF corresponded to 3 x 109 laboratory units.
489
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
TUMOR NECROSIS FACTOR
IL-1 Assay IL-1 titer in mouse plasma were determined by the assay of thymocyte comitogenic response [20]. In brief, thymocytes from C3HkIeJ mice were suspended at a concentration of 1 x 107 cells/ml in RPMI 1640 medium containing 5% fetal calf serum, 5 x 10-5 M of 2-mercaptoethanol, 100 units/ml of penicillin, and 100 pg/ml of streptomycin. 100 p1 of the cell suspension was seeded into 96-well multiplates, and mixed with 50 pl of the prediluted plasma or standard rmIL-la. Then, 50 p1 of phytohemagglutinin-P (final 1 pl/ml) was added and the cells were cultured for 48 h at 37°C. Six hours before harvest, the cells were pulsed with 0.5 pCi of [3H]thymidine. The cells were harvested and radioactivities were counted by a liquid scintillation system (Aloka, Tokyo, Japan). The amount of IL-1 in the samples was calculated by using a curve prepared with the standard rmIL-la (specific activity; 108 units/mg) and expressed as U/ml. Statistical Analysis The data expressed as mean values f SE were statistically analyzed by means of Student’s t test; p < 0.05 was the criterion of significance.
RESULTS Plasma TNF and IL-1 Levels in GalN/LPS-Treated Mice Neither TNF nor IL-I activity could be detected in plasma 1, 3, 6, 12 and 24 h after the injection of GalN (10 mg/mouse) and LPS (10 ng/mouse). Neutralization of Anti-mTNF and Anti-mIL-la A 15000 dilution of anti-mTNF neutralized the cell toxicity induced by lo4 units of rmTNF in the TNF bioassay. A 1:lOOO dilution of anti-mIL-1 a neutralized the thymocyte proliferation induced by 1 unit of rmIL-la in the IL-1 bioassays.
490
NAGAKAWA ET A L . 4000
3000 L L r
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
fr
2000
1000
0 0 1
3
6
12
Tlme after LPS injection (h)
FIG 1. Time-courses of TNF and IL-1 levels in the plasma of Corynebacterium parvum (1 mg/mouse)-LPS (1 pg/mouse)treated mice. The levels of TNF (a), IL-1 ( 0 ) were determined 1, 3, 6, 12 and 24 h after the injection of LPS (n=5 at each time).
To examine the in vivo neutralizing effects of anti-mTNF and anti-mIL- 1a , cytokine generation was elicited by i.v. injection of LPS (1 pg/mouse) in mice primed with C. parvum (1 mglmouse). A large amount of TNF and IL-1 were observed in plasma within 6 h of LPS injection. The increase in plasma TNF activity preceded that in plasma IL-1 activity: the peak activities of TNF and IL-1 were reached at 1 h and 3 h, respectively (FIG 1). TABLE 1 shows the in vivo neutralizing effects of anti-mTNF and anti-mIL-la against both plasma TNF and IL-1 activities in C parvum-LPS-treated mice. The levels of TNF and IL-1 were determined at 1 h and 3 h after LPS injection. Treatment with 1:3 dilution of anti-mTNF 30 min before LPS injection significantly reduced the plasma TNF activity at 1 h and showed a tendency to reduce IL-1 activity at 1 h and 3 h. Pretreatment with 1:3 dilution of anti-mIL-la significantly reduced IL-1 activity at 3 h (p < 0.05, versus serum control at each point) but showed no effect on TNF activity.
491
TUMOR NECROSIS FACTOR
TABLE 1 Neutralizing effects of anti-mTNF and anti-mIL- 1a against plasma TNF and IL-1 activities in C. parvurn-LPS-treated mice Treatment
Dose
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
(dilution) 1 h after LPS iniection Serum control 1:3 Anti-mTNF 1:3 Anti-mIL-la 1:3
.. .
No. of animals 8
8 8
TNF activity (x lo3 U/ml)
2888 f 278 1323 f 118 a ) 2775 f 360
3 h after JPS i n g u m Serum control Anti-mTNF Anti-mIL-la
1:3 1:3 1:3
8
8 8
1.5 f 0.5 0.6 f 0.1 4.8 f 2.2
IL-1 activity (U/ml)
694 f 290 192 f 15 4 0 6 f 123
1135 f 306 514f 95 337 f 600)
0) Significantly different from serum control at each point (p < 0.05, Student’s t test).
Effects of Anti-mTNF a n d Anti-IL-la on GalN/LPS Hepatitis TABLE 2 shows the effects of anti-mTNF and anti-mIL-la on GalN/LPS hepatitis in mice. The injection of LPS (10 ng/mouse) in combination with GalN (10 mg/mouse), neither of which had a n y effect individually, caused hepatic damage at 24 h. Treatment with anti-mTNF 30 min before GalN/LPS injection significantly protected mice from hepatic damage. Pretreatment with anti-mIL-1 a s h o w e d an apparent tendency to reduce the hepatic damage but the effect was not statistically significant. Hepatotoxicity of T N F and/or IL-la in GalN-Sensitized Mice As shown in TABLE 3, the injection GalN (10 mg/mouse) plus rmTNF (100 ng/mouse) elicited hepatic damage in mice. In contrast, the injection of GalN plus rmIL-la (1 ng/mouse) did not elicit hepatic damage. However, treatment with rmIL-1 a (1 ng/mouse) enhanced the hepatotoxic effect of rmTNF (100 ng/mouse) in GalN-sensitized mice (FIG 2).
TABLE 2 Effects of passive immunization against mouse TNF and. IL-la on galactosamineLPS hepatitis in mice.
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
Treatment
Dose Number (dilution) of animals
4 Saline alone Galactosamine alone 4 LPS alone 4 ine/J .PS - t r e u Saline control 6 Serum control 1:3 6 Anti-mTNF 1:30 6
1:lO 1:3 Anti-mIL-la
1:30 1:lO 1 :3
6 6 6 6 6
Plasma ALT (Ub-1
Plasma AST (U/L)
50 f 3 52 f 1 59 f 3
28 f 0 22 f 1 24 f 1
853 f 886 f 43 f 68 f 22f 1099 f 872 f 431 f
131 211 3 b) 24 b) 1b) 314 595 193
451 487 72 73
58 445 629 217
f f f f f f f f
94 134 50)
10 a ) 4b) 112 435 80
~~
b , Significantly different from saline control: p c 0.05, and p c 0.01, respectively (Student's f test).
TABLE 3 Hepatotoxicity of LPS, rmTNF or rmIL-la in combination with galactosamine in mice. Treatment
Dose (/mouse)
Saline alone Galactosamine alone 10 mg LPS alone 10 ng rmTNF alone 100 ng (3x105 U) r m I L - l a alone 1 ng ( 100 U) Galactosamone (10 mdmousebtreated 0.1 n g LPS 1 ng 10 n g rmTNF 1 ng (3x103 U) 10 ng (3x104 U) 100 ng (3x105 U) rmIL- 1 a 0.01 ng (1 U) 0.1 ng (10 U) 1 ng (100 U)
n
Plasma ALT (U/L)
3 3 3 3 3
22f 20 f 33 f 21 f 33 f
4 4 4 4 4 4 4 4 4
21 f 498 f 2585 f 22 f 25 f 1852 f 25 f 20 f 20f
1 1 4 1
15
Plasma AST (U/L)
56f 52f 120 f 62f 127f
8 2 38 18 80
1 58f 3 327 280f 173 355 1351 f 201 2 53f 3 3 60f 6 409 881 f 150 1 76f 16 2 49f 1 1 54f 5
A mixture of galactosamine (10 mg/mouse) and various recombinant murine TNF-a (rmTNF) or recombinant ( r m I L - l a ) was intravenously injected into Balb/c mice. plasma aminotransferase activities are expressed as mean
doses of LPS, murine IL-la The values of f SE.
-
TUMOR NECROSIS FACTOR ALTrctlvlty
2
z a 22
z
Immunopharmacology and Immunotoxicology Downloaded from informahealthcare.com by Cornell University on 12/15/14 For personal use only.
U
t
-1
!! 3000 0
P 4 0.05
ASTrctlvtty
+
T
I ;
-
i
E
P < 0.05
m
+P
