- Email: [email protected]
Serum interleukin-1 (IL-1) activity in alcoholic hepatitis
Life Sciences, Vol. 39, pp. 1479-1485 Printed in the U.S.A.
SERUM INTERLEUKIN-I
Pergamon Journals
(IL-I) ACTIVITY
IN ALCOHOLIC HEPATITIS
Craig J. McClain, Donald A. Cohen, Charles A. Dinarello, Joseph G. Cannon, Steven I. Shedlofsky, and Alan M. Kaplan Departments of Medicine and Medical Microbiology and Immunology, Lexington VA Medical Center and University of Kentucky, Lexington, KY 40536 and Tufts University School of Medicine, Boston, MA (Received in final form July 25, 1986) Surmary Interleukin-I (IL-I) is a monoklne which has been demonstrated to produce a variety of seemingly diverse metabolic events including fever, neutrophilia, anorexia, altered mineral metabolism, muscle catabolism, and fibroblast proliferation. Because many of the clinical features of alcoholic hepatitis are metabolic abnormalities that have been shown to be caused by IL-I, we questioned whether patients with alcoholic hepatitis had elevated serum levels of IL-I. Six patients with alcoholic hepatitis had serum IL-I activity measured by the thymocyte costimulator assay after serum inhibitors were removed. Their values were compared to those of 6 age and sex-matched healthy controls. Patients with alcoholic hepatitis had markedly elevated serum IL-I activity, with the integrated value of all fractions having serum IL-I activity being 9.8 times that of controls. IL-I activity in serum from alcoholic hepatitis patients also was blocked by antibody to IL-I. We conclude that patients with alcoholic hepatitis have increased serum IL-I activity which may play a role in certain of the metabolic complications of alcoholic hepatitis. Alcoholic hepatitis is an inflammatory disease of the liver that occurs in 10-25% of chronic alcoholics (i-3). Presenting features are often non-specific and may include anorexia, muscle wasting, fever, hypozincemia, neutrophilia, and neutrophil accumulation in the liver (I-4). The cytokine, interleukin-i (IL-I), is a recently described monokine which has been shown to produce a variety of metabolic responses including fever, neutrophilia, anorexia, altered mineral metabolism including hypozincemia, muscle catabolism, and fibroblast proliferation (5-7). Because many of the presenting features of alcoholic hepatitis are similar to those abnormalities that can be induced experimentally by IL-I, we questioned whether serum IL-I levels were elevated in patients with alcoholic hepatitis, and whether IL-I may be at least partially responsible for some of the metabolic abnormalities observed in alcoholic hepatitis. The purpose of this study was to determine serum IL-I levels in patients with well documented alcoholic hepatitis. Methods Patients. Six patients hospitalized with alcoholic hepatitis had serum specimens drawn between 8:00 and I0:00 AM in the fasting state. Serum was
0024-3205/86
$3.00 + .00
1480
Interleukin-i and Alcoholic Hepatitis
Vol. 39, No. 16, 1986
frozen at -20 ° until assayed for IL-I. The clinical profile of these patients is shown in the table. Five of the six patients had biopsy documentation of their alcoholic hepatitis, and the sixth had clinical criteria of alcoholic hepatitis as described by the International Liver Study Group (8). The sixth patient did not undergo liver biopsy because of the presence of tense ascites. One patient died from alcoholic hepatitis. Liver biochemlstries and serum zinc levels were performed the same day that serum for IL-I activity was obtained (Table I). Liver biochemical profiles were determined in our hospital clinical laboratory. Serum zinc was assayed by atomic absorption spectrophotometry as described previously (9).
TABLE 1 Clinical Profile of Patients Age Sex Serum zinc ug/dl Serum albumin g/dl SGOT U/L Bilirubin mg/dl
(nl (nl (nl (nl
70-120) 3.5-5.0) < 20) < 1.0)
43 + Male 38 + 2.6 + 89 + 14 +
4 6 .2 13 2
Serum IL-I Assay. IL-activity was separated from serum inhibitors by gel chromatography as described by Cannon and Dinarello (i0). Briefly, 0.25 ml of serum was loaded onto a i x 30 cm column containing Sephadex G-50 equilibrated with RPMI 1640 medium, containing 2-mercaptoethanol (50 uM), glutamine (200 uM) and gentamicin (40 ug/ml). One ml fractions were collected and assayed for IL-I activity in the thymocyte assay described below. It is difficult to detect IL-I activity in whole serum of febrile patients unless serum inhibitors are removed by gel filtration (i0). Furthermore, normal resting individuals do not have detectable serum IL-I and thus constitute a negative control. Positive controls consisted of normal serum containing an added amount of human monocyte-derived IL-I. In order to verify that the active serum fractions in the thymocyte assay were due to IL-I, neutralization of activity was performed with antl-human monocyte IL-I. IL-I activity was determined by the standard thymocyte costimulator assay as described by Mizel et al (12). Briefly, sample supernatants were added in triplicate to 96 well plates in 0.i ml aliquots and were sterilized in the plate with 5000 fads from a 137Cs irradiator. Thymocytes from endotoxin-resistant C3H/HeJ mice (1.5 x 106 cells/well) were then added to each well in RPMI 1640 medium containing 20% fetal calf serum, 2-mercaptoethanol (50 uM) and PHA (1:160) in a volume of 0.i ml. Plates were cultured for 72 hrs. at 370C and were pulsed with 2 uCi of 3H-methyl-thymidine during the last 4 hrs. of culture. Cells were harvested and pro[iferation was determined by the extent of thymidine incorporation. Results The results of the IL-I assay are expressed as the integrated value of all serum fractions having IL-I activity compared to control (stimulation index=patient IL-I activity/control IL-I activity). The mean stimulation index for these patients was markedly elevated at 9.8 (Figure I). Figure 2 shows IL-I activity from one individual patient compared to normal human serum to which human monocyte IL-I had been added. The IL-I activity in this alcoholic
Vol. 39, No. 16, 1986
Interleukin-1
and Alcoholic Hepatitis
1481
hepatitis patient chromatographed in a pattern similar to that seen in normal serum to which human monocyte IL-I had been added. The activity in lower molecular weight fractlons observed in Figure i is presumably due to cleavage products of IL-I that can occur both in vivo and in vitro during storage of the serum. However, these cleavage products possess biologic activity such as thymocyte proliferation, fever and muscle proteolysis (13). IL-I activity from alcoholic hepatitis patients was blocked by antibody to IL-I (data not shown), further documenting that we were, indeed, measuring IL-I activity in these patients. This anti-IL-1 has no effect on IL-2.
,6I
oalcoholic •, n o r m a l
(n=6) (n=6)
14 ¢,)
o, - 1 2 x
E
10
Q. 0
8 . ~
6
=
4
_J
15
20 fraction
25
30
number
FIG.
1
IL-I activity in fractionated serum from six patients with alcoholic hepatitis (ALD)° The integrated IL-I activity (area under the curve for all fractions having activity) in ALD patients is 9.8 times that seen in fractionated serum from normal controls. Each point represents the mean value for 6 patients or their respective controls.
1482
Interleukin-i
0 0 0
50
N
40 i
and Alcoholic Hepatitis
0 0
Vol. 39, No. 16, 1986
0 0
T
1.O
,
.8
1 i
$ .6
30 1<
• ALD patient
E o
T
I I
20
I I I I I
10 /
J
'1
o normal
serum
+
.4
IL-1
\\
I
/
10
\~o~x~
r~
O cO
6 d
.2
15
20 Fraction
25
30
number
FIG. 2 Detection of IL-I activity in a single alcoholic hepatitis (ALD) patient. Serum (0.25 ml) was fractionated on sephadex G-50 and I ml fractions were collected. Fractions eluting after blue dextran were assayed for IL-I activity in the thymocyte assay. No activity was detected in patient serum before fractlonation, and fractionated normal serum had no activity. A positive control consisted of normal serum with the addition of human monocyte IL-I. Background proliferation of thymocytes was 1390 cpm. IL-I activity in fractionated ALD serum eluted in a pattern similar to normal monocyte IL-I.
Discussion Interleukin-i is a polypeptide mediating several components of the acute phase response (5-7). Other terms that previously have been used synonomously with IL-I include endogenous pyrogen, leukocyte endogenous mediator, lymphocyte activating factor, B-cell activating factor, thymocyte proliferation factor, helper peak I and mononuclear cell factor (5-7,14). It is now clear that these all represent either the same molecule or a family of closely related molecules (14). IL-I can be produced in nearly all tissues containing mononuclear phagocytes including blood monocytes, Kupffer cells, splenic macrophages, and peritoneal macrophages (5-7). In addition, IL-I is also produced by non-phagocytic cells such as keratinocytes and astrocytes. Inducers of IL-I are multiple and include a variety of micro-organisms, bacterial endotoxins, inflammatory agents such as bile salts and CSa (5). IL-I not only has important irmnunologic properties, but also produces a series of diverse metabolic events. IL-I clearly causes fever and is synonomous with the older term endogenous pyrogen (15). The mechanism for this
Vol. 39, No. 16, 1986
Interleukin-i
and Alcoholic Hepatitis
1483
fever is thought to be due to IL-I stimulation of prostaglandin E-2 production in the hypothalmus (15-16). IL-I causes muscle catabolism, also possibly through increased prostaglandin E-2 production (17). Indeed, muscle catabolism induced by IL-I in an in vitro muscle assay system can be blocked by [ndomethacin, a cyclo-oxygenase inhibitor (17). Studies by Clowes et al. demonstrated a serum factor in patients who were septic or who had skeletal trauma which, when added to an in vitro muscle assay system, caused muscle breakdown (18). This factor subsequently was shown to be a cleavage product of IL-I (13). IL-I causes alterations in the plasma levels of certain minerals, including depression of the serum zinc level (5-7,19). IL-I administration to animals causes neutrophilia, with release of mature neutrophils from the bone marrow (5-7,20). Lastly, IL-I recently has been demonstrated to cause anorexia in rats (21). All of these metabolic abnormalities caused by IL-I may occur in patients with alcoholic hepatitis. It previously had been nearly impossible to measure IL-I in human plasma or serum using the thymocyte cost imulator assay because of circulating inhibitor(s) to IL-I (12,22). However, a recently described method for removing inhibitor(s) allowed investigators to detect plasma IL-I activity in febrile patients and in women after ovulation (12). Using this technique, we observed elevated levels of serum IL-I in selected patients with clinically evident alcoholic hepatitis. Several possible mechanisms exist for increased IL-I activity in alcoholic hepatitis. Endotoxin is an extremely potent inducer of IL-I production in experimental animals and in man (5). Recently, it has been shown that alcoholics have increased intestinal permeability to normally "non-absorbable" compounds, and alcohol ingestion in experimental animals has been associated with increased intestinal permeability to a variety of macromolecules (23-26). Abstinence from alcohol decreased permeability of the intestine to these compounds. In vitro studies using intestinal biopsies from alcoholics and healthy controls also showed significantly increased uptake of small molecules such as cyanocobalamin and EDTA in alcoholics (23). Furthermore, high titers to enteric bacteria have been reported in patients with liver disease compared to controls without liver disease (27,28). Portal vein endotoxemia, as determined by the Limulus test, also has been reported in alcoholic cirrhotics (29,30). Thus, there is a body of evidence to suggest that the intestinal tract in the alcoholic is more permeable to a variety of normally non-absorbed materials. Intestinally absorbed materials such as endotoxins may chronically stimulate IL-I release from Kupffer cells or other monocytes/macrophages. Metabolites of arachadonic acid appear to be major regulators of IL-I production, and alterations in fatty acid metabolism are well documented in alcoholic liver disease (31-32). Alterations in the hormone profile in alcoholic liver disease may influence IL-I production. For example, estrogens and progesterone both stimulate IL-I production, and the mild increase in body temperature during the luteal phase of the menstrual cycle correlates with increased IL-I prodution (12,33). Lastly, alcohol or acetaldehyde may directly influence IL-I production. Thus, there are multiple potential causes for increased IL-I in alcoholic hepatitis that warrant further investigation. In summary, our data demonstrate elevated serum IL-I activity in selected patients with clinically evident alcoholic hepatitis, and we suggest that increased IL-I activity may play at least a partial etiological role in certain of the clinical and laboratory abnormalities observed in alcoholic hepatitis such as hypozincemia, fever, neutrophilia, anorexia, and muscle wasting.
1484
Interleukin-i and Alcoholic Hepatitis
Vol. 39, No. 16, 1986
References i. 2. 3. 4. 5. 6. 7. 8.
9. i0. Ii. 12. 13. 14.
15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.
D. ZAKIM, T.D. BOYER, Hepatology - A Textbook of Liver Disease pp. 739-789 W.B. Saunders, Company, Philadelphia (1982). H.J. FALLON, Diseases of the Liver Fifth Edition, L. Schiff and E.R. Schiff eds., pp. 693-708, J.B. Lippincott, Company, Philadelphia (1982). M.F. SORRELL and R.K. ZETTERMAN, Practical Gastro. 6 : 3 6 - 4 2 (1982). C.J. McCLAIN and L-C SU, Alc. Clin. Exp. Res. 7: 5-~0 (1983). C.A. DINARELLO, Rev. Inf. Dis. 6 : 5 1 - 9 5 (1984)T C.A. DINARELLO, N. Eng[. J. Med. 3 1 1 : 1 4 1 3 - 1 4 1 8 (1984). R.F. KAMPSCHMIDT, J. Leukocyte Biology 3 6 : 3 4 1 - 3 5 5 (1984). Fogarty International Center Proceedings no. 22: Chapter I: Hepatitis. (571.1). Alcoholic hepatitis. In: Leevy CM, Popper H, Sherlock S, eds. Diseases of the liver and biliary tract. Standardization of nomenclature, diagnostic criteria and diagnostic methdology. DHEW publication no. (NIH) 76-725. Washington, D.C.: U.S. Government Printing office, p. 6 (1976). C.J. McCLAIN, D.H. VAN THIEL, S. PARKER, L.K. BADZIN and H. GILBERT, Alc. Clin. Exp. Res. 3 : 1 3 5 - 1 4 1 (1979). J.G. CANNON and C.A. DINARELLO, Science 227: 1247-1249 (1985). C.A. DINARELLO, L. RENFER and S.M. WOLFF, Proc. Natl. Acad. Sci. 74: 4624-4627 (1977). S.B. MIZEL, J.J. OPPENHEIM and D.L. ROSENSTREICH, J. Immunol. 120: 1497-1503 (1978). C.A. DINARELLO, G.H.A. CLOWES, H. GORDON, C.A. SARAVIS and S.M. WOLFF, J. Immunol. 133: 1332-1338 (1984). L.A. ARDEN, T.K. BRUNNER, J-C CEROTTINI, J-M DAYER, A.L. DE WECK, C.A. DINARELLO, G. DI SABATO, J.J. FARRAR, I. GERY, S. GILLIS, R.E. HANDSCHUMACHER, C.S. HENNEY, M.K. HOFFMANN, W.J. KOOPMAN, S.M. KRANE, L.B. LACHMAN, I. LEFKOWITS, R.I. MISHELL, S.B. MIZEL, J.J. OPPENHEIM, V. PAETKAU J. PLATE, M. ROLLINGHOFF, D. ROSENSTREICH, A.S. ROSENTHAL, L.J. ROSENWASSER, A. SCHIMPL, H.S. SHIM, P.L. SIMON, K.A. SMITH, H. WAGNER, J.D. WATSON, E. WECKER, D.D. WOOD, J. Immunol. 123: 2928-2929 (1979). C.A. DINARELLO and S.M. WOLFF, Am. J. Med. 7 2 : 7 9 9 - 8 1 9 (1982). H.A. BERNHEIM, T.M. GILBERT and J.T. STITT, J. Physiol. 3 0 1 : 6 9 - 7 8 (1980). V. BARACOS, H.P. RODEMANN, C.A. DINARELLO and A.L. GOLDBERG, N. Engl. J. Med. 3 0 8 : 5 5 3 - 5 5 8 (1983). G.H. CLOWES, B.C. GEORGE, C.A. VILLEE and C.A. SARAVIS, N. Engl. J. Med. 3 0 8 : 5 4 5 - 5 5 2 (1983). R.F. KAMPSCHMIDT and M. MESECHER, Proc. Soc. Exp. Biol. Med. 179: 197-200 (1985). R.F. KAMPSCHMIDT and H.F. UPCHURCH, Proc. Soc. Exp. Biol. Med. 155: 89-93 (1977). D.M. McCARTHY, M.J. KLUGER and A.J. VANDER, J. Leukocyte Biology 37: 725-726 (1985). C.A. DINARELLO, L.J. ROSENWASSER and S.M. WOLFF, J. Immunol. 127: 2517-2519 (1981). I. BJARNASON, K. WARD and T.J. PETERS, Lancet 28: 179-182 (1984). F.A. WILSON and A.M. HOYUMPA, Gastroenterology 7 6 : 3 6 8 - 4 0 3 (1979). J.S. SANDHU and D.R. FRASER, Clin. Sci. 63: 311-316 (1982). B.S. WORTHINGON, L. MESEROLE and J.A. SYROTUCK, Dig. Dis. Sci. 23: 23-32 (1978). D.R. TRIGER, M.H. ALP and R. WRIGHT, Lancet 1 : 6 0 - 6 3 (1972). M. BJORNEBOE, H. PRYTZ and F. ORSKOV, Lancet i: 58-60 (1972). G.A. DEGENSHEIN, P.H. KOZINN, A.I. JACOB, P.K. GOLDBERG and N. BLOOM, Gastroenterology 72: 1268-1270 (1977). B.K. LIEHR, H. PRYTZ, J. HOEST-CHRISTENSEN and B. KORNER, Scand. J. Gastro. 2 : 8 5 7 - 8 6 3 (1976).
Vol. 39, No. 16, 1986
31. 32. 33.
Interleukin-i
and Alcoholic Hepatitis
1485
S.L. KUNKEL and S.W. CHENSUE, Biochem. Biophys. Res. Comm. 128:892-897 (1985). S.B. JOHNSON, E. GORDON, C. McCLAIN, G. LOW and R.T. HOLMAN, Proc. Natl. Acad. Sci. 82: 1815-1818 (1985). A. FLYNN, Lymphokine Res. 3: 1 (1984).
SERUM INTERLEUKIN-I
Pergamon Journals
(IL-I) ACTIVITY
IN ALCOHOLIC HEPATITIS
Craig J. McClain, Donald A. Cohen, Charles A. Dinarello, Joseph G. Cannon, Steven I. Shedlofsky, and Alan M. Kaplan Departments of Medicine and Medical Microbiology and Immunology, Lexington VA Medical Center and University of Kentucky, Lexington, KY 40536 and Tufts University School of Medicine, Boston, MA (Received in final form July 25, 1986) Surmary Interleukin-I (IL-I) is a monoklne which has been demonstrated to produce a variety of seemingly diverse metabolic events including fever, neutrophilia, anorexia, altered mineral metabolism, muscle catabolism, and fibroblast proliferation. Because many of the clinical features of alcoholic hepatitis are metabolic abnormalities that have been shown to be caused by IL-I, we questioned whether patients with alcoholic hepatitis had elevated serum levels of IL-I. Six patients with alcoholic hepatitis had serum IL-I activity measured by the thymocyte costimulator assay after serum inhibitors were removed. Their values were compared to those of 6 age and sex-matched healthy controls. Patients with alcoholic hepatitis had markedly elevated serum IL-I activity, with the integrated value of all fractions having serum IL-I activity being 9.8 times that of controls. IL-I activity in serum from alcoholic hepatitis patients also was blocked by antibody to IL-I. We conclude that patients with alcoholic hepatitis have increased serum IL-I activity which may play a role in certain of the metabolic complications of alcoholic hepatitis. Alcoholic hepatitis is an inflammatory disease of the liver that occurs in 10-25% of chronic alcoholics (i-3). Presenting features are often non-specific and may include anorexia, muscle wasting, fever, hypozincemia, neutrophilia, and neutrophil accumulation in the liver (I-4). The cytokine, interleukin-i (IL-I), is a recently described monokine which has been shown to produce a variety of metabolic responses including fever, neutrophilia, anorexia, altered mineral metabolism including hypozincemia, muscle catabolism, and fibroblast proliferation (5-7). Because many of the presenting features of alcoholic hepatitis are similar to those abnormalities that can be induced experimentally by IL-I, we questioned whether serum IL-I levels were elevated in patients with alcoholic hepatitis, and whether IL-I may be at least partially responsible for some of the metabolic abnormalities observed in alcoholic hepatitis. The purpose of this study was to determine serum IL-I levels in patients with well documented alcoholic hepatitis. Methods Patients. Six patients hospitalized with alcoholic hepatitis had serum specimens drawn between 8:00 and I0:00 AM in the fasting state. Serum was
0024-3205/86
$3.00 + .00
1480
Interleukin-i and Alcoholic Hepatitis
Vol. 39, No. 16, 1986
frozen at -20 ° until assayed for IL-I. The clinical profile of these patients is shown in the table. Five of the six patients had biopsy documentation of their alcoholic hepatitis, and the sixth had clinical criteria of alcoholic hepatitis as described by the International Liver Study Group (8). The sixth patient did not undergo liver biopsy because of the presence of tense ascites. One patient died from alcoholic hepatitis. Liver biochemlstries and serum zinc levels were performed the same day that serum for IL-I activity was obtained (Table I). Liver biochemical profiles were determined in our hospital clinical laboratory. Serum zinc was assayed by atomic absorption spectrophotometry as described previously (9).
TABLE 1 Clinical Profile of Patients Age Sex Serum zinc ug/dl Serum albumin g/dl SGOT U/L Bilirubin mg/dl
(nl (nl (nl (nl
70-120) 3.5-5.0) < 20) < 1.0)
43 + Male 38 + 2.6 + 89 + 14 +
4 6 .2 13 2
Serum IL-I Assay. IL-activity was separated from serum inhibitors by gel chromatography as described by Cannon and Dinarello (i0). Briefly, 0.25 ml of serum was loaded onto a i x 30 cm column containing Sephadex G-50 equilibrated with RPMI 1640 medium, containing 2-mercaptoethanol (50 uM), glutamine (200 uM) and gentamicin (40 ug/ml). One ml fractions were collected and assayed for IL-I activity in the thymocyte assay described below. It is difficult to detect IL-I activity in whole serum of febrile patients unless serum inhibitors are removed by gel filtration (i0). Furthermore, normal resting individuals do not have detectable serum IL-I and thus constitute a negative control. Positive controls consisted of normal serum containing an added amount of human monocyte-derived IL-I. In order to verify that the active serum fractions in the thymocyte assay were due to IL-I, neutralization of activity was performed with antl-human monocyte IL-I. IL-I activity was determined by the standard thymocyte costimulator assay as described by Mizel et al (12). Briefly, sample supernatants were added in triplicate to 96 well plates in 0.i ml aliquots and were sterilized in the plate with 5000 fads from a 137Cs irradiator. Thymocytes from endotoxin-resistant C3H/HeJ mice (1.5 x 106 cells/well) were then added to each well in RPMI 1640 medium containing 20% fetal calf serum, 2-mercaptoethanol (50 uM) and PHA (1:160) in a volume of 0.i ml. Plates were cultured for 72 hrs. at 370C and were pulsed with 2 uCi of 3H-methyl-thymidine during the last 4 hrs. of culture. Cells were harvested and pro[iferation was determined by the extent of thymidine incorporation. Results The results of the IL-I assay are expressed as the integrated value of all serum fractions having IL-I activity compared to control (stimulation index=patient IL-I activity/control IL-I activity). The mean stimulation index for these patients was markedly elevated at 9.8 (Figure I). Figure 2 shows IL-I activity from one individual patient compared to normal human serum to which human monocyte IL-I had been added. The IL-I activity in this alcoholic
Vol. 39, No. 16, 1986
Interleukin-1
and Alcoholic Hepatitis
1481
hepatitis patient chromatographed in a pattern similar to that seen in normal serum to which human monocyte IL-I had been added. The activity in lower molecular weight fractlons observed in Figure i is presumably due to cleavage products of IL-I that can occur both in vivo and in vitro during storage of the serum. However, these cleavage products possess biologic activity such as thymocyte proliferation, fever and muscle proteolysis (13). IL-I activity from alcoholic hepatitis patients was blocked by antibody to IL-I (data not shown), further documenting that we were, indeed, measuring IL-I activity in these patients. This anti-IL-1 has no effect on IL-2.
,6I
oalcoholic •, n o r m a l
(n=6) (n=6)
14 ¢,)
o, - 1 2 x
E
10
Q. 0
8 . ~
6
=
4
_J
15
20 fraction
25
30
number
FIG.
1
IL-I activity in fractionated serum from six patients with alcoholic hepatitis (ALD)° The integrated IL-I activity (area under the curve for all fractions having activity) in ALD patients is 9.8 times that seen in fractionated serum from normal controls. Each point represents the mean value for 6 patients or their respective controls.
1482
Interleukin-i
0 0 0
50
N
40 i
and Alcoholic Hepatitis
0 0
Vol. 39, No. 16, 1986
0 0
T
1.O
,
.8
1 i
$ .6
30 1<
• ALD patient
E o
T
I I
20
I I I I I
10 /
J
'1
o normal
serum
+
.4
IL-1
\\
I
/
10
\~o~x~
r~
O cO
6 d
.2
15
20 Fraction
25
30
number
FIG. 2 Detection of IL-I activity in a single alcoholic hepatitis (ALD) patient. Serum (0.25 ml) was fractionated on sephadex G-50 and I ml fractions were collected. Fractions eluting after blue dextran were assayed for IL-I activity in the thymocyte assay. No activity was detected in patient serum before fractlonation, and fractionated normal serum had no activity. A positive control consisted of normal serum with the addition of human monocyte IL-I. Background proliferation of thymocytes was 1390 cpm. IL-I activity in fractionated ALD serum eluted in a pattern similar to normal monocyte IL-I.
Discussion Interleukin-i is a polypeptide mediating several components of the acute phase response (5-7). Other terms that previously have been used synonomously with IL-I include endogenous pyrogen, leukocyte endogenous mediator, lymphocyte activating factor, B-cell activating factor, thymocyte proliferation factor, helper peak I and mononuclear cell factor (5-7,14). It is now clear that these all represent either the same molecule or a family of closely related molecules (14). IL-I can be produced in nearly all tissues containing mononuclear phagocytes including blood monocytes, Kupffer cells, splenic macrophages, and peritoneal macrophages (5-7). In addition, IL-I is also produced by non-phagocytic cells such as keratinocytes and astrocytes. Inducers of IL-I are multiple and include a variety of micro-organisms, bacterial endotoxins, inflammatory agents such as bile salts and CSa (5). IL-I not only has important irmnunologic properties, but also produces a series of diverse metabolic events. IL-I clearly causes fever and is synonomous with the older term endogenous pyrogen (15). The mechanism for this
Vol. 39, No. 16, 1986
Interleukin-i
and Alcoholic Hepatitis
1483
fever is thought to be due to IL-I stimulation of prostaglandin E-2 production in the hypothalmus (15-16). IL-I causes muscle catabolism, also possibly through increased prostaglandin E-2 production (17). Indeed, muscle catabolism induced by IL-I in an in vitro muscle assay system can be blocked by [ndomethacin, a cyclo-oxygenase inhibitor (17). Studies by Clowes et al. demonstrated a serum factor in patients who were septic or who had skeletal trauma which, when added to an in vitro muscle assay system, caused muscle breakdown (18). This factor subsequently was shown to be a cleavage product of IL-I (13). IL-I causes alterations in the plasma levels of certain minerals, including depression of the serum zinc level (5-7,19). IL-I administration to animals causes neutrophilia, with release of mature neutrophils from the bone marrow (5-7,20). Lastly, IL-I recently has been demonstrated to cause anorexia in rats (21). All of these metabolic abnormalities caused by IL-I may occur in patients with alcoholic hepatitis. It previously had been nearly impossible to measure IL-I in human plasma or serum using the thymocyte cost imulator assay because of circulating inhibitor(s) to IL-I (12,22). However, a recently described method for removing inhibitor(s) allowed investigators to detect plasma IL-I activity in febrile patients and in women after ovulation (12). Using this technique, we observed elevated levels of serum IL-I in selected patients with clinically evident alcoholic hepatitis. Several possible mechanisms exist for increased IL-I activity in alcoholic hepatitis. Endotoxin is an extremely potent inducer of IL-I production in experimental animals and in man (5). Recently, it has been shown that alcoholics have increased intestinal permeability to normally "non-absorbable" compounds, and alcohol ingestion in experimental animals has been associated with increased intestinal permeability to a variety of macromolecules (23-26). Abstinence from alcohol decreased permeability of the intestine to these compounds. In vitro studies using intestinal biopsies from alcoholics and healthy controls also showed significantly increased uptake of small molecules such as cyanocobalamin and EDTA in alcoholics (23). Furthermore, high titers to enteric bacteria have been reported in patients with liver disease compared to controls without liver disease (27,28). Portal vein endotoxemia, as determined by the Limulus test, also has been reported in alcoholic cirrhotics (29,30). Thus, there is a body of evidence to suggest that the intestinal tract in the alcoholic is more permeable to a variety of normally non-absorbed materials. Intestinally absorbed materials such as endotoxins may chronically stimulate IL-I release from Kupffer cells or other monocytes/macrophages. Metabolites of arachadonic acid appear to be major regulators of IL-I production, and alterations in fatty acid metabolism are well documented in alcoholic liver disease (31-32). Alterations in the hormone profile in alcoholic liver disease may influence IL-I production. For example, estrogens and progesterone both stimulate IL-I production, and the mild increase in body temperature during the luteal phase of the menstrual cycle correlates with increased IL-I prodution (12,33). Lastly, alcohol or acetaldehyde may directly influence IL-I production. Thus, there are multiple potential causes for increased IL-I in alcoholic hepatitis that warrant further investigation. In summary, our data demonstrate elevated serum IL-I activity in selected patients with clinically evident alcoholic hepatitis, and we suggest that increased IL-I activity may play at least a partial etiological role in certain of the clinical and laboratory abnormalities observed in alcoholic hepatitis such as hypozincemia, fever, neutrophilia, anorexia, and muscle wasting.
1484
Interleukin-i and Alcoholic Hepatitis
Vol. 39, No. 16, 1986
References i. 2. 3. 4. 5. 6. 7. 8.
9. i0. Ii. 12. 13. 14.
15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.
D. ZAKIM, T.D. BOYER, Hepatology - A Textbook of Liver Disease pp. 739-789 W.B. Saunders, Company, Philadelphia (1982). H.J. FALLON, Diseases of the Liver Fifth Edition, L. Schiff and E.R. Schiff eds., pp. 693-708, J.B. Lippincott, Company, Philadelphia (1982). M.F. SORRELL and R.K. ZETTERMAN, Practical Gastro. 6 : 3 6 - 4 2 (1982). C.J. McCLAIN and L-C SU, Alc. Clin. Exp. Res. 7: 5-~0 (1983). C.A. DINARELLO, Rev. Inf. Dis. 6 : 5 1 - 9 5 (1984)T C.A. DINARELLO, N. Eng[. J. Med. 3 1 1 : 1 4 1 3 - 1 4 1 8 (1984). R.F. KAMPSCHMIDT, J. Leukocyte Biology 3 6 : 3 4 1 - 3 5 5 (1984). Fogarty International Center Proceedings no. 22: Chapter I: Hepatitis. (571.1). Alcoholic hepatitis. In: Leevy CM, Popper H, Sherlock S, eds. Diseases of the liver and biliary tract. Standardization of nomenclature, diagnostic criteria and diagnostic methdology. DHEW publication no. (NIH) 76-725. Washington, D.C.: U.S. Government Printing office, p. 6 (1976). C.J. McCLAIN, D.H. VAN THIEL, S. PARKER, L.K. BADZIN and H. GILBERT, Alc. Clin. Exp. Res. 3 : 1 3 5 - 1 4 1 (1979). J.G. CANNON and C.A. DINARELLO, Science 227: 1247-1249 (1985). C.A. DINARELLO, L. RENFER and S.M. WOLFF, Proc. Natl. Acad. Sci. 74: 4624-4627 (1977). S.B. MIZEL, J.J. OPPENHEIM and D.L. ROSENSTREICH, J. Immunol. 120: 1497-1503 (1978). C.A. DINARELLO, G.H.A. CLOWES, H. GORDON, C.A. SARAVIS and S.M. WOLFF, J. Immunol. 133: 1332-1338 (1984). L.A. ARDEN, T.K. BRUNNER, J-C CEROTTINI, J-M DAYER, A.L. DE WECK, C.A. DINARELLO, G. DI SABATO, J.J. FARRAR, I. GERY, S. GILLIS, R.E. HANDSCHUMACHER, C.S. HENNEY, M.K. HOFFMANN, W.J. KOOPMAN, S.M. KRANE, L.B. LACHMAN, I. LEFKOWITS, R.I. MISHELL, S.B. MIZEL, J.J. OPPENHEIM, V. PAETKAU J. PLATE, M. ROLLINGHOFF, D. ROSENSTREICH, A.S. ROSENTHAL, L.J. ROSENWASSER, A. SCHIMPL, H.S. SHIM, P.L. SIMON, K.A. SMITH, H. WAGNER, J.D. WATSON, E. WECKER, D.D. WOOD, J. Immunol. 123: 2928-2929 (1979). C.A. DINARELLO and S.M. WOLFF, Am. J. Med. 7 2 : 7 9 9 - 8 1 9 (1982). H.A. BERNHEIM, T.M. GILBERT and J.T. STITT, J. Physiol. 3 0 1 : 6 9 - 7 8 (1980). V. BARACOS, H.P. RODEMANN, C.A. DINARELLO and A.L. GOLDBERG, N. Engl. J. Med. 3 0 8 : 5 5 3 - 5 5 8 (1983). G.H. CLOWES, B.C. GEORGE, C.A. VILLEE and C.A. SARAVIS, N. Engl. J. Med. 3 0 8 : 5 4 5 - 5 5 2 (1983). R.F. KAMPSCHMIDT and M. MESECHER, Proc. Soc. Exp. Biol. Med. 179: 197-200 (1985). R.F. KAMPSCHMIDT and H.F. UPCHURCH, Proc. Soc. Exp. Biol. Med. 155: 89-93 (1977). D.M. McCARTHY, M.J. KLUGER and A.J. VANDER, J. Leukocyte Biology 37: 725-726 (1985). C.A. DINARELLO, L.J. ROSENWASSER and S.M. WOLFF, J. Immunol. 127: 2517-2519 (1981). I. BJARNASON, K. WARD and T.J. PETERS, Lancet 28: 179-182 (1984). F.A. WILSON and A.M. HOYUMPA, Gastroenterology 7 6 : 3 6 8 - 4 0 3 (1979). J.S. SANDHU and D.R. FRASER, Clin. Sci. 63: 311-316 (1982). B.S. WORTHINGON, L. MESEROLE and J.A. SYROTUCK, Dig. Dis. Sci. 23: 23-32 (1978). D.R. TRIGER, M.H. ALP and R. WRIGHT, Lancet 1 : 6 0 - 6 3 (1972). M. BJORNEBOE, H. PRYTZ and F. ORSKOV, Lancet i: 58-60 (1972). G.A. DEGENSHEIN, P.H. KOZINN, A.I. JACOB, P.K. GOLDBERG and N. BLOOM, Gastroenterology 72: 1268-1270 (1977). B.K. LIEHR, H. PRYTZ, J. HOEST-CHRISTENSEN and B. KORNER, Scand. J. Gastro. 2 : 8 5 7 - 8 6 3 (1976).
Vol. 39, No. 16, 1986
31. 32. 33.
Interleukin-i
and Alcoholic Hepatitis
1485
S.L. KUNKEL and S.W. CHENSUE, Biochem. Biophys. Res. Comm. 128:892-897 (1985). S.B. JOHNSON, E. GORDON, C. McCLAIN, G. LOW and R.T. HOLMAN, Proc. Natl. Acad. Sci. 82: 1815-1818 (1985). A. FLYNN, Lymphokine Res. 3: 1 (1984).