Vol. 169, No. 2, 1990 June 15, 1990
BIOCHEMICAL
ACUTE EFFECTS METABOLISM
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 623-628
OF INTERLEUKIN la AND 6 ON INTERMEDIARY IN FRESHLY ISOLATED RAT HEPATOCYTES
W.J. Vaartjes, C.G.M.
de Haas and M. Houweling
Laboratory of Veterinq Biochemistry, Utrecht University, P.O. Box 80.176, 3508 TD Utrecht, The Netherlands Received
April
17,
1990
SUMMARX Using hepatocytes in suspension, freshly isolated from adult male fed rats, we studied the acute influence of recombinant human interleukins la, 2 and 6 on glycogen and fatty acid metabolism. By far the largest effects were observed with interleukin-la: short incubations (up to 60 min) sufficed to depress glycogen synthesis in a dose-dependent manner, while the rates of glycogenolysis and glycolysis were increased as indicated by the release of glucose and lactate. Interleukin-6 acted similarly, though being much less effective on a molar basis, whereas interleukin-2 only caused a small increase in lactate production. In hepatocytes from 24h-starved rats interleukin-la caused a minor stimulation of gluconeogenesis. Although neither fatty acid synthesis nor oxidation of fatty acids in quiescent hepatocytes from fed rats was significantly affected by interleukins, interleukin-la was able to cause appreciable inhibition of fatty acid synthesis in hepatocytes from regenerating liver (isolated 22h after partial hepatectomy). It is concluded (i) that interleukins, in particular interleukin-la, acutely promote hepatic glucose release, and (ii) that transition of adult hepatocytes from a quiescent into a proliferatory state allows the occurrence of rapid effects of interleukin-la on fatty acid metabolism. 01990 Academic Press, Inc.
Both interleukin-1 (1,2) and interleukind (3) have been implicated in a variety of biological processes in different target cells. They have in common, however, their involvement in the synthesis of acute phase proteins. The secretion of these plasma proteins represents a hepatic response to sepsis and inflammatory processes. Interleukin-6 was reported to induce a wide spectrum acute phase response, whereas interleukin-1 stimulates generation of a limited subset of the acute phase proteins (4,5). In this connection isolated hepatocytes provide a relevant model system to study metabolic changes connected with the acute phase response, firstly because most of the acute phase proteins are exclusively synthesized by this cell type (6), and secondly because hepatocytes occupy a central position in the whole-body metabolism of glucose and fatty acids. Interleukin-2 was included in our experiments for control purposes. Apart from a stimulatory effect on the growth of glial cells (7) interleukin-2 is commonly supposed to exert its main effects on lymphoid cell types. 0006-291X40 623
$1.50
Copyrighr 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol. 169, No. 2, 1990
Materials
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
and Methods
Human recombinant ILla (rhIGla) was purchased from Janssen Biochimica (Beerse, Belgium); spec. act. 1.3 x 10’ U/m#. Human recombinant IL2 (rhIL-2; > 2 x lo6 U/mg) and IL6 (rhIG6; > 2 x 10 U/mg) were from Boehringer (Mannheim, F.R.G.). Other samples of rhIL-6 (2 x lo6 U/mg), kindly donated by Drs. LA Aarden (Amsterdam) and J. Content (Bruxelles), gave essentially the same results as the commercial IL6 preparation. Bovine serum albumin and collagenase type I were obtained from Sigma (St. Louis, U.S.A.); other enzymes and biochemicals were from Boehringer. Radioactive compounds were obtained from the Radiochemical Centre (Amersham, U.K.). Freshly isolated hepatocytes (8) from adult male Wistar rats (175-225 g), ad libitum fed a standard pelleted diet or (in case of gluconeogenesis) starved for 24 h, were used in suspension. In case of regenerating liver (Table II), ad libitum fed male Wistar rats (150-200 g) were used that had been kept under an inverted light cycle (dark between 06.00 a.m. and 18.00 p.m.) for at least two weeks. The hepatocyte isolation procedure was carried out at t = 22 h after either two-third partial hepatectomy or the appropriate control for surgical stress (sham operation), exactly as described earlier (9). The basic incubation mixture consisted of Krebs-Ringer bicarbonate buffer (pH 7.4) with 2.5 mM CaCl,, and l%(w/v) defatted and dialysed bovine serum albumin; cell concentration, 3-4 mg Lowry protein/ml. External glucose concentrations were: 30 mM (glycogen synthesis), 10 mM (fatty acid synthesis and oxidation) or zero (glycogen breakdown, gluconeogenesis). Triplicate incubations (final volume, 2.0 ml) were carried out at 37OC in a metabolic shaker (80 osc/min) in 25-ml Erlenmeyer flasks under an atmosphere of O.&O, (19:1, v/v). Fatty acid synthesis was monitored as incorporation of 3H,0 (0.5 mCi/ml) into total saponifiable fatty acids (8); glycogen synthesis as incorporation of [U“C]glucose (30 mM, 0.01 Ci/mol) into ethanol-precipitable glycogen (10); glycogen breakdown as release of glucose and lactate in the absence of added glucose; gluconeogenesis as glucose formation from 10 mM L-lactate (+ 1.5 mM pyruvate); and fatty acid oxidation as conversion of [l-14C]oleate (0.5 mM, 0.05 Ci/mol into 14C0, and acid-soluble radioactive products (mainly ketone bodies). The 14C0, released was trapped in 0.1 ml 6M KOH present in center wells. Standard methods were used for enzymatic assays of glucose (11) and lactate (12). Statistical analysis was performed using paired t-testing.
Results and Discussion Because profound changes in glucose homeostasis are a clinical hallmark of animals suffering from sepsis or trauma (13), we first investigated whether exposure of hepatocytes to interleukins would rapidly affect glucose metabolism in this cell type. The results shown in Table I indicate that rhIGla exerted a fast inhibitory effect on de novo glycogen synthesis and that, in the absence of added glucose, rh&la enhanced breakdown of endogenously present glycogen. Although the major part of glycogen glucose is released to the medium, some of it is channeled towards glycolysis. As judged from the accumulation of lactate, a rough index of the glycolytic rate, this latter pathway was rapidly stimulated by rhIL-la as well. In terms of glycogen synthesis and glycogenolysis we observed similar, though much smaller, effects with rhIL6, while lactate production also showed a tendency to increase after addition of this latter cytokine (see Table I). Ritchie (14) recently 624
Vol.
169, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table I Effect of interleukins on glucoseand fatty acid metabolismin isolated rat hepatocytes + Interleukin (Percentageof control without interleukin)
Metabolic parameter
rhICla (100 u/m Glycogen synthesis
(n = 5)
Glycogen breakdown - Glucose release - Lactate release
(n = 3)
Gluconeogenesis
(n = 4)
112rt
Fatty acid synthesis
(n = 4)
955
Fatty acid oxidation - 14C02production - acid-solubleproducts
(n = 3)
rhIL2 (100 u/w
71 AZ 6 *** 146 + 11 ** 132-c 8**
96 + 11
rhIG6 W-Jo u/m 89+
5’
104a 1122
5 5’
5’
972
2
98*
7
7
97-c
9
1032
4
89 + 13 101 + 4
97 k 28 103 ? 4
114? 6* 110 -c 7
94 k 17 loo? 5
Freshly isolated hepatocyteswere incubated with or without interleukins as describedin Materials and Methods. Incubation time, 60 min; except for fatty acid oxidation (40 min) and glycogen breakdown (30 min, from the time-interval t = lo-40 rnin). Control (100%) values were: 140 5 49 nmol glucose incorporated/h.mg protein (glycogen synthesis);glucoseand lactate release,482 ? 40 and 458 + 122 nmol/h.mg$rotein, respectively; 453 -C 36 run01glucoseproduced/h. mg protein (gluconeogenesis); 39 + 3 run01 H,O inc./h. mg protein (fatty acid synthesis);11.1 ? 1.4 and 27.8 & 6.5 nmol oleate utilized/h. mg protein (14C0, production and formation of acid-solubleproducts, respectively). Shown are means + SD. of triplicate incubations from at least 3 different hepatocyte preparations (number of preparationsin parentheses). Versus control: * P < 0.05; ** P < 0.02; *** P < 0.005
reported a more pronounced release of glucose from
prelabeled
glycogen pools
during the incubation of cultured hepatocytes with rhIL-6, but this effect required a
2-h lag period. In our hands rhIG2 did affect neither glycogen synthesis nor glycogen degradation to a significant extent, although it slightly promoted lactate accumulation (Table I). As visualized in Fig. 1 for the process of glycogen synthesis, the metabolic effects of rhILla were not only fast, but also dose-dependent. On the other hand, the small response to rhIL6 (cf. Table I) did not change as a function of rhIL6 concentration (Fig. 1). Inhibition of glycogen synthesis by rhILl-a was not likely to be due to cytotoxic effects of this cytokine, as the rate of fatty acid synthesis remained virtually unchanged (Table I). Moreover, incubations with rhIL-la
at 400
U/ml showed no decrease of cell viability (trypan blue exclusion) after 60 min (data not shown). In addition, we examined the ability of interleukins to affect gluconeogenesis from lactate in hepatocytes from 24-h starved rats. At the given incubation con625
Vol.
169,
No.
2, 1990
BIOCHEMICAL
AND
0
BIOPHYSICAL
200 [IL-la]
COMMUNICATIONS
400 W/ml1
,
I
0
2000 [IL-6
RESEARCH
]
I 4000 W/ml)
Fi . 1. Influence of interleukin concentration on the rate of glycogen synthesis in y isolated rat hepatocytes. Glycogen synthesis was monitored as incorporation of [U-‘4C]glucose (see Materials and Methods). Incubation time, 60 min. Values shown are means f S.D. for triplicate incubations in a representative experiment. Closed symbols, + rhILla; open symbols, + rhIG6.
-%fresh
caused a significant stimulation of this biosynthetic pathway (Table I). A slightly more pronounced stimulation by rhILla (up to 125% of control values) was observed in preliminary experiments (not shown) using Lalanine (5 mM) rather than lactate as gluconeogenic precursor. RhIL-la-induced stimulation of alanine uptake (15) may contribute to this difference. These results point to a rapid release of glucose by hepatocytes as a consequence of IL1 and IL6 secretion. In doing so these interleukins, especially rhIL la, act in concert with other secretion products of Kupffer cells such as prostaglandins Da (16) and F2= (17). Inasmuch as hepatocytes determine the blood glucose level, this glucose release could explain the transient hyperglycemia reported to occur during sepsis and shock (13). The hypoglycemia developing at a later stage (13) may be due to depletion of initial glycogen stores and/or increased glucose consumption in peripheral tissues. This hypoglycemia would also imply that in the long term the modest stimulation of hepatic gluconeogenesis by IL-1 (Table I) cannot meet the actual glucose requirement of the body. However, as discussed by Evans et al. (2), further research in this area is needed in view of conflicting reports on long-term effects of individual cytokines on glucose homeostasis. As to the influence of interleukins on fatty acid metabolism, rhIGla caused a small inhibition of [l-‘4C]oleate oxidation as far as 14C02 production was concerned. However, this effect, being diagnostic for depressed Krebs cycle activity, did not attain significance within the 4Omin incubation period (Table I). In addition, as indicated by the amount of acid-soluble radioactivity, ketone body formation was not changed by rhIGla. The other interleukins showed no appreciable effect on oleate oxidation either. Furthermore Table I demonstrates that under our ditions
only rhILla
626
Vol.
169, No. 2, 1990
Rapid effect of interleukin-la
BIOCHEMICAL
Sham-operation
RESEARCH COMMUNICATIONS
Table II on fatty acid synthesis in isolated hepatocytes from regenerating rat liver Additions
Partial hepatectomy
AND BIOPHYSICAL
Rate of fatty acid synthesis (run01 3H20 incorporated/h. mg protein)
None + rhIGla
24.6 + 1.0
None + rhLla
22.0 SC1.6 23.2 2 3.7
19.0 f 1.5 * (77%) (105%)
Rat hepatocytes were isolated at t=22 h post surgery (either 2/3 partial hepatectomy or sham-operation), and were incubated with 3H,0 and with or without rhIGla (125 U/ml). Incubation time, 60 min. For further details, see Material and Methods. Data shown represent means + S.D. of triplicate incubations from 3 different hepatocyte preparations for each surgical procedure. In parentheses, percentage of corresponding control value. Versus control: * P < 0.01
experimental conditions the de nova synthesis of fatty acids was not affected by any of the interleukins. This latter observation is in line with data published by ArgilCs et al. (18). Using an in vivo approach these authors found no effect (within 5 h) of IL1 on hepatic or adipose tissue lipogenesis in fed rats. Interestingly, the condition of the liver cells may affect the results obtained with interleukins. This became apparent when we used IL-la in studies with hepatocytes from partially hepatectomized and sham-operated rats. Adult rat hepatocytes are generally in a quiescent state. However, after partial hepatectomy a synchronized wave of DNA synthesis and subsequent mitosis develops in the liver remnants within 20-30 h post surgery, thereby allowing comparison of proliferating hepatocytes (partial hepatectomy) and non-proliferating hepatocytes (sham operation). As clearly demonstrated in Table II, ILla is indeed able to rapidly inhibit fatty acid synthesis in proliferating hepatocytes (22 h post surgery) to a significant extent. Although this dependency on the proliferatory state is difficult to interpret for the time being, the effect of IGla may be connected with the optimal DNA synthesis occurring between 20 and 24 h after partial hepatectomy (19). In fact, at other time-points (4 and 48 h after partial hepatectomy) the inhibition by IL-la was less prominent (data not shown). It remains to be seen whether metabolic effects of IL6 are also affected by the proliferatory state of hepatocytes. In conclusion, we have presented evidence here that IL1 and IL6 exert acute effects on intermediary metabolism in adult, non-proliferating hepatocytes. Their effects on glycogen metabolism favour the release rather than storage of glucose. Only IL-la stimulated gluconeogenesis, whereas fatty acid metabolism was affected by none of the interleukins. Yet a rapid effect of ILla on fatty acid synthesis could be observed if proliferating hepatocytes were used. Current investigations are 627
Vol.
169,
No.
BIOCHEMICAL
2, 1990
aimed to study long-term as the underlying
metabolic
AND
BIOPHYSICAL
effects of IL1
RESEARCH
COMMUNICATIONS
and IL-6 on hepatocytes as well
signaling mechanisms.
Acknowlednments These investigations were supported in part by the Netherlands Foundation for Chemical Research (SON) with financial aid from the Netherlands Organization for Scientific Research (NWO). References Dinarello, CA. (1989) Adv. Immunol. 44, 153-205. Evans, R.D., Argiles, J.M. and Williamson, D.H. (1989) Clin. Sci. 77, 357-364. 3. Heimich, P.C., Castell, J.V. and Andus, T. (1990) Biochem. J. 265, 621636. 4. Gauldie, J., Richards, C., Harnish, D., Lansdorp, P. and Baumann, H. (1987) Proc. Natl. Acad. Sci. USA 84, 7251-7255. 5. Andus, T., Geiger, T., Hirano, T., Kishimoto, T., Tran-Thi, T.-A, Decker, K. and Heimich, P.C. (1988) Eur. J. Biochem. 173, 287-293. 6. Koj, A. (1985) in: The Acute-Phase Response to Injury and Infection (Gordon, AH. and Koj, A, eds.), Vol. 10, pp. 139-232, Elsevier, Amsterdam. Benveniste, E.N. and Merrill, J.E. (1986) Nature 321, 610-613. ii: Beynen, A.C., Vaartjes, W.J. and Geelen, M.J.H. (1979) Diabetes 28, 828-835. 9. Houweling, M., Vaartjes, W.J. and van Golde, L.M.G. (1989) Biochem. Biophys. Res. Comrnun. 158, 294-301. 10. Hassid, W.Z. and Abraham, S. (1957) in: Methods in Enzymology (Colowick, S.P. and Kaplan, N.O., eds.), Vol. III, pp. 37-38, Academic Press, New York. 11. Slein, M.W. (1963) in: Methods of Enzymatic Analysis (Bergmeyer, H.U., ed.), pp. 117-123, Academic Press, New York. 12. Hohorst, H.J. (1974) in: Methoden der enzymatischen Analyse H.U., ed.), Vol. 2, pp. 1425-1429, Verlag Chemie, w&y?gx ;:
13. Wichterman, K.A., Baue, A.E. and Chaudry, I.H. (1980), J. Surg. Res. 29, 189-201 14. Ritchie, D.G. (1990) Am. J. Physiol. 258, E57-E64. 15. Roh, M.S., Moldawer, L.L., Ekman, LG., Dinarello, CA., Bistrian, B.R., Jeevanandam, M. and Brennan, M.F. (1986) Metabolism 35, 419-424. 16. Casteleijn, E., Kuiper, J., van Rooij, H.C.J., Kamps, J.A.A.M., Koster, J.F. and van Berkel, T.J.C. (1988) Biochem. J. 250, 77-80. 17. Athari, A. and Jungermann, K. (1989) Biochem. Biophys. Res. Commun. 163, 1235-1242. 18. Argiles, J.M., Lopez-Soriano, F.J., Evans, R.D. and Williamson, D.H. (1989) 259, 673-678. 19. McGowan, J.A. and Fausto, N. (1978) Biochem. J. 170, 123-127.
628
BIOCHEMICAL
ACUTE EFFECTS METABOLISM
AND BIOPHYSICAL RESEARCH COMMUNICATIONS Pages 623-628
OF INTERLEUKIN la AND 6 ON INTERMEDIARY IN FRESHLY ISOLATED RAT HEPATOCYTES
W.J. Vaartjes, C.G.M.
de Haas and M. Houweling
Laboratory of Veterinq Biochemistry, Utrecht University, P.O. Box 80.176, 3508 TD Utrecht, The Netherlands Received
April
17,
1990
SUMMARX Using hepatocytes in suspension, freshly isolated from adult male fed rats, we studied the acute influence of recombinant human interleukins la, 2 and 6 on glycogen and fatty acid metabolism. By far the largest effects were observed with interleukin-la: short incubations (up to 60 min) sufficed to depress glycogen synthesis in a dose-dependent manner, while the rates of glycogenolysis and glycolysis were increased as indicated by the release of glucose and lactate. Interleukin-6 acted similarly, though being much less effective on a molar basis, whereas interleukin-2 only caused a small increase in lactate production. In hepatocytes from 24h-starved rats interleukin-la caused a minor stimulation of gluconeogenesis. Although neither fatty acid synthesis nor oxidation of fatty acids in quiescent hepatocytes from fed rats was significantly affected by interleukins, interleukin-la was able to cause appreciable inhibition of fatty acid synthesis in hepatocytes from regenerating liver (isolated 22h after partial hepatectomy). It is concluded (i) that interleukins, in particular interleukin-la, acutely promote hepatic glucose release, and (ii) that transition of adult hepatocytes from a quiescent into a proliferatory state allows the occurrence of rapid effects of interleukin-la on fatty acid metabolism. 01990 Academic Press, Inc.
Both interleukin-1 (1,2) and interleukind (3) have been implicated in a variety of biological processes in different target cells. They have in common, however, their involvement in the synthesis of acute phase proteins. The secretion of these plasma proteins represents a hepatic response to sepsis and inflammatory processes. Interleukin-6 was reported to induce a wide spectrum acute phase response, whereas interleukin-1 stimulates generation of a limited subset of the acute phase proteins (4,5). In this connection isolated hepatocytes provide a relevant model system to study metabolic changes connected with the acute phase response, firstly because most of the acute phase proteins are exclusively synthesized by this cell type (6), and secondly because hepatocytes occupy a central position in the whole-body metabolism of glucose and fatty acids. Interleukin-2 was included in our experiments for control purposes. Apart from a stimulatory effect on the growth of glial cells (7) interleukin-2 is commonly supposed to exert its main effects on lymphoid cell types. 0006-291X40 623
$1.50
Copyrighr 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol. 169, No. 2, 1990
Materials
BIOCHEMICAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
and Methods
Human recombinant ILla (rhIGla) was purchased from Janssen Biochimica (Beerse, Belgium); spec. act. 1.3 x 10’ U/m#. Human recombinant IL2 (rhIL-2; > 2 x lo6 U/mg) and IL6 (rhIG6; > 2 x 10 U/mg) were from Boehringer (Mannheim, F.R.G.). Other samples of rhIL-6 (2 x lo6 U/mg), kindly donated by Drs. LA Aarden (Amsterdam) and J. Content (Bruxelles), gave essentially the same results as the commercial IL6 preparation. Bovine serum albumin and collagenase type I were obtained from Sigma (St. Louis, U.S.A.); other enzymes and biochemicals were from Boehringer. Radioactive compounds were obtained from the Radiochemical Centre (Amersham, U.K.). Freshly isolated hepatocytes (8) from adult male Wistar rats (175-225 g), ad libitum fed a standard pelleted diet or (in case of gluconeogenesis) starved for 24 h, were used in suspension. In case of regenerating liver (Table II), ad libitum fed male Wistar rats (150-200 g) were used that had been kept under an inverted light cycle (dark between 06.00 a.m. and 18.00 p.m.) for at least two weeks. The hepatocyte isolation procedure was carried out at t = 22 h after either two-third partial hepatectomy or the appropriate control for surgical stress (sham operation), exactly as described earlier (9). The basic incubation mixture consisted of Krebs-Ringer bicarbonate buffer (pH 7.4) with 2.5 mM CaCl,, and l%(w/v) defatted and dialysed bovine serum albumin; cell concentration, 3-4 mg Lowry protein/ml. External glucose concentrations were: 30 mM (glycogen synthesis), 10 mM (fatty acid synthesis and oxidation) or zero (glycogen breakdown, gluconeogenesis). Triplicate incubations (final volume, 2.0 ml) were carried out at 37OC in a metabolic shaker (80 osc/min) in 25-ml Erlenmeyer flasks under an atmosphere of O.&O, (19:1, v/v). Fatty acid synthesis was monitored as incorporation of 3H,0 (0.5 mCi/ml) into total saponifiable fatty acids (8); glycogen synthesis as incorporation of [U“C]glucose (30 mM, 0.01 Ci/mol) into ethanol-precipitable glycogen (10); glycogen breakdown as release of glucose and lactate in the absence of added glucose; gluconeogenesis as glucose formation from 10 mM L-lactate (+ 1.5 mM pyruvate); and fatty acid oxidation as conversion of [l-14C]oleate (0.5 mM, 0.05 Ci/mol into 14C0, and acid-soluble radioactive products (mainly ketone bodies). The 14C0, released was trapped in 0.1 ml 6M KOH present in center wells. Standard methods were used for enzymatic assays of glucose (11) and lactate (12). Statistical analysis was performed using paired t-testing.
Results and Discussion Because profound changes in glucose homeostasis are a clinical hallmark of animals suffering from sepsis or trauma (13), we first investigated whether exposure of hepatocytes to interleukins would rapidly affect glucose metabolism in this cell type. The results shown in Table I indicate that rhIGla exerted a fast inhibitory effect on de novo glycogen synthesis and that, in the absence of added glucose, rh&la enhanced breakdown of endogenously present glycogen. Although the major part of glycogen glucose is released to the medium, some of it is channeled towards glycolysis. As judged from the accumulation of lactate, a rough index of the glycolytic rate, this latter pathway was rapidly stimulated by rhIL-la as well. In terms of glycogen synthesis and glycogenolysis we observed similar, though much smaller, effects with rhIL6, while lactate production also showed a tendency to increase after addition of this latter cytokine (see Table I). Ritchie (14) recently 624
Vol.
169, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Table I Effect of interleukins on glucoseand fatty acid metabolismin isolated rat hepatocytes + Interleukin (Percentageof control without interleukin)
Metabolic parameter
rhICla (100 u/m Glycogen synthesis
(n = 5)
Glycogen breakdown - Glucose release - Lactate release
(n = 3)
Gluconeogenesis
(n = 4)
112rt
Fatty acid synthesis
(n = 4)
955
Fatty acid oxidation - 14C02production - acid-solubleproducts
(n = 3)
rhIL2 (100 u/w
71 AZ 6 *** 146 + 11 ** 132-c 8**
96 + 11
rhIG6 W-Jo u/m 89+
5’
104a 1122
5 5’
5’
972
2
98*
7
7
97-c
9
1032
4
89 + 13 101 + 4
97 k 28 103 ? 4
114? 6* 110 -c 7
94 k 17 loo? 5
Freshly isolated hepatocyteswere incubated with or without interleukins as describedin Materials and Methods. Incubation time, 60 min; except for fatty acid oxidation (40 min) and glycogen breakdown (30 min, from the time-interval t = lo-40 rnin). Control (100%) values were: 140 5 49 nmol glucose incorporated/h.mg protein (glycogen synthesis);glucoseand lactate release,482 ? 40 and 458 + 122 nmol/h.mg$rotein, respectively; 453 -C 36 run01glucoseproduced/h. mg protein (gluconeogenesis); 39 + 3 run01 H,O inc./h. mg protein (fatty acid synthesis);11.1 ? 1.4 and 27.8 & 6.5 nmol oleate utilized/h. mg protein (14C0, production and formation of acid-solubleproducts, respectively). Shown are means + SD. of triplicate incubations from at least 3 different hepatocyte preparations (number of preparationsin parentheses). Versus control: * P < 0.05; ** P < 0.02; *** P < 0.005
reported a more pronounced release of glucose from
prelabeled
glycogen pools
during the incubation of cultured hepatocytes with rhIL-6, but this effect required a
2-h lag period. In our hands rhIG2 did affect neither glycogen synthesis nor glycogen degradation to a significant extent, although it slightly promoted lactate accumulation (Table I). As visualized in Fig. 1 for the process of glycogen synthesis, the metabolic effects of rhILla were not only fast, but also dose-dependent. On the other hand, the small response to rhIL6 (cf. Table I) did not change as a function of rhIL6 concentration (Fig. 1). Inhibition of glycogen synthesis by rhILl-a was not likely to be due to cytotoxic effects of this cytokine, as the rate of fatty acid synthesis remained virtually unchanged (Table I). Moreover, incubations with rhIL-la
at 400
U/ml showed no decrease of cell viability (trypan blue exclusion) after 60 min (data not shown). In addition, we examined the ability of interleukins to affect gluconeogenesis from lactate in hepatocytes from 24-h starved rats. At the given incubation con625
Vol.
169,
No.
2, 1990
BIOCHEMICAL
AND
0
BIOPHYSICAL
200 [IL-la]
COMMUNICATIONS
400 W/ml1
,
I
0
2000 [IL-6
RESEARCH
]
I 4000 W/ml)
Fi . 1. Influence of interleukin concentration on the rate of glycogen synthesis in y isolated rat hepatocytes. Glycogen synthesis was monitored as incorporation of [U-‘4C]glucose (see Materials and Methods). Incubation time, 60 min. Values shown are means f S.D. for triplicate incubations in a representative experiment. Closed symbols, + rhILla; open symbols, + rhIG6.
-%fresh
caused a significant stimulation of this biosynthetic pathway (Table I). A slightly more pronounced stimulation by rhILla (up to 125% of control values) was observed in preliminary experiments (not shown) using Lalanine (5 mM) rather than lactate as gluconeogenic precursor. RhIL-la-induced stimulation of alanine uptake (15) may contribute to this difference. These results point to a rapid release of glucose by hepatocytes as a consequence of IL1 and IL6 secretion. In doing so these interleukins, especially rhIL la, act in concert with other secretion products of Kupffer cells such as prostaglandins Da (16) and F2= (17). Inasmuch as hepatocytes determine the blood glucose level, this glucose release could explain the transient hyperglycemia reported to occur during sepsis and shock (13). The hypoglycemia developing at a later stage (13) may be due to depletion of initial glycogen stores and/or increased glucose consumption in peripheral tissues. This hypoglycemia would also imply that in the long term the modest stimulation of hepatic gluconeogenesis by IL-1 (Table I) cannot meet the actual glucose requirement of the body. However, as discussed by Evans et al. (2), further research in this area is needed in view of conflicting reports on long-term effects of individual cytokines on glucose homeostasis. As to the influence of interleukins on fatty acid metabolism, rhIGla caused a small inhibition of [l-‘4C]oleate oxidation as far as 14C02 production was concerned. However, this effect, being diagnostic for depressed Krebs cycle activity, did not attain significance within the 4Omin incubation period (Table I). In addition, as indicated by the amount of acid-soluble radioactivity, ketone body formation was not changed by rhIGla. The other interleukins showed no appreciable effect on oleate oxidation either. Furthermore Table I demonstrates that under our ditions
only rhILla
626
Vol.
169, No. 2, 1990
Rapid effect of interleukin-la
BIOCHEMICAL
Sham-operation
RESEARCH COMMUNICATIONS
Table II on fatty acid synthesis in isolated hepatocytes from regenerating rat liver Additions
Partial hepatectomy
AND BIOPHYSICAL
Rate of fatty acid synthesis (run01 3H20 incorporated/h. mg protein)
None + rhIGla
24.6 + 1.0
None + rhLla
22.0 SC1.6 23.2 2 3.7
19.0 f 1.5 * (77%) (105%)
Rat hepatocytes were isolated at t=22 h post surgery (either 2/3 partial hepatectomy or sham-operation), and were incubated with 3H,0 and with or without rhIGla (125 U/ml). Incubation time, 60 min. For further details, see Material and Methods. Data shown represent means + S.D. of triplicate incubations from 3 different hepatocyte preparations for each surgical procedure. In parentheses, percentage of corresponding control value. Versus control: * P < 0.01
experimental conditions the de nova synthesis of fatty acids was not affected by any of the interleukins. This latter observation is in line with data published by ArgilCs et al. (18). Using an in vivo approach these authors found no effect (within 5 h) of IL1 on hepatic or adipose tissue lipogenesis in fed rats. Interestingly, the condition of the liver cells may affect the results obtained with interleukins. This became apparent when we used IL-la in studies with hepatocytes from partially hepatectomized and sham-operated rats. Adult rat hepatocytes are generally in a quiescent state. However, after partial hepatectomy a synchronized wave of DNA synthesis and subsequent mitosis develops in the liver remnants within 20-30 h post surgery, thereby allowing comparison of proliferating hepatocytes (partial hepatectomy) and non-proliferating hepatocytes (sham operation). As clearly demonstrated in Table II, ILla is indeed able to rapidly inhibit fatty acid synthesis in proliferating hepatocytes (22 h post surgery) to a significant extent. Although this dependency on the proliferatory state is difficult to interpret for the time being, the effect of IGla may be connected with the optimal DNA synthesis occurring between 20 and 24 h after partial hepatectomy (19). In fact, at other time-points (4 and 48 h after partial hepatectomy) the inhibition by IL-la was less prominent (data not shown). It remains to be seen whether metabolic effects of IL6 are also affected by the proliferatory state of hepatocytes. In conclusion, we have presented evidence here that IL1 and IL6 exert acute effects on intermediary metabolism in adult, non-proliferating hepatocytes. Their effects on glycogen metabolism favour the release rather than storage of glucose. Only IL-la stimulated gluconeogenesis, whereas fatty acid metabolism was affected by none of the interleukins. Yet a rapid effect of ILla on fatty acid synthesis could be observed if proliferating hepatocytes were used. Current investigations are 627
Vol.
169,
No.
BIOCHEMICAL
2, 1990
aimed to study long-term as the underlying
metabolic
AND
BIOPHYSICAL
effects of IL1
RESEARCH
COMMUNICATIONS
and IL-6 on hepatocytes as well
signaling mechanisms.
Acknowlednments These investigations were supported in part by the Netherlands Foundation for Chemical Research (SON) with financial aid from the Netherlands Organization for Scientific Research (NWO). References Dinarello, CA. (1989) Adv. Immunol. 44, 153-205. Evans, R.D., Argiles, J.M. and Williamson, D.H. (1989) Clin. Sci. 77, 357-364. 3. Heimich, P.C., Castell, J.V. and Andus, T. (1990) Biochem. J. 265, 621636. 4. Gauldie, J., Richards, C., Harnish, D., Lansdorp, P. and Baumann, H. (1987) Proc. Natl. Acad. Sci. USA 84, 7251-7255. 5. Andus, T., Geiger, T., Hirano, T., Kishimoto, T., Tran-Thi, T.-A, Decker, K. and Heimich, P.C. (1988) Eur. J. Biochem. 173, 287-293. 6. Koj, A. (1985) in: The Acute-Phase Response to Injury and Infection (Gordon, AH. and Koj, A, eds.), Vol. 10, pp. 139-232, Elsevier, Amsterdam. Benveniste, E.N. and Merrill, J.E. (1986) Nature 321, 610-613. ii: Beynen, A.C., Vaartjes, W.J. and Geelen, M.J.H. (1979) Diabetes 28, 828-835. 9. Houweling, M., Vaartjes, W.J. and van Golde, L.M.G. (1989) Biochem. Biophys. Res. Comrnun. 158, 294-301. 10. Hassid, W.Z. and Abraham, S. (1957) in: Methods in Enzymology (Colowick, S.P. and Kaplan, N.O., eds.), Vol. III, pp. 37-38, Academic Press, New York. 11. Slein, M.W. (1963) in: Methods of Enzymatic Analysis (Bergmeyer, H.U., ed.), pp. 117-123, Academic Press, New York. 12. Hohorst, H.J. (1974) in: Methoden der enzymatischen Analyse H.U., ed.), Vol. 2, pp. 1425-1429, Verlag Chemie, w&y?gx ;:
13. Wichterman, K.A., Baue, A.E. and Chaudry, I.H. (1980), J. Surg. Res. 29, 189-201 14. Ritchie, D.G. (1990) Am. J. Physiol. 258, E57-E64. 15. Roh, M.S., Moldawer, L.L., Ekman, LG., Dinarello, CA., Bistrian, B.R., Jeevanandam, M. and Brennan, M.F. (1986) Metabolism 35, 419-424. 16. Casteleijn, E., Kuiper, J., van Rooij, H.C.J., Kamps, J.A.A.M., Koster, J.F. and van Berkel, T.J.C. (1988) Biochem. J. 250, 77-80. 17. Athari, A. and Jungermann, K. (1989) Biochem. Biophys. Res. Commun. 163, 1235-1242. 18. Argiles, J.M., Lopez-Soriano, F.J., Evans, R.D. and Williamson, D.H. (1989) 259, 673-678. 19. McGowan, J.A. and Fausto, N. (1978) Biochem. J. 170, 123-127.
628
