J Biochem. 109, 507-513 (1991)

Metabolic Deviation of Mouse Liver by RhILl-a or RhTNF/Cachectin1 Kiichi Imamura,2 Ziyuan Wang, Kikuno Murayama-Oda, Hee-Kyoung Kim, and Takehiko Tanaka Department of Nutrition and Physiological Chemistry, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka 565 Received for publication, December 3, 1990

We have investigated the mechanisms of metabolic deviation in tumor-bearing hosts on the basis of the idea that one of the main causes of it is metabolic abnormalities or widespread direct or indirect modification of gene expression of the host by physiologically active factors derived from the tumor (2-5), because it can not be explained as due solely to malnutrition (6-8). Our previous findings and various other lines of evidence (4-13) support the feasibility of the above hypothesis for the cause of the metabolic deviation. In addition, endogenous factors that are produced in excess or inappropriately by host tissues as a result of interaction of these host tissues with invaders such as cancer and microorganisms are also strong candidates for mediators of metabolic deviation in tumor-bearing hosts. Cytokines such as TNF/cachectin and ELI are especially strong candidates because they are suggested to cause a wide variety of metabolic derangements when produced in excess or inappropriately (7, 14-18). Recently, there have been many reports on the effects of cytokines on metabolism (14-18) and on the relationship between cytokines and cachexia (7, 14, 16, 19). However, there have been few 1 This investigation was supported by a grant from the Ministry of Education, Science and Culture of Japan. Part of this investigation has already been reported at the 60th and 62nd Annual Meetings of the Japanese Biochemical Society (Ref. 1). 1 To whom correspondence should be addressed. Abbreviations: ODC, ornithine decarboxylase; ODC factor, ornithine decarboxylase-inducing factor; PK, pyruvate kinase; LDH, L-lactate dehydrogenase; Con A, concanavalin A; DTT, dithiothreitol, rhILl, recombinant human interleukin 1; rhTNF, recombinant human tumor necrosis factor; CX, cycloheximide; Act-D, actinomycin D; I M., indomethacin; FCS, fetal calf serum; i.p., intraperitoneal(-ly).

Vol. 109, No. 4, 1991

studies on the effects of cytokines, especially ILl, on the metabolism of the liver, which is the central metabolic organ of the body. In this study we examined the effects of the rhILl-a (recombinant human ILl-a), and rhTNF/cachectin (recombinant human TNF/cachectin) on ornithine decarboxylase (ODC) [EC] and pyruvate kinase [EC] isozymes in mouse Liver. ODC is the first, and rate-limiting enzyme in polyamine synthesis (20) and its activity is rapidly induced by various growth stimuli, due to its rapid turnover rate (21-24). Therefore, its induction is a sensitive marker of the effect of cytokines. Pyruvate kinase is a rate-limiting enzyme in the glycolytic pathway and its isozymes, the L- and M2-types, are good markers of differentiation and dedifferentiation, respectively, of liver function (2-4, 25). We found that administration of rhTLl-a- to normal mice increased the activity of ODC and its mRNA level, and that of M2 type PK (prototype isozyme) with decrease in PK-L (differentiated type of isozyme) in the liver. RhILl-a also induced ODC in primary cultures of mouse hepatocytes. In addition, we found that actinomycin D had a paradoxical effect (superinduction) (26-28) on liver ODC in mice treated with rhILl-a and that this effect was partially due to stabilization of the ODC enzyme. These results suggest that rhlLla induced deviation of liver metabolism by modification of gene expression, resulting in a state similar to that in a host bearing an Ehrlich ascites tumor. On the basis of these results, we speculate that ILl is an endogenous mediator in the development of cancer cachexia.


Downloaded from https://academic.oup.com/jb/article-abstract/109/4/507/820937 by University of Otago user on 19 January 2019

In this work deviation of liver metabolism by cytokines, especially recombinant human interleukin \-a (rhILl-a), was investigated. Administration of rhILl-a or recombinant human tumor necrosis factor (rhTNF/cachectin) to normal mice resulted in rapid, dosedependent induction of high liver ornithine decarboxylase (ODC) activity. The effects of these cytokines on liver ODC were not indirect effects mediated by eicosanoids. The induction of liver ODC by rhILl-a was at least partly a direct effect on hepatocytes, and was due to increase in de novo synthesis of the enzyme protein after increase in ODC mRNA. No specific protein was required for increase in the level of ODC-mRNA. On ILl treatment, actinomycin D caused superinduction of liver ODC, which was at least partly due to increased stability of the ODC enzyme, because actinomycin D doubled the apparent half-life (from 50 to 95 min). Daily administration of 2X10 3 U of rhILl-a to mice for 3 days also caused decrease in the level of the differentiated type of pyruvate kinase isozyme (PK-L) and marked increase in that of the prototype isozyme (PK-M2) in the liver, but did not cause significant change in the isozyme patterns of the kidney, thymus, and spleen. RhILl-a also induced hypertrophy of the spleen. These results indicate that rhILl-a causes metabolic deviation of the liver similar to that in tumor-bearing hosts.


K. Imamura et al.


Materials—DL-[l-uC]Ornithine hydrochloride (40-60 mCi/mmol) was from RCC Amersham, U.K. Recombinant human interleukin 1-a (rhlLl-a-) and recombinant human TNF/cachectin (rhTNF/cachectin, PT-050) were kindly supplied by Dainippon Pharmaceutical (Suita, Osaka) ODC-cDNA in pBR322 (clone F2) was a generous gift from Dr. C. Kahana (Weizmann Institute) (29). Animals and Treatments—Male ICR mice weighing 2535 g were supplied by Japan Charles River (Atsugi), and were given intraperitoneal (i.p.) injections of rhlLl-a, rhTNF/cachectin, or saline. They were killed by decapitation, and their tissues were promptly excised and washed with cold saline. Samples of tissues for RNA preparation were rapidly frozen on dry ice and stored at — 80'C until use. Tissue extracts for assays of ODC and pyruvate kinase were prepared as described previously (30, 31). Primary Cell Culture—Mouse liver parenchymal cells were isolated as described previously (30). Isolated mature hepatocytes ( l x l O 5 cells/cm2) were cultured in 60 mm Corning plastic dishes in Dulbecco's modified Eagle's MEM in the presence of 10% fetal calf serum (FCS) at 37'C under a humidified atmosphere of 5% CO2 in air for 24 h. Then the culture medium was replaced by FCS-free Dulbecco's modified Eagle's MEM, and culture was continued for 20 h. Assays of Ornithine Decarboxylase and Pyruvate Kinase Isozymes—ODC activity was determined by measuring liberation of CO2 from DL-[l-14C]ornithine as described previously (30, 32). PK activity was assayed at 21*C by the NADH-LDH coupling method as described previously (25, 31). The PK isozymes L and M2 in the tissue extracts were assayed differentially in the presence and absence of M2-IgG and their concentrations were calculated by using the formula described previously (31, 33). Protein was determined by the procedure of Lowry et al. (34) with bovine serum albumin as a standard. Poly(A+)-RNA Preparation and ODC-mRNA Content —The ODC-cDNA insert was purified by gel electrophoresis on 4% polyacrylamide gel after cleavage of the plasmid pBR322 containing the mouse kidney ODC-cDNA (clone F2) (29) with Pstl and electrocution on DE-82 ion exchange paper (35). The cDNA fragment was radiolabeled

acti vlty


a >• X

a a 300 E

1 /

> \ \ \\ \ \ I1 / \

27 O 200 U Q at a 100



/ / / / /



Effect of rhILl-a and rhTNF/Cachectin on ODC Activity in Mouse Live)—Figure 1 (A and B) shows the time courses of induction of liver ODC after administration of 2,000 U of rhILl-a (0.08//g/mouse) and 8,000 U of rhTNF/cachectin (2.76//g/mouse), respectively, to mice. The ODC activities in the liver reached a maximum about 4 and 6 h after the injection and then decreased rapidly to the basal level, respectively. These time courses of ODC mduction are very similar to those observed with most other inducers examined (20-22). Figure 2 shows the dose-response relation for ODC induction by rhlLl-a-; rhlLl-a 1 caused a detectable increase in ODC activity at 200 U (0.008 fi g/mouse) and had nearly a maximal effect at 20,000 U (0.8//g/mouse). Does rhILl-a Exert a Direct or Indirect Influence on Mouse Liver?—Kuiper et al (37) recently found that prostaglandins produced by non-hepatocytes Kupffer's and endothelial cells) in the liver are involved in the rapid stimulation of ODC activity in the liver of rats after injection of phorbol ester. Therefore, we examined whether the effect of rULl-a on the activity of liver ODC in mice was also mediated by eicosanoids derived from nonhepatocytes and other cells. As shown in Table I, induction of ODC in mouse liver by rhlLl-Q- was not blocked by a preinjection of the mouse with indomethacin (6-9 mg/kg body weight), an inhibitor of eicosanoid synthesis, indicating that the effect of rhlLl-a- on ODC activity in mouse liver is not mediated by eicosanoids. The effect of rhTNF/





/ / / /

1 4

£ 300

o o


i 2




^ «


Time after injection (h)


c u O 2






Time after injection (h)

Fig 1. Time coarse of ODC induction by rhILl-o and rhTNF/cachectin. In A, 2,000 UofrhILl-•





X o XI

< •>


• —•




S. 200


. /

o c







o 0

Concentration of r h I L l - a (U/mouse)

Fig. 2. Effect of rhlLl-a concentration on induction of liver ODC. The indicated concentrations of rhILl-a were injected l p into mice The mice were killed 4.5 h later and their liver ODC activity was assayed. Values are means ±SDs for four mice.

TABLE I Effect of indomethacin on ODC induction in mouse liver by rhILl-a or rhTNF/cachectin. Indomethacin (I M , 9 mg/ kg body weight) was injected l.p. into mice 1 h before rhILl-or (2,000 U/mouse) in experiment A or rhTNF/cachectin (10,000 U/mouse) in experiment B. The nnimnla were killed 4h after injection of the cytokine. Data are means ±SDs for four experiments.


1 10 102 103 104

Concentration of rhILl-a (U/ml)

Fig. 3 Dose response to rhTLl-a for induction of ODC in cultured hepatocytes. Isolated mature hepatocytes (10° cells/cm') were cultured in Dulbecco's modified Eagle's MEM supplemented with 10% fetal calf serum for 24 h as described in "MATERIALS AND METHODS." After culture of the cells for an additional 20 h in the same medium without FCS, the indicated concentrations of rhILl - a were added to the culture medium and culture was continued for 4 h. Then the hepatocytes were harvested and their ODC activity was measured. (O) and (•) are values for hepatocytes isolated from different mice.

ODC Activity (pmol C0i-h~'-mg protein"1)

Expt. A 1) rhILl-ff 2)IM.+rhILl-a 3) 10% ethanol 4) I.M. Expt. B 1) rhTNF/cachectin 2) I.M + rhTNF/cachectin 3) 10% ethanol 4) I.M.

748.85 ± 93.56 676.70± 133.00 12 30± 14.65 22.45± 13 35 488.72± 90.53 462.25±186.25 12.30± 14.65 22 47± 23.35


Vol. 109, No. 4, 1991



x g 100

E % o" a u

g 1

cachectin on ODC activity in the liver was also not mediated by eicosanoids (Table I). Next we examined whether rhILl-ar has a direct action on mouse liver to induce liver ODC. Primary cultures of mouse hepatocytes are a simple system and useful to study the direct effect of rhILl-ar on metabolism in liver with minimal interference by other factors. Thus, primary cultures of hepatocytes were used to examine whether rhILl-ar exerts a direct effect on mouse liver. Isolated mature hepatocytes were cultured in Dulbecco's modified Eagle's MEM supplemented with 10% fetal calf serum for 24 h as described under "MATERIALS AND METHODS." The cells were next cultured for an additional 20 h in the same medium without FCS, and then the effect of rhILl-a on ODC of the cells was examined. Figure 3 shows the dose-response relationship for induction of ODC in the cultured hepatocytes by rhILl-ar. Increase in ODC activity was detectable with 0.1 U (4 pg) of rhTLl-or/ml of the culture medium, and was mnyim^l at 10-100 U of rhlLl-or/ml of the medium, then rapidly decreased to the basal level at higher concentrations, suggesting that high concentrations were toxic. Figure 4 shows the time course of ODC induction in cultured hepatocytes on addition of 50




1 2 3 it 5 6 Time after addition (h)

Fig. 4 Time course of ODC induction in cultured mouse hepatocytes. Isolated mature hepatocytes were cultured under the same conditions with and without FCS as for Fig. 3, and then rhlLl-a' was added at 50 U/ml of medium The hepatocytes were harvested at the times indicated and their ODC activity was measured.

U of rhlLl-ff/ml of culture medium. ODC activity reached a maximum about 3 h after addition of rhILl-a, and then gradually decreased. These results suggest that at least rhTTil-ar has a direct action on mouse liver to induce liver ODC. The induction time was rather shorter, and the concentration of rULl-a- required for the induction was much lower in cultured hepatocytes in comparison with those of the in vivo experiment. This discrepancy may be due to the fact that in the in vitro experiment, cultured hepatocytes are able to directly contact rhILl-a molecules

Downloaded from https://academic.oup.com/jb/article-abstract/109/4/507/820937 by University of Otago user on 19 January 2019

a 6u

K. Imamura et al.

510 TABLE II. Effects of cycloheximide and actlnomycin D on induction of mouse liver ODC by rhILl-a. Mice were treated with rhILl-g (2,000 U/mouse) at zero time. In experiment A, cycloheximide (CX, 0.1 mg/g body weight), or in experiment B, actlnomycin D (act-D, 2 //g/body weight) was also injected at the indicated time The animals were killed at 4 h after rhILl - a treatment and their liver ODC activity was assayed Values are means±SDs for four experiments. ODC Activity (pmol COi'h"''ing protein"1)

410.75±101 13 7.54± 2.22 140 95± 40.21

+ CX(2h) + CX(3h)

Expt. B rhILl-a alone + Act-D ( - l h ) + Act-D (1 h) + Act-D (2 h)

410.75 + 101.13 10.17± 2 04 492.91 ± 97.23 573 36±111.35

o °•£ ra % O

* 300 7 *„ o u 200

." 1 "c


o 1










Time after Injection (h) 5


Fig. 5. Northern blot analysis of ornlthlne decarboxylase mKNA after Induction by rhILl-o. Poly(A+) RNA was isolated from the liver 0 (lane 2), 2 (lane 3), 3 (lane 4), 4 (lane 5), 6 (lane 6), 8 (lane 7), and 10 (lane 8) h after administration of 2,000 UrhILl-


In this work deviation of liver metabolism by cytokines, especially recombinant human interleukin 1-alpha (rhIL1-alpha), was investigated. Administrat...
4MB Sizes 0 Downloads 0 Views