Clinical Science (1993)
85, 679-685 (Printed in Great Britain)
679
lndomethacin stimulates basal glucose production in
humans without changes in concentrations of
glucoregulatory hormones
E. P. M. CORSSMIT', J. A. ROMIJN's*E. ENDERT', H. P. SAUERWEIN'*'
Departments of 'Endocrinology and %ternal Medicine, Academic Medical Centre, University of
Amsterdam, Amsterdam, The Netherlands
(Received 8 February/l6 July 1993; accepted
23 July 1993)
~
1. To investigate whether indomethacin affects basal
glucose production, we measured hepatic glucose production in six healthy postabsorptive subjects on two
occasions: once after administration of indomethacin
(150mg orally) and once after administration of
placebo.
2. Glucose production was measured by primed,
continuous infusion of [3-'H]-glucose.
3. Indomethacin administration resulted in an
increase in glucose production from 10.9 (SEM 0.3)
pmol min-' kg-' to a maximum of 16.5 (SEM 1.6)
pmol min-' kg-' (P<0.05) within 1 h, whereas in
the control experiment glucose production declined
gradually (P <0.01) (P <0.05 indomethacin versus
control). There were no differences in plasma concentrations of insulin, C-peptide and counterregulatory hormones between the two experiments.
4. Since indomethacin administration resulted in an
increase in glucose production in the absence of any
changes in concentrations of glucoregulatory hormones, we conclude that indomethacin stimulates
hepatic glucose production through other mechanisms.
-
INTRODUCTION
Hepatic glucose production (HGP) is regulated
by the interplay of several mechanisms. Major
factors involved in the regulation of HGP are levels
of glucoregulatory hormones, substrate concentrations and glucose itself [l-31.
In the liver, there is intensive interaction between
Kupffer cells and hepatocytes, and animal data in
uitro suggest that products of Kupffer cells also
influence HGP. In these studies, it has been shown
that stimulated Kupffer cells produce prostaglandins
[4], which in turn stimulate hepatic glycogenolysis
[S-71. Other Kupffer cell products are cytokines
[4, 81, which also affect glucose production [9-151.
Since Kupffer cells produce both prostaglandins and
cytokines, these cells could be involved in the regulation of HGP in uiuo. Unfortunately, it is hardly
possible to evaluate the paracrine interaction
between Kupffer cells and hepatocytes in uiuo.
Because Kupffer cells and circulating monocytes are
considered to be part of the monocyte-macrophage
system and are derived from the same bone marrow
precursor [161, and because Kupffer cells and circulating monocytes react similarly with respect to
lipopolysaccharide-(LPS) stimulated production of
cytokines after indomethacin [17-19], it is possible
that changes in secretory activity of circulating
monocytes reflect to some extent changes in Kupffer
cell activity. Since indomethacin inhibits prostaglandin synthesis and also modulates cytokine production, it may be useful as a tool to evaluate the
paracrine regulation of glucose production.
Previous studies on the effects of prostaglandins
on glucose metabolism in uiuo have focused on the
effects of prostaglandins on hormone secretion or
hormone-regulated glucose production [20, 213. The
effects of indomethacin on basal glucose production
have not been studied. The effect of indomethacin
on basal glucose concentration has been described
in only one report. In that study plasma glucose
concentration increased after indomethacin administration to humans [22]. The explanation for this
increase remains uncertain, because neither HGP
nor counterregulatory hormones were measured.
To evaluate the effects of indomethacin on glucose production, we conducted a study in six
healthy subjects. We studied glucose production,
glucoregulatory hormones and LPS-stimulated
cytokine production ex uiuo in whole blood in six
postabsorptive healthy subjects after oral administration of indomethacin (150mg) and, on another
occasion, during a control study.
METHODs
Subjects
We studied six healthy female subjects aged 28
(SEM 1) years. Their height was 175 (SEM 1)cm,
weight 63 (SEM 2)kg, percentage body fat 27.5
Key words: cytokina. glucose metabolism, prostaglandins.
Abbreviations: FFA, free fatty acids; HGP, hepatic glucose production; 11-1, interleukid; lL4, interleukid; LPS. lipopolpaccharide;TNF, tumour necrosis factor.
Correspondence: D r E. P. M. Corssmit. Department of Endocrinology (F-5), Academic Medical Centre, Meibergdreef 9, I105 AZ Amsterdam, The Nnherlands.
680
E. P. M. Commit
(SEM 1) 7; (measured by a body impedance analyser: BIA 109 Akern, Florence, Italy) [23] and
body mass index 21 (SEM 1 ) kg/m2. Medical history
and physical examination were completely normal
in all subjects. No subject had a family history of
diabetes mellitus. All subjects used oral contraceptives (sub-50 contraceptives). On both occasions
subjects were studied on day 12 after starting a new
pill strip. Pregnancy was excluded by a pregnancy
test before the studies were performed. No other
medication was used. The subjects were consuming
a weight-maintaining diet of at least 250g of carbohydrates for 3 days before the study. Oral informed
consent was obtained from all subjects. The studies
were approved by the Institutional Ethics and
Isotope Committees.
Study design
Each subject served as her own control and
completed two study protocols separated by an
interval of at least 1 month. On one occasion, the
subjects were studied after taking indomethacin
( 1 50 mg orally; Bufa BV, Uitgeest, the Netherlands)
and on another occasion after taking placebo
(control study). The studies were carried out in
random order. Subjects were studied in the postabsorptive state, after a 14h fast. A 19-gauge catheter was inserted in a forearm vein for infusion of
[3-3H]gIucose. Another 19-gauge catheter was
inserted retrogradely into a wrist vein of the contralateral arm, which was maintained at 65°C in a
thermoregulated plexiglass box (for sampling of
arterialized venous blood) [24]. The catheters were
kept patent by a slow saline drip.
After baseline samples were taken for determination
of
background
activity, a
primed
(120nCi/kg), continuous (1.5nCimin-'kg-') infusion of [3-3H]glucose (Amersham, Den Bosch, The
Netherlands) was started 2 h before indomethacin
was given ( t = - 2 h) and continued throughout the
study. After a 2 h equilibration period, four plasma
samples were collected with intervals of 5 min for
determination of baseline [3-3H]glucose specific
activity.
At t=Oh, after 2 h infusion of [3-3H]glucose,
indomethacin or placebo was administrated. Blood
samples for the measurement of glucose specific
activity and plasma concentrations of insulin, Cpeptide, glucagon, adrenaline, noradrenaline, cortisol and growth hormone were collected simultaneously at regular intervals from the beginning
(t=Oh) until the end of the study ( t = 6 h ) . Samples
for the measurement of plasma glucose concentration were obtained every 15 min throughout the
study. Concentrations of alanine, lactate, glycerol
and free fatty acids (FFA) were measured at the
beginning ( t = 0 h), 45 min after administration of indomethacin/placebo (t=0.75h) and at the end of
the study ( t = 6 h ) . Blood samples for measurement
of LPS-stimulated production of tumour necrosis
et
al.
factor (TNF), interleukin-1 (IL-1) and interleukin-6
(IL-6) ex uiuo were obtained at the beginning
(t=Oh), 0.5, 0.75, 1.25, 1.75, 2.75 and 6 h after
indomethacin/placebo administration.
Assays
All measurements were performed in duplicate
and all samples from each individual subject were
analysed in the same run. Plasma glucose was
measured by the glucose oxidase method (Beckman
Glucose Analyzer, Beckman Instruments Inc.,
Mijdrecht, The Netherlands). Plasma glucose specific activity was measured as described elsewhere
[25, 261. Plasma insulin was measured by a commercial r.i.a. (Pharmacia Diagnostics, Uppsala,
Sweden), plasma C-peptide was measured by a commercial r i a . (RIA-Matt, Malinckrodt Diagnostica,
Dietzenbach, Germany), plasma glucagon by an
r i a . (Daiichi Radioisotope Labs, Tokyo, Japan;
glucagon antiserum elicited in guinea pigs against
pancreatic specific glucagon; cross-reactivity with
glucagon-like substances of intestinal origin less
than l%), plasma catecholamines by h.p.1.c. and
electrochemical detection, after purification on
Biorex 70 and concentration by solvent extraction
[27], plasma cortisol by fluorescence polarization
immunoassay on technical device X (Abbott
Laboratories, Chicago, IL, U.S.A.) and plasma
growth hormone by an immunoradiometric assay
(Nichols Institute, Los Angeles, CA, U.S.A.). Plasma
alanine was determined by an amino acid analyser
(Chromocon 500, Kontron, Italy), and plasma lactate and glycerol by enzymic methods (Boehringer
Mannhein, Almere, The Netherlands) on a Cobas
Bio Centrifugal analyser. Serum FFA were measured by an enzymic method (NEFAC; Wako
Chemicals GmbH, Neuss, Germany).
Cytokine production ex uiuo was measured in
whole blood, to avoid extensive manipulation with
cells involved with the separation of monocytes.
Production of cytokines in whole blood was corrected for the number of monocytes, the main cells
producing cytokines in blood [28]. Leucocyte
counts and leucocyte differentiation were determined
in blood anticoagulated with EDTA (potassium salt)
with the use of a flow cytometer (Technicon H I
system; Technicon Instruments, Tarrytown, NY,
U.S.A.). Blood ( 1 5 ml) for determination of cytokine
production ex uiuo was obtained by direct venepuncture in sterile syringes and immediately transferred to 50 ml tubes containing 750 i.u. of heparin
(Organon Teknika, Boxtel, The Netherlands) and
45 pg of aprotinin (Boehringer Mannheim, Almere,
The Netherlands). After gentle mixing the blood was
divided between three tubes. Tube A (unstimulated
cytokine concentration) containing loop1 of PBS
(buffered NaCI, pH 7.4) was centrifuged immediately
and the plasma was frozen at -70°C. Tube B
(control) containing loop1 of PBS was incubated at
37°C for 22 h, and after centrifugation the plasma
lndomethacin stimulates glucose production
681
Indomethacin
(IMmg)
was stored at -70°C. Tube C containing 1OOpl
(50pg) of LPS (Escherichia coli K-235, Sigma
Chemical Company, St Louis, MO, U.S.A.; 1Opg of
LPS/ml of whole blood) was treated identically with
tube B.
TNF was determined by e.1.i.s.a. [29] (antibodies
kindly donated by W. A. Buurman) (detection level
in our laboratory 30pg/ml), IL-1 by a commercial
immunoradiometric assay (Medgenix, Brussels,
Belgium) (detection level 50pg/ml) and IL-6 by a
commercial e.1.i.s.a. (CLB, Amsterdam, The
Netherlands) (detection level 50 pg/ml).
Calculations and statistics
Glucose output was calculated by the steady-state
(baseline samples) and by non-steady state equations of Steele [30] in their derivative form. The
effective volume of distribution for glucose was
assumed to be 165ml/kg. The results are presented
as means & SEM. Data within and between experiments were analysed by analysis of variance and
Fishers least significant difference test for multiple
comparison, as indicated. Statisticai significance was
set at P<O.O5.
i
I
RESULTS
*
1
Glucose metabolism
.
Baseline plasma glucose concentration and HGP
were not different between the two studies. In the
control experiment plasma glucose concentration
and HGP declined gradually in all subjects from 4.6
(SEM 0.1) to 4.1 (SEM O.l)mmol/l (or by 11%)
(P<O.Ol) and from 2.1 (SEM 0.1) to 1.7 (SEM
0.02)pmolmin-'kg-' (or by 18%) (P<O.Ol), respectively (Fig. 1).
In the indomethacin experiment, plasma glucose
concentration and H G P increased transiently in all
subjects (P<O.O5) from 4.8 (SEM 0.2) to a maximum of 6.3 (SEM 0.3)mmol/l (or by 31%) (P<O.O5
versus control) and from 10.9 (SEM 0.3) to a
maximum of 16.5 (SEM 1.6)pmolmin-'kg-' (or by
51%) (P<0.05 versus control), respectively after 1.6
(SEM 0.2) h and 1.1 (SEM 0.2)h, respectively.
Plasma glucose concentration subsequently returned
to baseline values; HGP initially decreased to below
control values (P <0.05 versus control) and subsequently returned to baseline values.
control). Plasma concentrations of glucagon, adrenaline and noradrenaline did not change and were
not different between the two studies. Plasma cortisol concentration declined in both experiments
(P < 0.01) (not significant, indomethacin versus
control). Plasma growth hormone concentration
increased from 1.8 (SEM 0.6) to 11.5 (SEM 4.4) pg/l
(P<O.O5) in the control experiment and from 1.6
(SEM 0.5) to 13.7 (SEM 2.7)pg/l (P<0.05) in the
indomethacin experiment (not significant, indomethacin versus control).
Hormones
Substrates
Baseline plasma concentrations of insulin, Cpeptide and counterregulatory hormones were not
different between the two studies (Figs 2 and 3).
Plasma concentrations of insulin did not change
and were not different between the two studies. In
both experiments, plasma concentrations of Cpeptide declined throughout the experiment
(P <0.05) (not significant, indomethacin versus
Baseline plasma concentrations of substrates and
FFA were not different between the two studies
(Table 1). Plasma alanine concentrations declined in
both experiments (P<0.05) (not significant, indomethacin versus control experiment). Plasma concentrations of lactate and glycerol did not change
and were not different between the two experiments.
Although serum FFA concentrations increased in
'0
I
0
i
2
3
Time (h)
4
5
6
Fig. I. Eficctr of indomethacin ( 0 )and of placebo (0)
on plasma
glucose concentration and HGP. Statistical significance: *P <0.05
compared with placebo. Values are means SEM.
+
E. P. M. Commit et al.
602
P
ol,
0'1
1
0
I
2
3
4
5
6
Time (h)
Fig. 2. Effects of indomethacin ( 0 )and of placebo (0)
on plasma
insulin and C-peptide concentrations. Values are means fSEM.
loo0 I
c
the control experiment (P<0.05) and did not
change in the indomethacin experiment, there was
no significant difference between the two
experiments.
R
=
Cytokines
Leucocyte and monocyte counts were not different between the two studies. Cytokine concentrations in supernatants of unstimulated whole
blood (tubes A and B) were below detection levels
in both control and indomethacin experiments.
In the control experiment, LPS-stimulated T N F
and IL-6 production did not change; IL-1 production declined by 26% after 6 h (P<O.O5) (Fig. 4).
Although after indomethacin administration, LPSstimulated TNF, IL-1 and IL-6 production all
increased transiently ( P < 0.05), only LPS-stimulated
TNF production was significantly higher after indomethacin administration ( P < 0.05).
DISCUSSION
Indomethacin administration resulted in a transient increase in HGP of -51% above basal values.
This stimulatory effect of indomethacin on HGP
was not associated with any changes in the plasma
concentrations of insulin, C-peptide or other glucoregulatory hormones. Therefore, from the data
obtained in our study, we conclude that indomethacin stimulates H G P through other mechanisms than by changes in secretion of glucoregulatory hormones.
Theoretically, the possibility cannot be excluded
that indomethacin affects the volume of distribution
250 I
2o
E
R
E
1
5
0
0
1
2
3
4
5
6
Time (h)
Fig. 3. Effects of indomethacin ( 0 )and of placebo (0)
on plumr
concentrations of gluugon, adrenaline, noradrenaline, cortisol and
growth hormone. Values are means SEM.
+
of [3-3H]glucose, however, since this would affect
tracer and tracee similarly, this would not influence
glucose specific activity and, as a consequence, the
calculation of HGP. Although indomethacin may
reduce splanchnic blood flow [31, 321, no relationship has been shown between physiological changes
in splanchnic blood flow and HGP. Even when
indomethacin would reduce splanchnic blood flow it
seems rather unlikely that a decrease in blood flow
stimulates HGP.
The data on the effects of indomethacin on
peripheral glucose uptake are conflicting. Some
studies documented an inhibitory effect of indomethacin on insulin-stimulated glucose uptake in
lndomethacin stimulates gluccse production
Table 1. Plasma concentrations of alanine, lactate, glycerol and FFA
at t=Oh, and 0.75h and 6h after indomethacin or placebo
administration. Valuer are means f SEM. Statistical significance: *P ~ 0 . 0 5
compared with time Oh.
Time (h) ... 0
0.75
6
230f16
257f18
221 f 1 7
241+26
182f12*
201fIF
Lactate concn. (mmol/l)
Control
lndomethrin
0.66 f 0.06
0.58 & 0.09
0.66 f 0.04
0.61 fO.08
0.59 kO.05
0.56 f 0.07
Glycerol concn. (mmol/l)
Control
lndomethrin
0.10 f 0.01
0.10~0.01
0.08 f 0.01
0.22f 0.03
0.21 f 0.02
0.28 f 0.06
Alanine concn. (pnol/l)
Control
lndomethrin
FFA concn. (g/l)
Control
indomethrin
0. I9 f 0.02
loo
0
1
683
;
3,
1
1
0.12 f0.02
0.13k0.02 0.11 fO.01
0.30 fO.O5*
0.24 f 0.01
muscle/adipose tissue [33, 341, whereas other authors reported no effect of indomethacin on insulinstimulated glucose uptake [35]. Consequently, the
possibility that the rise in plasma glucose concentration is also the result of a decrease in peripheral uptake in addition to a change in hepatic
glucose production cannot be excluded.
After the increase in HGP in the indomethacin
experiment, HGP initially decreased to below
control values. This decrease in HGP seems to be
the result of an autoregulatory effect of plasma
glucose concentration on HGP [3], because insulin
secretion, reflected in insulin concentrations and
C-peptide concentrations, did not increase, even
though the plasma glucose concentration was
elevated. The fact that insulin secretion did not
increase despite an increase in plasma glucose concentration could be the result of an inhibitory effect
of indomethacin on glucose-stimulated insulin secretion [36].
The dose of indomethacin used in our experiments has been shown to reduce tissue concentrations of prostaglandin synthetase, and to
decrease plasma and urinary metabolic products of
prostaglandin El and prostaglandin E, in animals
and man [37]. To evaluate the effects of prostaglandins on HGP, infusion of prostaglandins in
humans may be a more direct approach than the
administration of an inhibitor of prostaglandin synthesis. However, we choose not to infuse prostaglandins of the E-series for two reasons. First, since
prostaglandins are rapidly metabolized in the lung,
infusions of these agents may not result in the
desired concentration at the tissue level [38].
Secondly, prostaglandin E, infusion in dogs has
been shown to decrease gut blood flow, which might
alter glucose kinetics [39].
In the present study, we administered indomethacin, a prostaglandin synthesis inhibitor. This
0
1
i
1
1,
O
1
1
2
3
4
5
6
Time (h)
Fig. 4. Effects of indomethacin ( 0 ) and of placebo ( 0 )on
LPSrtimulated cytokim production (corrected for monoqte
number). Statistical significance: *P <0.05 compared with placebo. Values
are mernsfSEM.
resulted in an increase in HGP. Conversely, Jauch
et al. [40] found an inhibitory effect of bradykinin
infusion (which stimulates tissue prostaglandin
synthesis) on HGP in humans in the absence of
changes in insulin or glucagon [40]. These data are
opposite to those obtained in uitro, where prostaglandins stimulate hepatic glycogenolysis [5-71. The
discrepancy of effects of prostaglandins in uitro and
modulation of prostaglandin synthesis by indomethacin or, as in the study of Jauch et a1 [40], by
bradykinin in v i m , could be the result of other
effects of indomethacin and bradykinin besides inhibition and stimulation, respectively, of prostaglandin
synthesis. Although indomethacin also impairs both
the cyclic AMP-dependent and independent forms
of protein kinase C41-431 and since both glucagon
and adrenaline mediate HGP in part by activation
684
E. P. M. Commit et al
of the adenylate cyclase-cyclic AMP complex [44],
indomethacin could be able to interfere with the
cyclic AMP response to these hormones. Since this
results in inhibition of glucagon- and adrenalinestimulated glucose production rather than in augmentation of HGP, this mechanism can not explain
the effects observed in the present study. Insulin
also exerts some of its effects by increasing protein
kinase C after binding to its receptor. However, this
activation was only found in myocytes [45] and not
in hepatocytes [46]. In contrast, however, Miller
et al. [47] observed impaired insulin-induced suppression of H G P by indomethacin in dogs. Therefore, we can not exclude the possibility that indomethacin decreases hepatic insulin sensitivity resulting in increased HGP, although the magnitude of
the response makes this possibility not very likely.
In this study data were obtained in women on
oral contraceptives. Theoretically, the use of these
agents could have influenced our results. However,
since every subject served as her own control, using
oral contraceptives during both studies, the difference between both studies can not be explained by
the use of oral contraceptives. Moreover, we have
shown that there are no differences in basal glucose
production during 22 h of starvation between men,
women during different phases of the menstrual
cycle and women on oral contraceptives [48].
Besides its effects on prostaglandin synthesis,
indomethacin also modulates cytokine production
[ 17-19], probably indirectly by inhibiting synthesis
of prostaglandins, known for their inhibitory effect
on LPS-stimulated cytokine production [ 17, 18).
Kupffer cells produce several cytokines [4, 81. The
effects of cytokines on glucose production, with
reports of inhibitory and stimulatory effects, are not
well elucidated due to a very complicated interaction of the different cytokines and classical hormones, and species differences [IS-151. In the present
study, we measured production of cytokines by
circulating cells ex uiuo, by methods described
above. Since Kupffer cells and circulating monocytes
react similarly with respect to LPS-stimulated
cytokine production after indomethacin [ 17-19],
monocytes could possibly provide insight into
hepatic paracrine regulation after indomethacin
administration in uiuo. Our data on the production
of cytokines by circulating cells ex uiuo, presumably
mainly monocytes, show that after indomethacin
administration, LPS-stimulated production of T N F
is significantly increased, in line with the results of
Endres et al. [19]. Interestingly, the increase in
LPS-stimulated production of T N F preceded the
rise in HGP. The influence of indomethacin on
HGP could be explained by its inhibition of prostaglandin synthesis, its effects on cytokine production,
especially on T N F production, or a combination of
both.
Since in our study indomethacin increased HGP
independent of glucoregulatory hormones, it is
likely that other, possibly paracrine, factors are
involved. In this respect Kupffer cells are probably
important mediators. These cells produce prostaglandins as well as cytokines after stimulation,
mediators known for their influence on hepatic
glucose metabolism, at least in uitro. However, since
it is hard to evaluate the paracrine interaction of
Kupffer cells and hepatocytes in humans in uiuo, this
remains an interesting speculation.
In conclusion, after indomethacin administration,
HGP increases in the absence of changes in concentrations of glucoregulatory hormones, concomitantly with an increase in LPS-stimulated cytokine
production in whole blood. The increase in H G P
after indomethacin may be related to hepatic paracrine mechanisms, e.g. alterations in hepatic prostaglandin and/or cytokine production.
ACKNOWLEDGMENTS
This study was supported financially by the
Dutch Diabetes Foundation. We are indebted to
the Laboratory of Endocrinology for technical
assistance.
REFERENCES
I . Felig P. Sherwin R. Carbohydrate homeostasis, liver and diabetes. Prog Liver
Dis. 1976 5 149-71.
2. Jahoor F. Peters EJ, Wolfe RR. The relationship between gluconeogenic
substrate supply and glucose production in humans. Am J Physiol 19w; W:
E288-96.
3. Muller MJ, Moring J, Seitz HJ. Regulation of hepatic glucose output by glucose
in vivo. Metab Clin Exp 1988; 37: 55-60,
4. Decker K. Biologically active product of stimulated liver macrophages
(KupHer cells). Eur J Biochem 1990; I9k 24S-61.
5. Casteleiin E. Kuiper J, van Rwij HC. Kamps JA, Koster JF. van Berkel TJ,
Hormonal control of glycogenolysis in parenchymal liver cells by KupHer and
endothelial liver cells. J Biol Chem 1988; 263: 2699-703.
6 . Casteleiin E. Kuiper J, van Rwij HC. K a t e r JF. van Berkel TJ. Conditioned
media of Kupffer and endothelial liver cells influence protein phosphorylation
in parenchymal liver cells. Involvement of prostaglandins. Biochem J 1988; Zst:
601-5
7 Kuiper J, Ziilstra FJ. Kamps ]A, van Berkel TJ, Cellular communication inside
the liver. Binding conversion and metabolic eflectr of prostaglandin D, on
parenchymal liver cells. Biochem J 1989; 261: 19W.01.
8 Magilavy DB. Rothstein JL. Spontaneous production of tumor necrosis factor
z by KupHer cells of MRL/lpr mice. J Exp Med 1988; I& 789-94.
9. Van der Poll T. Romiin ]A. Endert E, Borm JJJ, Buller HR. Sauerwein HP.
Tumor necrosis factor mimics the metabolic response to acute infection in
healthy humans. Am J Physiol 1991; 261: E457-65.
10. Role AM, Conyers RAJ. Bais R. Gamble JR. Vadas MA. The eHects of
recombinant tumour necrosis factor (cachectin) on metabolism in isolated rat
adipocyte. hepatocyte and muscle preparations. Biochem J 1987; 247: 789-92.
I I. Lang CH, Dobrescu C, Bagby GJ. Tumor necrosis factor impairs insulin action
on peripheral glucose disporal and hepatic glucose output. Am J Physiol 1992;
2% E57-64).
12. Besedovsky HO. del Rey A. Interleukin-I and glucose homeostasis: an
example of the biological relevance of immune-neuroendocrine interactions.
Horm Res 1989; 31: 949.
13. Hill MR, Stith RD. McCallum RE. lnterleukin I: a regulatory role in
glucocorticoid-regulated hepatic metabolism. J lmmunol 1986; 137: 858-62.
lndomethacin stimulates glucose production
14. Vurtjes WJ, de Huc CGM, Houweling M. Acute effects of interleukin la and
6 on intermediary metabolism in freshly isolated rat hepatocytes. Biochem
Biophys Res Commun 1990; I* 623-8.
IS. Ritchie DG. lnterleukin 6 stimulates hepatic glucose release from prelabeled
glycogen pools. Am ] Physiol 1990; W: E57-64.
16. Gale RP. Sparker RS, Golde DW. Bone marrow origin of hepatic macrophages
(Kupffer cells) in humans. Science (Washington D.C.) 1987; II:937-8.
17. Shirahama M. Ishibushi H, Tsuchiya Y. et al. Kupffer cells may autoregulate
interleukin I production by producing interleukin I inhibitor and
prostaglandin E.l Sund J lmmunol 1988; Uk 719-25.
18. Gllery MP. Mangino MJ, Kamei T, Flye MW. Interleukin-6 production by
endotoxin-stimulated Kupffer cells is regulated by prostaglandin E,. J Surg Res
1990; Q 523-7.
19. Endrcr S, Cannon JG, Ghorbani R. In vitro production of IL I/?, IL la,TNF
and IL 1 in healthy subjects: distribution, effect of cyclwxygenase inhibition
and evidence of independent gene regulation. Eur J lmmunol 1983; I9
2327-33.
20. Bnrr EP, Garrity MJ. Effect of E-series prostaglandins on cyclic
AMPdependent and independent hormon-s’imulated glycogenolysis in
hepatocytes. Diabetes 1985; 34 2914.
21. Levine RA, Schwamel EH, Jr. Prostaglandin El inhibition of glucagoninduced
hepatic gluconeogenesis and cyclic adenosine Y,S’monophosphate
accumulation. Biochem Pharmacol 1980; 29: 6814.
22. SyvPahti EKG. The effect of indomethacin on serum growth hormone,
immunoreactive insulin and blood glucose levels of young adult males. lnt J
Clin Pharmacol 1974; 10: I 11-16.
U. Lukaski HC, Bolonchuk WW, Hall C8, Siders WA. Validation of tetrapolar
bioelectrid impedance method t o asses human body composition. J Appl
Physiol 1986, 6& 1327-32.
24. McGuire EAH, Helderman JH. Tobin ID. Andres R, Berman M. Effects of
arterial versus venous sampling on analysis of glucose kinetics in man. J Appl
Physiol 1976 41: 565-73.
25. Romijn ]A. Godfried MH, Hommes MJT, Endert E, Sauerwein HP. Decreased
glucose oxidation during short-term starvation. Metab Clin Exp 1990; 39:
525-30.
31. Konturek SJ, Bnozowrki T, Majka J. Drozdowin 0, Stachura ). Adaptation d
the gastric mucosa to stress. Role of prostaglandin and epidermal growth
hctor. Scand J Gastroenterol 1992; 193 (Suppl.): 39-45.
32. Bruix 1, Bosch J, Knvetz D, Mastai R, Rods J. Effects of prostaglandin
on on systemic and hepatic hemodynamics in patients with cirrhosis of
the liver. Gastroenterology 1985. 8& 430-5.
33. Richelsen B, Hjollund E, Pedenen 0, Sorensen NS. Effects of prostaglandin
E.l indomethacin and adenosine deaminase on basal and insulin-stimulated
glucose metabolism in human adipocytes. Biochim Biophys Acta 1985;
35%.
34. Segal S, Blair A, Weinberg A. In vitro effects of salicylate on carbohydrate
metabolism. Metab Clin Exp 1960; 9 1033-46.
35. Zonano A, Balon TW, Jakubowski JA, Goodman MN, Deykin D, Duderman
NB. Effects of insulin and prior exercise on prostaglandin release from
perfused rat muscle. Biochem J 1986: UD: 43743.
36. Widstrom A. Influence of indomethacin on glucose-induced insulin response in
normal mac-role of prostaglandins in the rapid insulin release? Horm Metab
R ~ S1 9 n ;9: 171-5.
37. Flower RJ. Drugs which inhibit prostaglandin biorynthesis. Pharmacol Rev
1974; 1:33-67,
38. Ferreira SH, Vane JR. Prostaglandins: their disappearance from and release
into the circulation. Nature (London) 1967; 216: 868-73.
39. Levebvre PJ. Luyckx AS. Stimulation of insulin secretion after prostaglandin
PGE, in the anesthetized dog. Biochem Pharmacol 1973; B 1773-9.
40. Jauch KW, Hart1 WH, Georgieff M, Wolfe RR. Diene GJ. Gunther B.
Lowdose bradykinin infusion reduces endogenous glucose production in
surgical patients. Metab Clin Exp 1988; n: 185-90.
41. Goueli SA. Ahmed K. lndomethacin and inhibition of protein kinase reactions.
Nature (London) 1980, M: 171-1.
41. Kantor HS, Hampton M. lndomethacin in submicromolar concentrations
inhibits cyclic AMPdependent protein kinase. Nature (London) 1987; 276:
841-2.
43. Ganguli S. Sterman 8, Harris PS. Sinha MK, Banach WJ. Modulation of hepatic
protein kinase activity by indomethacin. Clin Exp 1984; 3 3 845-51.
44. Bitensky MW. Russell V, Robertson W. Evidence for separate epinephrine and
26. Romijn ]A. Endert E. Sauerwein HP. Glucose and fat metabolism during
short-term starvation in cirrhosis. Gastroenterology 1991; 100: 731-7.
17. Smedes F, K m k JC. Poppe H. Sample and fast solvent extraction system for
selective and quantitative isolation of adrenaline, noradrenaline and dopamine
from plasma and urine. J Chromatogr 1982; UI: 25-39.
28. Allen IN, Henyk DJ. Allen ED, Wevers MD. Human whole blood
interleukiwl-/? production: kinetics, cell source, and comparison with TNFu.
] Lab Clin Med 1992 119: 538-46.
29. Engelberts I, Moller A, Schoen GJM, van der Linden CJ. Buurman WA.
Evaluation of measurement of human TNF in plasma by ELISA. Lymphokine
Cytokine Res 1991; 10: 69-76.
30. Stele R. Influences of glucose loading and of injected insulin on hepatic
glucose output. Ann NY Acad Sci 1959;
420-30.
685
45.
46.
47.
48.
glucagon responsive adenyl cyclase systems in rat liver. Biochem Biophys Res
Commun 1968; 31: 706-12.
Cooper DR. Konda TS. Standaert ML, Davis JS, Pollet RJ. Farese RV. Insulin
increases membrane and cytosolic protein kinase C activity in BC3H-I
myocytes. J Biol Chem 1987; 262: 3633-9.
Taylor D, Uhing RJ. Blackmore PF, Prpic V, Exton JH. Insulin and epidermal
growth factor do not affect phosphoinositide metabolism in rat liver
membranes and hepatocytes. J 8iol Chem 1985; W. 101 1-14.
Miller ID, Ganguli S, Sperling MA. lndomethacin and salicylate modulate effect
of insulin on glucose kinetics in dogs. Am ] Physiol 1985; ulk E648-55.
Stouthard JML, Corssmit EPM, Romijn JA, Endert E, Sauerwein HP.
Sexdifferences in adaptation t o short-term starvation. Neth J Med 1993; 41:
A9H.