Hypothyroidism Minimizes Liver Damage and Improves Survival
in Rats With Thioacetamide Induced Fulminant Hepatic Failure
RAFAEL BRUCK,1 RAN OREN,1 HAIM SHIRIN,1 HUSSEIN AEED,1 MOSHE PAPA,1 ZIPORA MATAS,2 LILIANA ZAIDEL,3
YONA AVNI,1 AND ZAMIR HALPERN1
Recent data from animal studies suggest that induced
hypothyroidism prevents the hyperdynamic circulation in
portal vein ligated rats, liver cirrhosis in rats chronically
treated with thioacetamide (TAA), and immune-mediated
acute liver injury induced in mice by concanavalin A.
Therefore, the aim of this present study is to determine
whether hypothyroidism would likewise prevent fulminant
hepatic failure (FHF) in rats. FHF was induced by 3
consecutive ip injections of TAA (400 mg/kg) at 24-hour
intervals. Hypothyroidism was induced in rats by either
methimazole (MMI) or propylthiouracil (PTU) and surgical
thyroidectomy and was confirmed by elevated serum thy-
roid stimulating hormone levels. Serum levels of liver
enzymes, blood ammonia, and prothrombin time were
significantly lower in all 3 groups of hypothyroid rats. The
stage of hepatic encephalopathy (HE) and the survival rates
were significantly improved in the hypothyroid rats (P F
.01); the histologic examination of their livers showed less
necrosis and inflammation (P F .01). In the hypothyroid
rats, the serum levels of malondialdehyde 48 hours after
thioacetamide (TAA) administration were lower than in
control rats (P F .01). Exogenous supplementation of
hypothyroid rats with L-thyroxine started 48 hours before
TAA administration abrogated the protective effects of
hypothyroidism. The serum levels of tumor necrosis factor
alfa (TNF-a), interleukin (IL) 2 and IL-6 after 24 hours
were slightly lower in the hypothyroid rats, but the adminis-
tration of soluble receptor of TNF (10-1,000 mg/rat) did not
prevent the induction of fulminant liver failure by TAA.
Oxygen extraction, studied in isolated perfused liver prepa-
ration, was significantly lower in livers of hypothyroid rats
(P F .01). These results suggest that induced hypothyroid-
ism decreases the development of liver injury in a rat model
of FHF. The mechanism may involve diminished oxidative
cell injury caused by decreased oxygen utilization and
hypometabolism associated with hypothyroidism. (HEPATOL-
OGY 1998;27:1013-1020.)
Abbreviations: FHF, fulminant hepatic failure; HE, hepatic encephalopathy; TAA,
thioacetamide; TNF-a, tumor necrosis factor alfa; IL, interleukin; MMI, methimazole;
PTU, propylthiouracil; s-TNF-R, soluble tumor necrosis factor receptor.
From the 1Departments of Gastroenterology, 2Biochemistry, and 3Pathology, The E.
Wolfson Medical Center, Holon, Israel.
Received February 6, 1997; accepted November 20, 1997.
Presented at the Digestive Disease Week held in San Francisco, California, May 18-24,
1996, and was published in an abstract form: Gastroenterology 1966;110:1159A.
Address reprint requests to: Rafael Bruck, M.D., Department of Gastroenterology, The
E. Wolfson Medical Center, Holon 58100, Israel. Fax: 972-3-5035111.
Copyright r 1998 by the American Association for the Study of Liver Diseases.
0270-9139/98/2704-0017$3.00/0
1013
Several lines of evidence suggest that thyroid status may
affect the induction and clinical course of both animals and
humans with various liver diseases. Drugs used for the
treatment of portal hypertension, such as beta adrenergic
blocking agents, have also proved useful in controlling the
cardiovascular manifestations of thyrotoxicosis. Moreover,
propylthiouracil (PTU), a commonly used drug for the
treatment of hyperthyroidism, was proposed for the manage-
ment of patients with alcoholic liver disease.1 Data from
recent studies suggest that induced hypothyroidism prevents
the development of liver injury in several animal models. In a
rat model of portal vein ligation, hypothyroidism caused
amelioration of the hyperdynamic circulation followed by
reduction of the portal pressure.2 Hypothyroidism, induced
either medically or surgically, prevented liver cirrhosis in rats
chronically-treated with thioacetamide (TAA),3 and immune-
mediated T cell-dependent acute liver injury in mice induced
by the lectin concanavalin A.4
Fulminant hepatic failure (FHF) is a rare but severe
complication of acute hepatitis. FHF is characterized by
massive hepatic necrosis and encephalopathy and carries a
very high mortality. Viral hepatitis, drugs, and hepatotoxic
chemical-induced liver injury account for most cases of
FHF.5,6 Although a wide variety of medical therapies, such as
benzodiazepine antagonists,7 L-dopa and branched chain
amino acids,8 and prostaglandin E1,9 as well as extracorporeal
perfusion techniques,10 have been used for the management
of this ominous condition, very few therapies have been
evaluated in controlled clinical trials.11,12 The only treatment
of proven efficacy for those patients is emergency liver
transplantation.13,14
Recently, a rat model of TAA-induced FHF has been
described. Following 2 to 3 consecutive doses of TAA, rats
develop FHF characterized by massive liver necrosis, rapid
neurologic deterioration, and death caused by severe encepha-
lopathy and brain edema.15,16
In the present study, we demonstrate that hypothyroidism
induced either pharmacologically or by surgical thyroidec-
tomy, inhibited the development of TAA-induced fulminant
liver failure in rats.
MATERIALS AND METHODS
Materials
Animals. Male Wistar rats (range, 250-300 g), obtained from
Tel-Aviv University Animal Breeding Center (Tel Aviv, Israel), were
kept in the animal breeding house of the Wolfson Medical Center
and fed a Purina chow ad libitum. Animals were kept in a 12-hour
light-dark cycle at constant temperature and humidity.
1014 BRUCK ET AL.
Methods
Induction of Hypothyroidism. Hypothyroidism was induced by the
administration of either methimazole (MMI) 0.04% (Taro, Herzlia,
Israel) or propylthiouracil (PTU) 0.05% (Teva, Netanya, Israel) in
drinking water for 3 weeks. Surgical thyroidectomy was performed
with the animals under chloral hydrate anesthesia (400 mg/kg) 3
weeks before the induction of FHF. To confirm that correction of
hypothyroidism would reverse its protective effects on the liver, 3
groups of hypothyroid rats (induced by PTU, MMI, or thyroidec-
tomy) were supplemented with thyroxine (eltroxin, Glaxo, C.T.S.,
Petah Tikva, Israel) 5 µg/day by gavage, started 24 hours before TAA
and continued during the 3 days of the study. All rats had free access
to tap water during the week before the beginning of the study.
Induction of Fulminant Hepatic Failure. For induction of FHF, rats
were given ip injections of TAA (Sigma Chemical Co., St. Louis,
MO), 400 mg/kg, three times with a 24-hour interval, as previously
described.15,16 Control rats were treated with ip injections of NaCl
0.9%. Supportive therapy by subcutaneous administration of 5%
dextrose (25 mL/kg) and NaCl 0.9% with potassium (20 mEq/L)
every 12 hours were administered to avoid weight loss, hypoglyce-
mia, and renal failure, as previously described.17
Evaluation of Liver Injury. Four hours following the third TAA
injection, blood samples were drawn for analysis of aminotransfer-
ase levels, serum glucose and bilirubin, prothrombin time, Interna-
tional Neutralization Ratio, and blood ammonia to evaluate the
degree of liver failure. Commercial enzyme-linked immunosorbant
assay kits were used according to the manufacturer’s specifications
to determine the serum levels of tumor necrosis factor alfa (TNF-a),
interleukin (IL)-2, and interleukin (IL)-6 (Genzyme Corp., Cam-
bridge, MA). The concentrations of malondialdehyde were mea-
sured as previously described.18,19
Effect of Soluble Receptor of TNF-a. Recombinant human soluble
TNF receptor (sTNF-R; Interpharm, Israel), was produced in
chinese hamster ovaries cells and purified by immunoaffinity
column, using monoclonal antibodies to the sTNF-R1. Purity .95%
was verified by sodium dodecyl sulfate-polyacrylamide gel electro-
phoresis and size exclusion high-performance liquid chromatogra-
phy. sTNF-R at TNF:sTNF-R molar ratio of 1:103, 1:104, and 1:105
(10, 100, and 1,000 µg/rat), was administered subcutaneously to
each rat 16 hours and 1 hour before each of the TAA treatments and
24 hours afterwards, as previously described.20
Hepatic Encephalopathy and Survival. The stage of hepatic encepha-
lopathy (HE) and survival of the rats in the different treatment
groups were determined 4 hours after the third TAA injection. The
stage of HE was determined by the following neurobehavioral scale:
stage 1, lethargy; stage 2, mild ataxia; stage 3, lack of spontaneous
movement and loss of righting reflex, but still responsive; and stage
4, coma and lack of response to pain.21
For liver histopathology analysis, midsections of the left lobes of
the liver were processed for light microscopy. This processing consisted
of fixing the specimen in a 5% neutral formol solution, embedding
the specimens in paraffin, slicing sections to 5-µm thick, and staining the
sections with hematoxylin and eosin. The tissue slices were scanned
and scored semiquantitatively by two expert pathologists who were
not aware of sample assignment to experimental groups. The degree
of inflammation and necrosis were expressed as the mean of 10
different fields within each slide that had been classified on a scale of
0 to 3 (normal, 0; mild, 1; moderate, 2; and severe, 3).
Measurement of O2 Extraction in Isolated Perfused Rat Liver. The
surgical procedures were essentially performed as described previ-
ously.22 Briefly, the pancreaticoduodenal branch of the portal vein
was ligated and the bile duct, portal vein, and inferior vena cava
were cannulated under chloral hydrate anesthesia. The liver was
transferred into a heated perfusion chamber maintained at 37°C and
perfused at a constant rate of 40 mL/min with Krebs-Ringer-
Bicarbonate buffer containing 5.5 mmol/L glucose and gassed with
95% O2/5% CO2. With the use of a single pass system, O2 extraction
by the rat liver was measured after 30 minutes of equilibration,
expressed as a percentage and calculated as:
HEPATOLOGY April 1998
O2 inflow 2 O2 outflow
Extraction rate (%) 5 3 100
O2 inflow
Experimental Design. Five groups of rats were studied, as follows:
control groups: 1) normal controls: 3 NaCl 0.9% injections; 2) FHF
controls: 3 TAA injections in 24-hour interval; 3) TAA 1 sTNF-R
(10-1,000 µg/rat); hypothyroid: 4) TAA 1 MMI; 5) TAA 1 PTU; 6)
TAA 1 surgical thyroidectomy; 7) TAA 1 PTU 1 eltroxin 5 µg/day
started 24 hours before TAA; 8) TAA 1 MMI 1 eltroxin given as
described; and 9) TAA 1 ST 1 eltroxin.
Statistical Analysis. The data are presented as the means 6 SEM for
liver enzymes. All other data are presented as means 6 SD. The
significance of differences among different groups was determined
by ANOVA followed by a post-hoc test.
RESULTS
Induction of Hypothyroidism
The induction of hypothyroidism was confirmed by ele-
vated serum levels of TSH. The administration of PTU and
MMI, as well as surgical thyroidectomy, were each followed
by a significant elevation of serum TSH (7.8 6 0.9, 7.6 6 0.8,
and 6.9 6 0.9 µIU/mL, respectively) as compared with the
untreated control groups (0.31 6 0.05 and 0.32 6 0.04
µIU/mL, respectively).
Effect of TAA on Liver Enzymes, Prothrombin Time,
and Blood Ammonia
Rats were injected with 3 doses of TAA at 24-hour intervals
and bled 4 hours after the third injection; the serum levels of
hepatic enzymes, bilirubin and glucose, as well as of prothrom-
bin time and blood ammonia were then analyzed. Severe liver
injury, as manifested by elevation of serum aspartate amino-
transferase, alanine aminotransferase, and lactate dehydroge-
nase levels was observed 24 and 52 hours after TAA adminis-
tration (Table 1 and Fig. 1). The serum levels of bilirubin and
alkaline phosphatase were not significantly elevated in re-
sponse to TAA administration in both the hypothyroid and
control groups (data not shown).
Prothrombin time (and International Neutralization Ratio)
were markedly prolonged in the TAA-treated rats 24 and 52
hours after TAA administration (53.5 6 5.0 and 82.4 6 8.0
seconds, respectively) compared with pretreatment levels
(18.2 6 3.8 seconds) (Fig. 2), although bleeding phenomena
were not observed in the TAA treated rats. Blood ammonia
levels in the TAA treated group was elevated fivefold over
control untreated rats (8.5 6 2.1 vs. 1.7 6 0.2 µg/mL, P ,
.001) (Fig. 3).
Inhibition of TAA-Induced Liver Injury by Hypothyroidism
In all 3 groups of hypothyroid rats, the serum levels of liver
enzymes were significantly decreased; although in the thyroid-
ectomized rats the liver enzymes were slightly higher than in
the animals with the medically induced hypothyroidism
(Table 1 and Fig. 1). The prolonged prothrombin time
observed in the TAA-treated rats was almost fully corrected in
all 3 groups of hypothyroid rats (Fig. 2, P , .001). Blood
ammonia, measured 4 hours after the third TAA injection,
was threefold lower in the hypothyroid rats than in the
euthyroid, TAA-treated rats (Fig. 3, P , .001).
Reversal of Hypothyroidism by Thyroxin Administration
In the 3 groups of rats where the hypothyroidism was
corrected by exogenous supplementation with thyroxin 5
HEPATOLOGY Vol. 27, No. 4, 1998 BRUCK ET AL. 1015

µg/day started 24 hours before the first injection of TAA, the

liver enzymes and blood ammonia were not different than in
control rats treated with TAA only (Table 1; Fig. 3). The
results of this experiment support the hypothesis that the
protective effects of hypothyroidism on the insulted liver
were not caused by drug interactions but rather may be
attributed to consequences of the hypothyroid state.

Serum Cytokine Levels

The serum levels of TNF-a, IL-2, and IL-6 measured 2 and
6 hours after the first injection of TAA were undetectable.
Proinflammatory cytokines were first measurable in the
serum of the TAA-injected rats after 24 hours and were
significantly lower in the hypothyroid TAA-treated group
compared with rats treated with TAA alone (Table 2, P , .05).
At this time point, liver damage was already established as
assessed by the elevated levels of liver enzymes (Table 1 and
Fig. 1). In addition, the levels of all 3 cytokines reached low
levels in all groups, including the control rats. These findings
indicate that the release of those cytokines may be a second-
ary event that occurred in response to the products of cell
lysis, after hepatic damage had already been established.

Effect of Soluble Receptor of TNF-a

To further investigate the role of TNF-a in TAA-induced
FHF, the in vivo protective effects of recombinant prepara-
tions of TNF sTNF-R were assessed in rats in response to TAA
administration. sTNF-R, at a molar range of 1:103, 1:104, and
1:105 to TNF-a (10, 100, and 1,000 µg/rat, respectively, based FIG. 1. Effect of hypothyroidism on serum levels of liver enzymes (A)
on TNF-a serum value measured in mice two hours after alanine aminotransferase and (B) aspartate aminotransferase 24 and 52 hours
concanavalin A inoculation), had no beneficial effects on the after the first TAA injection. Mean 6 SD (n512) in TAA alone; and TAA 1
release of aminotransferases (Table 1), blood ammonia levels MMI and TAA 1 PTU (n55) in the thyroidectomy group. *P , .05; **P ,
.01 compared with TAA alone.
(Fig. 3) or the survival of the TAA-treated rats. Thus, the
administration of sTNF-R, that prevents immune-mediated
hepatic damage in mice in response to concanavalin A23 and
TABLE 1. Effect of Hypothyroidism on Serum Levels of Liver Enzymes in toxic liver injury in rats induced by CCl4,20 had no protective
TAA-Induced FHF effect in this model of TAA-induced FHF.
AST ALT LDH Serum and Hepatic Levels of Malondialdehyde
n Hours (IU/L) (IU/L) (IU/L)
Fifty-two hours after the first TAA injection, the serum
Normal 5 52 160 6 22 45 6 6 1476 6 182
malondialdehyde levels in the rats treated with TAA only
TAA alone 6 24 3780 6 425 1540 6 220 15,305 6 2780
TAA alone 12 52 6176 6 307 3260 6 396 17,033 6 1833
increased from 0.83 6 0.09 to 3.06 6 0.6 nmol/mL compared
TAA 1 MMI 12 52 2008 6 270** 1669 6 166** 4868 6 699* with 1.86 6 0.4 and 1.77 6 0.5 nmol/mL in the MMI and the
TAA 1 PTU 12 52 1595 6 286** 1185 6 159** 4176 6 481** PTU treated rats, respectively, P , .01 (data not shown).
TAA 1 ST 5 52 3972 6 267** 2213 6 119* 9828 6 1280* Likewise, in the TAA-treated hypothyroid rats hepatic MDA
TAA 1 PTU levels were significantly lower than in the euthyroid rats
1 Elt 3 52 5017 6 408 1503 6 174 13,137 6 1327 (17.5 6 3.5 and 16.7 6 4.2 nmol/g wet tissue in the MMI and
TAA 1 ST 1 Elt 3 52 5293 6 895 2560 6 306 13,310 6 1875 PTU groups vs. 32 6 5.8 nmol/g wet tissue in the livers of rats
TAA 1 sTNF-R treated with TAA only, P , .01), and the administration of
100 µg/rat 5 52 5150 6 830 2920 6 344 15,560 6 3840 sTNF-R had no effect on the increased hepatic levels of
1000 µg/rat 5 52 7220 6 1310 2970 6 465 14,864 6 3584
malondialdehyde in TAA treated rats.
NOTE. Means 6 SEM. Hepatic enzymes determined after 3 ip injections of
400 mg/kg TAA in 24-hour intervals, 52 hours after the first administration. HE
TAA alone, liver enzymes were determined 24 hours after a single ip injection Four hours after the third injection, all rats in the
of 400 mg/kg TAA. L-Thyroxin 5 µg/day, administered to correct hypothyroid- TAA-treated group were in stage 3 to 4 HE.21 The level of HE
ism, started 24 hours before the induction of FHF. Note that serum in all groups of the hypothyroid rats was significantly lower
aminotransferase levels in rats that were pretreated with s-TNF-R, and in
(Table 3).
hypothyroid rats that their hypothyroidism was corrected by supplementa-
tion with exogenous thyroxine, is not different from TAA-treated control Survival
rats.
Abbreviations: ST, surgical thyroidectomy; Elt, L-thyroxin. Short-Term. To determine the effect of induced hypothyroid-
*P 5 .001 compared to TAA. ism on the survival of rats with TAA-induced FHF, control
**P , .001 compared with TAA alone (52 h). and hypothyroid rats (PTU, MMI, and surgical thyroidec-
1016 BRUCK ET AL. HEPATOLOGY April 1998

TABLE 2. Effect of Hypothyroidism on Cytokine Release

in TAA-Induced FHF
TNF-a (pg/mL) IL-2 (pg/mL) IL-6 (pg/mL)
Hours
After TAA Cont MMI Cont MMI Cont MMI

2 0 0 0 0 0 0
6 0 0 0 0 0 0
24 2 6 0.4 0.4 6 0.1* 6.8 6 1.1* 0.4 6 0.1* 10 6 1.6 2 6 0.4*
48 0 0 0 0 0 0

NOTE. Mean 6 SD (n 5 4). TAA 400 mg/kg injected ip. Cytokine levels in
normal untreated rats were undetectable in all time points (not shown in the
table).
*P , .01 compared with control.

FIG. 2. Effect of hypothyroidism on (A) prothrombin time and (B)
International Neutralization Ratio in TAA-induced fulminant liver failure in
24 and 52 hours. The prolonged prothrombin time (and International
Neutralization Ratio) in the TAA-treated euthyroid rats was partially
corrected in all 3 groups of hypothyroid rats. Mean 6 SD (n58) in TAA
alone; and TAA 1 MMI and TAA 1 PTU (n 5 5 ) in the thyroidectomy group.
**P , .01 compared with TAA alone.
tomy) that received 3 injections of 400 mg/kg TAA, were
followed after the last TAA dose. Sevety-two hours after the
third TAA injection only 24% 6 8.9% of 25 control rats (TAA
only) survived, whereas 100% of the hypothyroid rats treated
with PTU and MMI and 80% of ST rats were alive (Table 4).
The survival rate of rats treated with s-TNF-R (30% 6 14.1%)
was not different than that of TAA treated euthyroid rats
(Table 4).
Long-Term. Hypothyroid (MMI and PTU) and 10 control
rats that received 2 doses of 300 mg/kg TAA,15 were followed
for up to 10 days after the induction of FHF. By the end of the
follow-up period, only 20% died in the hypothyroid (one
PTU and one MMI) rats. Liver histology in the surviving rats
10 days after the induction of FHF was normal. In the
TAA-treated euthyroid rats, 70% died during the 72 hours
after the first TAA injection and the rest survived the
follow-up period. The survival rate in this control group is
very similar to the data of a previous study that characterized
the TAA-induced FHF as a model of HE.15 In this study, where
two doses of 300 mg/kg TAA were administered, 77% of the
control rats were dead before 72 hours following the first TAA
dose, while the rest of the rats survived.15
Liver Histopathology. Histopathologic examination of liver
specimens taken 24 and 52 hours after the first TAA injection
showed less necrosis (P , .01) and inflammation (P , .05) in
the livers of the hypothyroid rats compared to control rats
treated with TAA only (Table 5, Fig. 4A-4F). However,
moderate inflammatory changes were present also in the
TAA-treated hypothyroid livers, indicating lesser liver injury
in those rats. These inflammatory changes may be consistent
with the moderate increase of hepatic enzymes and prothrom-
bin time observed in the hypothyroid rats as well.
O2 Extraction by Isolated Perfused Rat Liver. O2 extraction by
isolated perfused rat liver of the hypothyroid rats was
decreased to 39% 6 7% compared with 82% 6 12% in control
euthyroid rats (Fig. 5, P , .001).

TABLE 3. Effect of Hypothyroidism on HE in TAA-Induced FHF

Group Grade of Encephalopathy

TAA alone 3-4

TAA 1 PTU 1 Elt 3-4
TAA 1 ST 1 Elt 3-4
TAA 1 sTNF-R 3-4
TAA 1 PTU 1-2
TAA 1 MMI 1-2
TAA 1 ST 1-2

FIG. 3. Effect of hypothyroidism on blood ammonia in TAA-induced
fulminant liver failure. Blood ammonia was significantly lower in all 3 groups
of hypothyroid TAA-treated rats. Note that in rats that were pretreated with
s-TNF-R, and in hypothyroid rats that were supplemented with L-thyroxin
before TAA administration (TAA 1 ELT), to correct hypothyroidism, the
high blood ammonia is not different from TAA-treated control rats. Mean 6
SD (n58). **P , .001 compared with TAA alone.
NOTE. L-thyroxin 5 µg/day, administered to correct hypothyroidism,
started 24 h before the induction of FHF. HE was evaluated in all treatment
groups 4 hours following the 3rd injection of TAA. n 5 12 in TAA alone;
TAA 1 MMI and TAA 1 PTU; n 5 5 in TAA 1 ST group, TAA 1 sTNF-R
(100 or 1,000 µg/rat); and n 5 3 in TAA 1 PTU 1 Elt, and TAA 1 ST 1 Elt.
Abbreviations: ST, surgical thyroidectomy; Elt, L-thyroxin.
HEPATOLOGY Vol. 27, No. 4, 1998 BRUCK ET AL. 1017

TABLE 4. Effect of Hypothyroidism on Survival in TAA-Induced FHF is somewhat less impressive than in the rats with the
No. TAA Only TAA 1 PTU TAA 1 MMI TAA 1 ST TAA 1 sTNF-R drug-induced hypothyroidism. Therefore, our studies do not
Rats (%) (%) (%) (%) (%) entirely exclude that the beneficial effect of hypothyroidism
5 20 100 100 80 20 on the insulted liver could be augmented by other actions
5 20 100 100 40 induced by the anti-thyroid drugs, such as suppression of the
5 20 100 100 microsomal flavin adenine dinucleotide-containing mo-
5 20 100 100 nooxigenases in the liver by MMI.25 Other effects of anti-
5 40 100 100 80 thyroid drugs, such as alteration of hepatic glutathione
Mean 24 100* 100* 80 30 content or kinetics, should also be considered. It has been
SD 8.9 0 0 0 14.1
demonstrated in a recent study in rats that although TAA
NOTE. Survival was recorded in all treatment groups up to 72 hours administration had no effect on the total hepatic glutathione
following the 3rd injection of TAA. content, it changed the oxidative status of glutathione,
Abbreviations: ST, surgical thyroidectomy. inducing a significant increase in glutathione disulfide levels,
*P , .01. and a glutathione-dependent mechanism has been suggested
as responsible for the protection of S-adenosyl-L-methionine
against TAA hepatotoxicity.27 Nevertheless, in a recent study
DISCUSSION
from our lab, the continuous administration of the glutathi-
The present study was undertaken to examine whether one donor N-acetylcysteine, before and during the 48 hours
hypothyroidism that prevents liver damage in several animal of TAA administration had no beneficial effects on either liver
models could also be protective in a model of FHF induced by function tests or survival of rats with TAA-induced hepatic
TAA. This model was characterized previously by clinical, failure.28 Because acute administration of PTU in rats can
biochemical, and histologic methods, and it proved to be a increase portal blood flow, independent of its effect on
reliable and satisfactory model of FHF and HE.15,16,24 Hypothy- thyroid function,29 the inhibition of FHF by MMI and in the
roidism, regardless of the mode of induction, essentially thyroidectomized rats likewise excluded the possibility that
inhibited the development of FHF in this rat model. The the beneficial effect of hypothyroidism in this model was
ominous manifestations of FHF, including severe coagulopa- caused by a direct effect of PTU on the liver.
thy, high grade HE and high mortality rate, were prevented. The mechanism(s) responsible for the inhibition of fulmi-
Consistent with these findings, liver histology in all groups of nant hepatitis in rats by hypothyroidism are not clear.
hypothyroid rats showed significantly less hepatic necrosis, Immunologic factors should be considered, as studies using
although substantial infiltration of liver tissue with inflamma- the TAA model have shown strong features of inflammation
tory cells was still observed (Table 5, Fig. 4), which is and cellular infiltration in the pericentral areas of livers from
consistent with moderate elevations of serum aminotransfer-
TAA-treated rats,24 and in a rat model of chronic TAA
ase levels in the hypothyroid groups.
administration, immune cells are involved in the induction of
TAA is a potent hepatotoxin in rats that acts via the
liver cirrhosis by TAA.30 Furthermore, hypothyroidism pre-
hepatocyte mono-oxigenase cytochrome system. The active
vents liver injury in a model of immune-mediated concanava-
metabolites responsible for hepatotoxicity of TAA are those
lin A-induced acute hepatitis in mice which is associated with
derived from TAA S-oxide, the product of oxidation of TAA
by the flavin adenine dinucleotide-monooxigenase system.25 significantly reduced serum levels of TNF-a in the hypothy-
Free radicals are generated by this oxidative pathway, causing roid mice.4 Several lines of evidence suggest that the thyroid
lipid peroxidation and hepatocyte damage.26 status may have immunomodulatory effects: decreased thy-
To exclude the possibility of drug interaction between TAA roid function is associated with reduced CD41 T lymphocytes
and the anti-thyroid drugs used in the study, hypothyroidism activation, increased number and activation of CD81 cells
was induced also by surgical thyroidectomy. The results in and decreased soluble IL-2 receptors.31 In rats and mice,
this group of rats also showed improvement in liver function MMI-induced hypothyroidism suppressed the expression of
which is similar to those of the rats with drug-induced TNF gene in peritoneal macrophages32,33 and reduced alveo-
hypothyroidism, suggesting that the hypothyroid status itself, lar macrophage production under the stimulation of lipopoly-
and not drug interaction, inhibited the development of FHF saccharide.34 In a recent study, the administration of soluble
in TAA-treated hypothyroid rats. However, the correction of receptor of TNF that neutralizes circulating serum TNF-a,
liver function in the group of rats undergoing thyroidectomy prevented acute liver injury in rats which was induced by the
hepatotoxin CCl4.20 Nevertheless, the cytokine response in
TAA-induced FHF was not characterized in previous studies,
TABLE 5. Effect of Hypothyroidism on Liver Histology and, therefore, the role of TNF-a and other proinflammatory
in TAA-Induced FHF cytokines as mediators of liver injury was not determined.
Inflammation (0-3) Necrosis (0-3) To address this issue, we measured the serum levels of
TAA alone 2.4 6 0.4 2.4 6 0.5
TNF-a, IL-2, and IL-6 for 2, 6, 24, and 48 hours following
TAA 1 MMI 1.6 6 0.3* 0.3 6 0.1** TAA administration in hypothyroid and normal rats. The
TAA 1 PTU 1.4 6 0.3* 0.4 6 0.1** lower serum levels of TNF-a and the other cytokines in the
TAA 1 sTNF-R hypothyroid compared with the euthyroid rats suggest that
100 µg/rat 2.3 6 0.5 2.0 6 0.4 the suppression of cytokine release might have a role in the
1,000 µg/rat 2.5 6 0.6 2.2 6 0.5 prevention of FHF by hypothyroidism. However, the increase
NOTE. Mean 6 SD (n 5 5). Rats were sacrificed 52 hours after the first in the serum levels of TNF-a occurred late (TNF-a levels
TAA injection. were measurable not earlier than 24 hours after TAA injec-
*P , .05 compared with TAA alone. tion) and reached low levels of only 2 pg/mL in control rats,
**P , .01. 250-fold lower than the serum TNF concentrations observed
1018 BRUCK ET AL. HEPATOLOGY April 1998

FIG. 4. Effect of hypothyroidism on liver histology in TAA-induced acute hepatic failure. Rats were sacrificed and livers fixed 24 and 52 hours after the first
TAA injection. (A and B) Liver section from a rat treated only with TAA, showing diffuse centrilobular necrosis and severe inflammatory reaction. (C and D)
TAA and hypothyroidism induced by PTU. Note that no significant hepatic necrosis is present. (E and F) TAA and hypothyroidism induced by MMI. Although
some portal and pericentral inflammatory changes are present, no substantial liver necrosis is observed. (G and H) TAA and hypothyroidism induced by
thyroidectomy. Inflammatory infiltration is more intense [h], and small areas of hepatic necrosis can be observed around the central vein [g], however, no
extensive necrosis is present (compared to TAA alone, A and B). (Hematoxylin and eosin; original magnification 380.)
HEPATOLOGY Vol. 27, No. 4, 1998
FIG. 5. Effect of hypothyroidism on oxygen extraction in isolated
perfused rat liver. Oxygen extraction by the liver was significantly reduced in
hypothyroid (methimazole-treated) rats compared with control liver from
euthyroid rats (82% 6 12% vs. 39 6 7%). Mean 6 SD (n55); **P , .001.
in mice with immune-mediated liver injury induced by
concanavalin A.35 Likewise, the serum levels of IL-2 and IL-6
measured 24 hours after TAA administration were low as
well, whereas severe liver damage has already been confirmed
(Table 2). Moreover, the administration of high doses of
soluble TNF receptor, which neutralizes circulating serum
TNF-a, failed to prevent TAA-induced FHF in rats (Table 1,
3, and 5; Fig. 3). Therefore, it seems that the suppression of
TNF-a and other proinflammatory cytokines do not play a
key role in the prevention of severe liver damage by hypothy-
roidism.
In this model of FHF, cell necrosis probably has the
following two major components: 1) metabolism of TAA to
generate reactive radicals leading to oxidative cell damage;
and 2) the secondary inflammatory response to the products
of cell lysis. Hypothyroidism may, therefore, protect the liver
by inhibiting the generation of free-oxygen radicals, causing
cell necrosis via oxidation of cellular proteins, DNA, and
lipids.26 To determine whether hypothyroidism decreases
lipid peroxidation, we measured the serum and hepatic levels
of malondialdehyde in response to TAA.24 In control euthy-
roid rats, a marked increase in serum malondialdehyde
concentrations was observed 52 hours after the initiation of
TAA treatment, indicating lipid peroxidation.36 In contrast, in
the hypothyroid rats there was only slight increase in the
serum levels of malondialdehyde. Similar results were ob-
tained from measurement of hepatic malondialdehyde concen-
trations, suggesting that the increased serum malondialde-
hyde levels indicate hepatocyte oxidative damage, and not
lipid peroxidation, of extrahepatic tissues. These findings
suggest that the hypothyroid state is protective in FHF and
may minimize oxidative damage to the hepatocytes in AA-
treated rats. In a rat model of chronic TAA ingestion, liver
cirrhosis was completely prevented by hypothyroidism.3
Similar to TAA-induced FHF, in this chronic model hepatic
damage also results from toxic oxygen species causing
chronic liver injury and leads to the development of fibrosis,
probably through the products of lipid peroxidation.30,37,38 In
hyperthyroid rats, generalized hypermetabolism and in-
creased hepatocyte oxygen demand lead to an accelerated
development of TAA-induced liver cirrhosis and portal hyper-
BRUCK ET AL. 1019
tension.3 Thus, under conditions of hyperthyroidism, the
liver is particularly susceptible to injury.39-41 It has been
shown in a previous study that hypothyroidism which is
induced by PTU, as well as by surgical thyroidectomy,
protected rat livers from galactosamine-induced necrosis.
This preventive effect has been ascribed to cellular hypome-
tabolism, although the exact mechanism of ‘‘hepatic protec-
tion’’ was not elucidated.39 Therefore, it appears that hypome-
tabolism and decreased hepatocyte oxygen demand, associated
with hypothyroidism, may be protective for the insulted liver.
This is supported by the decreased oxygen extraction in the
isolated perfused liver of hypothyroid rats observed in our
studies and also by the prevention of liver damage by
hypothyroidism in several experimental models based on
mechanisms different than toxic liver injury, i.e., immune-
mediated hepatitis4 or mechanically-induced liver damage
such as portal vein and bile-duct ligation.2,3 Altogether, the
findings of our studies suggest that decreased thyroid func-
tion may be beneficial for the insulted liver, regardless of the
mechanisms involved in the initiation of liver injury.
REFERENCES
1. Orrego H, Blake JE, Blendis LM, Compton KV, Israel Y. Long-term
treatment of alcoholic liver disease with propylthiouracil. N Engl J Med
1987;317:1421-1427.
2. Oren R, Hilzenrat N, Maaravi Y, Yaari A, E. Sikuler. Hemodynamic effects
of hypothyroidism induced by methimazole in normal and portal
hypertensive rats. Dig Dis Sci 1995,40:1941-1945.
3. Oren R, Dotan I, Papa M, Maaravi Y, Aeed H, Barg J, Zaidel L, et al.
Inhibition of experimentally induced liver cirrhosis in rats by hypothy-
roidism. HEPATOLOGY 1996;24:419-423.
4. Oren R, Shirin H, Dotan I, Aeed H, Zaidel L, Matas Z, Halpern Z.
Hypothyroidism inhibits experimentally-induced acute T cell dependent
hepatic damage in mice [Abstract]. J Hepatol 1995;23(1):213A.
5. O’Grady J, Schalman SW, Williams R. Acute liver failure: redefining the
syndrome. Lancet 1993;324:273-275.
6. Lee WR. Acute liver failure. N Engl J Med 1993;329:1862-1872.
7. Grimm G, Ferenci P, Katzenshlager R, Madl C, Schneeweiss B, Laggner
AN, Lenz K, et al. Improvement of hepatic encephalopathy with
flumazenil. Lancet 1988;2:1392-1394.
8. Mehta JM, Karamarkar SG, Pimparkar BD, Sheth UK. Levodopa in the
treatment of hepatic coma due to fulminant hepatic failure. J Postgrad
Med 1976;22:32-36.
9. Sinclair SB, Greig PD, Blendis LM, Abecassis M, Roberts EA, Phillips MJ,
Cameron R, et al. Biochemical and clinical response of fulminant viral
hepatitis to administration of prostaglandin E. J Clin Invest 1989;84:1063-
1069.
10. Denis J, Opolon P, Delorme ML, Granger A, Darnis F. Long-term
extra-corporeal assistance by continuous haemofiltration during fulmi-
nant hepatic failure. Gastroenterol Clin Biol 1979;3:337-348.
11. Berk P, Goldberg J. Charcol hemoperfusion plus Ca change, plus c’est la
même chose. Gastroenterology 1988;94:1228-1230.
12. O’Grady JG, Gimson AES, O’Brien CJ, Pucknell A, Hughes RD, Williams
R. Controlled trials of charcoal hemoperfusion and prognostic factors in
fulminant hepatic failure. Gastroenterology 1988;94:1186-1192.
13. Bernuau J, Benhamou JP. Fulminant and subfulminant liver failure. In:
McIntyre N, Benhamou JP, Bircher J, Rizzeto M, Rodes J, eds. Oxford
Textbook of Clinical Hepatology. Oxford: Oxford University Press,
1991;942.
14. Bismuth H, Samuel D, Gugenheim J, Castaing D, Bernuau J, Rueff B,
Benhamou JP. Emergency liver transplantation for fulminant hepatitis.
Ann Intern Med 1987;107:337-341.
15. Zimmermann C, Ferenci P, Pifl C, Yurdaydin C, Ebner J, Lassmann H,
Roth E, et al. Hepatic encephalopathy in thioacetamide-induced acute
liver failure in rats: characterization of an improved model and study of
amino acid-ergic neurotransmission. HEPATOLOGY 1989;9:595-601.
16. Larsen FS, Knudsen GM, Paulson OB, Vilstrup H. Cerebral blood flow
autoregulation is absent in rats with thioacetamide-induced hepatic
failure. J Hepatol 1994;21:491-495.
17. Geller D. An improved rat model of hepatic encephalopathy due to
fulminant hepatic failure: the role of supportive therapy. In: Soeters PB,
Wilson JHP, Meijer AJ, Holm E, eds. Advances in Ammonia Metabolism
1020 BRUCK ET AL.
and Hepatic Encephalopathy. Amsterdam: Elsevier Science Publishers,
1988:213-217.
18. Adams JD, Lauterburg BH, Mitchell JR. Plasma glutathione disulfide in
the rat. Regulation and response to oxidative stress. J Pharmacol Exp
Ther 1983;227:749-754.
19. Niehaus WG, Samuelsson JR, Wills ED. Lipid peroxide formation in
microsomes. Biochem J 1969;113:315-341.
20. Czaja MJ, Xu J, Alt A. Prevention of carbon tetrachloride-induced liver
injury by soluble tumor necrosis factor receptor. Gastroenterology
1995;108:1849-1854.
21. Mullen KD. Evaluation of the suitability of galactosamine-induced
fulminant hepatic failure as a model of hepatic encephalopathy in the rat
and rabbit. In: Soeters PB, Wilson JHP, Meijer AJ, Holm E, eds. Advances
in Ammonia Metabolism and Hepatic Encephalopathy. Amsterdam:
Elsevier Science Publishers, 1988:205-211.
22. Corasanti JG, Smith ND, Gordon ER, Boyer JL. Protein kinase C agonists
inhibit bile secretion independently of effects on the microcirculation in
the isolated perfused rat liver. HEPATOLOGY 1989;10:8-13.
23. Bruck R, Shirin H, Hershkoviz R, Kenet G, Aeed H, Matas Z, Zaidel L, et
al. Analysis of Arg-Gly-Asp mimetics and soluble receptor of tumour
necrosis factor as therapeutic modalities for concanavalin A induced
hepatitis in mice. GUT 1997;40:133-138.
24. Diez-Fernandez C, Bosca L, Fernandez-Simon L, Alvarez A, Cascales M.
Relationship between genomic DNA ploidy and parameters of liver
damage during necrosis and regeneration induced by thioacetamide.
HEPATOLOGY 1993;18:912-918.
25. Chieli E, Mavaldi G. Role of microsomal FAD-containing mo-
nooxigenase in the liver toxicity of thioacetamide S-oxide. Toxicology
1984;31:41-52.
26. Moore M, Thor H, Moore G, Nelson S, Moldeus P, Orrenius S. The
toxicity of acetaminophen and N-acetyl-p-benzoquinone imine in iso-
lated hepatocytes is associated with thiol depletion and increased
cytosolic Ca21. J Biol Chem 1985;260:13035-13040.
27. Mesa ML, Carrizosa R, Martinez-Honduvilla C, Benito M, Fabregat I.
Changes in rat liver gene expression induced by thioacetamide: protec-
tive role of S-adenosyl-L-methionine by a glutathione-dependent mecha-
nism. HEPATOLOGY 1996;23:600-606.
28. Bruck R, Aeed H, Shirin H, Matas Z, Zaidel L, Avni Y, Halpern Z.
Prevention of fulminant hepatic failure in rats by the hydoxyl radical
scavenger dimethylsulfoxide [Abstract]. HEPATOLOGY 1997;26:340A.
HEPATOLOGY April 1998
29. Kawasaki T, Carmichael FJ, Giles G, Saldivia V, Israel Y, Orrego H. Effects
of propylthiouracil and methimazole on splanchnic hemodynamics in
awake and unrestrained rats. HEPATOLOGY 1989;10:273-278.
30. Nuber R, Teutsch HF, Sasse D. Metabolic zonation in thioacetamide-
induced liver cirrhosis. Histochemistry 1980;69:277-278.
31. Volpe R. Evidence that the immunosuppressive effects of antithyroid
drugs are mediated through actions on the thyroid cell, modulating
thyrocyte-immunocyte signaling: a review. Thyroid 1994;4:217-223.
32. Liu WK, Tsui KW, Wong CC. Repressed activity of peritoneal macro-
phages of hypothyroid mice. Virchows Arch B Cell Pathol Incl Mol
Pathol 1993;63:131-136.
33. Liu WK. Expression of tumor necrosis factor and c fos genes in
peritoneal macrophages of hypothyroid mice. Inflammation 1993;17:217-
225.
34. Liu WK, Ng TB. Effect of methimazole induced hypothyroidism on
alveolar macrophages. Virchows Arch B Cell Pathol Inc Mol Pathol
1991;60:21-26.
35. Mizuhara H, O’Neill E, Seki N, Ogawa T, Kusunoki C, Otsuka K, Satoh S,
et al. T cell activation-associated hepatic injury: mediation by tumor
necrosis factors and protection by interleukin 6. J Exp Med 1994;179:
1529-1537.
36. Meister A. Glutathione metabolism and its modification. J Biol Chem
1988;263:17205-17208.
37. Muller D, Sommer M, Kretzchmar M, Zimmerman T, Buko VU,
Lukivaskaya O, Dargel R. Lipid peroxidation in thioacetamide-induced
macronodular rat liver cirrhosis. Arch Toxicol 1991;65:199-203.
38. Houglum K, Breener DA, Chojkier M. d-Alpha-tocopherol inhibits
collagen alpha 1 (I) gene expression in cultured human fibroblasts.
Modulation of constitutive collagen gene expression by lipid peroxida-
tion. J Clin Invest 1991;87:2230-2235.
39. Schwedes U, Wdowinski JM, Siede WH, Vardarli I, Schmidt R, Usadel
KH. Effect of hypometabolism on cell injury. Klin Wochenschr 1986;
64(suppl 7):146-148.
40. Carmichael FJ, Orrego H, Saldivia V, Israel Y. Effect of propylthiouracil
on the ethanol-induced increase in liver oxygen consumption in awake
rats. HEPATOLOGY 1993;18:415-421.
41. Romney JS, Chan J, Carr FE, Mooradin AD, Wong NCW. Identification
of the thyroid hormone-responsive messenger RNA spot 11 as apolipo-
protein A1 messenger RNA and effects of the hormone on the promotor.
Mol Endocrinol 1992;6(6):943-950.