thyroid hormone preconditioning: protection against ischemia-reperfusion liver injury in the rat

Thyroid Hormone Preconditioning: Protection AgainstIschemia-Reperfusion Liver Injury in the Rat

Virginia Fernandez,1 Ivan Castillo,1 Gladys Tapia,1 Pamela Romanque,1,2 Sebastian Uribe-Echevarrıa,3 Mario Uribe,3

Denise Cartier-Ugarte,1 Gonzalo Santander,1 Marıa T. Vial,4 and Luis A. Videla1

Recently, we reported that oxidative stress due to 3,3�,5-triiodothyronine (T3)-inducedcalorigenesis up-regulates the hepatic expression of mediators promoting cell protection. Inthis study, T3 administration in rats (single dose of 0.1 mg/kg intraperitoneally) inducedsignificant depletion of reduced liver glutathione (GSH), with higher protein oxidation, O2

consumption, and Kupffer cell function (carbon phagocytosis and carbon-induced O2 up-take). These changes occurred within a period of 36 hours of T3 treatment in animalsshowing normal liver histology and lack of alteration in serum AST and ALT levels. Partialhepatic ischemia-reperfusion (IR) (1 h of ischemia via vascular clamping and 20 h reperfu-sion) led to 11-fold and 42-fold increases in serum AST and ALT levels, respectively, andsignificant changes in liver histology, with a 36% decrease in liver GSH content and a 133%increase in that of protein carbonyls. T3 administration in a time window of 48 hours wassubstantially protective against hepatic IR injury, with a net 60% and 90% reduction in liverGSH depletion and protein oxidation induced by IR, respectively. Liver IR led to decreasedDNA binding of nuclear factor-�B (NF-�B) (54%) and signal transducer and activator oftranscription 3 (STAT3) (53%) (electromobility shift assay), with 50% diminution in theprotein expression of haptoglobin (Western blot), changes that were normalized by T3

preconditioning. Conclusion: T3 administration involving transient oxidative stress in theliver exerts significant protection against IR injury, a novel preconditioning maneuver thatis associated with NF-�B and STAT3 activation and acute-phase response. (HEPATOLOGY

2007;45:170-177.)

Ischemia-reperfusion (IR) liver injury is associatedwith temporal occlusion of the hepatic pedicle duringliver surgery, hypoperfusion shock, and graft failure

after liver transplantation.1 Experimental studies have de-fined two phases underlying IR injury with developmentof hepatocellular necrosis and liver inflammation. The

initial phase (within 4 h of reperfusion) involves the acti-vation of Kupffer cells resulting in the production andrelease of reactive oxygen species (ROS) and proinflam-matory cytokines, with the consequent up-regulation ofintercellular adhesion molecule, vascular adhesion mole-cule, and C-X-C chemokines that promote neutrophilrecruitment and directed migration into the liver.1,2 Thesecond phase occurring between 6 to 24 hours after reper-fusion is characterized by further activation of Kupffercells and that of recruited neutrophils, which extravasateand adhere to endothelial cell lining causing parenchymalcell injury by ROS and protease release and phagocytosis.3

IR also involves microcirculatory disturbances leading tounderperfused areas in the liver that may further enhancethe injury.4 Thus, liver IR injury is a complex phenome-non involving several contributory factors, making it dif-ficult to attain effective protection by targeting individualmechanisms.1 In this respect, the development of hepaticpreconditioning has been extensively explored in recentyears as a protective strategy to enhance the resistance ofliver cells to IR events.1,5 This has been achieved by pre-exposing the liver to conditions triggering a mild oxida-

Abbreviations: APP, acute-phase protein; APR, acute-phase response; GSH, re-duced glutathione; IR, ischemia-reperfusion; NF-�B, nuclear factor-�B; ROS, re-active oxygen species; STAT3, signal transducer and activator of transcription 3; T3,3,3�,5-triiodothyronine; TNF-�, tumor necrosis factor-�.

From the 1Molecular and Clinical Pharmacology Program, Institute of Biomed-ical Sciences, Faculty of Medicine, University of Chile, Santiago, Chile; the 2HealthSciences Faculty, Diego Portales University, Santiago, Chile; the 3ExperimentalSurgery Laboratory, Department of Surgery, Del Salvador Hospital, Faculty ofMedicine, University of Chile; and the 4Pathological Anatomy Unit, San BorjaArriaran Hospital, Santiago, Chile.

Received July 3, 2006; accepted September 14, 2006.Supported by grant 1050131 from FONDECYT, Chile.Address reprint requests to: Luis A. Videla, Programa de Farmacologıa Molecular

y Clınica, Instituto de Ciencias Biomedicas, Facultad de Medicina, Universidad deChile, Santiago, Chile. E-mail: [email protected]; fax: (56)-2-7372783.

Copyright © 2006 by the American Association for the Study of Liver Diseases.Published online in Wiley InterScience (www.interscience.wiley.com).DOI 10.1002/hep.21476Potential conflict of interest: Nothing to report.

170

tive stress status, as shown for transient ischemia,6

hyperthermia,7 or pro-oxidant conditions induced by theredox cycling agent doxorubicin,8 the model oxidant tert-butyl hydroperoxide,9 hyperbaric oxygen therapy,10 andozone.11 Oxidative stress preconditioning may involvegene reprogramming through activation of specific tran-scription factors that control the expression of proteinsrelated to cell protection and survival.5,12

Thyroid hormone, or 3,3�,5-triiodothyronine (T3), isrequired for the normal function of most tissues, withmajor effects on energy metabolism and metabolic rate.Acceleration of O2 consumption in the liver by T3 en-hances ROS production with concomitant antioxidantdepletion, thus inducing oxidative stress, a redox imbal-ance that also involves increased respiratory burst activityin Kupffer cells.13 T3-induced calorigenesis is accompa-nied by redox up-regulation of the hepatic expression ofcytokines (tumor necrosis factor-� [TNF-�], IL-10, IL-1�),14 enzymes (inducible NO synthase, manganese su-peroxide dismutase),15,16 antiapoptotic proteins (Bcl-2),16 and acute-phase proteins (APP) (haptoglobin,�-fibrinogen).17 These responses involve TNF-�/nuclearfactor-�B (NF-�B) or IL-6/signal transducer and activa-tor of transcription 3 (STAT3) cascades14-17 and may rep-resent adaptive mechanisms to re-establish redoxhomeostasis and promote cell survival under conditionsof ROS toxicity secondary to T3-induced oxidativestress.13 In view of these considerations, the objective ofthis study was to test the hypothesis that preconditioningthe liver by T3 administration to rats could protect theliver against subsequent IR injury. For this purpose, wefirst established the optimal conditions for the precondi-tioning maneuver by subjecting rats to a single dose of T3,followed by assessment of parameters related to calorigen-esis, liver oxidative stress, Kupffer cell functioning, andliver damage for up to 72 h after treatment. Next, westudied the effects of T3 treatment on subsequent hepaticIR injury, the results of which were correlated withNF-�B and STAT3 activation and the acute-phase re-sponse (APR) of the liver.

Materials and Methods

T3 Treatment and Model of Partial Hepatic IRInjury in the Rat. Male Sprague-Dawley rats (BioterioCentral, ICBM, Faculty of Medicine, University ofChile) weighing 170-200 g were housed on a 12-hourlight/dark cycle and were provided with rat chow andwater ad libitum. Animals received a single intraperito-neal dose of 0.1 mg of T3/kg body weight or equivalentvolumes of hormone vehicle (0.1 N NaOH, controls) andstudies were performed at the indicated times after treat-

ment. Thyroid hormone-induced calorigenesis was as-sessed by the rectal temperatures of the animals measuredwith a thermocouple (Cole-Parmer Instrument Com-pany, Chicago, IL). Blood samples were taken from thetail vein for the measurement of serum T3 (MonobindInc., Lake Forest, CA) and TNF-� (Biosource Interna-tional, Camarillo, CA) levels via ELISA.

Rats were anesthetized with intraperitoneal (1 mL/kg)zolazepam chlorhydrate (25 mg/mL) and tiletamine chlor-hydrate (25 mg/mL) (Zoletil 50; Virbac S/A, Carros,France), and IR was induced by temporarily occludingthe blood supply to the left lateral and median lobes of theliver by means of a Schwartz clip (Fine Science Tools Inc.,Vancouver, British Columbia, Canada), as previously de-scribed.18 Control animals were subjected to anesthesiaand sham laparotomy. Experimental groups included eu-thyroid and T3-treated rats subjected to either sham lap-arotomy or IR (Fig. 1). Liver samples were taken from themedial lobes at the end of the reperfusion period (Fig. 1)and immediately frozen in liquid nitrogen or fixed inphosphate-buffered formalin, embedded in paraffin, andstained with hematoxylin-eosin. Experimental animalprotocols and animal procedures complied with theGuide for the Care and Use of Laboratory Animals (Na-tional Academy of Sciences, NIH Publication 86-23, re-vised 1985).

Parameters Related to Oxidative Stress, KupfferCell Function, and Liver Injury. In anesthetized ani-mals, livers were perfused in situ with a cold solutioncontaining 150 mM KCl and 5 mM Tris (pH 7.4) toremove blood; total reduced glutathione (GSH) equiv-alents,19 protein oxidation,20 and total protein con-tent21 were measured. Kupffer cell functioning wasassessed in the isolated perfused rat liver by the deter-mination of rates of colloidal carbon uptake, in addi-tion to carbon-induced O2 uptake and total O2

Fig. 1. Experimental protocol for T3 preconditioning. Animals weregiven either T3 vehicle or a single dose of T3 (0.1 mg/kg) at time zero.At 48 hours after treatment, groups of control rats and T3-treatedanimals were subjected to sham operation or to 1 hour ischemia(light gray bars) followed by 20 hours reperfusion (dark gray bars),thus conforming four experimental groups. Blood and liver sampleswere obtained at 69 hours.

HEPATOLOGY, Vol. 45, No. 1, 2007 FERNANDEZ ET AL. 171

consumption measured polarographically, as describedpreviously.22 In all experiments, the severity of liverdamage was determined by measuring serum ALT andAST levels and by performing liver light microscopy.

NF-�B and STAT3 Electromobility Shift Assay.Nuclear protein extracts from liver samples were pre-pared according to Deryckere and Gannon.23 The sam-ples were subjected to electromobility shift assay forassessment of NF-�B and STAT3 DNA binding usingthe NF-�B probe 5�-GAT CTC AGA GGG GACTTT CCG AG-3� or the STAT3 probe 5�-GTC GACATT TCC CGT AAA TCG TCG A-3� (InvitrogenLife Technologies, Carlsbad, CA), labeled with �-32P-dCTP using the Klenow DNA Polymerase Fragment I(Invitrogen Corp., Carlsbad, CA), as described previ-ously.14-16 The specificity of the reaction was deter-mined by a competition assay using 100-fold molarexcess of unlabeled DNA probes. Samples were loadedon nondenaturating 6% polyacrylamide gels and rununtil the free probe reached the end of the gel; NF-�Band STAT3 bands were detected by autoradiographyand quantified by densitometry using Scion Image(Scion Corp., Frederick, MD).

Western Blot Analysis of Haptoglobin. Liver sam-ples (100-500 mg) frozen in liquid nitrogen were ho-mogenized and suspended in a buffer solution pH 7.9containing 10 mM HEPES, 1 mM EDTA, 0.6% Non-idet P-40, 150 mM NaCl, and protease inhibitors (1mM phenylmethylsulfonyl fluoride, 1 �g/mL aproti-nin, 1 �g/mL leupeptin, and 1 mM orthovanadate).Soluble protein fractions (25 �g) were separated on12% polyacrylamide gels using SDS-PAGE24 andtransferred to nitrocellulose membranes,25 which wereblocked for 1 hour at room temperature with TBS-containing 5% nonfat dry milk. The blots were washedwith TBS-containing 0.1% Tween 20 and hybridizedwith rabbit polyclonal antibodies for human haptoglo-bin (Dako Corp., Carpinteria, CA). In all determina-tions, mouse monoclonal antibody for rat �-actin(ICN Biomedicals, Inc., Aurora, OH) was used as in-ternal control. After extensive washing, the antigen–antibody complexes were detected using horseradishperoxidase labeled goat anti-rabbit IgG or goat anti-mouse IgG and a SuperSignal West Pico Chemilumi-nescence kit detection system (Pierce, Rockford, IL).

Statistical Analyses. Values shown represent themean � SEM for the number of separate experimentsindicated. One-way ANOVA and the Newman-Keulstest assessed the statistical significance of differences be-tween mean values. A P value of less than 0.05 was con-sidered significant.

ResultsRats subjected to a single dose of T3 achieved a

significant elevation in serum T3 levels within 24hours, returning to control values at 36 hours (Fig.2A). Under these conditions, T3 administration led to(1) a calorigenic response evidenced by significant in-creases in the rectal temperature of the animals thatlasted for 34 hours, leveling off at later times (Fig. 2B),and (2) oxidative stress in the liver within 24 hours,characterized by GSH depletion at 24 hours (Fig. 2C)and higher protein carbonylation at 12 hours (Fig.2D), changes that occurred without major alteration inserum levels of AST (Fig. 2E) and ALT (Fig. 2F).

Kupffer cell function was continuously monitoredbased on the uptake of colloidal carbon and the carbon-induced O2 uptake by the isolated perfused rat liver.22

T3 administration elicited a significant 37% enhance-ment in the basal rate of O2 consumption of the liverover control values, 24 hours after hormone treatment(Fig. 3A). This effect was observed concomitantly withincreases of 136% and 48% (P � 0.05) in the carbon-induced O2 uptake (Fig. 3B) and in the rate of carbonuptake (Fig. 3C), respectively, with a net 62% eleva-tion in the respective oxygen/carbon uptake ratios (Fig.3D). All these changes induced by T3 returned towardcontrol values at 48 hours after hormone treatment(Fig. 3), coinciding with the normalization of the se-rum levels of T3 and T3-induced calorigenesis, liverGSH depletion, and hepatic protein oxidation (Fig. 2).Therefore, a period of T3 preconditioning of 48 hourswas chosen for subsequent studies in the hepatic IRinjury model.

The hepatic IR model used consisted of 1 hour ofpartial hepatic ischemia via vascular clamping followedby reperfusion for 20 hours (Fig. 1), which achievedminimal mortality but extensive liver injury. This wasevidenced by significant 11-fold and 42-fold increasesin serum AST (Fig. 4A) and ALT (Fig. 4B) levels ob-served in the IR group compared with sham-operatedanimals, respectively. Compared with T3-treatedsham-operated rats, the T3-IR group exhibited 4.6-foldand 11.5-fold increases in serum AST (Fig 4A) andALT (Fig. 4B) levels, thus leading to net reductions of61% and 68% (P � 0.05) in relation to the IR group,respectively. In agreement with these data, control an-imals exhibited normal liver morphology (Fig. 4C),whereas IR resulted in substantial distortion of liverarchitecture, with extensive areas of hepatocyte necro-sis in midzone reaching pericentral and periportal ar-eas, neutrophil infiltration in midzone and portaltracts, and slight enlargement of portal tracts and cen-

172 FERNANDEZ ET AL. HEPATOLOGY, January 2007

tral venules (Fig. 4D). On the contrary, the livers ofT3-treated rats subjected to either sham operation (Fig.4E) or IR (Fig. 4F) showed normal architecture, withminimal central venule and portal tract enlargement

and absence of necrosis. Fibrosis was not observed inany of the experimental groups studied.

Hepatic IR injury was observed concomitantly with a36% decrease in the content of hepatic GSH (Fig. 5A)

Fig. 3. Basal rate of (A) O2 consumption, (B) carbon-induced O2 uptake, (C) carbon uptake, and (D) the respective oxygen/carbon uptake ratios in thelivers of control rats and animals given T3 after 24 hours and 48 hours of hormone administration. Data are expressed as the mean � SEM for five to sixanimals per group. (B) Carbon-induced O2 uptake was assessed via the integration of the area under the O2 consumption curves between 30-45 minutesof colloidal carbon infusion (0.5 mg/mL) into isolated perfused livers, and expressed as �mol O2/g liver.22 (C) Rates of carbon uptake were calculated frominfluent minus effluent concentration differences, referred to the perfusion flow, and expressed as mg/g liver/min.22 (D) Oxygen/carbon uptake ratios werecalculated from the carbon-induced O2 uptake (�mol O2/g liver) and the integrated carbon uptake curves (mg/g liver), and expressed as �mol O2/mgcarbon. aP � 0.05 versus control values. bP � 0.05 versus T3-treated animals at 24 hours. cP � 0.05 versus T3-treated animals at 48 hours.

Fig. 2. Time course study of the effects of T3administration on (A) serum T3 levels, (B) rectaltemperature, (C) liver GSH content, (D) liverprotein carbonyl content, and serum (E) ASTand (F) ALT levels in the rat. T3 (0.1 mg/kgintraperitoneally) or T3 vehicle (controls) weregiven at time zero (arrows). Data are expressedas the mean � SEM for 3 to 16 animals pergroup. aP � 0.05 compared with the respectivetime controls or with average control data attime zero.


and 133% increase in that of protein carbonyls, as indi-cation of the oxidative stress status developed. Thesechanges were significantly attenuated by T3 precondition-ing, as evidenced by the 60% and 90% reduction in GSHdepletion and protein oxidation, respectively, when dif-ferences between IR and sham-operated groups werecompared with those between T3-treated rats with orwithout IR (Fig. 5). In addition, IR led to 116% enhance-ment in the serum TNF-� levels (P � 0.05) over controlvalues in unpreconditioned rats, an effect that was abol-ished by T3 preconditioning [control-sham, 26.7 � 4.2(n � 7) pg/mL; control-IR, 57.7 � 5.2 (n � 5); T3-sham,35.6 � 1.7 (n � 5); T3-IR, 32.8 � 4,5 (n � 8)].

IR in unpreconditioned animals induced a significant54% diminution in liver NF-�B DNA binding, as as-sessed via electromobility shift assay, compared with val-ues in sham-operated rats (Fig. 6A). Although T3

administration in sham-operated rats led to a decrease inNF-�B activation, IR in T3 preconditioned rats resultedin a 37% (P � 0.05) enhancement in the nuclear bindingof the transcription factor over values in unprecondi-tioned animals (Fig. 6A). In addition, hepatic STAT3activation was reduced by 53% (P � 0.05) in unprecon-ditioned rats subjected to IR when compared with thesham-operated group, whereas the T3-IR group showed alower decrease (35%; P � 0.05) over values observed inT3-treated sham-operated animals (Fig. 6A). Interest-ingly, values for both NF-�B and STAT3 DNA bindingin the T3-IR group are comparable to those in the controlsham-operated group (Fig. 6A).

The hepatic expression of the APP haptoglobin wasassessed via Western blot analysis. The data presented inFig. 6B show that IR elicited a 50% diminution in theprotein expression of haptoglobin compared with the

Fig. 4. Serum (A) AST and (B) ALT levels,and liver histology after hepatic IR injury inunpreconditioned and T3 preconditioned rats.Representative liver sections from (C) a control-sham operated rat, (D) a control animal sub-jected to 1 hour of ischemia followed by 20hours of reperfusion, (E) a T3-treated sham-operated rat, and (F) a T3-treated animal sub-jected to IR (hematoxylin-eosin–stained liversections), as shown in Fig. 1. The values inpanels A and B are expressed as the mean �SEM for 4 to 11 different rats. aP � 0.05versus control sham-operated rats. bP � 0.05versus control animals subjected to IR. cP �0.05 versus T3-treated sham-operated rats.dP � 0.05 versus T3-treated animals subjectedto IR. (Original magnification �20.)

Fig. 5. (A) Liver GSH content and (B) proteincarbonyl content after hepatic IR injury in unpre-conditioned and T3 preconditioned rats. Data areexpressed as the mean � SEM for 4 to 14 differentanimals per group. aP � 0.05 versus control sham-operated rats. bP � 0.05 versus control animalssubjected to IR. cP � 0.05 versus T3-treated sham-operated rats. dP � 0.05 versus T3-treated animalssubjected to IR.


sham-operated group, an effect that declined to a 15%reduction upon T3 preconditioning. Liver expression ofthe APP �-fibrinogen was also examined in this study;however, it was drastically augmented by the surgical pro-cedures used (untreated controls, 21 � 6 [n � 3] arbitraryunits; sham-operated rats, 77 � 6 [n � 5]; P � 0.05),thus hindering the assessment of the net effects of IR, T3

preconditioning, or both (data not shown).

DiscussionDevelopment of oxidative stress is associated with a

wide spectrum of cellular responses, depending on the celltype, the level of ROS and/or reactive nitrogen speciesachieved, and the duration of the exposure.26 Previouswork by our group has shown that concurrence of pro-oxidant stimuli in T3-treated animals exhibiting a sus-tained enhancement in the oxidative stress status inducesor exacerbates liver injury due to potentiation of the pro-oxidant state of the organ.27-29 The data presented in thisstudy indicate that the calorigenic response induced bythe administration of a single dose of T3 to fed rats leads toa transient increase in the oxidative stress status of theliver. This redox imbalance occurs within 24 hours of T3

administration under conditions of higher rates of hepaticO2 consumption, with significant GSH depletion andprotein oxidation enhancement. In addition to thesechanges that are presumed to occur primarily at the pa-renchymal cell level, Kupffer cells may play a contributoryrole, as evidenced by the enhancement in both carbonphagocytosis and the related respiratory burst activity ob-served 24 hours after T3 treatment. The latter process,which is mainly due to the activity of the ROS generatorNADPH oxidase with a smaller contribution by NO syn-thase,30 returned toward control values 48 hours after T3

treatment, concomitant with the other oxidative stress-related parameters studied. Collectively, these data indi-cate that T3 administration may represent apreconditioning stimulus, considering (1) the productionof a temporary and reversible redox imbalance within atime window of 48 hours, which is devoid of hepatotox-icity as revealed by the occurrence of normal liver histol-ogy and lack of significant changes in serum ALT andAST levels, and (2) the enhancement in the activity ofKupffer cells, macrophages known to play a key role in thehomeostatic response to liver injury.31

The present study demonstrates that the administra-tion of a single dose of T3 to rats significantly diminisheshepatocellular injury induced by IR when given 48 hoursbefore the IR protocol. T3 preconditioning is related tosuppression of the TNF-� response and the oxidativestress component induced by IR, as evidenced by normal-ization of IR-induced GSH depletion and protein oxida-

Fig. 6. (A) Liver NF-�B and STAT3 DNA binding on electromobility shiftassay after hepatic IR injury in unpreconditioned and T3 preconditionedrats. Top autoradiographs represent lanes containing 8 �g nuclearprotein from control sham-operated rats (group a) in competition exper-iments without (�) and with 100-fold excess of the respective oligonu-cleotides (Com). Bottom autoradiographs correspond to nuclear extracts(8 �g protein) from an animal of each experimental group studied. Bargraphs correspond to densitometric quantification of relative NF-�B andSTAT3 DNA binding, expressed as the mean � SEM for three to sevendifferent rats. (B) Liver haptoglobin expression evaluated via Westernblotting after hepatic IR injury in unpreconditioned and T3-preconditionedrats. Representative blots of haptoglobin and �-actin protein expressionare shown, using 25 �g of soluble protein from a different rat of eachgroup studied. Bar graphs correspond to the respective densitometricquantification expressed as haptoglobin/�-actin ratios to compare lane-to-lane equivalency in total protein content. Values shown are expressedas the mean � SEM for five to seven different rats. aP � 0.05 versuscontrol sham-operated rats. bP � 0.05 versus control animals subjectedto IR. cP � 0.05 versus T3-treated sham-operated rats. dP � 0.05 versusT3-treated animals subjected to IR.


tion upon hormone pretreatment. Under theseconditions, cell protection by T3 may involve preventionof ROS-dependent oxidative deterioration of biomol-ecules. Consistent with this, the administration of theantioxidant N-acetylcysteine before the T3-precondition-ing phase increased serum ALT levels by more than three-fold after the IR protocol, thus diminishing thehepatoprotection afforded by T3 (unpublished data). Inaddition, T3 may revert changes in signal transductionand gene expression underlying liver injury induced byIR,26,32 which are characterized by depression in hepaticNF-�B or STAT3 DNA binding and in the expression ofthe APP haptoglobin. The APR is a major pathophysio-logical reaction mainly focused on the liver, in whichnormal homeostatic mechanisms are replaced by new setpoints contributing to defensive or adaptive capabilities.33

Thus, T3 preconditioning significantly recovers the levelof NF-�B and STAT3 activation reduced by IR to thatobserved in the liver of control animals, with up-regula-tion of the APR, effects that may be ascribed to theKupffer cell–dependent expression and release of IL-1�,TNF-�, and IL-6 elicited by T3 administration.14-17 IL-6binding to receptor � subunit leads to homodimerizationof glycoprotein 130 (gp130) and activation of Janus ki-nase, STAT3 phosphorylation and dimerization, andtranslocation of the dimers to the nucleus where they bindto response elements in the promoter regions of type IAPP genes (haptoglobin) and type II APP genes.34 Inaddition, IL-1� and TNF-� up-regulate hepatic type IAPP expression, either through NF-�B and/or activatorprotein-1 signaling,35 thus providing a point of cross-talkwith the IL-6/gp130/STAT3 pathway.17,33 Consistentwith the data presented in this study, administration ofthyroxine (T4) has been shown to increase the hepaticsynthesis of the APP �1-acid glycoprotein, �2-(acutephase) globulin, haptoglobin,36 and ceruloplasmin,37 al-though the mechanisms involved were not studied. Hap-toglobin, angiotensinogen, fibrinogen, complementcomponent 4A, and serum amyloid A protein precursorare also up-regulated by T3 in the HepG2 cell line.38 Inaddition, T3 administration leads to posttranscriptionalup-regulation of the hepatic expression of the APP ferritinthrough enhancement of the iron-induced displacementof rat and human iron regulatory protein from the iron-responsive element present in ferritin messenger RNA.39

Thus, up-regulation of the hepatic APR in T3 precondi-tioning may represent a defense mechanism against thedeleterious effects of IR-induced oxidative stress, redirect-ing gene expression into a pattern fulfilling antioxidant,transport, coagulation, and immune functions, theformer being associated with induction of haptoglobin,ceruloplasmin, and ferritin by thyroid hormones.33,36-39

In conclusion, the results of this study indicate thattransient oxidative stress induced by T3 plays a major roleas a regulatory mediator in signaling processes, leading tosignificant prevention of liver injury associated with IR.T3 preconditioning is related to a gain of liver cell signal-ing functions represented by recovery of NF-�B andSTAT3 DNA binding and APR, which are lost during IR.These responses may protect the liver against IR-inducedoxidative stress by re-establishing redox homeostasis, al-though actions involving other physiological functionscannot be discarded. Of particular interest is the action ofT3 as a primary hepatic mitogen in vivo inducing theprocess of direct hyperplasia,40 which may compensate forliver cells lost due to IR-induced hepatocellular necrosis,through signaling mechanisms involving NF-�B andSTAT3 activation.41 This aspect, and the involvement ofsignaling cascades underlying NF-�B, activator pro-tein-1, and STAT3 activation in T3 preconditioning arecurrently under study in our laboratory, considering itsclinical potential. The latter view is based on the fact thatT3 is a natural occurring molecule and is a widely used andwell-tolerated therapeutic agent whose side effects can bereadily controlled.41,42

References1. Jaeschke H. Molecular mechanisms of hepatic ischemia-reperfusion injury

and preconditioning. Am J Physiol Gastrointest Liver Physiol 2003;284:G15-G26.

2. Colletti LM, Kunkel SL, Walz A, Burdick MD, Kunkel RG, Wilke CA, etal. The role of cytokine networks in the local liver injury following hepaticischemia/reperfusion in the rat. HEPATOLOGY 1996;23:506-514.

3. Cutrın JC, Boveris A, Zingaro B, Corvetti G, Poli G. In situ determinationby surface chemiluminescence of temporal relationships between evolvingwarm ischemia-reperfusion injury in rat liver and phagocyte activation andrecruitment. HEPATOLOGY 2000;31:622-632.

4. Vollmar B, Glasz J, Leiderer R, Post S, Menger M. Hepatic microcircula-tory perfusion failure is a determinant of liver dysfunction in warm isch-emia-reperfusion. Am J Pathol 1994;145:1421-1431.

5. Selzner N, Rudiger H, Graf R, Clavien PA. Protective strategies againstischemic injury of the liver. Gastroenterology 2003;125:917-936.

6. Peralta C, Hotter G, Closa D, Gelpi E, Bulbena O, Rossello-Catafau J.Protective effect of preconditioning on the injury associated to hepaticischemia-reperfusion in the rat: role of nitric oxide and adenosine. HEPA-TOLOGY 1997;25:934-937.

7. Terajima H, Enders G, Thiaener A, Hammer C, Kondo T, Thiery J, et al.Impact of hyperthermic preconditioning on postischemic hepatic micro-circulation disturbances in an isolated perfusion model of rat liver. HEPA-TOLOGY 2000;31:407-415.

8. Ito K, Ozasa H, Sanada K, Horikawa S. Doxorubicin preconditioning: aprotection against rat hepatic ischemia-reperfusion injury. HEPATOLOGY

2000;31:416-419.9. Rudiger HA, Graf R, Clavien PA. Sub-lethal oxidative stress triggers the

protective effects of ischemic preconditioning in the mouse liver. J Hepatol2003;39:972-977.

10. Yu SY, Chiu JH, Yang SD, Yu HY, Hsied CC, Chen PJ, et al. Precondi-tioned hyperbaric oxygenation protects the liver against ischemia-reperfu-sion injury in rats. J Surg Res 2005;128:28-36.

11. Ajamieh HH, Menendez S, Martinez-Sanchez G, Candelario-Jalil E, Re L,Giuliani A, et al. Effects of ozone oxidative preconditioning on nitric oxide


generation and cellular redox balance in a rat model of hepatic ischaemia-reperfusion. Liver Int 2004;24:55-62.

12. Cutrın JC, Perrelli MG, Cavalieri B, Peralta C, Rosello-Catafau J, Poli G.Microvascular dysfunction induced by reperfusion injury and protectiveeffect of ischemic preconditioning. Free Radic Biol Med 2002;33:1200-1208.

13. Fernandez V, Tapia G, Varela P, Romanque P, Cartier-Ugarte D, VidelaLA. Thyroid hormone-induced oxidative stress in rodents and humans: acomparative view and relation to redox regulation of gene expression.Comp Biochem Physiol, Part C 2006;146:231-239.

14. Tapia G, Fernandez V, Varela P, Cornejo P, Guerrero J, Videla LA. Thy-roid hormone-induced oxidative stress triggers nuclear factor-�B and cy-tokine gene expression in rat liver. Free Radic Biol Med 2003;35:257-265.

15. Fernandez V, Tapia G, Varela P, Videla LA. Redox regulation of thyroidhormone-induced Kupffer cell-dependent I�B-� phosphorylation in rela-tion to inducible nitric oxide synthase expression. Free Radic Res 2005;39:411-418.

16. Fernandez V, Tapia G, Varela P, Castillo I, Mora C, Moya F, et al. Redoxup-regulated expression of rat liver manganese superoxide dismutase andBcl-2 by thyroid hormone is associated with inhibitor of I�B-� phosphor-ylation and nuclear factor-�B activation. J Endocrinol 2005;186:539-547.

17. Tapia G, Fernandez V, Pino C, Ardiles R, Videla LA. The acute-phaseresponse of the liver in relation to thyroid hormone-induced redox signal-ing. Free Radic Biol Med 2006;40:1628-1635.

18. Gonzalez-Flecha B, Cutrın JC, Boveris A. Time course and mechanism ofoxidative stress and tissue damage in rat liver subjected to in vivo ischemia-reperfusion. J Clin Invest 1993;91:456-464.

19. Tietze F. Enzymic method for quantitative determination of nanogramamounts of total and oxidized glutathione: applications to mammalianblood and other tissues. Anal Biochem 1969;27:287-290.

20. Reznic AZ, Packer L. Oxidative damage to proteins: spectrophotometricmethod for carbonyl assay. Methods Enzymol 1994;233:357-363.

21. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurementwith the Folin phenol reagent. J Biol Chem 1951;193:265-275.

22. Tapia G, Pepper I, Smok G, Videla LA. Kupffer cell function in thyroidhormone-induced liver oxidative stress in the rat. Free Radic Res 1997;26:267-279.

23. Deryckere F, Gannon F. A one-hour minipreparation technique for extrac-tion of DNA-binding proteins from animal tissues. Biotechniques 1994;16:405.

24. Laemmli UK. Cleavage of structural proteins during the assembly of thehead of bacteriophage T4. Nature 1970;227:680-685.

25. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteinsfrom polyacrylamide gels to nitrocellulose sheets: procedure and someapplications. Proc Natl Acad Sci U S A 1979;76:4350-4354.

26. Droge W. Free radicals in the physiological control of cell function. PhysiolRev 2002;82:47-95.

27. Troncoso P, Smok G, Videla LA. Potentiation of ischemia-reperfusionliver injury by hyperthyroidism in the rat. Free Radic Biol Med 1997;23:19-25.

28. Boisier X, Schon M, Sepulveda A, Basualdo A, Cornejo P, Bosco C, et al.Derangement of Kupffer cell functioning and hepatotoxicity in hyperthy-roid rats subjected to acute iron overload. Redox Rep 1999;4:243-250.

29. Simon-Giavarotti KA, Giavarotti L, Gomes LF, Lima AF, Veridiano AM,Garcia EA, et al. Enhancement of lindane-induced liver oxidative stressand hepatotoxicity by thyroid hormone is reduced by gadolinium chloride.Free Radic Res 2002;36:1033-1039.

30. Wang JF, Komarov P, de Groot H. Luminol chemiluminescence in ratmacrophages and granulocytes: the role of NO, O2

�/H2O2, and HOCl.Arch Biochem Biophys 1993;304:189-196.

31. Tsukamoto H. Redox regulation of cytokine expression in Kupffer cells.Antioxid Redox Signal 2002;4:741-748.

32. Poli G, Leonarduzzi G, Biasi F, Chiarpotto E. Oxidative stress and cellsignalling. Curr Med Chem 2004;11:1163-1182.

33. Samols D, Agrawal A, Krushner I. Acute phase proteins. In: OppenheimJJ, Feldman M, eds. Cytokine reference on-line. London: Academic Press,2003.

34. Heinrich PC, Behrmann I, Hann S, Hermanns HM, Muller-Newen G,Schaper F. Principles of interleukin (IL)-6-type cytokine signaling and itsregulation. Biochem J 2003;374:1-20.

35. Baud V, Karin M. Signal transduction by tumor necrosis factor and itsrelatives. Trends Cell Biol 2001;11:372-377.

36. Griffin EE, Miller LL. Effects of hypothyroidism, hyperthyroidism, andthyroxine on net synthesis of plasma proteins by the isolated perfused ratliver. J Biol Chem 1973;248:4716-4723.

37. Fitch CA, Song Y, Leverson CW. Developmental regulation of hepaticceruloplasmin mRNA and serum activity by exogenous thyroxine anddexamethasone. Proc Soc Exp Biol Med 1999;221:27-31.

38. Lin K-H, Lee H-Y, Shih C-H, Yen C-C, Chen S-L, Yang R-C, et al. Plasmaprotein regulation by thyroid hormone. J Hepatol 2003;179:367-377.

39. Leedman PJ, Stein AR, Chin WW, Rogers JT. Thyroid hormone modu-lates the interaction between iron regulatory proteins and the ferritinmRNA iron-responsive element. J Biol Chem 1996;271:12017-12023.

40. Francavilla A, Carr BI, Azzarone A, Polimeno L, Wang Z, Van Thiel DH,et al. Hepatocyte proliferation and gene expression induced by triiodothy-ronine in vivo and in vitro. HEPATOLOGY 1994;20:1237-1241.

41. Columbano A, Shinozuka H. Liver regeneration versus direct hyperplasia.FASEB J 1996;10:1118-1128.

42. Oren R, Dabeva MD, Karnezis AN, Petrov PM, Rosencrantz R, SandhuJP, et al. Role of thyroid hormone in stimulating liver repopulation in therat by transplanted hepatocytes. HEPATOLOGY 1999;30:903-913.


thyroid hormone preconditioning: protection against ischemia-reperfusion liver injury in the rat

Documents