Experimental and Molecular Pathology 92 (2012) 97–104

Contents lists available at SciVerse ScienceDirect

Experimental and Molecular Pathology

j ourna l homepage: www.e lsev ie r .com/ locate /yexmp

Attenuation of endoplasmic reticulum stress using the chemical chaperone4-phenylbutyric acid prevents cardiac fibrosis induced by isoproterenol

Pedro Ayala a, José Montenegro a, Raúl Vivar a, Alan Letelier a, Pablo Aránguiz Urroz a,Miguel Copaja a, Deisy Pivet a, Claudio Humeres a, Rodrigo Troncoso a, José Miguel Vicencio a,Sergio Lavandero a,b, Guillermo Díaz-Araya a,⁎a FONDAP CEMC, Centro de Estudios Moleculares de la Célula, Facultad de Ciencias Químicas y Farmacéuticas, Universidad de Chile, Santiago, Chileb Instituto de Ciencias Biomédicas, Facultad de Medicina, Universidad de Chile, Santiago, Chile

Abbreviations: ER, Endoplasmic reticulum; UPR, UnIsoproterenol; 4-PBA, 4-Phenylbutyric acid; ECM, Extrac⁎ Corresponding author at: Facultad de Ciencias Químic

de Chile, Sergio Livingstone 1007, Santiago 8380492, ChilE-mail address: gadiaz@ciq.uchile.cl (G. Díaz-Araya)

0014-4800/$ – see front matter © 2011 Elsevier Inc. Alldoi:10.1016/j.yexmp.2011.10.012

a b s t r a c t

a r t i c l e i n f o

Article history:Received 28 July 2011Available online 10 November 2011

Keywords:4-Phenylbutyric acidEndoplasmic reticulum stressHeartIsoproterenolFibrosis

Increasing evidence indicates that endoplasmic reticulum (ER) stress is involved in various diseases. Inthe human heart, ischemia/reperfusion has been correlated to ER stress, and several markers of the unfoldedprotein response (UPR) participate during cardiac remodeling and fibrosis. Here, we used isoproterenol (ISO)injection as a model for in vivo cardiac fibrosis. ISO induced significant cardiomyocyte loss and collagendeposition in the damaged areas of the endocardium. These responses were accompanied by an increase inthe protein levels of the luminal ER chaperones BIP and PDI, as well as an increase in the UPR effectorCHOP. The use of the chemical chaperone 4-phenylbutyric acid (4-PBA) prevented the activation of theUPR, the increase in luminal chaperones and also, leads to decreased collagen deposition, cardiomyocyteloss into the damaged zones. Our results suggest that cardiac damage and fibrosis induced in vivo by thebeta-adrenergic agonist ISO are tightly related to ER stress signaling pathways, and that increasing the ERluminal folding capacity with exogenously administrated 4-PBA is a powerful strategy for preventing thedevelopment of cardiac fibrosis. Additionally, 4-PBA might prevent the loss of cardiomyocytes. Our datasuggests that the attenuation of ER stress pathways with pharmacological compounds such as the chemicalchaperone 4-PBA can prevent the development of cardiac fibrosis and adverse remodeling.

© 2011 Elsevier Inc. All rights reserved.

Introduction

A variety of human diseases involve endoplasmic reticulum (ER)stress, a condition characterized by an accumulation of unfolded pro-teins in the ER lumen (Hotamisligil, 2010; Kaufman, 1999; Oyadomariet al., 2001). The activation of a mechanism known as the unfoldedprotein response (UPR) attempts to reduce the amount of misfoldedproteins by increasing the production of the ER chaperones such asBIP/Grp78, Grp94 and PDI, which optimize protein folding (Patil andWalter, 2001). However, the persistent accumulation of misfoldedproteins leads to cellular dysfunction and cell death. In the humanheart, myocardial ischemia is a severe trauma for cardiac cells. As aresult, ischemia causes extensive biochemical changes and one ofthem is stress affecting the ER (Scarabelli and Gottlieb, 2004). Severalstudies have correlated high levels of ER stress with myocardialdamage (Belmont et al., 2010; Brodsky, 2007; Glembotski, 2008;

folded protein response; ISO,ellular matrix.as y Farmacéuticas, Universidade. Fax: +56 2 978 2912..

rights reserved.

Okada et al., 2004; Thuerauf et al., 2006), whereas others suggestthat ER stress might protect the heart, and even foster the hypertrophicgrowth of themyocardium (Barnes and Smoak, 2000; Mao et al., 2006).Irrespective of such discrepancies, the involvement of ER stress incardiac fibrosis in particular, has not been extensively studied.

The experimental administration of isoproterenol (ISO) constitutesa well-established model for in-vivo study of an acute hyperadrenergicstate that it is accompanied by cardiac myocyte necrosis which hap-pens to lead to tissue repair and consequent fibrosis (Benjamin etal., 1989; Grimm et al., 1998; Rona, 1985). The pathogenesis of thecatecholamine-induced myocardial necrosis is multifactorial (Benjaminet al., 1989; Chagoya de Sánchez et al., 1997; Díaz-Muñoz et al., 2006;Rona, 1985). The necrosis is due to catecholamine-induced intracellularCa2+ overloading of cardiac myocytes, including their mitochondria.There follow the induction of oxidative stress and the opening ofthe mitochondrial inner membrane permeability transition pore withensuing organellar degeneration. Myocyte necrosis follows and isreplaced by collagen. An imbalance between synthesis/degradation ofextracellularmatrix (ECM) proteins finally results in excessive accumu-lation of fibrillar collagen.

High levels of type I collagen, the main fibrillar collagen found incardiac fibrosis, stiffen the ventricles and impede both contraction

98 P. Ayala et al. / Experimental and Molecular Pathology 92 (2012) 97–104

and relaxation, impairing the electrical coupling of cardiomyocytesand the global cardiac function (Swynghedauw, 1999). The synthesisof type I collagen is tightly regulated by ER chaperones; thereforeproper luminal folding of its triple helix is crucial for the generationof the cardiac ECM. The precursor procollagen is flanked by globularN- and C-terminal peptides, and a variety of different chaperone mol-ecules are involved in its folding. Some of them are general ER chap-erones (calnexin, BIP, GRP94 and PDI), whereas others are collagen-specific (HSP47 and prolyl-4-hydroxylase) (Lamande and Bateman,1999). Therefore, cardiac fibrosis apparently is tightly related withER folding capacity and an imbalance between synthesis and foldingcapacity could trigger ER stress and treatments leading to enhancedER folding capacity might restore proper collagen ECM balance andavoid cardiac fibrosis.

Sodium 4-phenylbutyrate (4-PBA) is a low-molecular weight fattyacid and a non-toxic pharmacological compound that is currentlyapproved for its clinical use in pathologic disorders of the urea cyclebecause of its properties as an ammonia scavenger (Maestri et al.,1996). In addition, 4-PBA is a weak histone deacetylase inhibitorand a transcriptional activator of β and γ globins (Perlmutter, 2002).4-PBA displays low toxicity and provides protection against variousnoxious stimuli (Ozcan et al., 2006), thus it has been proposed forthe treatment of cystic fibrosis, sickle cell disease and cancer (Doveret al., 1994; Goh et al., 2001; Zeitlin et al., 2002). Importantly, 4-PBAcan act as a chemical chaperone in the ER, because its physicochemicalproperties allow the stabilization of peptide structures, improvingthe luminal folding capacity and the traffic of aberrant proteins (deAlmeida et al., 2007; Vilatoba et al., 2005). Thus, the use of 4-PBAmay provide a therapeutic approach for blocking the pathologicprocess induced by ISO. However, no pharmacologic approach forthe treatment of cardiac fibrosis due to ER stress has been reportedto our knowledge. In this work, we aimed at this hypothesis and eval-uated the effect of the chemical chaperone 4-PBA upon cardiac dam-age, as well as the mechanisms underlying these effects in a modelof cardiac fibrosis induced by ISO. We present novel and interestingdata that supports the cardioprotective actions of increased bufferingof misfolded proteins on cardiac fibrosis.

Materials and methods

Reagents

The following reagents were acquired from Sigma Chemical Co(St. Louis, MO, USA): trypan blue, 4-phenylbutyric acid. Trypsin/EDTA, pre-stained molecular weight standard and fetal bovine serum(FBS) were purchased from Gibco BRL (Carlsbad, CA, USA). All organicand inorganic compounds were purchased from Merck (Darmstadt,Germany). The enhanced chemo-luminescence reagent was pur-chased from PerkinElmer Life Sciences, Inc (Boston, MA, USA). Theprimary antibody for PDI and BIP was purchased from Cell SignalingTechnology (Boston, MA, USA). CHOP antibody was acquired fromSanta Cruz Biotechnology (Santa Cruz, CA, USA).

Animals

Animal handlings were conducted according to the AnimalWelfare Regulations of the University of Chile. The investigationconformed to the Guide for the Care and Use of Laboratory Animalspublished by the National Institutes of Health (NIH publicationNo. 85–23, revised 1996).

Experimental groups

Cardiac fibrosis was induced by ISO administration as previouslydescribed (Grimm et al., 1998). Briefly, rats received a single subcuta-neous injection of (±) isoproterenol hemisulfate (ISO; Sigma, St.

Louis, MO), 50 mg/kg body weight. We recognize that small and re-petitive doses or isoproterenol delivery by osmotic pump is morefrequently used, however, with the single doses of isoproterenol wehad a rapid and strong cardiac damage. A matched group of controlrats received saline injection. The chemical chaperone (4-PBA sodiumsalt, 80 mg/kg, was dissolved in 0.9% saline) was administrated viasubcutaneous injection, 2 h before isoproterenol injection, and for10 days after. Controls were given similar volumes of saline. Animalswere killed at 9 h, 2, 5 and 10 days after ISO injection. On the day ofthe experiment, hearts were rapidly removed and washed in cold0.9% saline, freed of connective tissue, and placed on ice. The leftand right ventricles were separated, washed extensively with salineto remove all contaminating blood, dried, and weighed. The tissueswere quick-frozen in liquid nitrogen and stored at −80 °C until ana-lyzed for western blot. Hearts were also fixed in neutrally bufferedformalin for collagen and morphometric analysis.

Collagen deposition

A 2-mm thick coronal section was taken from the equator of eachheart and fixed in neutrally buffered formalin. Formalin-fixed sec-tions were dehydrated through a graded series of alcohol and xyleneand embedded in paraffin. Paraffin sections (5 mm thick) werestained with hematoxylin and eosin for histological evaluation. Sec-tions from the myocardium located at the equator of the left ventriclewere stained with Masson's Trichrome for determination of intersti-tial collagen. Each section was viewed in its entirety by a single inves-tigator who was unaware of the nature of the experimental groups.ISO-induced fibrosis was located primarily within the endocardium,with extensions of fibrous tissue reaching the midwall.

Soluble collagen quantification

For soluble collagen quantification, heart tissue homogenized ali-quots were taken and samples were treated with “soluble collagenassay” Sircol® (Biocolor, Ireland) according to the manufacturer'sinstructions.

Immunohistochemistry

Left ventricle (LV) sample tissue was fixed with silane, dried anddewaxed to water, and cuts were washed with TBS/Tween-20 (TBS-T)and blocked with 3% horse serum in TBST. The samples were incubatedwith anti-BIP 1:50, anti-PDI 1:50 and developed with VectastainPK7100 kit (Vector, California, USA). The cell nuclei are contrastedwith Mayer hematoxylin, washed with water and turned to blue andthe sectionswere dehydrated to a battery of ethanol, rinsedwith xyleneand permanently mounted with hydrophobic medium.

Preparation of total cell extract

Left ventricle sample was homogenized in cold lysis buffer(20 mM HEPES, pH 7.5, 150 mM NaCl, 1% Triton-X100, 10% glycerol,1 mM EDTA, 10 mM sodium diphosphate, 100 mM NaF, 17.5 mMbeta-glycerophosphate, 1 mM phenylmethylsulfonyl fluoride, leu-peptin 2 μg/mL; 10 mM aprotinin; 1 mM PMSF and 100 μM Na3VO4).Homogenized solutions were centrifuged at 10,000 RPM during15 min at 4 °C. Supernatants were recovered and protein concentra-tion was determined by a Bradford assay. Proteins were denaturizedinto SDS-PAGE buffer 4× (glycerol 20 mL, 2-mercaptoethanol 10 mL,SDS 5 g, Tris-base 1.51 g, bromophenol blue 0.01 g, water q. s. 100 mL,pH adjusted to 6.8). Samples were heated to 95 °C for 5 min beforeelectrophoresis.

99P. Ayala et al. / Experimental and Molecular Pathology 92 (2012) 97–104

Western blot

Lysates were then separated by 10% SDS-PAGE and samples weretransferred to nitrocellulose membranes (Bio-Rad). Membranes wereblocked with 5% dry milk in TBS-T and incubated with anti-BIP, anti-PDI, anti β-tubulin, and anti-CHOP antibodies. Membranes were in-cubated with HRP-conjugated anti-rabbit or anti-mouse IgG. Imageanalysis was performed using Image J (U.S. National Institutes ofHealth, Bethesda, MD, USA).

Statistical analysis

All data are expressed as means±S.D. The differences in eachparameter were evaluated by a one-way ANOVA of the increase ordecrease of each variable measured. One-way ANOVA was followed

Fig. 1. Cardiac fibrosis induced by isoproterenol is prevented by 4-PBA. Panel A) LV endomtreated with saline or isoproterenol ISO (50 mg/kg) at the indicated times, and then stainestains the nuclei. Panel B) LV endomyocardial histological samples were obtained from adthen were stained with trichrome masson staining to visualize collagen deposition (bluemale SD saline-treated rats or treated with ISO and/or 4-PBA (80 mg/kg) for 10 days. Samp(blue staining). Panel D) LV endomyocardial samples were obtained from adult male Spraguehomogenized. Soluble collagen was quantified from the tissue homogenized using the Sircol®(**pb0.01 vs. control, #pb0.05 vs. ISO group).

by the Tukey test to compare the effect of different conditions onthese parameters. Significance was accepted at pb0.05.

Results

4-PBA attenuates myocardial fibrosis induced by isoproterenol

To establish our cardiac injury and fibrosis method, we studied thecardiac histology of ISO-treated Sprague–Dawley adult male rats, in atime-dependent manner. After 9 h of ISO-treatment to the animals,there were no visible differences on cardiac tissue histology, however,from 2 days after injection we observed clear areas of tissue damagewith cardiomyocyte loss and their replacement by granulation tissue(enclosed area by arrow heads), which isolates cardiomyocytes (seeFig. 1A at 2 days (2d) and at 5 days (5d)). Granulation tissue was

yocardial histological samples were obtained from adult male Sprague–Dawley ratsd with hematoxylin and eosin (H/E). Pink color indicates cardiac myocytes, blue colorult male Sprague–Dawley rats treated with saline or ISO at the indicated times, andstaining). Panel C) LV endomyocardial histological samples were obtained from adultles were then stained with trichrome masson staining to visualize collagen deposition–Dawley saline-treated rats or treated with ISO and/or 4-PBA (80 mg/kg) for 10 days andmethod. Data are expressed as mean±SEM of at least three independent experiments

Fig. 2. Isoproterenol increases the cardiac expression of ER folding chaperones. Western blot samples were obtained from homogenized LV of adult male Sprague–Dawley ratstreated with saline or isoproterenol (ISO; 50 mg/kg) at the indicated times. Panel A) Representative blot with an anti-PDI antibody (upper part), and densitometry analysisof all performed blots (lower part). Panel B) Representative blot with an anti-BIP antibody (upper part), and densitometry analysis (lower part). Panel C) Immunohistochemistryanalysis of BIP and PDI from endomyocardial samples obtained from adult male Sprague–Dawley rats treated with saline or isoproterenol (ISO, 50 mg/kg) at the indicated times(***pb0.001 and *pb0.05 vs. controls).

100 P. Ayala et al. / Experimental and Molecular Pathology 92 (2012) 97–104

characterized by early cellular infiltration evidentiated by an increasein the nuclei density and later by extracellular matrix deposition. Inorder to evaluate if ISO-administration had an effect on the deposi-tion of interstitial collagen, we performed trichrome masson staining

on histology samples. Collagen deposition (revealed by blue staining)was higher in ISO groups than in controls from 5 days to 10 days aftertreatment (Fig. 1B). These results indicate that 10 days in our modelwas sufficient for inducing in vivo cardiac fibrosis. Therefore, in

101P. Ayala et al. / Experimental and Molecular Pathology 92 (2012) 97–104

order to evaluate the effects of 4-PBA on collagen deposition, westudied the cardiac tissue histology after 10 days of ISO-treatment incombination with 4-PBA (pretreated by 2 h and present during ISOtreatment). Importantly, 4-PBA pretreatment prevented cardiac dam-age induced by ISO, as well as the interstitial collagen deposition (seereduced blue staining in Fig. 1C). Collagen content was additionallyanalyzed by the picrosirius red method. By this approach, 4-PBAalso demonstrated to prevent the increase of collagen content in-duced by ISO at 10 days. No effect of 4-PBA alone on collagen contentwas observed (Fig. 1D).

Fig. 3. Effect of 4-PBA upon levels of cardiac ER folding chaperones. Western blot samples weredose of isoproterenol (ISO; 50 mg/kg/), and/or 4-PBA (80 mg/kg/10 days) as indicated. Panel Aall performed blots (lower part). Panel B) Representative blot with an anti-BIP antibody (upperBIP and PDI from endomyocardial samples obtained from adult male Sprague–Dawley rats tr(*pb0.05 vs. controls; #pb0.01 and ##pb0.01 vs. ISO group).

ISO-treatment leads to increased levels of ER stress markers BIP and PDI

We evaluated the protein levels of PDI and BIP after ISO adminis-tration by western blot analysis, as well as by immunohistochemistryfor their cardiac tissue localization. At 5 and 10 days of ISO treatment,PDI protein levels in the hearts of the treated rats were significantlyhigher than in control animal; however BIP protein levels were signif-icantly increased only after 10 days (Figs. 2A–B). By immunohisto-chemistry, we found that ISO induces an increase in the expressionof PDI and BIP. However, this technique provided more information

obtained from homogenized LV of adult male Sprague–Dawley rats treated with a single) Representative blot with an anti-PDI antibody (upper part), and densitometry analysis ofpart), and densitometry analysis (lower part). Panel C) Immunohistochemistry analysis ofeated with isoproterenol (ISO; 50 mg/kg) and/or 4-PBA (80 mg/kg/10 days) as indicated

Fig. 4. Effect of 4-PBA upon ER stress. Western blot samples were obtained from homoge-nized LV of adult male Sprague–Dawley rats treated with isoproterenol (ISO; 50 mg/kg)and/or 4-PBA (80 mg/kg). Panel A) Representative blot with an anti-CHOP antibody(upper part), and densitometry analysis of all performed blots (lower part) of ISO treatedrats at indicated times. Panel B) Representative blot with an anti-CHOP antibody (upperpart), and densitometry analysis (lower part) of rats pretreated with 4-PBA and ISO by9 h (***pb0.001 vs. control; **pb0.05 vs. control and #pb0.05 vs. ISO group).

102 P. Ayala et al. / Experimental and Molecular Pathology 92 (2012) 97–104

about the distribution of both markers within the samples. From day2, PDI was initially located in granulation tissue within the area ofdamaged tissue. However, from day 5 it was also located in neighbor-ing cardiomyocytes to the damaged area, while at day 10 it waslocated mostly in the healing site (see arrow heads in Fig. 2C). Con-cerning the protein BIP, we observed a significant increase at day10, the staining was more highly pronounced in the healing site(see arrow heads in Fig. 2C). Therefore, the data from immunohisto-chemistry experiments correlates with the data from western blots,indicating that there is a delayed increase of BIP with respect to PDIat 5 days, but at 10 days both increases were significant, and locatedin the myocardial injury area.

4-PBA attenuates cardiac PDI and BIP levels after ISO treatment

Based on the observation that 4-PBA prevents the cardiac collagendeposition induced by ISO in the cardiac damaged area, at 10 days oftreatment (maximum increase in both proteins), we also monitoredthe effect of 4-PBA administration on PDI and BIP expression levels.4-PBA treatment abolished the increase on PDI and BIP protein levelsobserved after 10 days of ISO treatment (Figs. 3A–B). Of note, 4-PBAby itself did not significantly modify PDI and BIP expression levels.After 10 days of treatment, our immunohistochemistry results showedthat 4-PBA prevented the increase in PDI, as well as in BIP immunore-activity induced by ISO (Fig. 3C). These results strongly suggest that4-PBA attenuates the need of the UPR to induce the increasedexpression of the chaperones BIP and PDI; therefore, 4-PBA mightbe acting upstream ER stress pathways thus preventing the activa-tion of the UPR.

Effect of ISO on CHOP expression levels is prevented by 4-PBA

In order to evaluate a downstream effector of the UPR we moni-tored CHOP protein levels by western blot. At 9 h of treatment ofthe animals with ISO, we observed a significant increase in CHOP pro-tein levels, which returned to control levels at 2 days, and were main-tained until day 10 (Fig. 4A). In concordance with our previousresults, the administration of 4-PBA prevented the increase in CHOPexpression induced by ISO. Of note, the treatment with 4-PBA aloneinduced no significant changes on CHOP protein levels (Fig. 4B),which is consistent with the general lack of adverse effects found onthe literature for this compound.

Discussion

We demonstrate that ISO, a non-selective beta-adrenergic agonistleads to cardiac fibrosis in vivo, and the dose used was in according tothose described before (Feng and Li, 2010; Grimm et al., 1998). Theappearance of granulation tissue correlated with increased collagendeposits on the injured areas of the endocardium and importantly,ISO induced the expression of PDI and BIP, and both proteins reachedthe highest level of expression at 10 days of treatment, coincidingwith the presence of cardiac fibrosis. Via immunohistochemistry,our results show that the expression of PDI and BIP was dependenton both time of ISO treatment, and the cell type. PDI increased fromday 2, initially localized in the granulation tissue, probably in infiltrat-ing cells, while at day 5 it was located in neighboring cardiac myo-cytes to the area of tissue damage. These results suggest that theearly expression of PDI could be linked to ER stress, since at thistiming cardiac fibrosis has not yet been fully developed. Supportingthis, there is evidence that PDI is expressed in infarcted zones inhuman hearts and not in non-ischemic areas (Severino et al., 2007).In the same work, these authors showed that over expression of PDIis protective against cardiomyocyte apoptosis (Severino et al., 2007).Based on literature, PDI expression is most likely to be protectiveagainst cardiac damage, as its cytoprotective effect has been described

via an interaction with the redox-balancing protein SOD (Becker et al.,1999), preventing post-ischemic insults in the heart and diminishinginflammation and apoptosis (Qi et al., 2004). However, from our resultswe cannot conclude that this increased expression could be linked toprotection of these cells. On the other hand, the expression levels ofBIP increased late with respect to those of PDI, and were locatedmainlyin zones of tissue damage and cardiac fibrosis, and not in cardiac myo-cytes. These results suggest that BIP levels could be increased due tothe enhanced synthesis of collagen, which correlates with its functionas specific chaperone in collagen synthesis (Lamande and Bateman,1999).

In the ER stress signaling cascade, the chaperones BIP and PDI arefirst elements, which means that these chaperones participate in pro-tein folding and consequently, they are responsible of activating thethree UPR sensors PERK, ATF6 and IRE1 (Maattanen et al., 2010).One common element after the activation of these three sensors isthe expression of the protein CHOP, which has been documentedto mediate apoptosis after ER stress and to participate in severalER-stress related diseases (Oyadomari and Mori, 2004). Our resultsshowed an early and transitory increase of CHOP, a common effectorof the main three signaling pathways of the UPR. At 9 h of treatments,the protein CHOP, reached its maximal levels, returning to baseline

103P. Ayala et al. / Experimental and Molecular Pathology 92 (2012) 97–104

from day 2. This result agrees with the work of Szegezdi et al. (2006),who described that ischemia/reperfusion in vitro leads to increasedCHOP levels in cardiac myocytes, responsible of triggering apoptosis.Our results strongly suggest that the UPR is activated early in theheart after ISO treatment, leading to the expression of CHOP and acti-vation of transcriptional responses to upgrade the buffering status ofmisfolded proteins in the lumen of the ER by BIP and PDI expression.Additionally, CHOP might be involved in the activation of apoptoticsignaling, responsible of the cardiac loss observed from day 2 of treat-ments. Collectively, these results confirm that ER stress and UPR areinduced in the heart of ISO-treated rats. In the heart, in vivo modelshave demonstrated that ischemia/reperfusion induces ER stress (Kimet al., 2008; Liu et al., 2008; Martindale et al., 2006; Qi et al., 2007). Sim-ilarly, heart ischemia activates mechanisms that induce ER stress andthe UPR (Martindale et al., 2006; Thuerauf et al., 2006). ER stress inthe heart has been associated with discordant results. Some studiesshow an association between high ER stress with myocardial damage.For instance, ER stress is activated in chronic inflammatory damage inthe hearts of transgenic mice that overexpress MCP-1 protein, leadingto heart failure (Azfer et al., 2006). In contrast, other researchers sug-gest that ER stressmay protect the heart, and even promotemyocardialhypertrophic growth (Nickson et al., 2007; Toth et al., 2006). However,our results agree with the first notion, showing that the damage causedby the administration of ISO, and thus the development of cardiacfibrosis is strongly associated with ER stress.

The dose of 4-PBA that was used in this work is within the normalrange of use of this chaperone. The dose for models in vivo rangesfrom 20 to 120 mg/kg/day (Qi et al., 2004; Vilatoba et al., 2005).Additionally, this drug has been described to have little adverseeffects, and to be protective against several diseases, which is consis-tent with the cardioprotective role that we observed here upon ISO-induced cardiac fibrosis (de Almeida et al., 2007; Goh et al., 2001;Ozcan et al., 2006; Qi et al., 2004; Rishikof et al., 2004; Vilatoba etal., 2005). Via immunohistochemistry, our results show that 4-PBAprevented the increase in PDI expression that was observed in thefibrotic and damaged areas, as well as in neighboring cardiomyocytes.This could indicate that in the presence of 4-PBA, ISO leads to a lowerdegree of ER stress, as evidenced by the lower expression of PDI.This decreased expression of luminal chaperones correlates withdecreased expression of the UPR effector CHOP, which might contrib-ute to explain the lack of cell death and cardiac cell loss. This resultagrees with those indicating that 4-PBA prevents the activation ofthe UPR and the expression of CHOP, increasing the resistance tothe apoptotic effects of ER stress (Qi et al., 2004). It is possiblethat under enhanced ER luminal conditions of protein buffering andfolding; there is no need of activating UPR-mediated defense mecha-nisms of survival, or further apoptotic signaling. In this context, ithas been described that in a model of cerebral ischemia, 4-PBA(120 mg/kg) decreases the size of infarcted area (Qi et al., 2004). Inthe same study, it was determined by TUNEL that in the brains ofrats treated with 4-PBA, a significant decrease in the number of apo-ptotic cells in the damaged region was observed. Similarly, there wasa decrease in all markers of endoplasmic reticulum stress, suggestingthat this protection is due to the ability of 4-PBA to act as a chemicalchaperone and to modulate the UPR. On the other hand, our resultsshow that in the damaged areas the decrease in the expression ofPDI was linked to a reduction in the secretion of collagen. Thus, inaddition to the loss of cardiac myocytes induced by ISO, the furtherreplacement of the scarred tissue by collagen fibers was reduced bythe treatment with 4-PBA.

4-PBA is a histone deacetylase inhibitor, and this class of drugs hasshown protective effects against fibrotic disorders (Pang and Zhuang,2010), cardiac fibrosis and remodeling (Iyer et al., 2010; Tsung-Minget al., 2007), and recent reports have shown renoprotective effectsof 4-PBA, by suppressing oxidative stress by attenuating endoplasmicreticulum (ER) stress (Luo et al., 2010). The protective effect of 4-PBA

upon cardiac damage has been shown in the heart against adriamycintoxicity in a mouse model (Daosukho et al., 2007). However, there arenot published data about other effects of 4-PBA on cardiovascular func-tion and structure. Thus, the results that we observe here can beexplained by several hypotheses, including: a) 4-PBA can avoid the ac-tivation of ER stress and UPR through increased folding capacity, as itwas able to prevent the expression of luminal chaperones required forthe synthesis of collagen (PDI) and CHOP expression level, and b) 4-PBA is able to decrease the production of collagen in fibroblasts. Accord-ingly, Rishikof et al. (2004), showed that 4-PBA reduced the secretion ofcollagen in lung fibroblasts, and also it shows antifibrotic effects (Pangand Zhuang, 2010). c) Finally, because 4-PBA could be a protective fac-tor against cell death by necrosis/apoptosis ISO-induced, by decreasingoxidative stress, the decrease in collagen deposition may also be due tolower damaged areas to repair. Accordingly, it has been previouslyshown that ISO induces death by necrosis and apoptosis in heart tissue(Benjamin et al., 1989; Galvez et al., 2005; Shizukuda et al., 1998). Over-all, our results show that cardiac fibrosis induced by ISO is likely to bemediated by ER stress. Therefore, the chemical chaperone 4-PBA pre-vents the harmful effects of cardiac fibrosis at the level of ER stressand activation of the UPR, at the level of collagen secretion and at thelevel of cardiomyocyte loss.

Conclusions

In concordance with the involvement of ER stress induced by ISO,the in vivo administration of 4-PBA, a chemical chaperone thatimproves the ER luminal folding capacity, completely suppressed car-diac fibrosis, collagen deposition, activation of ER stress and cardiacmyocyte loss. These results indicate that ER stress is deeply connectedto cardiac damage, and that the prevention of stress affecting the ER isa powerful strategy for new clinical approaches for the treatment ofcardiac fibrosis.

Conflict of interest statement

The authors declare that there are no conflicts of interest.

Acknowledgments

P.A. holds a doctoral fellowship from MECESUP, Chile. R.V. andP.A.U. hold a doctoral fellowship from CONICYT, Chile. R.T. is a FONDAPpost-doctoral fellow. This work was supported by FONDECYT grant1100443 to G.D.A., as well as, by FONDAP grant 15010006 to S.L.

References

Azfer, A., Niu, J., Rogers, L.M., Adamski, F.M., Kolattukudy, P.E., 2006. Activation ofendoplasmic reticulum stress response during the development of ischemic heartdisease. The American Journal of Physiology 291, H1411–H1420.

Barnes, J.A., Smoak, I.W., 2000. Glucose-regulated protein 78 (GRP78) is elevated inembryonic mouse heart and induced following hypoglycemic stress. Anatomyand Embryology 202, 67–74.

Becker, L.B., vanden Hoek, T.L., Shao, Z.H., Li, C.Q., Schumacker, P.T., 1999. Generation ofsuperoxide in cardiomyocytes during ischemia before reperfusion. The AmericanJournal of Physiology 277, H2240–H2246.

Belmont, P.J., Chen, W.J., San Pedro, M.N., Thuerauf, D.J., Gellings Lowe, N., Gude, N.,Hilton, B., Wolkowicz, R., Sussman, M.A., Glembotski, C.C., 2010. Roles for endo-plasmic reticulum-associated degradation and the novel endoplasmic reticulumstress response gene Derlin-3 in the ischemic heart. Circulation Research 106,307–316.

Benjamin, I.J., Jalil, J.E., Tan, L.B., Cho, K., Weber, K.T., Clark, W.A., 1989. Isoproterenol-induced myocardial fibrosis in relation to myocyte necrosis. Circulation Research65, 657–670.

Brodsky, J.L., 2007. The protective and destructive roles played by molecular chaperonesduring ERAD (endoplasmic-reticulum-associated degradation). The BiochemicalJournal 404, 353–363.

Chagoya de Sánchez, V., Hernández-Muñoz, R., López-Barrera, F., Yañez, L., Vidrio, S.,Suárez, J., Cota-Garza, M.D., Aranda-Fraustro, A., Cruz, D., 1997. Sequential changesof energy metabolism andmitochondrial function in myocardial infarction inducedby isoproterenol in rats: a long-term and integrative study. Canadian Journal ofPhysiology and Pharmacology 75, 1300–1311.

104 P. Ayala et al. / Experimental and Molecular Pathology 92 (2012) 97–104

Daosukho, C., Chen, Y., Noel, T., Sompol, P., Nithipongvanitch, R., Velez, J.M., Oberley,T.D., St Clair, D.K., 2007. Phenylbutyrate, a histone deacetylase inhibitor, protectsagainst adriamycin-induced cardiac injury. Free Radical Biology & Medicine 15,1818–1825.

de Almeida, S.F., Picarote, G., Fleming, J.V., Carmo-Fonseca, M., Azevedo, J.E., de Sousa,M., 2007. Chemical chaperones reduce endoplasmic reticulum stress and preventmutant HFE aggregate formation. The Journal of Biological Chemistry 282,27905–27912.

Díaz-Muñoz, M., Alvarez-Pérez, M.A., Yáñez, L., Vidrio, S., Martínez, L., Rosas, G., Yáñez,M., Ramírez, S., de Sánchez, V.C., 2006. Correlation between oxidative stress andalteration of intracellular calcium handling in isoproterenol-induced myocardialinfarction. Molecular and Cellular Biochemistry 289, 125–136.

Dover, G.J., Brusilow, S., Charache, S., 1994. Induction of fetal hemoglobin production insubjects with sickle cell anemia by oral sodium phenylbutyrate. Blood 84, 339–343.

Feng, W., Li, W., 2010. The study of ISO induced heart failure rat model. Experimentaland Molecular Pathology 88, 299–304.

Galvez, A.S., Fiedler, J.L., Ocaranza, M.P., Jalil, J.E., Lavandero, S., Diaz-Araya, G., 2005.Perindopril regulates beta-agonist-induced cardiac apoptosis. Journal of CardiovascularPharmacology 46, 255–261.

Glembotski, C.C., 2008. The role of the unfolded protein response in the heart. Journalof Molecular and Cellular Cardiology 44, 453–459.

Goh, M., Chen, F., Paulsen, M.T., Yeager, A.M., Dyer, E.S., Ljungman, M., 2001. Phenylbu-tyrate attenuates the expression of Bcl-X(L), DNA-PK, caveolin-1, and VEGF inprostate cancer cells. Neoplasia 3, 331–338.

Grimm, D., Elsner, D., Schunkert, H., Pfeifer, M., Griese, D., Bruckschlegel, G., Muders, F.,Riegger, G.A., Kromer, E.P., 1998. Development of heart failure following isoproter-enol administration in the rat: role of the renin–angiotensin system. CardiovascularResearch 37, 91–100.

Hotamisligil, G.S., 2010. Endoplasmic reticulum stress and the inflammatory basis ofmetabolic disease. Cell 140, 900–917.

Iyer, A., Fenning, A., Lim, J., Le, G.T., Reid, R.C., Halili, M.A., Fairlie, D.P., Brown, L., 2010.Antifibrotic activity of an inhibitor of histone deacetylases in DOCA-salt hyperten-sive rats. British Journal of Pharmacology 159, 1408–1417.

Kaufman, R.J., 1999. Stress signaling from the lumenof the endoplasmic reticulum: coordina-tion of gene transcriptional and translational controls. Genes & Development 13,1211–1233.

Kim, D.S., Ha, K.C., Kwon, D.Y., Kim,M.S., Kim, H.R., Chae, S.W., Chae, H.J., 2008. Kaempferolprotects ischemia/reperfusion-induced cardiac damage through the regulation ofendoplasmic reticulum stress. Immunopharmacology and Immunotoxicology 30,257–270.

Lamande, S.R., Bateman, J.F., 1999. Procollagen folding and assembly: the role of endo-plasmic reticulum enzymes and molecular chaperones. Seminars in Cell & Devel-opmental Biology 10, 455–464.

Liu, X.H., Zhang, Z.Y., Sun, S., Wu, X.D., 2008. Ischemic postconditioning protects myo-cardium from ischemia/reperfusion injury through attenuating endoplasmic retic-ulum stress. Shock 30, 422–427.

Luo, Z.F., Feng, B., Mu, J., Qi, W., Zeng, W., Guo, Y.H., Pang, Q., Ye, Z.L., Liu, L., Yuan, F.H.,2010. Effects of 4-phenylbutyric acid on the process and development of diabeticnephropathy induced in rats by streptozotocin: regulation of endoplasmic reticu-lum stress-oxidative activation. Toxicology and Applied Pharmacology 246, 49–57.

Maattanen, P., Gehring, K., Bergeron, J.J., Thomas, D.Y., 2010. Protein quality control inthe ER: the recognition of misfolded proteins. Seminars in Cell & DevelopmentalBiology 21, 500–511.

Maestri, N.E., Brusilow, S.W., Clissold, D.B., Bassett, S.S., 1996. Long-term treatment ofgirls with ornithine transcarbamylase deficiency. The New England Journal ofMedicine 335, 855–859.

Mao, C., Tai, W.C., Bai, Y., Poizat, C., Lee, A.S., 2006. In vivo regulation of Grp78/BiP tran-scription in the embryonic heart: role of the endoplasmic reticulum stress re-sponse element and GATA-4. The Journal of Biological Chemistry 281, 8877–8887.

Martindale, J.J., Fernandez, R., Thuerauf, D., Whittaker, R., Gude, N., Sussman, M.A.,Glembotski, C.C., 2006. Endoplasmic reticulum stress gene induction and protec-tion from ischemia/reperfusion injury in the hearts of transgenic mice with atamoxifen-regulated form of ATF6. Circulation Research 98, 1186–1193.

Nickson, P., Toth, A., Erhardt, P., 2007. PUMA is critical for neonatal cardiomyocyte apoptosisinduced by endoplasmic reticulum stress. Cardiovascular Research 73, 48–56.

Okada, K., Minamino, T., Tsukamoto, Y., Liao, Y., Tsukamoto, O., Takashima, S., Hirata,A., Fujita, M., Nagamachi, Y., Nakatani, T., Yutani, C., Ozawa, K., Ogawa, S.,

Tomoike, H., Hori, M., Kitakaze, M., 2004. Prolonged endoplasmic reticulum stressin hypertrophic and failing heart after aortic constriction: possible contributionof endoplasmic reticulum stress to cardiac myocyte apoptosis. Circulation 110,705–712.

Oyadomari, S., Mori, M., 2004. Roles of CHOP/GADD153 in endoplasmic reticulumstress. Cell Death and Differentiation 11, 381–389.

Oyadomari, S., Takeda, K., Takiguchi, M., Gotoh, T., Matsumoto, M., Wada, I., Akira,S., Araki, E., Mori, M., 2001. Nitric oxide-induced apoptosis in pancreatic betacells is mediated by the endoplasmic reticulum stress pathway. Proceedingsof the National Academy of Sciences of the United States of America 98,10845–10850.

Ozcan, U., Yilmaz, E., Ozcan, L., Furuhashi, M., Vaillancourt, E., Smith, R.O., Gorgun, C.Z.,Hotamisligil, G.S., 2006. Chemical chaperones reduce ER stress and restore glucosehomeostasis in a mouse model of type 2 diabetes. Science 313, 1137–1140.

Pang, M., Zhuang, S., 2010. Histone deacetylase: a potential therapeutic target forfibrotic disorders. The Journal of Pharmacology and Experimental Therapeutics335, 266–272.

Patil, C., Walter, P., 2001. Intracellular signaling from the endoplasmic reticulum to thenucleus: the unfolded protein response in yeast and mammals. Current Opinion inCell Biology 13, 349–355.

Perlmutter, D.H., 2002. Chemical chaperones: a pharmacological strategy for disordersof protein folding and trafficking. Pediatric Research 52, 832–836.

Qi, X., Hosoi, T., Okuma, Y., Kaneko, M., Nomura, Y., 2004. Sodium 4-phenylbutyrateprotects against cerebral ischemic injury. Molecular Pharmacology 66, 899–908.

Qi, X., Vallentin, A., Churchill, E., Mochly-Rosen, D., 2007. deltaPKC participates in theendoplasmic reticulum stress-induced response in cultured cardiac myocytes andischemic heart. Journal of Molecular and Cellular Cardiology 43, 420–428.

Rishikof, D.C., Ricupero, D.A., Liu, H., Goldstein, R.H., 2004. Phenylbutyrate decreases type Icollagen production in human lung fibroblasts. Journal of Cellular Biochemistry 91,740–748.

Rona, G., 1985. Catecholamine cardiotoxicity. Journal of Molecular and CellularCardiology 17, 291–306.

Scarabelli, T.M., Gottlieb, R.A., 2004. Functional and clinical repercussions of myocyteapoptosis in the multifaceted damage by ischemia/reperfusion injury: old andnew concepts after 10 years of contributions. Cell Death and Differentiation 11,S144–S152.

Severino, A., Campioni, M., Straino, S., Salloum, F.N., Schmidt, N., Herbrand, U., Frede, S.,Toietta, G., Di Rocco, G., Bussani, R., Silvestri, F., Piro, M., Liuzzo, G., Biasucci, L.M.,Mellone, P., Feroce, F., Capogrossi, M., Baldi, F., Fandrey, J., Ehrmann, M., Crea, F.,Abbate, A., Baldi, A., 2007. Identification of protein disulfide isomerase as a cardio-myocyte survival factor in ischemic cardiomyopathy. Journal of the AmericanCollege of Cardiology 50, 1029–1037.

Shizukuda, Y., Buttrick, P.M., Geenen, D.L., Borczuk, A.C., Kitsis, R.N., Sonnenblick, E.H.,1998. Beta-adrenergic stimulation causes cardiocyte apoptosis: influence of tachy-cardia and hypertrophy. The American Journal of Physiology 275, H961–H968.

Swynghedauw, B., 1999. Molecular mechanisms of myocardial remodeling. PhysiologicalReviews 79, 215–262.

Szegezdi, E., Duffy, A., O'Mahoney, M.E., Logue, S.E., Mylotte, L.A., O'Brien, T., Samali, A.,2006. ER stress contributes to ischemia-induced cardiomyocyte apoptosis. Biochemicaland Biophysical Research Communications 349, 406–1411.

Thuerauf, D.J., Marcinko, M., Gude, N., Rubio, M., Sussman, M.A., Glembotski, C.C., 2006.Activation of the unfolded protein response in infarcted mouse heart and hypoxiccultured cardiac myocytes. Circulation Research 99, 275–282.

Toth, A., Jeffers, J.R., Nickson, P., Min, J.Y., Morgan, J.P., Zambetti, G.P., Erhardt, P., 2006.Targeted deletion of Puma attenuates cardiomyocyte death and improves cardiacfunction during ischemia–reperfusion. The American Journal of Physiology 291,H52–H60.

Tsung-Ming, L., Mei-Shu, L., Nen-Chung, C., 2007. Inhibition of histone deacetylase onventricular remodeling in infarcted rats. The American Journal of Physiology 293,H968–H977.

Vilatoba, M., Eckstein, C., Bilbao, G., Smyth, C.A., Jenkins, S., Thompson, J.A., Eckhoff,D.E., Contreras, J.L., 2005. Sodium 4-phenylbutyrate protects against liver ischemiareperfusion injury by inhibition of endoplasmic reticulum-stress mediated apopto-sis. Surgery 138, 342–351.

Zeitlin, P.L., Diener-West, M., Rubenstein, R.C., Boyle, M.P., Lee, C.K., Brass-Ernst, L.,2002. Evidence of CFTR function in cystic fibrosis after systemic administration of4-phenylbutyrate. Molecular Therapy 6, 119–126.