RESEARCH ARTICLE

Multidrug Resistance-Associated Protein 4Expression in Ammonia-Treated Cultured

Rat Astrocytes and Cerebral Cortex ofCirrhotic Patients with Hepatic

Encephalopathy

Markus S. J€ordens, Verena Keitel, Ayse Karababa, Irina Zemtsova, Holger Bronger,

Dieter H€aussinger, and Boris G€org

Hepatic encephalopathy (HE) is a neuropsychiatric syndrome frequently accompanying liver cirrhosis and reflects the clinicalmanifestation of a low grade cerebral edema associated with cerebral oxidative/nitrosative stress. The multidrug resistance-associated protein (Mrp) 4 is an export pump which transports metabolites that were recently suggested to play a major rolein the pathogenesis of HE such as neurosteroids and cyclic nucleotides. We therefore studied Mrp4 expression changes inammonia-exposed cultured astrocytes and postmortem human brain samples of cirrhotic patients with HE. NH4Cl increasedMrp4 mRNA and protein levels in astrocytes in a dose- and time-dependent manner up to threefold after 72 h of exposureand concurrently inhibited N-glycosylation of Mrp4 protein. Upregulation of Mrp4 mRNA and protein as well as impaired N-glycosylation of Mrp4 protein by ammonia were sensitive towards the glutamine-synthetase inhibitor L-methionine-S-sulfoxi-mine and were not induced by CH3NH3Cl (5 mmol/L). Upregulation of Mrp4 mRNA required ammonia-induced activation ofnitric oxide synthases or NADPH oxidase and p38MAPK-dependent activation of PPARa. Inhibition of Mrp4 by ceefourin 1 syn-ergistically enhanced both, inhibition of astrocyte proliferation as well as transcription of the oxidative stress surrogate markerheme oxygenase 1 by forskolin (10 mmol/L, 72 h) or NH4Cl (5 mmol/L, 72 h) in cultured rat astrocytes. Increased Mrp4 mRNAand protein levels were also found in postmortem brain samples from patients with liver cirrhosis with HE but not in thosewithout HE. The data show that Mrp4 is upregulated in HE, which may be relevant for the handling of neurosteroids andcyclic nucleotides in response to ammonia.

GLIA 2015;63:2092–2105Key words: hepatic encephalopathy, Mrp4, ammonia, glutamine, astrocytes

Introduction

Hepatic encephalopathy (HE) defines a neuropsychiatric

syndrome accompanying acute and chronic liver disease.

Currently, HE is seen as a clinical manifestation of a low-

grade cerebral edema (H€aussinger et al., 1994, 2000) which is

associated with oxidative/nitrosative stress (G€org et al.,

2013b; H€aussinger and G€org, 2010). The increased formation

of reactive oxygen and nitrogen species (RNOS) triggers a

series of functional consequences such as tyrosine nitration of

proteins (Schliess et al., 2002), oxidation of RNA (G€org

et al., 2008), activation of Zn21-dependent gene transcription

(Kruczek et al., 2009) and astrocyte senescence (G€org et al.,

2015). Such alterations may affect synaptic plasticity, impair

neurotransmission and disturb oscillatory networks in the

brain, which finally accounts for symptoms of HE (G€org

et al., 2013b; Timmermann et al., 2003).

View this article online at wileyonlinelibrary.com. DOI: 10.1002/glia.22879

Published online June 23, 2015 in Wiley Online Library (wileyonlinelibrary.com). Received Mar 16, 2015, Accepted for publication June 8, 2015.

Address correspondence to Dieter H€aussinger, Universit€atsklinikum D€usseldorf, Klinik f€ur Gastroenterologie, Hepatologie und Infektiologie, Moorenstrasse 5, D-

40225 D€usseldorf, Germany. E-mail: haeussin@uni-duesseldorf.de

From the Clinic for Gastroenterology, Hepatology, and Infectious Diseases, Heinrich-Heine University, D€usseldorf, Germany

Additional Supporting Information may be found in the online version of this article.

2092 VC 2015 Wiley Periodicals, Inc.

The multidrug resistance-associated protein (Mrp) 4 is a

member of the ATP-binding cassette subfamily C (Abcc) and

a transmembrane export pump which transports a variety of

substrates such as organic anions, cyclic adenosine/guanosine

monophosphate (cAMP/cGMP), steroids, glutathione, prosta-

noids, and others (Russel et al., 2008). Mrp4 is particularly

expressed in tissues with barrier function, such as liver, intes-

tine, kidneys, and placenta (Russel et al., 2008). In brain,

Mrp4 is strongly expressed in brain capillaries at the luminal

membrane of endothelial cells constituting the blood brain

barrier, at the blood cerebrospinal fluid barrier and in astro-

cytes (Nies et al., 2004). The physiological functions of Mrp4

in brain are poorly understood but may involve export of tox-

ins (Banerjee et al., 2014), control of intracellular cyclic

nucleotide levels (Chen et al., 2001) and clearance of PGE2

from brain parenchyma across the blood brain barrier into

blood vessels (Akanuma et al., 2010).

Interestingly, a recent transcriptome analysis on human

postmortem brain tissue revealed that upregulated genes in

patients with liver cirrhosis and HE are significantly enriched

for associations with cellular responses to toxins (G€org et al.,

2013a). Moreover, several Mrp4 substrates were suggested to

play a major role in the pathophysiology of HE such as neu-

rosteroids (Ahboucha et al., 2005, 2006; Keitel et al., 2010),

glutathione (Hilgier et al., 2010), cyclic guanosine monophos-

phate (cGMP) (Hermenegildo et al., 2000; Montoliu et al.,

2010) or prostanoids (Br€uck et al., 2011; Chung et al., 2001;

G€org et al., 2010a; Lachmann et al., 2013; Rodrigo et al.,

2010). We therefore studied the effects of ammonia on Mrp4

expression in cultured astrocytes and Mrp4 expression in post-

mortem human brain samples from patients with liver cirrho-

sis with or without hepatic encephalopathy. The results

suggest that ammonia activates transcription of Abcc4 and ele-

vates Mrp4 mRNA levels in a L-methionine-S-sulfoximine-,

oxidative/nitrosative stress-, p38MAPK- and PPARa-dependent

manner in cultured astrocytes and concurrently upregulates

Mrp4 protein levels and inhibits Mrp4 N-glycosylation. Phar-

macological inhibition of Mrp4 enhanced upregulation of the

oxidative stress surrogate marker heme oxygenase 1 as well as

inhibition of astrocyte proliferation by ammonia and forsko-

lin. Increased Mrp4 mRNA and protein levels were also

found in postmortem brain samples from patients with liver

cirrhosis and HE but not in those without HE.

Materials and Methods

MaterialsPD98056 was from Calbiochem (Merck Millipore, Darmstadt, Ger-

many). SP600125 was from Tocris Biosciences (Wiesbaden-Norden-

stadt, Germany). Rat monoclonal antibodies against Mrp4 were

from Enzo (Life Sciences L€orrach, Germany). L-methionine-S-sulfox-

imine (MSO), GW6471 (N-((2S)-2-(((1Z)-1-Methyl-3-oxo-3-(4-

(trifluoromethyl) phenyl)prop-1-enyl)amino)-3-(4-(2-(5-methyl-2-

phenyl-1,3-oxazol-4-yl)ethoxy)phenyl) propyl)propanamide), Nx-

nitro-L-arginine methyl ester (L-NAME), uric acid, rabbit polyclonal

antibodies against PPARa, goat polyclonal antibodies against Mrp4

and saponin were from Sigma-Aldrich (Deisenhofen, Germany).

NH4Cl, SB203580 (4-(4’-fluorophenyl)-2-(4’-methylsulfinylphenyl)- 5-

(4’-pyridyl)-imidazole) and CH3NH3Cl were from Merck-Millipore

(Darmstadt, Germany). Cell culture media and Hoechst34580 were

from Life Technologies (Invitrogen, Karlsruhe, Germany). Fetal calf

serum was from LONZA GmbH (Cologne, Germany). The monoclo-

nal antibody against 3-glycerophosphate-dehydrogenase (GAPDH) was

from Biodesign International (Saco, USA). HRPOD-coupled secondary

antibodies against rat and mouse IgG were from DAKO international

(Hamburg, Germany) or BioRad Laboratories (Munich, Germany),

respectively. The monoclonal antibody against 78 kDa glucose-regulated

protein (Grp78) was from BD/Transduction Laboratories (Heidelberg,

Germany). Monoclonal antibodies against 8OH(d)G were from QED

Biosciences (San Diego, CA). Monoclonal antibodies against actin were

from Abcam (Cambridge, UK).

Preparation, Culture, and Experimental Treatmentof Rat AstrocytesAstrocytes were prepared from rat cerebrocortical hemispheres of

newborn Wistar rats (E1-3) as described in Schliess et al. (2002) and

cultured for 4 to 8 weeks before experiments were carried out as

described. When indicated, astrocytes were pre-treated with L-methi-

onine-S-sulfoximine (MSO, 3 mmol/L), apocynin (300 mmol/L),

Nx-nitro-L-arginine methyl ester (L-NAME, 1 mmol/L), SB203580

(10 mmol/L), SP600125 (100 mmol/L), or GW6471 (10 mmol/L)

for 30 min before cells were exposed to NH4Cl (5 mmol/L) for

72 h or were left untreated. All inhibitors remained present in the

cell culture media during the whole experiment.

Abbreviations

Abcc4 ATP-binding cassette subfamily C member 4cAMP cyclic adenosine monophosphatecGMP cyclic guanosine monophosphateER endoplasmic reticulumGAPDH glyceraldehyde-3-phosphate dehydrogenasegrp78 glucose-regulated protein 78HE hepatic encephalopathyHPRT1 hypoxanthine ribosyl transferase 1HO-1 heme oxygenase 1JNK1,2 c-jun N-terminal kinase 1, 2MAPK mitogen-activated protein kinaseMrp4 multidrug resistance-associated protein 4MSO L-methionine-S-sulfoximineNADPH nicotinamide adenine dinucleotide phosphate;NF-jB nuclear factor-jBNO nitric oxideONOO- peroxynitritep38MAPK p38 mitogen-activated protein kinasePPARa peroxisome proliferator-activated receptor a

RNOS reactive nitrogen and oxygen speciesROS reactive oxygen speciessGC soluble guanylate cyclaseSMRT silencing mediator of retinoid and thyroid hormone receptorsSR-SIM super-resolution structured illumination microscopyTIRFM total internal reflection microscopy

J€ordens et al.: Mrp4 Expression Changes in Hepatic Encephalopathy

November 2015 2093

Postmortem Human Brain TissuePostmortem human brain tissue used for microarray analysis in a

recent study (G€org et al., 2013a) was obtained from autopsies of

control subjects free from any neurological disease and patients with

liver cirrhosis with accompanying HE. Tissue was provided by the

body donor program of the Department of Anatomy of the Univer-

sity of D€usseldorf, Germany. Additional tissue from three patients

with liver cirrhosis without HE was obtained from the Australian

Brain Donor Programs NSW Tissue Resource Centre. Information

on patient histories, demographics, causes of death, etiologies of cir-

rhosis and comorbidities can be found in (G€org et al., 2013a) and

in Supporting Information Tables 1 and 2.

Postmortem human brain tissue used for analysis of Mrp4

expression by Western blot was obtained from autopsies of seven con-

trol subjects, four patients with liver cirrhosis without HE, and eight

patients with liver cirrhosis with accompanying HE. Tissue was pro-

vided by the Australian Brain Donor Programs NSW Tissue Resource

Centre. Information on patients histories, demographics, causes of

death, liver and brain pathologies, etiologies of cirrhosis and comor-

bidities can be found in Supporting Information Tables 3 and 4.

Real-Time PCR AnalysisReal-time PCR analysis was carried out as described previously

(G€org et al., 2006). Total RNA was isolated using the RNAeasy

mini kit (Qiagen, Hilden, Germany) according to the manufacturer�s

protocol. RNA was quantified with a spectrophotometer (Nano-

Drop1000 System, Thermo Scientific, Wilmington). First strand

cDNA was synthesized from RNA using the QuantiTect Reverse

Transcription Kit (Qiagen, Hilden, Germany). Gene expression levels

were quantified using real-time SYBRVR Green PCR on a 7500 real-

time PCR system (Applied Biosystems). The following PCR-primer

sequences were used in the present study: Mrp1, for 5�-GGACT

TGGTTCTCAAGCACATAAAT-3�; rev 5�-CAACCTTTTCTCCAC

CCTCAAT-3; Mrp4, for 5�-CACCTTGGAGAGGAGTTGCAA-3�;

rev 5�-AAGGCTTCCGAGCGTCCTT-3; Mrp5, for 5�-ACCAGG

CTCATCCTGTCCATC-3�; rev 5�-ACGAAGGCTGGTCCACTG

AA-3; HO-1, for 5�-CGGCCCTGGAAGAGGAGA TAG-3�; rev 5�-

CGATGCTCGGGAAGGTGAAAA-3�; rat HPRT1, for 5�-TGCTCG

AGAT GTCATGAAGGA-3�; rev 5�-CAGAGGGCCACAATGT-

GATG-3�.

Data for each gene were produced in duplicate and mean cycle

numbers for the target amplification were subtracted from mean

cycle numbers of the house-keeping gene (HPRT1) for the respective

sample.

AgilentTM Microarray AnalysisGene array analysis was carried out with postmortem human brain

tissue taken from intersection parietal to occipital cortex in a recent

study published in (G€org et al., 2013a). For patients histories, see

Supporting Information Tables 1 and 2. Technical information on

AgilentTM Whole Human Microarray analysis can be found in G€org

et al. (2013a). Gene array data was deposited at the public genomic

data repository “Gene Expression Omnibus” (GEO, accession no.:

GSE41919, URL: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc

= GSE41919) from the National Center for Biotechnology Informa-

tion (NCBI) (G€org et al., 2013a).

Western Blot AnalysisWestern blot analysis was performed as described before (Schliess

et al., 2002). In brief, proteins were purified from cultured astrocytes

or from postmortem human brain tissue taken from the fusiform

gyrus and protein content was determined by the BioRad protein

assay (BioRad, Munich, Germany). For patients histories, see Sup-

porting Information Tables 3 and 4. After gel electrophoresis and

semi-dry protein transfer, the nitrocellulose membranes were blocked

using bovine serum albumin (BSA, 10%, 30 min) before the mem-

brane was incubated with antibodies against multidrug resistance

protein 4 (Mrp4, mAb, 1:5,000), glyceraldehyde-3-phosphate dehy-

drogenase (GAPDH, mAb, 1:5,000), glucose-regulated protein 78

(grp78; mAb, 1:5,000), or actin (mAb, 1:5,000), respectively. For

detection of primary antibodies, blots were incubated with horserad-

ish peroxidase-coupled anti-IgG antibodies diluted 1:10,000 at room

temperature for 2 h. Blots were washed extensively and peroxidase

activity was detected using Western-LightningTM chemiluminescence

reagent plus (Perkin Elmer, Waltham). Images were acquired using

the Kodak Image Station 4000MM and chemiluminescence was ana-

lyzed by densitometric analysis using the Kodak MI software

(v.4.0.3).

Super-Resolution Structured IlluminationMicroscopyFor super-resolution structured illumination microscopy (SR-SIM),

astrocytes were grown on customized microscope cover glasses

(112 mm, 0.170 6 0.005 mm thickness; Karl Hecht GmbH & Co

KG, Sondheim, Germany). SR-SIM was performed using the

ELYRA PS.1 microscope (ZEISS, Oberkochem, Germany). At the

end of the experimental treatment, astrocytes were washed in PBS

and fixed with paraformaldehyde (2%) and glutaraldehyde (2%) for

3 min. Cells were washed thrice with PBS and incubated in PBS

containing 5% BSA for 30 min at room temperature. Immunostain-

ing was performed with primary antibodies against Mrp4 (pAb,

1:100) and grp78 (mAb, 1:100) diluted in PBS containing 10%

bovine serum albumin (BSA). Cells were incubated for 2 h at room

temperature, washed thrice in PBS and incubated again for 2 h at

room temperature with Hoechst34580 (1:10,000) and secondary

antibodies (1:200, Jackson Corp., West Grove). Cover glasses were

mounted on objective holder using Fluoromount-G (Southern Bio-

tech, Birmingham) and images were acquired and processed accord-

ing to the manufacturer�s instruction.

Total Internal Reflection MicroscopyFor total internal reflection microscopy (TIRFM), astrocytes were

grown on IbidiVR dishes (m-Dish35 mm, high, glass bottom). At the

end of the experimental treatment astrocytes were fixed and Mrp4

was immunostained as described above. TIRFM was performed with

the ELYRA PS.1 microscope (ZEISS, Oberkochen, Germany) using

an oil immersion objective a-Plan FLUAR 1003, 1.45 numerical

aperture. The TIRF angle was adjusted to allow efficient excitation

2094 Volume 63, No. 11

of the fluorochromes by the evanescent wave to about 120 nm above

the specimen holder.

Semiquantitative Measurement of AstrocyteProliferationAstrocyte proliferation was measured as described recently (G€org

et al., 2015). In brief, proliferation was estimated by fluorimetric

measurement of DNA content (Fluoscan Ascent FL, Thermo Elec-

tron Cooperation). For this, astrocytes seeded on 24 wells were

treated as indicated or were left untreated for 72 h. DNA was

stained using Hoechst34580 (1:10,000, 15 min) and Hoechst34580

fluorescence was measured fluorimetrically (excitation: 380 nm,

emission: 460 nm). For each experimental condition DNA content

was measured in quadruplicate and mean fluorescence intensities

were corrected for background fluorescence. Fluorescence intensity

found under the respective experimental condition is given relative

to the untreated control.

Analysis of ResultsExperiments on cultured astrocytes were carried out with at least

three independent astrocyte preparations. Results are expressed as

mean values 6 SEM and compared using two-sided Student’s t-test

or one-way analysis of variance (ANOVA) followed by Tuke�ys or

Dunnett�s multiple comparison post hoc tests, where appropriate

(GraphPad Prism; GraphPad, La Jolla; Excel for Windows; Micro-

soft, Redmond). P values �0.05 were considered significant.

Results

Effect of Ammonia on Mrp4 mRNA Levels inCultured Rat AstrocytesMultidrug resistance proteins 4 mRNA levels were analyzed

by real-time PCR in cultured rat astrocytes exposed to

NH4Cl (5 mmol/L) for 24, 48, and 72 h. As shown in Fig.

1, NH4Cl (5 mmol/L, 72 h) decreased Mrp4 mRNA levels

by about 50% and 25% after 24 h and 48 h, respectively,

before mRNA levels increased to 2.5-fold of untreated con-

trols after 72 h.

NH4Cl (5 mmol/L, 72 h) did not upregulate the multi-

drug resistance protein (Mrp) isoforms 1 and 5 mRNA in

cultured astrocytes (Supp. Info. Fig. 1).

The results suggest that ammonia transiently decreases

Mrp4 mRNA levels at 24 and 48 h and upregulates Mrp4

mRNA levels 72 h after treatment.

Pharmacological Characterization of Ammonia-Induced Upregulation of Mrp4 mRNA in CulturedRat AstrocytesEffects of ammonia-induced pH-changes on Mrp4 mRNA levels

in astrocytes were analyzed using CH3NH3Cl (5 mmol/L) which

induces intracellular pH changes similar to NH4Cl (Nagaraja and

Brookes, 1998), but is metabolically inert. As shown in Fig. 2A,

CH3NH3Cl (5 mmol/L, 72 h) did not change Mrp4 mRNA lev-

els in astrocytes compared with controls.

A contribution of NH4Cl (5 mmol/L, 72 h)-dependent

glutamine synthesis to upregulation of Mrp4 mRNA was stud-

ied using the glutamine synthetase inhibitor L-methionine-S-

sulfoximine (MSO). As shown in Fig. 2A, NH4Cl (5 mmol/L,

72 h)-induced upregulation of Mrp4 mRNA was completely

prevented by MSO (3 mmol/L). Treatment of astrocytes with

MSO in the absence of NH4Cl did not affect Mrp4 levels

(0.94 6 0.16, n 5 6 independent experiments), indicating that

the observed effects are not due to the inhibitor himself.

Inhibition of NH4Cl-induced nitric oxide (NO) forma-

tion by the broad range NO-synthase inhibitor L-NAME (1

mmol/L) or of NADPH oxidase by apocynin (300 mmol/L)

did not prevent the NH4Cl (5 mmol/L, 72 h)-induced upreg-

ulation of Mrp4 mRNA (Fig. 2A). However, upregulation of

Mrp4 mRNA by NH4Cl (5 mmol/L, 72 h) was completely

abolished in presence of both inhibitors or the peroxynitrite

(ONOO-) scavenger uric acid (Fig. 2A).

Likewise, blocking proteasomal degradation using

MG132 (10 mmol/L) completely prevented NH4Cl (5 mmol/

FIGURE 1: Effects of NH4Cl on multidrug resistance-associated protein 4 mRNA expression in cultured rat astrocytes. Astrocytes weretreated with NH4Cl (5 mmol/L) or left untreated for the time periods indicated before RNA was isolated and Mrp4 mRNA expression lev-els were analyzed by real-time PCR. mRNA levels found in astrocytes treated with NH4Cl (5 mmol/L, 72 h) are given relative tountreated controls. *Statistically significantly different compared with the respective control.

J€ordens et al.: Mrp4 Expression Changes in Hepatic Encephalopathy

November 2015 2095

L)-induced upregulation of Mrp4 mRNA in cultured astro-

cytes (Fig. 2A).

The data suggest that ammonia upregulates Mrp4

mRNA in a L-methionine-S-sulfoximine sensitive manner and

requires activation of nitric oxide synthases and NADPH-

oxidase, formation of ONOO- and an intact proteasome.

Ammonia-induced oxidative/nitrosative stress activates

mitogen-activated protein kinases (MAPK) p38MAPK and c-

Jun N-terminal kinases (JNK) 1/2 (Dai et al., 2013; Jayaku-

mar et al., 2006; Schliess et al., 2002). As shown in Fig. 2B,

ammonia-induced upregulation of Mrp4 mRNA was insensi-

tive towards inhibition of JNK1/2 by SP600125 (100 mmol/

L), but was completely prevented by the p38MAPK inhibitor

SB203580 (10 mmol/L).

Abcc4 gene transcription is triggered by peroxisome

proliferator-activated receptor a (PPARa) (Moffit et al., 2006).

As shown in Fig. 2B, inhibition of PPARa activation by

GW6471 (10 mmol/L), but not inhibition of ammonia-induced

NF-jB activation (Schliess et al., 2002; Sinke et al., 2008) by

Bay11-7082 (10 mmol/L) prevented ammonia-induced upregu-

lation of Mrp4 mRNA (5 mmol/L, 72 h) in astrocytes.

The results suggest that ammonia enhances Abcc4 gene

transcription and upregulates Mrp4 mRNA in cultured rat

astrocytes in a p38MAPK- and PPARa-dependent manner.

Characterization of Ammonia-Induced Mrp4 ProteinExpression Changes in Cultured Rat AstrocytesTime- and concentration-dependent effects of ammonia on

Mrp4 protein levels were analyzed by Western blot analysis.

As shown in Fig. 3, Mrp4 protein levels were significantly ele-

vated in NH4Cl (5 mmol/L, 72 h)-exposed astrocytes by

about 2.5-fold. Already a NH4Cl concentration of 1 mmol/L

FIGURE 2: Characteristics of the NH4Cl-induced upregulation of Mrp4 mRNA in cultured rat astrocytes. Quantification of Mrp4 mRNAlevels by real-time PCR. (A, B) Astrocytes were exposed to NH4Cl (5 mmol/L, 72 h) or CH3NH3Cl (5 mmol/L, 72 h) or were left untreated(72 h). (A) Where indicated L-methionine-S-sulfoximine (MSO, 3 mmol/L), L-NAME (1 mmol/L), and/or apocynin (300 mmol/L), uric acid(200 mmol/L) or MG132 (10 mmol/L) or (B) SP600125 (100 mmol/L), SB203580 (10 mmol/L), GW6471 (10 mmol/L) or BAY117082 (10 mmol/L) was present during the whole experiment. mRNA levels found in astrocytes treated with NH4Cl (5 mmol/L) are given relative to therespective control (untreated or inhibitor only). *Statistically significantly different compared with the respective control. #Statisticallysignificantly different compared with NH4Cl. n.s.: not significantly different compared with the respective control.

2096 Volume 63, No. 11

(72 h) was sufficient to upregulate Mrp4 protein levels in cul-

tured astrocytes (Fig. 3D).

As shown in Fig. 3, the molecular mass of immunoreac-

tive Mrp4 was lowered in response to NH4Cl (1–5 mmol/L,

72 h; Fig. 3A,C) to similar extent as found after PNGase F

(100 U/mL, 6 h)-treatment of protein samples from astro-

cytes which were exposed to NH4Cl (5 mmol/L, 72 h) or left

untreated (Fig. 4A).

Both, NH4Cl (5 mmol/L, 72 h)-induced upregulation as

well as the decreased molecular mass of Mrp4 were fully pre-

vented by the glutamine synthetase inhibitor L-methionine-S-

sulfoximine (MSO, 3 mmol/L; Fig. 4B), and were not induced

by addition of CH3NH3Cl (5 mmol/L, 72 h; Fig. 4D).

These data suggest that ammonia upregulates Mrp4

protein levels and impairs N-glycosylation of Mrp4 in a

MSO-sensitive manner, but independently from intracellular

pH-changes.

A role of p38MAPK and PPARa for ammonia-induced

upregulation of Mrp4 protein in cultured rat astrocytes was

tested using the p38MAPK inhibitor SB203580 (10 mmol/L)

and the PPARa inhibitor GW6471 (10 mmol/L), which pre-

vents the release of PPARa-bound repressors such as the

silencing mediator of retinoid and thyroid hormone receptors

(SMRT). As shown in Fig. 5A, inhibition of p38MAPK or of

PPARa activation largely prevented upregulation of Mrp4

protein by NH4Cl (5 mmol/L) in astrocytes.

Activation of PPARa depends on the release (Daynes

and Jones, 2002) and proteasomal degradation (Dennis et al.,

2005) of repressors. An involvement of proteasomal degrada-

tion in ammonia-induced upregulation of Mrp4 protein was

tested using MG132 (10 mmol/L). As shown by Western blot

analysis, NH4Cl (5 mmol/L)-induced upregulation of Mrp4

protein (Fig. 5B) was prevented by MG132 (10 mmol/L).

Impaired protein N-glycosylation may indicate ER-

stress, which is characterized by upregulation of the chaperone

and ER-stress surrogate marker glucose-regulated protein 78

(grp78). As shown by Western blot analysis, both, the N-

glycosylation inhibitor tunicamycin as well as NH4Cl (5

FIGURE 3: Concentration and time dependence of NH4Cl-induced Mrp4 protein expression in cultured rat astrocytes. Astrocytes wereexposed to NH4Cl at the concentrations indicated for 72 h (A, B) or were exposed to NH4Cl (5 mmol/L) for the time periods indicated (C,D) or were left untreated before protein was isolated and Mrp4 protein expression was analyzed by Western blot. GAPDH served as load-ing control. (B, D) Densitometric quantification of Mrp4 expression. Mrp4 expression in NH4Cl-treated astrocytes is given relative to therespective untreated control. *Statistically significantly different compared with the respective control. n.s.: not significantly different.

J€ordens et al.: Mrp4 Expression Changes in Hepatic Encephalopathy

November 2015 2097

mmol/L, 72 h) strongly increased grp78 protein levels in a L-

methionine-S-sulfoximine (MSO, 3 mmol/L)-sensitive man-

ner in astrocytes (Fig. 6A).

Defective protein processing in the ER may interfere

with subcellular distribution of proteins which may suggest

impaired protein function. Therefore, grp78 and Mrp4 pro-

tein levels were analyzed in cultured rat astrocytes by super-

resolution structured illumination microscopy (SR-SIM). As

shown in Fig. 6B, anti-grp78 and anti-Mrp4 immunoreactiv-

ities were strongly increased by NH4Cl (5 mmol/L, 72 h).

Whereas under control conditions most of the anti-Mrp4

immunoreactivity was spatially restricted to patches, Mrp4

was strongly distributed throughout the soma and at the

plasma membrane in NH4Cl (5 mmol/L, 72 h)-exposed

astrocytes as shown by SR-SIM (Fig. 6B) and total internal

reflection microscopy (TIRFM; Supp. Info. Fig. 2).

The results suggest that ammonia induces ER-stress and

increases the Mrp4 immunoreactivity at the plasma mem-

brane of cultured rat astrocytes.

Effects of Mrp4 Inhibition on Proliferation andOxidative Stress in Forskolin- or NH4Cl-ExposedAstrocytesEffects of Mrp4 inhibition on the proliferation of cultured

astrocytes were measured using the recently established Mrp4

inhibitor ceefourin 1 (Cheung et al., 2014) which blocks

Mrp4 transport with high affinity and selectivity by an

unknown mechanism and fluorimetric quantification of

Hoechst34580 fluorescence as a measure for DNA-content as

described recently (G€org et al., 2015). As shown in Fig. 7A,

ceefourin 1 (10 mmol/L), forskolin (10 mmol/L), and NH4Cl

(5 mmol/L) significantly reduced the DNA content in astro-

cyte cultures by 35 to 45% compared with untreated con-

trols. DNA-content in astrocyte cultures further decreased

synergistically, when astrocytes were exposed to ceefourin 1

(10 mmol/L) plus forskolin (10 mmol/L) or to ceefourin 1 (10

mmol/L) plus NH4Cl (5 mmol/L) (Fig. 7A).

The data indicate that pharmacological inhibition of

Mrp4 enhances the antiproliferative effect of ammonia and

forskolin in cultured rat astrocytes.

Ammonia inhibits astrocyte proliferation through induc-

tion of oxidative stress (G€org et al., 2015). As shown by

immunofluorescence analysis, both, NH4Cl (5 mmol/L) and

forskolin (10 mmol/L, 72 h) strongly increased anti-8OH(d)G

immunoreactivity in astrocytes which serves as a biomarker

for oxidative stress (Fig. 7B). In order to test, whether inacti-

vation of Mrp4 impairs astrocyte proliferation in ammonia-

treated astrocytes through induction of oxidative stress,

mRNA levels of the oxidative stress surrogate marker heme

oxygenase 1 (HO-1) were measured. As shown in Fig. 7C

and in line with recent reports (Warskulat et al., 2002),

NH4Cl (5 mmol/L, 72 h) and forskolin (10 mmol/L, 72 h)

FIGURE 4: Glutamine synthesis- and pH-dependence of NH4Cl-induced upregulation of Mrp4 protein expression changes in cultured ratastrocytes. Astrocytes were exposed to NH4Cl (A–C) or CH3NH3Cl (C) for 72 h before protein was isolated and Mrp4 protein expression wasanalyzed by Western blot. (A) PNGase-treatment (100 units/mL, 6 h) of protein lysates taken from NH4Cl (5 mmol/L)-treated or untreatedastrocytes. The upper arrow points to fully N-glycosylated Mrp4. The lower arrow indicates Mrp4 with reduced molecular mass. (B) Effect ofL-methionine-S-sulfoximine (MSO, 3 mmol/L) on NH4Cl-induced Mrp4 protein expression. (C) GAPDH and actin served as loading controls.

2098 Volume 63, No. 11

upregulated HO-1 mRNA levels in cultured astrocytes (Fig.

7D). Inactivation of Mrp4 by ceefourin 1 synergistically

enhanced NH4Cl (5 mmol/L, 72 h)- or forskolin (10 mmol/

L, 72 h)-induced upregulation of HO-1 mRNA levels in cul-

tured astrocytes (Fig. 7C,D).

The data suggest that pharmacological inactivation of

Mrp4 enhances oxidative stress in ammonia- or forskolin-

treated cultured rat astrocytes.

Mrp4 mRNA and Protein Expression in PostmortemBrain Samples from Patients with Liver Cirrhosiswith or Without HEmRNA expression profiles of ATP-binding cassette transporter

subfamily C (Abcc) genes were analyzed in postmortem human

brain samples using Agilent Whole Human Genome Array tech-

nique of control patients and patients with liver cirrhosis with-

out or with HE as already described in (G€org et al., 2013a).

Whereas Abcc1 mRNA levels were reduced and Abcc2,

3, 5, 6 and Abcc8-13 mRNA expression levels remained

unchanged, Abcc4 mRNA levels were selectively elevated in

patients with liver cirrhosis with HE but not in those without

HE as compared with controls (Fig. 8A,B).

As shown by Western blot and in line with mRNA data

(Fig. 8A,B), also Mrp4 protein levels were significantly upreg-

ulated in cerebral cortex of patients with liver cirrhosis and

HE but not in those without HE when compared with con-

trols (Fig. 8C,D).

Discussion

The present study suggests that ammonia, a major toxin in

the pathogenesis of hepatic encephalopathy upregulates Mrp4

mRNA and protein (Figs. 1–3) due to glutamine formation

(Figs. 2A and 4B).

Interestingly, Mrp4 mRNA expression was downregu-

lated by ammonia at 24 h and this was mimicked by hypoos-

motic astrocyte swelling (205 mosmol/L, 24 h, data not

shown), which may suggest that ammonia-induced astrocyte

swelling transiently blocks Abcc4 gene transcription or

FIGURE 5: Role of p38MAPK, PPARa and proteosomal degradation for NH4Cl-induced upregulation of Mrp4 protein. (A and B) Analysisof Mrp4 protein expression by Western blot. Astrocytes were treated with NH4Cl (5 mmol/L, 72 h) or were left untreated in the pres-ence or absence of (A) GW6471 or SB203580 (10 mmol/L, each) or (B) MG132 (10 mmol/L). Protein was isolated and Mrp4 proteinexpression was analyzed by Western blot. GAPDH served as loading control. (C) Densitometric analysis of Mrp4 protein expression.Mrp4 protein levels were normalized to GAPDH expression levels. *Statistically significantly different compared with the respective con-trol. n.s.: not significantly different.

J€ordens et al.: Mrp4 Expression Changes in Hepatic Encephalopathy

November 2015 2099

destabilizes Mrp4 mRNA. However, ammonia-induced down-

regulation of Mrp4 mRNA was not accompanied by

decreased Mrp4 protein levels (Fig. 3D) probably due to the

large half-life of the Mrp4 protein as shown for other ABC

transporters (Kipp et al., 2001).

As opposed to ammonia, hypoosmotic astrocytes swelling

did not enhance Abcc4 transcription at 72 h (data not shown).

However, this finding does not rule out that ammonia-induced

astrocyte swelling contributes to upregulation of Mrp4 at 72 h

as astrocyte swelling is counteracted by a regulatory volume

decrease (Eriksson et al., 1992) and data on volume changes in

astrocytes exposed to hypoosmotic cell culture media for 72 h

is missing. It is also reasonable to speculate, that both,

swelling-dependent and -independent effects may contribute to

enhanced Abcc4 gene transcription by ammonia.

Cerebral glutamine formation is a key event in the

pathogenesis of hepatic encephalopathy and ammonia-

induced glutamine formation as well as glutamine per se were

FIGURE 6: Effect of NH4Cl on expression of the ER-stress surrogate marker grp78. (A) Analysis of grp78 protein expression by Westernblot. Astrocytes were treated with NH4Cl (5 mmol/L) or left untreated in the presence or absence of MSO (3 mmol/L) or were treatedwith tunicamycin (10 mmol/L) for 72 h before protein was isolated and grp78 protein expression was analyzed by Western blot. GAPDHserved as loading control. (B) Immunofluorescence analysis of Mrp4 and grp78 protein expression. Astrocytes were treated with NH4Cl(5 mmol/L) or were left untreated for 72 h before cells were fixed and Mrp4 and grp78 were analyzed by super-resolution structuredillumination microscopy (SR-SIM). Nuclei were counterstained using Hoechst34580.

2100 Volume 63, No. 11

shown to induce oxidative stress in astrocytes (H€aussinger

et al., 1994; Jayakumar et al., 2004; Murthy et al., 2001;

Schliess et al., 2002). In line with this, the present study

shows that enhanced transcription of Abcc4 requires both,

ammonia-induced activation of NO synthases (Kruczek et al.,

2011; Schliess et al., 2002) and NADPH oxidase (Jayakumar

et al., 2009; Reinehr et al., 2007) and involves the formation

of peroxynitrite (Fig. 2A) which was recently suggested to

play a major role in the pathogenesis of HE (Gimenez-Garz�o

et al., 2015; Schliess et al., 2002). As a consequence of

ammonia-induced nitric oxide, superoxide anion radical and

peroxynitrite formation, p38MAPK becomes activated (Jayaku-

mar et al., 2006; Schliess et al., 2002) and enhances Abcc4

gene transcription through activation of peroxisome

proliferator-activated receptor (PPAR) a (Figs. 2, 3, and 5A).

Interestingly, inhibition of the proteasome prevented

enhanced transcription of Abcc4 (Fig. 5B) which may indicate

that activation of PPARa requires the proteasomal degrada-

tion (Dennis et al., 2005) of repressors.

Ammonia also impaired N-glycosylation of Mrp4 (Figs.

3A,C and 4A) and induced endoplasmic reticulum (ER) stress

in a L-methionine-S-sulfoximine-dependent way (MSO, Fig.

4B), as evidenced by upregulation of glucose-regulated pro-

tein 78 (grp78, Fig. 6A). However, molecular mechanisms

FIGURE 7: Effect of Mrp4 inhibition on astrocyte proliferation and oxidative stress in forskolin- and NH4Cl-treated astrocytes. Astrocyteswere either treated with forskolin (10 mmol/L) or NH4Cl (5 mmol/L, 72 h) in the presence or absence of ceefourin 1 (10 mmol/L, 30 minpretreatment), or were left untreated (control). (A) Astrocyte proliferation was measured by fluorimetric quantification of Hoechst34580fluorescence as described in the materials and methods section. Fluorescence in experimentally-treated astrocytes is given relative tountreated controls. (B) Immunofluorescence analysis of oxidized DNA/RNA using an anti-8OH(d)G antibody in untreated (control), NH4Cl(5 mmol/L)-, DMSO-, or forskolin (10 mmol/L)-treated astrocytes (72 h). (C, D) Quantification of HO-1 mRNA levels by real-time PCR. HO-1 mRNA levels found in astrocytes treated with NH4Cl (5 mmol/L), forskolin (10 mmol/L) in absence or presence of ceefourin 1 (10 mmol/L) are given relative to untreated or DMSO-treated controls. *Statistically significantly different compared with the respective control.n.s.: not significantly different.

J€ordens et al.: Mrp4 Expression Changes in Hepatic Encephalopathy

November 2015 2101

FIGURE 8: Multidrug resistance-associated protein mRNA and protein expression changes in cerebral cortex of patients with liver cirrho-sis without or with HE. (A, B) Analysis of multidrug resistance-associated protein (Mrp) mRNA expression changes by Agilent WholeHuman Genome Array as performed in (G€org et al., 2013a). mRNA was isolated from the cerebral cortex of human postmortem brainbiopsies from eight controls, eight patients with liver cirrhosis with HE, and three patients with liver cirrhosis without HE (for patientshistories see Supp. Info. Tables 1 and 2). Gene expression levels were analyzed by Agilent Microarray as described in (G€org et al.,2013a) and in materials and methods. (A and B) Volcano plot analysis of Mrp isoform mRNA expression level changes in patients withliver cirrhosis without or with HE compared with controls. Each dot represents the mean expression change for the respective Mrp trans-porter isoform analyzed. The y-axis denotes the P values of the respective Mrp transporter isoform. For clarity reasons, only Mrp trans-porter expression changes associated with a P value <0.05 are labeled and marked by boxed areas. (C, D) Analysis of Mrp4 proteinexpression changes by Western blot analysis. Protein was isolated from the cerebral cortex of human postmortem brain biopsies fromseven controls, eight patients with liver cirrhosis with HE, and four patients with liver cirrhosis without HE (for patients histories seeSupp. Info. Tables 3 and 4). Mrp4 protein expression was analyzed by Western blot. GAPDH served as loading control. (D) Densitometricanalysis of Mrp4 protein expression. Mrp4 protein levels were normalized to GAPDH expression levels. *Statistically significantly differ-ent compared with the respective control. n.s.: not significantly different. [Color figure can be viewed in the online issue, which is avail-able at wileyonlinelibrary.com.]

2102 Volume 63, No. 11

underlying glutamine-mediated inhibition of Mrp4 N-

glycosylation and induction of ER-stress remain to be estab-

lished but may involve glutamine-dependent RNOS forma-

tion (Jayakumar et al., 2004).

Defective N-glycosylation in the ER may perturb cellu-

lar trafficking, distribution and function of proteins (Romero-

Fernandez et al., 2011). However, Mrp4 was homogenously

distributed throughout the soma and also expressed at the

plasma membrane in ammonia-exposed astrocytes (Fig. 6B;

Supp. Info. Fig. 2). Moreover, elevated cAMP levels inhibit

astrocyte proliferation (Gagelin et al., 1994, 1999) and inhi-

bition of Mrp4 transport activity by ceefourin 1 synergisti-

cally impaired astrocyte proliferation in ammonia-exposed

astrocytes (Fig. 7A) suggesting that hypoglycosylated Mrp4 is

still functionally active. In line with this, a recent study even

suggested enhanced drug transport by hypoglycosylated

Mrp4, which may promote tumor cell resistance against cyto-

static agents (Beretta et al., 2010). Interestingly, inhibition of

Mrp4 by ceefourin 1, neither activated Abcc4 gene transcrip-

tion per se, nor enhanced ammonia-induced upregulation of

Mrp4 mRNA (Supp. Info. Fig. 3) indicating that upregula-

tion of Mrp4 is not a consequence of impaired Mrp4 func-

tion. Therefore, upregulation of Mrp4 and impaired N-

glycosylation in ammonia-treated astrocytes may represent a

regulatory response in order to facilitate the export of com-

pounds known to be formed under the influence of ammo-

nia, such as prostanoids (G€org et al., 2010a), cGMP

(Hermenegildo et al., 2000), conjugated glutathione (Murthy

et al., 2000; Wegrzynowicz et al., 2007), or neurosteroids

(Ahboucha et al., 2005, 2006) which trigger the formation of

cAMP in cultured astrocytes (Keitel et al., 2010). In line with

this, raising intracellular cAMP levels through activation of

adenylate cyclase by forskolin significantly upregulated Mrp4

mRNA levels (Supp. Info. Fig. 4).

However, our data does not rule out, that hypoglycosy-

lation may affect the transport of individual Mrp4 substrates

in a different way.

In contrast to ammonia-exposed cultured astrocytes,

Mrp4 glycosylation was unchanged in cerebral cortex of cir-

rhotic patients with HE when compared with controls (Fig.

8C). This is not surprising, as ammonia levels which impair

Mrp4 N-glycosylation in cultured rat astrocytes (e.g. 0.5

mmol/L) are only rarely observed in peripheral blood of cir-

rhotic patients with HE. However, lack of Mrp4 hypoglycosy-

lation in human brain in HE as compared with rat astrocytes

may also be due to species differences, limited detection sensi-

tivity and cellular heterogeneity of Mrp4 expressing cells in

brain.

Interestingly, elevated intracellular cAMP levels block

cell cycle progression in astrocytes (Gagelin et al., 1994,

1999) and abrogate growth factor-dependent signaling in

astrocytes (Bayatti and Engele, 2001, 2002), which is also

observed in ammonia-treated astrocytes (G€org et al., 2015;

Sobczyk et al., 2015). Therefore, one may speculate that

upregulation of Mrp4-dependent cAMP export may counter-

act cAMP-induced inhibition of astrocyte proliferation invitro. Indeed, as shown in the present study, inactivation of

Mrp4 by ceefourin 1 strongly enhanced the antiproliferative

effects of forskolin and ammonia (Fig. 7A).

As ammonia blocks astrocyte proliferation through oxi-

dative stress, cAMP formation may contribute to ammonia-

induced RNOS formation. Indeed, raising intracellular cAMP

concentration by forskolin strongly enhanced RNA oxidation

in astrocytes indicative for oxidative stress (Fig. 7B). More-

over, blocking Mrp4-dependent export in ammonia- as well

as in forskolin-treated astrocytes strongly enhanced mRNA

levels of the oxidative stress surrogate marker heme oxygenase

1 (Fig. 7C,D). These results indicate that Mrp4 may counter-

act ammonia-induced RNOS-formation through stimulation

of cAMP and neurosteroid export. In line with this, activation

of the adenylate cyclase-coupled neurosteroid and bile acid

receptor TGR5 by the neurosteroid 5ß-pregnan-3a-ol-20-one

increased cAMP formation and induced oxidative stress in

astrocytes (Keitel et al., 2010).

In view of this, upregulation of Mrp4 by ammonia may

also enhance the export of neurosteroids from astrocytes. In

this regard it is interesting to note that downregulation of

TGR5 by ammonia in cultured astrocytes and in postmortem

brain tissue of cirrhotic patients with HE was suggested to

counteract neurosteroid-induced cAMP- and RNOS-

formation (Keitel et al., 2010).

Interestingly, Mrp4 mRNA and protein were selectively

upregulated in brain of cirrhotic patients with HE but not in

those without HE (Fig. 8). The reason for this is currently

unclear but according to the results of the present study may

involve oxidative/nitrosative stress. In line with this, bio-

markers for oxidative stress are selectively upregulated in brain

of cirrhotic patients with HE but not in those without HE

(G€org et al., 2010a,b). Our study also suggests an involve-

ment of peroxynitrite (ONOO-) in ammonia-induced upreg-

ulation of Mrp4 (Fig. 2A). ONOO- was already suggested to

play an important role in the pathogenesis of HE due to

inactivating glutamine synthetase by tyrosine nitration (G€org

et al., 2010b; Schliess et al., 2002). Moreover, a strong corre-

lation exists between enhanced 3-nitrotyrosine levels in

peripheral blood, which may serve as a footprint for ONOO-

formation, and cognitive impairment in HE (Gimenez-Garz�o

et al., 2015; Montoliu et al., 2011). However, whether

protein-tyrosine nitration is involved in activation of Abcc4gene transcription is presently unclear.

Interestingly, the Mrp4 substrate cGMP modulates cere-

bral processes which are impaired in hepatic encephalopathy

J€ordens et al.: Mrp4 Expression Changes in Hepatic Encephalopathy

November 2015 2103

and nitric oxide (NO)-inducible cGMP synthesis is enhanced

in cerebral cortex of cirrhotic patients who died in hepatic

coma (Rodrigo et al., 2004). Thus, upregulation of Mrp4

may represent a consequence of enhanced cGMP synthesis

facilitating its release. However, enhanced Mrp4-mediated

neurosteroid export may also further derange the glutamate-

NO-cGMP pathway through modulation of NMDA recep-

tors (Cauli et al., 2011). Supporting Information Fig. 5 sum-

marizes our current view on Mrp4 regulation by ammonia in

astrocytes.

It is important to note, that upregulation of Mrp4 (Fig.

8), PPARa, HO-1 (G€org et al., 2013a), other oxidative stress

markers (G€org et al., 2010b), biomarkers for senescence (G€org

et al., 2015), downregulation of TGR5 (Keitel et al., 2010) as

well as deranged cGMP formation (Rodrigo et al., 2004) in

brain is a feature of HE in patients with liver cirrhosis indicat-

ing a potential in vivo relevance of the present findings.

Further studies are required to illuminate the role of

Mrp4 for the pathogenesis of HE.

Acknowledgment

Grant sponsor: Deutsche Forschungsgemeinschaft through

Collaborative Research Center “Communication and Systems

Relevance in Liver Injury and Regeneration (D€usseldorf )”;

Grant number: SFB 974; Grant sponsor: University of Syd-

ney, National Health and Medical Research Council of Aus-

tralia, Schizophrenia Research Institute, National Institute of

Alcohol Abuse and Alcoholism, and NSW Department of

Health.

The authors are grateful for tissue provision by the Austra-

lian Brain Donor Programs NSW Tissue Resource Centre.

Expert technical assistance was provided by Torsten Janssen

and Ursula Kristek.

ReferencesAhboucha S, Layrargues GP, Mamer O, Butterworth RF. 2005. Increasedbrain concentrations of a neuroinhibitory steroid in human hepaticencephalopathy. Ann Neurol 58:169–170.

Ahboucha S, Pomier-Layrargues G, Mamer O, Butterworth RF. 2006.Increased levels of pregnenolone and its neuroactive metaboliteallopregnanolone in autopsied brain tissue from cirrhotic patients who diedin hepatic coma. Neurochem Int 49:372–378.

Akanuma S, Hosoya K, Ito S, Tachikawa M, Terasaki T, Ohtsuki S. 2010.Involvement of multidrug resistance-associated protein 4 in efflux transport ofprostaglandin E(2) across mouse blood-brain barrier and its inhibition by intra-venous administration of cephalosporins. J Pharmacol Exp Ther 333:912–919.

Banerjee M, Carew MW, Roggenbeck BA, Whitlock BD, Naranmandura H, LeXC, Leslie EM. 2014. A novel pathway for arsenic elimination: humanmultidrug resistance protein 4 (MRP4/ABCC4) mediates cellular export ofdimethylarsinic acid (DMAV) and the diglutathione conjugate ofmonomethylarsonous acid (MMAIII). Mol Pharmacol 86:168–179.

Bayatti N, Engele J. 2001. Cyclic AMP modulates the response of centralnervous system glia to fibroblast growth factor-2 by redirecting signallingpathways. J Neurochem 78:972–980.

Bayatti N, Engele J. 2002. Cyclic AMP differentially regulates the expressionof fibroblast growth factor and epidermal growth factor receptors in culturedcortical astroglia. Neuroscience 114:81–89.

Beretta GL, Benedetti V, Cossa G, Assaraf YG, Bram E, Gatti L, Corna E,Carenini N, Colangelo D, Howell SB, Zunino F, Perego P. 2010. Increasedlevels and defective glycosylation of MRPs in ovarian carcinoma cells resistantto oxaliplatin. Biochem Pharmacol 79:1108–1117.

Br€uck J, G€org B, Bidmon HJ, Zemtsova I, Qvartskhava N, Keitel V, KircheisG, H€aussinger D. 2011. Locomotor impairment and cerebrocortical oxidativestress in portal vein ligated rats in vivo. J Hepatol 54:251–257.

Cauli O, Gonz�alez-Usano A, Agust�ı A, Felipo V. 2011. Differential modulationof the glutamate-nitric oxide-cyclic GMP pathway by distinct neurosteroids incerebellum in vivo. Neuroscience 190:27–36.

Chen ZS, Lee K, Kruh GD. 2001. Transport of cyclic nucleotides and estradiol17-beta-D-glucuronide by multidrug resistance protein 4. Resistance to 6-mercaptopurine and 6-thioguanine. J Biol Chem 276:33747–33754.

Cheung L, Flemming CL, Watt F, Masada N, Yu DM, Huynh T, Conseil G, TivnanA, Polinsky A, Gudkov AV, Munoz MA, Vishvanath A, Cooper DM, HendersonMJ, Cole SP, Fletcher JI, Haber M, Norris MD. 2014. High-throughput screeningidentifies Ceefourin 1 and Ceefourin 2 as highly selective inhibitors of multidrugresistance protein 4 (MRP4). Biochem Pharmacol 91:97–108.

Chung C, Gottstein J, Blei AT. 2001. Indomethacin prevents thedevelopment of experimental ammonia-induced brain edema in rats afterportacaval anastomosis. Hepatology 34:249–254.

Dai H, Song D, Xu J, Li B, Hertz L, Peng L. 2013. Ammonia-induced Na, K-ATPase/ouabain-mediated EGF receptor transactivation, MAPK/ERK and PI3K/AKTsignaling and ROS formation cause astrocyte swelling. Neurochem Int 63:610–625.

Daynes RA, Jones DC. 2002. Emerging roles of PPARs in inflammation andimmunity. Nat Rev Immunol 2:748–759.

Dennis AP, Lonard DM, Nawaz Z, O’Malley BW. 2005. Inhibition of the 26Sproteasome blocks progesterone receptor-dependent transcription through failedrecruitment of RNA polymerase II. J Steroid Biochem Mol Biol 94:337–346.

Eriksson PS, Nilsson M, Wagberg M, R€onnb€ack L, Hansson E. 1992. Volumeregulation of single astroglial cells in primary culture. Neurosci Lett 143:1952199.

Gagelin C, Pierre M, Toru-Delbauffe D. 1994. Inhibition of G1 cyclinexpression and G1 cyclin-dependent protein kinases by cAMP in an astrocyticcell line. Biochem Biophys Res Commun 205:923–929.

Gagelin C, Toru-Delbauffe D, Gavaret JM, Pierre M. 1999. Effects of cyclicAMP on components of the cell cycle machinery regulating DNA synthesis incultured astrocytes. J Neurochem 73:1799–1805.

Gimenez-Garz�o C, Urios A, Agust�ı A, Gonz�alez-L�opez O, Escudero-Garc�ıa D,Escudero-Sanchis A, Serra MA, Giner-Dur�an R, Montoliu C, Felipo V. 2015. Iscognitive impairment in cirrhotic patients due to increased peroxynitrite andoxidative stress? Antioxid Redox Signal 22:871–877.

G€org B, Bidmon HJ, H€aussinger D. 2013a. Gene expression profiling in thecerebral cortex of patients with cirrhosis with and without hepaticencephalopathy. Hepatology 57:2436–2447.

G€org B, Bidmon HJ, Keitel V, Foster N, Goerlich R, Schliess F, H€aussinger D.2006. Inflammatory cytokines induce protein tyrosine nitration in ratastrocytes. Arch Biochem Biophys 449:104–114.

G€org B, Karababa A, Shafigullina A, Bidmon HJ, H€aussinger D. 2015.Ammonia-induced senescence in cultured rat astrocytes and in humancerebral cortex in hepatic encephalopathy. Glia 63:37-50.

G€org B, Morwinsky A, Keitel V, Qvartskhava N, Schr€or K, H€aussinger D.2010a. Ammonia triggers exocytotic release of L-glutamate from cultured ratastrocytes. Glia 58:691–705.

G€org B, Qvartskhava N, Bidmon HJ, Palomero-Gallagher N, Kircheis G, ZillesK, H€aussinger D. 2010b. Oxidative stress markers in the brain of patients withcirrhosis and hepatic encephalopathy. Hepatology 52:256–265.

G€org B, Qvartskhava N, Keitel V, Bidmon HJ, Selbach O, Schliess F,H€aussinger D. 2008. Ammonia induces RNA oxidation in cultured astrocytesand brain in vivo. Hepatology 48:567–579.

2104 Volume 63, No. 11

G€org B, Schliess F, H€aussinger D. 2013b. Osmotic and oxidative/nitrosativestress in ammonia toxicity and hepatic encephalopathy. Arch BiochemBiophys 536:1582163.

H€aussinger D, G€org B. 2010. Interaction of oxidative stress, astrocyteswelling and cerebral ammonia toxicity. Curr Opin Clin Nutr Metab Care 13:87–92.

H€aussinger D, Kircheis G, Fischer R, Schliess F, vom Dahl S. 2000. Hepaticencephalopathy in chronic liver disease: A clinical manifestation of astrocyteswelling and low-grade cerebral edema?. J Hepatol 32:1035–1038.

H€aussinger D, Laubenberger J, vom Dahl S, Ernst T, Bayer S, Langer M,Gerok W, Hennig J. 1994. Proton magnetic resonance spectroscopy studieson human brain myo-inositol in hypo-osmolarity and hepatic encephalopathy.Gastroenterology 107:1475–1480.

Hermenegildo C, Monfort P, Felipo V. 2000. Activation of N-methyl-d-aspartate receptors in rat brain in vivo following acute ammonia intoxication:characterization by in vivo brain microdialysis. Hepatology 31:709–715.

Hilgier W, Wegrzynowicz M, Ruszkiewicz J, Oja SS, Saransaari P, Albrecht J.2010. Direct exposure to ammonia and hyperammonemia increase theextracellular accumulation and degradation of astroglia-derived glutathionein the rat prefrontal cortex. Toxicol Sci 117:163–168.

Jayakumar AR, Panickar KS, Murthy CR Norenberg MD. 2006. Oxidativestress and mitogen-activated protein kinase phosphorylation mediateammonia-induced cell swelling and glutamate uptake inhibition in culturedastrocytes. J Neurosci 26:4774–4784.

Jayakumar AR, Rama Rao KV, Schousboe A, Norenberg MD. 2004.Glutamine-induced free radical production in cultured astrocytes. Glia 46:296–301.

Jayakumar AR, Rama Rao KV, Tong XY, Norenberg MD. 2009. Calcium in themechanism of ammonia-induced astrocyte swelling. J Neurochem 109(Suppl1):252–257.

Keitel V, G€org B, Bidmon HJ, Zemtsova I, Spomer L, Zilles K, H€aussinger D.2010. The bile acid receptor TGR5 (Gpbar-1) acts as a neurosteroid receptorin brain. Glia 58:1794–1805.

Kipp H, Pichetshote N, Arias IM. 2001. Transporters on demand: Intrahepaticpools of canalicular ATP binding cassette transporters in rat liver. J BiolChem 276:7218–7224.

Kruczek C, G€org B, Keitel V, Bidmon HJ, Schliess F, H€aussinger D. 2011.Ammonia increases nitric oxide, free Zn(21), and metallothionein mRNAexpression in cultured rat astrocytes. Biol Chem 392:1155–1165.

Kruczek C, G€org B, Keitel V, Pirev E, Kr€oncke KD, Schliess F, H€aussinger D.2009. Hypoosmotic swelling affects zinc homeostasis in cultured ratastrocytes. Glia 57:79–92.

Lachmann V, G€org B, Bidmon HJ, Keitel V, H€aussinger D. 2013. Precipitantsof hepatic encephalopathy induce rapid astrocyte swelling in an oxidativestress dependent manner. Arch Biochem Biophys 536:1432151.

Moffit JS, Aleksunes LM, Maher JM, Scheffer GL, Klaassen CD, Manautou JE.2006. Induction of hepatic transporters multidrug resistance-associated pro-teins (Mrp) 3 and 4 by clofibrate is regulated by peroxisome proliferator-activated receptor alpha. J Pharmacol Exp Ther 317:5372545.

Montoliu C, Cauli O, Urios A, ElMlili N, Serra MA, Giner-Duran R, Gonz�alez-Lopez O, Del Olmo JA, Wassel A, Rodrigo JM, Felipo V. 2011. 3-nitro-tyrosine as a peripheral biomarker of minimal hepatic encephalopathy inpatients with liver cirrhosis. Am J Gastroenterol. 106:1629–1637.

Montoliu C, Rodrigo R, Monfort P, Llansola M, Cauli O, Boix J, Elmlili N,Agusti A, Felipo V. 2010. Cyclic GMP pathways in hepatic encephalopathy.Neurological and therapeutic implications. Metab Brain Dis 25:39–48.

Murthy CR, Bender AS, Dombro RS, Bai G, Norenberg MD. 2000. Elevationof glutathione levels by ammonium ions in primary cultures of rat astrocytes.Neurochem Int 37:255–268.

Murthy CR, Rama Rao KV, Bai G, Norenberg MD. 2001. Ammonia-inducedproduction of free radicals in primary cultures of rat astrocytes. J NeurosciRes 66:282–288.

Nagaraja TN, Brookes N. 1998. Intracellular acidification induced by passiveand active transport of ammonium ions in astrocytes. Am J Physiol 274:C883–C891.

Nies AT, Jedlitschky G, K€onig J, Herold-Mende C, Steiner HH, Schmitt HP,Keppler D. 2004. Expression and immunolocalization of the multidrugresistance proteins, MRP1-MRP6 (ABCC1-ABCC6), in human brain. Neuro-science 129:349–360.

Reinehr R, G€org B, Becker S, Qvartskhava N, Bidmon HJ, Selbach O, HaasHL, Schliess F H€aussinger D. 2007. Hypoosmotic swelling and ammoniaincrease oxidative stress by NADPH oxidase in cultured astrocytes and vitalbrain slices. Glia 55:758–771.

Rodrigo R, Cauli O, Gomez-Pinedo U, Agusti A, Hernandez-Rabaza V, Garcia-Verdugo JM, Felipo V. 2010. Hyperammonemia induces neuroinflammationthat contributes to cognitive impairment in rats with hepatic encephalopathy.Gastroenterology 139:675–684.

Rodrigo R, Montoliu C, Chatauret N, Butterworth R, Behrends S, Del OlmoJA, Serra MA, Rodrigo JM, Erceg S, Felipo V. 2004. Alterations in solubleguanylate cyclase content and modulation by nitric oxide in liver disease.Neurochem Int 45:947–953.

Romero-Fernandez W, Borroto-Escuela DO, Alea MP, Garcia-Mesa Y, GarrigaP. 2011. Altered trafficking and unfolded protein response induction as aresult of M3 muscarinic receptor impaired N-glycosylation. Glycobiology 21:1663–1672.

Russel FG, Koenderink JB, Masereeuw R. 2008. Multidrug resistance protein4 (MRP4/ABCC4): A versatile efflux transporter for drugs and signallingmolecules. Trends Pharmacol Sci 29:200–207.

Schliess F, G€org B, Fischer R, Desjardins P, Bidmon HJ, Herrmann A,Butterworth RF, Zilles K, H€aussinger D. 2002. Ammonia induces MK-801-sensitive nitration and phosphorylation of protein tyrosine residues in ratastrocytes. FASEB J 16:739–741.

Sinke AP, Jayakumar AR, Panickar KS, Moriyama M, Reddy PV, NorenbergMD. 2008. NFkappaB in the mechanism of ammonia-induced astrocyte swel-ling in culture. J Neurochem 106:2302–2311.

Sobczyk K, Jordens MS, Karababa A, G€org B, H€aussinger D. 2015. Ephrin/ephrin receptor expression in ammonia-treated rat astrocytes and in humancerebral cortex in hepatic encephalopathy. Neurochem Res 40:274–283.

Timmermann L, Gross J, Butz M, Kircheis G, H€aussinger D, Schnitzler A.2003. Mini-asterixis in hepatic encephalopathy induced by pathologicthalamo-motor-cortical coupling. Neurology 61:689–692.

Warskulat U, G€org B, Bidmon HJ, M€uller HW, Schliess F, H€aussinger D. 2002.Ammonia-induced heme oxygenase-1 expression in cultured rat astrocytesand rat brain in vivo. Glia 40:324–336.

Wegrzynowicz M, Hilgier W, Dybel A, Oja SS, Saransaari P, Albrecht J. 2007.Upregulation of cerebral cortical glutathione synthesis by ammonia in vivo andin cultured glial cells: The role of cystine uptake. Neurochem Int 50:883–889.

J€ordens et al.: Mrp4 Expression Changes in Hepatic Encephalopathy

November 2015 2105