Ce

In iaIm spM

KlAnHaMiAnFro

Me

Ch

the

Tec

Un

Bilimsuteitiohesupocell damage but decreasing HSP expression, confirmingimpaired heat shock response. Using proteomics andbioinformatics, simultaneously activated apoptosis-relatedand inflammation-related pathways were identified ascandidate mechanisms. Testing the role of sterile inflam-mation, addition of necrotic cell material to mesothelialcelcepleaingincIL-resfrorelduonpape10

Peplamocothetioactheca

shorPDstranindarePDingco

tiotheinflammatory response and the regulatory elements of theheat shock response share common elements, intracel-lular cross talk between these pathways will very likelyplay a role as soon as both pathways are triggered. Thiscross talk indeed can be observed in other models andresulted in impairment of the heat shock response.10

The American Journal of Pathology, Vol. 178, No. 4, April 2011

Cop

Pub

DO

15ls increased, whereas addition of the interleukin-1 re-tor (IL-1R) antagonist anakinra to PDF decreased re-seof inflammatory cytokines.Additionof anakinradur-PDF exposure resulted in cytoprotection and

reased chaperone expression. Thus, activation of the1R plays a pivotal role in impairment of the heat shockponse of mesothelial cells to PDF. Danger signalsm injured cells lead to an elevated level of cytokineease associated with sterile inflammation, which re-ces expression of HSP and other cytoprotective chaper-es and exacerbates PDF damage. Blocking the IL-1R

In this study, we hypothesized that HSP expression inMCs after acute exposure to PDF was inadequate, andsearched for responsible pathways. With the aid of acombined proteomics and bioinformatics approach, we

Supported by the Else-Krner-Fresenius Stiftung (C.A.) and by the FWF(Austrian Science Fund) Project P18130-B13 (C.A.).

Accepted for publication December 30, 2010.

Supplemental material for this article can be found at http://ajp.amjpathol.org or at DOI: 10.1016/j.ajpath.2010.12.034.ll Injury, Repair, Aging, and Apoptosis

terleukin-1 Receptor-Medpairs the Heat Shock Reesothelial Cells

aus Kratochwill,* Michael Lechner,ton Michael Lichtenauer,* Rebecca Herzog,*ns Christian Lederhuber,* Christian Siehs,chaela Endemann,* Bernd Mayer,dreas Rizzi, and Christoph Aufricht*m the Department of Pediatrics and Adolescent Medicine,*

dical University of Vienna, Vienna; the Institute of Analytical

emistry and Food Chemistry, University of Vienna, Vienna;

Institute for Computer Languages, Vienna University of

hnology, Vienna; and the Institute for Theoretical Chemistry,

iversity of Vienna, Vienna, Austria

oincompatibility of peritoneal dialysis fluids (PDF)its their use in renal replacement therapy. PDF expo-

re harms mesothelial cells but induces heat shock pro-ns (HSP), which are essential for repair and cytoprotec-n. We searched for cellular pathways that impair theat shock response in mesothelial cells after PDF-expo-re. In a dose-response experiment, increasing PDF-ex-sure times resulted in rapidly increasing mesothelialthway might be useful in limiting damage duringritoneal dialysis. (Am J Pathol 2011, 178:15441555; DOI:.1016/j.ajpath.2010.12.034)

APedWaauf

44ted Inflammationonse of Human

ritoneal dialysis (PD) is a frequently used renal re-cement modality and an attractive alternative to he-dialysis. The main drawback of PD is the limited bio-mpatibility of commonly used PD fluids (PDF). Due tohigh concentration of glucose and glucose degrada-

n products resulting from heat sterilization, as well asidic lactate buffer, PDF exposure causes damage tomesothelial cells (MCs) that are lining the peritoneal

vity.13

Data available from in vitro and in vivo models of PDow that PDF exposure does not only result in cell deathdamage. We have recently shown that cell damage byF or components thereof leads to a complex cellularess response, where heat shock proteins (HSP) playimportant role but other biological processes are alsouced.4,5 HSP, a group of highly conserved proteins,some of the best studied mediators of cellular repair.F exposure is able to induce HSP expression depend-on the composition of the fluid6 and induced HSP

nfer cytoprotection to the stressed MCs.7,8

It has long been suspected that smoldering inflamma-n leads to a worse therapy outcome.9 Because bothsignaling pathways involved in the regulation of the

yright 2011 American Society for Investigative Pathology.

lished by Elsevier Inc. All rights reserved.

I: 10.1016/j.ajpath.2010.12.034ddress reprint requests to Christoph Aufricht, M.D., Department ofiatrics and Adolescent Medicine, Medical University of Vienna,ehringer Grtel 18-20, 1090 Vienna, Austria. E-mail: christoph.richt@meduniwien.ac.at.

idetodutesmefecsh

MSta(St

Ce

ImweProtur10anhuothtry

Ex

MCav(Cmecorin(PBforweToleaac

rolNeturM1prPDcoFothe(ifsufaccoorABsuanIL-(ELBe

Flo

ThasDicerusThtremeou(FAthewaInc

Pr

Ce101-mChfonPhtabSwmiforwaprusma

QuSu

FopeanThonweasTh(ILtemtur

W

EqPA12thinybyanmm

Inflammation Impairs Heat Shock Response 1545AJP April 2011, Vol. 178, No. 4ntified sterile inflammation as a candidate mechanismexplain increased susceptibility of MCs to PDF in-ced damage. The role of that pathway was furtherted in a novel bioassay by inducing or inhibiting injury-diated sterile inflammation. Finally, we evaluated ef-ts of specific anti-inflammatory agents on the heatock response of MCs exposed to PDF.

aterial and Methodsndard chemicals were purchased from Sigma-Aldrich. Louis, MO) if not specified otherwise.

ll Culture

mortalized human MCs (MeT-5A, ATCC CRL-9444)re maintained in 75 cm2 flasks (TPP; Techno Plasticsducts AG, Trasadingen, Switzerland) with M199 cul-e medium (M4530, Sigma-Aldrich) supplemented with% fetal calf serum, 5.96 g/L HEPES, 50 U/mL penicillin,d 50 g/mL streptomycin at 5% CO2 and 37C in amidified atmosphere. Medium was changed everyer day. When confluent, cells were passaged bypsinization to 6-well plates (TPP) for the experiments.

perimental Design

s near confluence were exposed to a commerciallyailable glucose-monomer/acidic lactate-based PDFAPD2; Fresenius, Bad Homburg, Germany) or shamdium changes using M199 culture medium as thentrol. After the exposition for 0.5 to 4 hours, cells weresed briefly with Dulbeccos phosphate buffered salineS, Sigma) and subsequently kept in culture mediuma recovery period of 16 hours. All incubation periodsre carried out at 5% CO2 in a humidified atmosphere.assess cell viability lactate dehydrogenase (LDH) re-se was determined using the TOX-7 assay kit (Sigma)cording to the manufacturers instructions.A series of experiments was carried out to evaluate thee of sterile inflammation in the in vitro model of PD.crotic cell material (NCM) was obtained from cell cul-e wells by mechanic homogenization of MCs in regular99 growth medium. Cells grown in adjacent wells weree-incubated with NCM for 24 hours before the standardF exposure experiments (1 hour PDF 16 hours re-very) or sham treatment was carried out as the control.r the experiments with blockage of specific receptors,pre-incubation medium, the NCM containing mediumapplicable), the PDF and the control medium werepplemented either with 100 g of the tumor necrosistor--mab (TNF--mab) infliximab (Remicade, Cento-r Ortho Biotech, Horsham, PA) per milliliter of medium,with 500 ng of the IL-1Ra anakinra (Kineret, Biovitrum, Stockholm, Sweden) per milliliter of medium. Mediumpernatants from the recovery phase were collectedd analyzed for secretion of the interleukins IL-6 and

8 using the enzyme-linked immunosorbent assayISA) kits (IL-6, eBioscience, San Diego, CA; IL-8,nder MedSystems, Vienna, Austria).

braSaanw Cytometry

e modality of cell death of MCs after PDF exposure wassessed using the Apoptosis Detection Kit II (Bectonkinson, San Jos, CA) according to the manufactur-s protocol. Exposure to ultraviolet light (254 nm) wased as the positive control for induction of apoptosis.e recovery time after the exposure to PDF or shamatment as the control was 2 hours for optimal assess-nt of apoptotic cell death. Measurements were carriedt on a fluorescent-activated cell sorting flow cytometerCSCalibur, Becton Dickinson). Statistical analysis ofobtained data was performed using the FlowJo soft-re (Tree Star Inc. Ashland, OR) and SPSS 16 (SPSS., Chicago, IL).

otein Sample Preparation

lls were washed three times (250 mmol/L sucrose,mmol/L Tris, pH 7) and lysed by incubation withL lysis solution (7 M urea, 2 M thiourea, 4% 3-[(3-olamidopropyl)dimethylammonio]1-propanesulate [CHAPS], 10 mmol/L DTT, 1 mmol/L EDTA, 0.5%armalyte 310 [GE Healthcare, Uppsala, Sweden], 1let of complete protease inhibitor [Roche, Basel,itzerland] per 100 mL) per 3 107 cells for 45nutes at 25C. The resulting lysates were centrifuged30 minutes (14,000 g, 10C) and the supernatants stored at 80C until further processing. Totalotein concentration of the samples was evaluateding the 2D-Qant kit (GE Healthcare) according to thenufacturers manual.

antification of Cytokine Release Into thepernatant

llowing PDF exposure as well as NCM treatments werformed analysis of the secretion of interleukin (IL)-6d IL-8 into the supernatant during the recovery phase.e supernatants were removed and immediately cooledice. Aliquots used for the assessment of LDH releasere stored separately from aliquots used for cytokinesessment to avoid repeated thawing of the samples.e cytokines were measured using dedicated ELISA kits6: eBioscience, San Diego, CA; IL-8: Bender MedSys-s GmbH, Vienna, Austria) according to the manufac-ers recommendations.

estern Blotting

ual amounts of protein lysates were separated by SDS-GE, on a Bio-Rad Criterion cell using Criterion precast.5% Tris-HCl gels (Bio-Rad, Hercules, CA) of 1-mmckness. Proteins were electroblotted onto polyvi-lidene difluoride membranes immediately after the runtank blotting using a Criterion blotting cell (Bio-Rad)d the according transfer buffer (200 mmol/L glycine, 25ol/L Tris base, 0.1% SDS, 20% methanol). The mem-

nes were blocked with 5% dry milk in Tris Bufferedline with Tween 20 and then incubated with the primarytibody [Hsp72 (HSPA1), SPA-810, Stressgen/Assay

DechseRaCaenninDoRaac

St

Stawe16t-temiantesforproWi

Tw

TwforingweanzammCoangcmusa racphV awe10glysta10ThphelereaforforIPGmowatheweanHeca

(Dmawis

Pr

TrypewitscDaacinsoanbymmatng37ge8.5ofdrire-sousC1

TOBiowitaccocincoOnMawawaSwkecludo

Bi

Thofproabteiamt-temagro

riv

1546 Kratochwill et alAJP April 2011, Vol. 178, No. 4signs, Ann Arbor, MI; -Tubulin, # 691261, MP Bio-emicals, Solon, OH] for 16 hours. After incubation withcondary, peroxidase-coupled antibodies (Polyclonalbbit Anti-Mouse Ig/HRP P0260, Dako Cytomation,rpinteria, CA) detection was accomplished by usinghanced chemiluminescence solution (Western Light-g reagent, Perkin Elmer, Boston, MA), and a Chemi-c XRS chemiluminescence detection system (Bio-d). The densitometric quantification of 1D bands wascomplished using the Bio-Rad QuantityOne software.

atistical Tests

tistical analysis of the data obtained from ELISA andstern blotting experiments was performed using SPSS. Values from different groups were compared usingsts or analysis of variance where appropriate with animum level of P 0.05 for significance. In case ofalysis of variance Tukeys HSD was used as posthoct. To account for the requirements of statistical testingnon-parametrically distributed data, especially in theteomics sections, significance was validated using alcoxon 2-sample test for these data.

o-Dimensional Gel Electrophoresis

o-dimensional gel electrophoresis (2DGE) was per-med using total protein from cell lysates of MCs follow-the workflow described recently.4 In brief, samples

re concentrated and desalted using a modified meth-ol/chloroform precipitation, reconstituted with solubili-tion buffer [7 M urea, 2 M thiourea, 4% CHAPS, 1ol/L EDTA, 30 mmol/L Tris/HCl pH 8.5, 1 tablet ofmplete Mini Protease Inhibitor (Roche) per 100 mL]d left at 48C overnight. Total protein amounts of 300per IPG strip (Immobiline DryStrip pH 310, linear, 18, GE Healthcare) were applied by rehydration loadinging a final volume of the rehydration mix of 340 L andehydration time of 16 hours. Isoelectric focusing wascomplished on a Multiphor II (GE Healthcare) electro-oresis unit by gradually increasing the voltage to 3500nd a constant phase for another 5.5 hours. The stripsre consecutively incubated for 2 15 minutes in-mL equilibration buffer (6 M urea, 2%w/v SDS, 25%w/vcerol, 3.3%v/v 50 mmol/L Tris/HCl buffer pH 8.8,ined with bromphenol blue) first supplemented with0 mg of DTT and then with 480 mg of 2-iodoacetamide.e second dimension SDS polyacrylamide gel electro-oresis (SDS-PAGE) was carried out on Multiphor IIctrophoresis units at 20C using Excelgel XL 12-14dy cast gels (GE Healthcare) at 100 V, 25 mA, 30 W60 minutes in the first phase and 800 V, 40 mA, 50 W3 hours in the second phase. After the first phase thestrip was removed and the anodic buffer strip was

ved to the position of the removed IPG strip. The runs stopped when the bromphenol blue stain reachedfront of the cathodic buffer strip. The protein spots

re visualized using Coomassie Brilliant Blue staining

d the gels were scanned on an Imagescanner II (GEalthcare) at 300 dpi, 16 bit gray-scale. Spot quantifi-tion was carried out using the Delta2D 3.6 software

intlowcoecodon GmbH, Greifswald, Germany) based on nor-lized relative spot volume (percent volume) on group-e warped images.

otein Identification by Mass Spectrometry

ptic in-gel digestion of the excised protein spots wasrformed following the Shevchenko standard protocolh slight modifications.11 Gel plugs were excised with aalpel, washed with water, water/acetonitrile (Merck,rmstadt, Germany) and 50 mmol/L NH4HCO3 buffer/etonitrile, altogether eight times. Afterward the proteinsthe gel plugs were reduced by adding a 10 mmol/Llution of dithiothreitol (in 50 mmol/L NH4HCO3 buffer)d incubating for 45 minutes at 56C and then alkylatedadding a 55 mmol/L solution of iodoacetamide (in 50ol/L NH4HCO3 buffer) and incubating over 30 minutes

25C in the dark. The proteins were digested with a 12.5/l solution of trypsin in 50 mmol/L NH4HCO3 buffer atC overnight. The cleaved peptides were eluted from thel pieces with 50 l of 25 mmol/L NH4HCO3 buffer (pH), 50 l of 5% formic acid (Merck) and 50 l 1:1 mixturethese solutions with acetonitrile. The pooled eluates wereed by vacuum centrifugation. Then the peptides weredissolved with 10 l 0.1% trifluoroacetic acid (TFA) andnicated briefly. The obtained solution was desalted bye of ZipTip columns (Millipore, Billerica, MA) containing8 material.MS and MS/MS analyses were performed on a MALDIF-TOF instrument (4700 Proteomics Analyzer; Appliedsystems Inc., Foster City, CA) which was equippedh a Nd:YAG laser (355 nm). For MS experiments theceleration voltage was 20 kV, for MS/MS 8 kV. Thellision gas was N2. As the matrix -cyano-4-hydroxy-namic acid (Sigma-Aldrich, Germany) was used at ancentration of 10 mg/mL in acetonitrile/0.1% TFA (3:2).-target elution with -C18 ZipTips was performed.ss accuracy was 50 ppm, 1 missed cleavage sites allowed. Finally, a combined search (MS MS/MS)s performed using the software MASCOT with theiss-Prot database. Known masses of contaminatingratin and autodigestion products of trypsin were ex-ded from the searches. Calibration for the spectra wasne by a solution of external peptide standards.

oinformatics Procedures

e core data set for performing bioinformatics analysisthe proteomic data were given by the list of identifiedteins showing statistically significant difference inundance (primary candidates), represented by all pro-ns in Supplementary Table S1 (available at http://ajp.jpathol.org). Significance was derived by applying ast for comparing the distribution of intensities oftched spots for the control group and PDF stressedup (P 0.05).A bioinformatics expansion of the experimentally de-ed protein list was performed.4,12 Studying the putative

eractions between candidates, and based thereon al-ing a functional interpretation of their relevance in thentext of cell stress, we generated protein interaction

neaconB.OPthedaneadinbefol

bioclagivmeproferbethepa

Re

Re

Cepreateobliarisrelingityinaleains

etrisoshletanoflowabthemafol

proexstr10inpic

ficcople

acregboendeteidapro0.0OPproboarydidamsuteron

atiwelarseresme(PsucaofHs

Figassdamto tnumsamcovper

Inflammation Impairs Heat Shock Response 1547AJP April 2011, Vol. 178, No. 4tworks (PIN) using the Online Predicted Human Inter-tion Database (OPHID).13 OPHID provides informationinteractions of the type protein A interacts with proteinProcessing a given candidate list (A, B, . . . , N) withHID results in a set of undirected graphs representinginteractions between the candidates. To enhance theta set for bioinformatics analysis, we extracted nextighbors of our primary candidates from the PIN andded these secondary candidates.14 A next neighbor Xthe PIN is defined as A-X-B, where A and B are mem-rs of the primary candidate list, and X links A and Blowing the interaction data as given in OPHID.For deriving a categorization of candidates in terms oflogical processes and pathways we used the PANTHERssification system.15 Next to a contextual grouping of aen protein list, PANTHER provides information on enrich-nt or depletion of given proteins with respect to distinctcesses or pathways by computing a 2 test (with Bon-roni correction for multiple testing) comparing the num-r of candidates belonging to a process or pathway andtotal number of proteins assigned to this process or

thway as given by the PANTHER classification.

sults

sponses of MCs to PDF Stress

llular viability, monitored by release of LDH, and ex-ssion of heat shock protein 72 (Hsp72) were evalu-d after exposure of MCs to PDF (CAPD2). It becomesvious that insults caused by PDF triggers the mesothe-l heat shock response. Although expression of Hsp72es above control level in early time points when LDHease is still low, Hsp72 levels rapidly drop with increas-exposure times, associated with exacerbated mortal-of MCs. The results shown in Figure 1 suggest andequate cellular heat shock response to PDF stress,ding to increased cell death potentially resulting fromufficient levels of Hsp72.The modality of cell death was analyzed by flow cytom-y. The results of propidium iodide/AnnexinV-fluoresceinthiocyanate double staining following PDF exposure oram treatment are representative for a number of non-hal (60 minutes) and lethal PDF exposure (120 minutes)d recovery (2 hours) times (Figure 2, A-C). Quantificationquadrant populations shows only a marginal shift to theer right quadrant on exposure to PDF (Figure 2D). Thesence of Annexin V positive viable cells, representingapoptotic population, leads to the conclusion that thejority of non-viable MCs underwent necrotic cell deathlowing acute exposure to pure PDF.To investigate the mechanisms of cell deterioration ateomics experiment was carried out at the sublethalposure time of 60 minutes. When the proteome of PDFessed MCs was compared to a control cell proteome,

1 protein spots were found significantly altered (P 0.05)5 control versus 5 PDF stressed gels. Robot-assistedking of these proteins spots led to 53 successful identi-

ind(nerrotheations with 49 unique proteins shown in Figure 3A. Themplete list of all identified protein spots is given in Sup-mentary Table S1 (available at http://ajp.amjpathol.org).For the analysis of pathways the number of proteinscessible in 2DGE experiments can by no means bearded as complete, as low abundant or membrane-und proteins as well as proteins which are not differ-tially expressed in their mode of action are hardlytected by this method. Therefore information on pro-n-protein interactions were obtained from the OPHIDtabase. Of the 49 primary candidates (MS identifiedteins showing significantly different abundance (P 5) in the two-dimensional gels), 11 were not listed inHID. Expanding this candidate list of the remaining 38teins represented in OPHID by using the next neigh-r approach resulted in additional 55 proteins (second-candidates), giving a cumulative number of 93 can-ates in total (see Supplementary Table S2 at http://ajp.jpathol.org). The PIN resting on this list was a largebgraph holding 80 protein nodes and 216 protein in-actions. The remaining 13 candidates were found ine additional subgraph, or did not show interactions.Certainly, next neighbor expansion is prone for gener-ng larger subgraphs. To test the impact of this effect,computed the mean number of protein nodes of thegest subgraphs derived on the basis of 100 randomlylected lists of 38 proteins, each expanded by theirpective next neighbors. This procedure resulted in aan number of 21.8 nodes (SD 14.5), significantly lower 0.05) than the number of protein nodes of the largestbgraph derived on the basis of the actual list of ourndidates, indicating an increased functional interplaythese proteins. In particular two proteins, namelyp72 (Swiss-Prot accession HSP71_HUMAN) and the

ure 1. PDF induces insufficient levels of HSP. Expression of Hsp72 asessed by Western blot analysis and LDH release as a parameter for cellularage over various exposure times. Hsp72 is shown in values normalizedhe expression of -tubulin to compensate for a potential decrease in cellber with increasing PDF exposure time and relative to control. Controlples underwent sham treatment with normal growth medium. The re-ery time was 16 hours for all samples. Mean values are presented ascent-fraction of the observed maximum and were computed from three

ependent biological experiments with three biological replicates each 9 per data point). LDH release of control samples was set to 0%. Ther bars represent the standard error. Expression of Hsp72 dropped belowdetection range (n.d.) in case of 4-hour exposure to PDF.

glustred

fie

FigMCabuomare

1548 Kratochwill et alAJP April 2011, Vol. 178, No. 4cose regulated protein Grp78 (GRP78_HUMAN) formong hubs in the largest subgraph, showing 28 proteinges each as shown in Figure 3B.On the basis of the expanded protein list we identi-d involvement of 19 of the MS-identified proteins in

ure 3. Proteomics and bioinformatics approach. A: A representative Cooma

s. Proteins were separated by isoelectric point (pI) in the first dimension and molecundance (P 0.05), where proteins could be identified by MS, are marked with arritted). B: Representation of the PIN resulting from the expanded primary candidate lidepicted using black circles and next neighbors (secondary candidates) are depictnificantly enriched (P 0.05) processes or path-ys using the PANTHER classification system. The listthese proteins and the expression profiles of thesociated protein spots are given in Table 1 alongth assignment to enriched processes.

Figure 2. Modality of cell death in ex-perimental PD. Results from flow cytom-etry experiments after exposure of MCsto control medium (A) or PDF (B, C).The modality of cell death was assessedusing standard propidium iodide (y axis)and AnnexinV-fluorescein isothiocya-nate (x axis) labeling. Thresholds wereset according to an unstained control(not shown) and applied to all samplesas shown in the representative panels.The mean values and standard devia-tions of quadrant populations presentedin D were computed from three inde-pendent biological experiments. PI, pro-pidium iodide; FITC, fluorescein isothio-cyanate.

iant blue stained two-dimensional electrophoresis gel of the proteome ofsigwaofaswi

ssie brill

lar weight (MW) in the second dimension. Spots with significantly alteredows and labeled by Swiss-Prot identifiers according to Table 1 (humanst on the basis of OPHID interaction data. Proteomic (primary) candidatesed using light gray triangles, the lines denote the interaction edge.

Proapstr

duplafroinvonewitan

likewaPDmasteares

St

Toforwe16suinfcafor4 (

stipena(ILTNtheIL-adtiopaeff

vopaseCoplipa

effdusteme

catovia

exFigviainranocHsat

meprostricathePDratgro

altedu0.0in

DiInpaduexpeleacetot

Ceto

Instrincabexbeexextalmibyopviaacapsushcewade

Inflammation Impairs Heat Shock Response 1549AJP April 2011, Vol. 178, No. 4The results on the pathway level are given in Table 2.teomic pathway analysis suggests involvement ofoptotic and inflammatory pathways in the mesothelialess response to PDF stress.Interestingly, the apoptotic pathway became relevante to the differential expression of mainly anti-apoptoticyers, which is in accordance with our data obtainedm flow cytometry. Indeed, the proteins identified by MSthe course of our 2DGE experiment, which were in-lved in this pathway or showed interactions with nextighbors involved in this pathway, were chaperonesh known apoptosis inhibiting function, namely Grp78d Hsp72.The enrichment of inflammation related pathways (Toll-receptor and chemokine, cytokine signaling path-

ys) in MCs seems to be a highly relevant finding, asF toxicity is known to induce sterile peritoneal inflam-tion. Based on this observation we hypothesize thatrile inflammation mediated by damaged MCs might bekey to the deleterious alterations of the heat shockponse after PDF exposure.

erile Inflammation

test this hypothesis we exposed MCs near confluence24 hours to NCM harvested from neighboring cell culturells by mechanic homogenization. After a recovery time ofhours secreted interleukins (IL-6 and IL-8) were mea-red in the recovery supernatant. The results show sterilelammatory stimulation of the exposed MCs with signifi-ntly elevated interleukin levels (P 0.01 for IL-6, P 0.05IL-8) relative to the untreated control, as shown in Figurewhite and gray bars).To further investigate the mechanism of inflammatorymulation as co-factor of MC injury we performed ex-riments using substances, which inhibit potential sig-ling receptors like the interleukin-1 receptor antagonist-1Ra, anakinra) and a monoclonal antibody againstF- (infliximab). The effects of these compounds onsecretion of the pro-inflammatory cytokines IL-6 and

8 are shown in Figure 4 as shaded bars. Whereas thedition of anakinra leads to significantly lowered secre-n of IL-6 (P 0.05) and to normal levels of IL-8 com-red to control, the addition of infliximab had no specificect.In the same PDF exposure system, where the in-lvement of sterile inflammation was shown on thethway level using proteomics methods, we assessedcretion of IL-6 and IL-8 into the medium supernatant.mpared to NCM treatment we found diminished am-tudes, but a similar pattern, suggesting similarthogenic mechanisms.As shown in Figure 5, anakinra was again able toectively block the sterile inflammation induced in MCsring PDF exposure, supporting the hypothesis thatrile inflammation was triggered by damaged MCs anddiated by IL-1R dependent signaling.Because the addition of anakinra demonstrated signifi-

nt reduction of inflammation inMCs aswell after exposureNCM and PDF, respectively, the effect of anakinra on thebility (LDH release) and the heat shock response (Hsp72

tosanompression) was evaluated in a time-course experiment.ure 6 shows the systematic effect of anakinra on MCbility and the heat shock response. The addition of anak-a delays cellular damage significantly (P 0.05 at 1, 2,d 3 hours of exposure time). As soon as cellular damagecurs despite the addition of anakinra, the expression ofp72 is maintained at significantly higher levels (P 0.052 and 3 hours of exposure time).Finally, we investigated the mode of action of anakinra-diated protection of MCs against PDF exposure in ourteomics model. Proteins with known association to PDFess, as proven by their involvement in enriched biolog-l processes or pathways (Table 1), were analyzed forir expression with or without the addition of anakinra toF using 2DGE as described before. The expressionios of these proteins and the associated P values for aupwise comparison are also given in Table 1.More than 30% of these candidates revealed a significantlyred abundance (P 0.05, expected 5%) in the gels pro-ced from the anakinra experiment. Significant effects (P 5) of anakinra on protein expression levels are visualizedFigure 7 in relation to the initial proteomics experiment.

scussionprevious work, we reported that the same bioincom-tible properties of PDF that induce MC injury also in-ce specific HSPs, which lead to repair and recovery inperimental PD.4,16,17 Consequently, the progressiveritoneal damage in the clinical setting of PD must atst, in part, be regarded as an imbalance betweenllular injury and cytoprotective processes following cy-oxic PDF exposure.

ll Damage and Repair Patterns in ResponsePDF Exposure

the first part of this study, we related cytoprotectiveess responses to cellular survival in MCs followingreasingly severe PDF stress. These results confirm theility of PDF to induce HSP but also demonstrate that thepression of HSP rapidly falls below the control levelyond a threshold of PDF stress, resulting in severeacerbation of cell death. This inadequately low HSPpression might be either passively caused by acciden-cell death or actively induced by cellular reprogram-ng. Whereas the former cell fate can only be improveddecreasing PDF cytotoxicity, the latter fate might been to alternate interventions. Programmed cell deaththe apoptosis pathway would represent the prototypetive cellular pathway. In immune cells, induction ofoptotic cell death via exposure to lethal cytokines re-lted in an almost complete suppression of the heatock response, mediated by caspase-dependent pro-sses.18 In our study, such active induction of apoptosiss not responsible for the exacerbated cell death, asmonstrated by the almost complete absence of apop-

is markers in the fluorescent-activated cell sortingalysis. Similar results have been previously reported inental-derived MCs following acute exposure to undi-

lutrelthaoldlarreg

bio

Ta ogical P

Wilc2-sa

te

Sig.PD

0.0.0.0.0.0.0.0.0.0.0.

0.0.0.

0.0.0.0.

0.

PANTHtificationstresseroups (c

ana

(P

mo

trea

Ta

bas

ran

pat

1550 Kratochwill et alAJP April 2011, Vol. 178, No. 4ed PDF.19 Together with the abrupt increase in LDHease, our data indicate alternate cellular processest sensitize MCs to PDF exposure beyond the thresh-. Combined cytotoxic insults and counteracting cellu-reactions likely involve complex, interdependentlyulated molecular mechanisms.20,21

The systems biology approach using proteomics andinformatics is particularly well suited to investigate

ble 1. Identified Proteins Involved in Significantly Enriched Biol

Coefficientvariation (%)*

Studentst-test

Protein NameRatioPDF

Groupcontrol

GroupPDF

P valuePDF

Heat shock 70 kDa protein 1 1.44 16 23 0.042Heat shock 70-kDa protein 1 1.43 12 12 0.002078-kDa glucose-regulated 1.37 6 4 0.000Heat-shock protein beta-1 1.34 7 7 0.001Heat-shock protein beta-1 0.65 22 11 0.015150-kDa oxygen-regulated protein 1.44 12 9 0.001Peptidyl-prolyl cis-trans isomerase A 0.81 13 6 0.023Macrophage migration inhibitory factor 0.7 19 11 0.019Calreticulin 1.17 10 5 0.019FK506-binding protein 1A 0.64 11 12 0.001Small glutamine-rich tetratricopeptide

repeat-containing protein0.57 13 49 0.055

Elongation factor 2 1.78 48 18 0.027Ubiquitin 1.6 27 27 0.045Ubiquinol-cytochrome-c reductase

complex core protein1.38 22 7 0.013

Cytosol aminopeptidase 1.27 14 7 0.02NEDD8 precursor 0.66 18 27 0.027Poly(rC)-binding protein 1 0.43 26 34 0.01Ubiquinol-cytochrome-c reductase

complex core protein 20.41 24 43 0.004

Protein disulfide-isomerase 1.12 5 6 0.018

Proteins are assigned to their biological processes as obtained from the*Coefficient variation: Relative standard deviation in percent of the quanRatio PDF: fold-change of spot abundance of significantly altered PDF-The column contains the P value of the t-test comparing the respective gkinra group n 6 per group).The column contains the level of significance of the Wilcoxon 2-sample tevalue less than the given number).Ratio anakinra: fold-change of spot abundance of cells treated with PDNumber of peptides matching in the peptide mass fingerprint based id**Mascot protein score: obtained from the combined (MSMS/MS data)Relative molecular mass of the protein as calculated from the amidifications.Calculated pI of the polypeptide as obtained from Swiss-Prot databasetment of MC with PDF and PDF supplemented with anakinra.

ble 2. Significantly Enriched Pathways after Next NeighborExpansion

49 Primary candidates secondarycandidates from next neighbor expansion

PathwayTotalno.*

Expectedno.

Observedno. P value

Apoptosis signaling pathway 131 0.53 15 1.78E-15Toll-like receptor signaling

pathway71 0.29 10 7.50E-11

Inflammation mediated bychemokine and cytokinesignaling pathway

315 1.28 7 4.40E-02

*Total number of genes in the given pathway in the PANTHER data-e.Expected number (resulting from sample size) of candidates in anyacstube

dom dataset.Observed number (after next neighbor expansion) of candidates.Bonferroni corrected P value for overrepresentation of the givenhway.ch complex interplay of cellular responses.12 With thisproach we could recently determine that acute PDFposure leads to reduction of unspecific biological pro-sses in favor of the MCs response to stress includingchanisms for repair, cell structure modification, signalnsduction, and carbohydrate metabolism.4,5 Enrichedlogical processes also confirmed cytoskeletal dam-e, epithelial to mesenchymal transition, and cellularodeling as mechanisms mediating MC injury. In oursent study, we applied specific pathway analysis us-a bioinformatically expanded list of proteomics can-ates.12 Pathway analysis is a tool to validate the bio-ical relevance of the experiment and to eliminate thection of false-positive candidates in every omics ex-riment before further interpretation.4 Our results dem-strate simultaneous activation of apoptosis-relatedthways and inflammation during recovery of MCs fromF exposure via database queries of the identified pro-ns. These data corroborate previous studies that havescribed apoptotic changes in MCs following either ex-sure to lethal cytokines, to toxic glucose degradationducts, or to PDF in modified PDF exposure systemsing Transwell culture apparatus.2224 However, the

rocesses

oxonmplest Coefficient variation (%)*

Studentst-test

Wilcoxon2-sample

test

levelF

Ratioanakinra

GroupPDF

Group PDF anakinra

P valueanakinra

Sig. levelanakinra

01 0.88 10 17 0.173005 0.79 19 18 0.072 0.05005 1.44 20 12 0.004 0.005005 0.47 11 23 0.000 0.005005 1.18 7 10 0.017 0.01005 0.95 25 27 0.75301 0.65 82 36 0.380025 0.82 78 55 0.66305 1.29 11 15 0.016 0.02500505 1.12 17 31 0.491

05 0.37 38 42 0.006 0.00505 2.17 58 46 0.047 0.0505 1.38 77 32 0.368

01 0.81 22 35 0.269025025 1.38 23 24 0.061 0.0501 1.38 77 32 0.368

01 1.16 14 7 0.051 0.05

(table continues)ER database.data within the given groups.

d cells versus control cells in the initial experiment.ontrol versus PDF group n 5 per group; PDF versus PDF containing

e respective groups with an identical n per group as mentioned before

ining anakinra versus PDF-stressed cells.ion.se search.sequence of the polypeptide without any co- or posttranslational

ns indicated as bold rows were found in significantly altered betweensuapexcemetrabioagrempreingdidlogfrapeonpaPDteidepoprous

st for th

F contaentificatdatabano acid

. Proteiute exposure and recovery model used in the presentdy is not ideal for studying apoptosis in MCs but haden established for the investigation of cellular stress

resseexanvetheanletbetotheofbysuknlecthosy

Po

Peheveasinginhthe

Ta

pNCBIacc. no

33033303330933153315

1052554784282

81122806449

193873117385

51056473850937385

5034

Inflammation Impairs Heat Shock Response 1551AJP April 2011, Vol. 178, No. 4ponses and short-term repair mechanisms.4,68 In thistting, the PDF exposure time chosen for the proteomicsperiments has previously been shown to be sublethal,d our fluorescent-activated cell sorting results also re-aled an almost absent rate of cell death.7 Accordingly,spectrum of identified proteins was predominantly

ti-apoptotic, reflecting an equilibrium of protective andhal processes in PDF stressed MCs, which has alsoen observed in related models.25 Hsp72 was reportedcounteract apoptosis by inhibition of oligomerization atlevel of the apoptosome, by inhibition of translocationthe apoptosis inducing factor (AIF) to the nucleus andnegative interaction with JNK.26,27 Hsp27 is known toppress apoptosis by Fas/FasL interaction.28 Less isown about the distinctive modes of action of the mo-ular chaperones Grp78 and HYOU1 (Orp150) al-ugh both show anti-apoptotic potential in other modelstems.29,30

tential Role of Inflammatory Processes

ritoneal MCs are known to express cytokines and ad-sion molecules, when stimulated.31,32 Our data re-aled involvement of novel players like HSPs that can besigned to immunity and defense processes, FK-bind-

ble 1. Continued

Numberof

eptides Mascotscore** MR

pISwiss-Prot entry

name

SwissProtprim. acc.

no.

26 667 70280.1 5.48 HSP71_HUMAN P0810726 1000 70280.1 5.48 HSP71_HUMAN P081075 102 72505.5 5.07 GRP78_HUMAN P110215 260 22825.5 5.98 HSPB1_HUMAN P047928 267 22825.5 5.98 HSPB1_HUMAN P047925 103 111494.3 5.16 HYOU1_HUMAN Q9Y4L19 388 18229 7.68 PPIA_HUMAN P629373 182 12639.3 7.74 MIF_HUMAN P14174

11 323 48282.9 4.29 CALR_HUMAN P277972 65 12000.1 7.88 FKB1A_HUMAN P629425 223 34269.8 4.81 SGTA_HUMAN O43765

10 173 96219.3 6.41 EF2_HUMAN P136393 96 8559.6 6.56 UBIQ_HUMAN P629884 92 48583.9 8.74 UQCR2_HUMAN P22695

18 516 56530 8.03 AMPL_HUMAN P288382 75 9066 7.98 NEDD8_HUMAN Q15843

10 602 37987.1 6.66 PCBP1_HUMAN Q153655 306 48583.9 8.74 UQCR2_HUMAN P22695

11 380 57453.7 4.73 PDIA1_HUMAN P07237protein (FKB1A), cyclophilin A (PPIA), and migrationibitory factor (MIF). FKB1A and PPIA are members ofimmunophilin family and are better known as re-

thi(usinjptors for the immunosuppressive drugs tacrolimusd cyclosporine.33 MIF represents a constitutively ex-essed cytokine that promotes transcription andnslation of immune and inflammatory genes by inhi-ion of degradation of specific mRNA and transcrip-n factors.34

Chronic peritoneal exposure to PDF has been sus-cted to be associated with sterile inflammatory pro-sses.9 The danger model (injury-induced inflamma-n)35 suggests sensing of cellular injury by MCs withbsequent induction of peritoneal sterile inflamma-n: indeed, extracts from NCM, used as danger sig-ls, recently induced cytokine and chemokine pro-ction in MCs via membrane-bound IL-1 receptor-1R).36 IL-1R signals through a conserved intracel-ar region, termed the Toll/interleukin-1 receptor (TIR)main, representing a relevant overlap with TLR andF signaling pathways.37,38

ll Damage and PDF Exposure Inducelammatory Signaling via IL-1R

the next part of this study, we adapted the experimentsm Eigenbrod et al36 by substituting NCM from macro-age and liver cell extracts with extracts from MCs. In

PANTHER process

Immunity anddefense

Stressresponse

Proteincomplexassembly

Proteinfolding

Proteinmetabolism

andmodification

1 1 1 1 11 1 1 1 11 1 1 1 11 1 1 11 1 1 11 11 1 11

1 11 11 1

111

1111

1ceanprtrabittio

pecetiosutionadu(ILluldoTN

CeInf

Infrophs approach, exposure of normal MCs to injured MCsing MC-derived NCM as surrogate for ambient cellury) induced sterile inflammation, which resulted in el-

evThthecoforbiocotraTNorinfsethereltonof

oufla

whalswatheresofartinfIL-redthaforim

Figtionsupas arelatatireperroeffeantwis(* P

FigNCtheenzconind6 p

1552 Kratochwill et alAJP April 2011, Vol. 178, No. 4ated levels of pro-inflammatory cytokines IL-6 and IL-8.e addition of anakinra led to a marked reduction ofse inflammation markers. The drug anakinra is a re-mbinant polypeptide that represents a shortened iso-m analog of the human IL-1R antagonist (IL-1Ra). Thelogical activity of anakinra derives from its ability tompetitively inhibit IL-1 binding to the IL-1R.39 In con-st, addition of a monoclonal antibody against humanF- (infliximab) did not alter the elevated levels of IL-6IL-8 after NCM exposure. This ineffective inhibition oflammation by blocking TNF- likely reflects the ab-nce of this ligand in the pure MC-derived NCM used forstimulation experiments. Certainly, TNF- and other

ure 4. Necrotic cell material (NCM) induces sterile inflammation. Secre-of the pro-inflammatory cytokines IL-6 (A) and IL-8 (B) into the mediumernatant after pre-incubation of MCs with NCM from homogenized cellsssessed by enzyme-linked immunosorbent assay. Mean values are showntive to control without NCM pretreatment (white bars) and are represen-ve for three independent biological experiments with two biologicallicates each (n 6 per data point). The error bars represent the standardr. NCM pre-treatment is shown as gray bars. Hatched bars show thects of the addition of the IL-1Ra anakinra (single hatched) and thei-TNF-mab infliximab (cross-hatched). Significant differences in group-e comparisons from analysis of variance tests are indicated with brackets 0.05, ** P 0.01).evant cytokines play a relevant role in intercellular peri-eal cross talk and in MC fate in more complex modelsthe in vivo situation of PD and peritonitis.25 However,

is sIL-1hatvarr data provide clear evidence for IL-1Rmediated in-mmation in MCs following NCM exposure.Based on this model, we searched for evidenceether IL-1R-dependent cellular signal transduction iso responsible for the activation of inflammatory path-ys in MCs following PDF exposure. Although sublethal,chosen PDF stress caused sufficient MC damage toult in the release not only of intracellular LDH but alsodanger signals into the supernatant. Similar to theificial addition of NCM, PDF was able to induce sterilelammation as evidenced by elevated levels of IL-6 and8. Again, the addition of anakinra led to a substantialuction of these inflammation markers, demonstratingt endogenous ligands of the IL-1R were responsiblethe activation of pro-inflammatory signaling in exper-ental PD.

ure 5. PDF induces a pattern of pro-inflammatory interleukins similar toM. Secretion of the pro-inflammatory cytokines IL-6 (A) and IL-8 (B) intomedium supernatant after exposure of MCs to PDF as assessed byyme-linked immunosorbent assay. Mean values are shown relative totrol without PDF exposure (white bars) and are representative for threeependent biological experiments with two biological replicates each (ner data point). The error bars represent the standard error. PDF exposure

hown as gray bars. Hatched bars show the effects of the addition of theRa anakinra (single hatched) and the anti-TNF-mab infliximab (cross-ched). Significant differences in group-wise comparisons from analysis ofiance tests are indicated with brackets (* P 0.05, ** P 0.01).

Cyas

InmatoxcyredmaeqstaizethaMCnopetor

spoumadeshinmepatheindce

Figgraor Pplereplogerrovarsupblonordecindbet

Figposaftefromthediffbetproprotwoicalrep

Figmemotion

Inflammation Impairs Heat Shock Response 1553AJP April 2011, Vol. 178, No. 4tokine Receptor Antagonists Might Be UsedPotential Cytoprotective Agents

the final part of this study, we related sterile inflam-tion to increased susceptibility of MCs to PDF cyto-icity. The addition of anakinra indeed resulted intoprotection of MCs, as demonstrated by markedlyuced MC damage with blockade of sterile inflam-tion. PDF exposure times could be doubled forual amounts of cell death. LDH release followingndard exposure of 1 hour was completely normal-d to control levels. These findings clearly indicatet IL-1Rmediated signal transduction sensitizeds to PDF cytotoxicity. The clinical relevance of thisvel pathomechanism was recently pointed out in aritonitis model, where high levels of pro-inflamma-y ligands in PD effluents could also be blocked

ure 6. Addition of anakinra enables higher expression of Hsp72. Theph shows the results of a time-course experiment with PDF (black bars)DF supplemented with the IL-1Ra anakinra (shaded bars). Control sam-s underwent sham treatment with normal growth medium. Data areresentative for three independent biological experiments with two bio-ical replicates each (n 6). Bars represent mean values relative to control;r bars are shown as standard error. A: LDH release was assessed afterious exposure times. Enzymatic measurements were carried out in theernatant after a recovery of 16 hours. B: Hsp72 was assessed by Westernt from total protein lysates after a recovery of 16 hours. Values are

malized to the expression of -tubulin to compensate for a potentialrease in cell number with increasing PDF exposure time. Asterisksicate time-points with significant differences in pair-wise comparisonsween PDF and PDF supplemented with anakinra (t-test; P 0.05).

simleapatexaecifically by IL-1b antibodies.40 It has been previ-sly shown that pro-inflammatory stimuli sensitize hu-n cells to subsequent insults, exacerbating cellularath.41 This phenomenon has been called the heatock paradox.42 Although the inflammatory stimulusthese experiments consisted of lipopolysaccharide-diated Toll-like receptor signaling, this signalingthway is known to overlap with the IL-1R pathway viaTIR domain, suggesting that IL-1 and LPS mightuce a common effector mechanism that sensitizells to subsequent stress.

ure 7. Proteomics evaluation of anakinra effects on MCs after PDF ex-ure. Expression profiles of proteins with known association to PDF stressr addition of the IL-1Ra anakinra to PDF. The abundance as measuredtwo-dimensional gel staining is given relative to control conditions of

initial experiment. All proteins presented in this graph show significanterences in spot abundance as well between control and PDF treatment asween PDF and PDF supplemented with anakinra (P 0.05). Expressionfiles are labeled with Swiss-Prot entry names as given in Table 1. The twofiles for Hsp27 (HSPB1_HUMAN) refer to two separate spots on the-dimensional gels. Data are representative for three independent biolog-experiments with two biological replicates each (n 6). Error bars

resent the SE.

ure 8. New concept of peritoneal dialysis fluid (PDF) stress. Scheme ofsothelial cell monolayer exposed to peritoneal dialysis PDF. The currentdel of cytotoxic damage is extended with the model of sterile inflamma-, which is revealed by pathway analysis and confirmed in experiments

ulating sterile inflammation. Danger signals, released from injured cells,d to sterile inflammation in adjacent cells via IL-1 receptor (IL-R)-mediatedhways. This sterile inflammation affects cytoprotective responses andcerbates cell death on PD fluid exposure.

meduanbioofcaexofofIL-suGrproglunaceshthetwtioknaltannonainfcegeLPsiothaogPD

deinDaeleilestetivexbiodablope

Re

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

1554 Kratochwill et alAJP April 2011, Vol. 178, No. 4To further deepen our understanding of molecularchanisms responsible for cytoprotection by re-ced cellular inflammation, we analyzed effects ofakinra on expression of proteins involved in relevantlogical processes during recovery from PDF. A thirdthe re-detected proteins demonstrated a signifi-ntly altered abundance (p 0.05) in the anakinraperiment, indicating the relevance of one or severalthese players in this phenomenon. The major subgroupproteins that were increased following blockade of1R consisted of cytoprotective molecular chaperonsch as Hsp27 (HSPB1), Grp78, and calreticulin.43

p78 and calreticulin are endoplasmic reticulum stressteins and are expected to be up-regulated by thecose-based PDF.17,44 It is conceivable that IL-1R sig-ling results in a dampening of these effectors of thellular stress response, and that this dampening (heatock paradox-like phenomenon) could be released byaddition of anakinra. The differential abundance of

o Hsp27 isoforms are caused by altered phosphoryla-n of this actin-binding chaperon, most likely due toown anakinra effects on p38 activation.45 Additionalered proteins are involved in protein synthesis (eEF2)d protein degradation (ubiquitin). These proteins havet previously been associated with IL-1R-mediated sig-ling but likely reflect the general effects of reducedlammation, such as improved protein turnover. A re-nt transcriptomics study comparing the globalnomic response to cells subjected to heat shock orS stress has also demonstrated overlapping expres-n patterns between these stimuli in a subset of genest has functional annotation to chemokine related biol-y, including IL-1, supporting indirectly our data in theF exposure model.46

In summary, this new concept of PDF stress allows tolineate the deleterious effect of sterile inflammationMCs after exposure to cytotoxic PDF (Figure 8).nger signals, released from injured MCs, lead to anvated level of cytokine release associated with ster-inflammation via IL-1R-mediated pathways. Thisrile inflammation reduces expression of cytoprotec-e chaperones and exacerbates MC damage on PDFposure. Our findings, based on proteomics andinformatics techniques as well as confirming stan-rd techniques, raise the possibility that agents thatck the actions of IL-1 might be useful in limitingritoneal damage during PD.

ferences

Topley N, Kaur D, Petersen MM, Jorres A, Williams JD, Faict D,Holmes CJ: In vitro effects of bicarbonate and bicarbonate-lactatebuffered peritoneal dialysis solutions on mesothelial and neutrophilfunction. J Am Soc Nephrol 1996, 7:218224Jorres A, Bender TO, Finn A, Witowski J, Frohlich S, Gahl GM, Frei U,Keck H, Passlick-Deetjen J: Biocompatibility and buffers: effect ofbicarbonate-buffered peritoneal dialysis fluids on peritoneal cell func-tion. Kidney Int 1998, 54:21842193

Witowski J, Topley N, Jorres A, Liberek T, Coles GA, Williams JD:Effect of lactate-buffered peritoneal dialysis fluids on human perito-neal mesothelial cell interleukin-6 and prostaglandin synthesis. Kid-ney Int 1995, 47:282293Kratochwill K, Lechner M, Siehs C, Lederhuber HC, Rehulka P, En-demann M, Kasper DC, Herkner KR, Mayer B, Rizzi A, Aufricht C:Stress responses and conditioning effects in mesothelial eells ex-posed to peritoneal dialysis fluid. J Proteome Res 2009, 8:17311747Lechner M, Kratochwill K, Lichtenauer A, Rehulka P, Mayer B, AufrichtC, Rizzi A: A proteomic view on the role of glucose in peritonealdialysis. J Proteome Res 2010, 9:24722479Arbeiter K, Bidmon B, Endemann M, Bender TO, Eickelberg O, Ruff-ingshofer D, Mueller T, Regele H, Herkner K, Aufricht C: Peritonealdialysate fluid composition determines heat shock protein expressionpatterns in human mesothelial cells. Kidney Int 2001, 60:19301937Bidmon B, Endemann M, Arbeiter K, Ruffingshofer D, Regele H,Herkner K, Eickelberg O, Aufricht C: Overexpression of HSP-72 con-fers cytoprotection in experimental peritoneal dialysis. Kidney Int2004, 66:23002307Endemann M, Bergmeister H, Bidmon B, Boehm M, Csaicsich D,Malaga-Dieguez L, Arbeiter K, Regele H, Herkner K, Aufricht C:Evidence for HSP-mediated cytoskeletal stabilization in mesothelialcells during acute experimental peritoneal dialysis. Am J PhysiolRenal Physiol 2007, 292:F4756Flessner MF: Inflammation from sterile dialysis solutions and thelongevity of the peritoneal barrier. Clin Nephrol 2007, 68:341348Hooper PL: Insulin Signaling. GSK-3, heat shock proteins and thenatural history of type 2 diabetes mellitus: a hypothesis. Metab SyndrRelat Disord 2007, 5:220230Shevchenko A, Wilm M, Vorm O, Mann M: Mass spectrometric se-quencing of proteins silver-stained polyacrylamide gels. Anal Chem1996, 68:850858Perco P, Rapberger R, Siehs C, Lukas A, Oberbauer R, Mayer G,Mayer B: Transforming omics data into context: bioinformatics ongenomics and proteomics raw data. Electrophoresis 2006, 27:26592675Brown KR, Jurisica I: Online predicted human interaction database.Bioinformatics 2005, 21:20762082Platzer A, Perco P, Lukas A, Mayer B: Characterization of protein-interaction networks in tumors. BMC Bioinformatics 2007, 8:224Mi H, Guo N, Kejariwal A, Thomas PD: PANTHER version 6: proteinsequence and function evolution data with expanded representationof biological pathways. Nucleic Acids Res 2007, 35:D247252Arbeiter K, Bidmon B, Endemann M, Ruffingshofer D, Mueller T,Regele H, Eickelberg O, Aufricht C: Induction of mesothelial HSP-72upon in vivo exposure to peritoneal dialysis fluid. Perit Dial Int 2003,23:499501Aufricht C, Endemann M, Bidmon B, Arbeiter K, Mueller T, Regele H,Herkner K, Eickelberg O: Peritoneal dialysis fluids induce the stressresponse in human mesothelial cells. Perit Dial Int 2001, 21:8588Schett G, Steiner CW, Groger M, Winkler S, Graninger W, Smolen J,Xu Q, Steiner G: Activation of Fas inhibits heat-induced activation ofHSF1 and up-regulation of hsp70. FASEB J 1999, 13:833842Alscher DM, Biegger D, Mettang T, van der Kuip H, Kuhlmann U, FritzP: Apoptosis of mesothelial cells caused by unphysiological charac-teristics of peritoneal dialysis fluids. Artif Organs 2003, 27:10351040Xu ZG, Kim KS, Park HC, Choi KH, Lee HY, Han DS, Kang SW: Highglucose activates the p38 MAPK pathway in cultured human perito-neal mesothelial cells. Kidney Int 2003, 63:958968Vargha R, Bender TO, Riesenhuber A, Endemann M, Kratochwill K,Aufricht C: Effects of epithelial-to-mesenchymal transition on acutestress response in human peritoneal mesothelial cells. Nephrol DialTransplant 2008, 23:34943500Amore A, Cappelli G, Cirina P, Conti G, Gambaruto C, Silvestro L,Coppo R: Glucose degradation products increase apoptosis of hu-man mesothelial cells. Nephrol Dial Transplant 2003, 18:677688Catalan MP, Subira D, Reyero A, Selgas R, Ortiz-Gonzalez A, EgidoJ, Ortiz A: Regulation of apoptosis by lethal cytokines in humanmesothelial cells. Kidney Int 2003, 64:321330Yang AH, Chen JY, Lin YP, Huang TP, Wu CW: Peritoneal dialysissolution induces apoptosis of mesothelial cells. Kidney Int 1997,51:12801288Santamaria B, Benito-Martin A, Ucero AC, Aroeira LS, Reyero A,

Vicent MJ, Orzaez M, Celdran A, Esteban J, Selgas R, Ruiz-Ortega M,Cabrera ML, Egido J, Perez-Paya E, Ortiz A: A nanoconjugate Apaf-1inhibitor protects mesothelial cells from cytokine-induced injury. PLoSONE 2009, 4:e6634

26. Ravagnan L, Gurbuxani S, Susin SA, Maisse C, Daugas E, ZamzamiN, Mak T, Jaattela M, Penninger JM, Garrido C, Kroemer G: Heat-shock protein 70 antagonizes apoptosis-inducing factor. Nat Cell Biol2001, 3:839843

27. Mosser DD, Caron AW, Bourget L, Denis-Larose C, Massie B: Role ofthe human heat shock protein hsp70 in protection against stress-induced apoptosis. Mol Cell Biol 1997, 17:53175327

28. Mehlen P, Schulze-Osthoff K, Arrigo AP: Small stress proteins asnovel regulators of apoptosis. Heat shock protein 27 blocks Fas/APO-1- and staurosporine-induced cell death J Biol Chem 1996,271:1651016514

29. Bando Y, Tsukamoto Y, Katayama T, Ozawa K, Kitao Y, Hori O, SternDM, Yamauchi A, Ogawa S: ORP150/HSP12A protects renal tubularepithelium from ischemia-induced cell death. FASEB J 2004, 18:14011403

30. Li J, Lee AS: Stress induction of GRP78/BiP and its role in cancer.Curr Mol Med 2006, 6:4554

31. Bender TO, Riesenhuber A, Endemann M, Herkner K, Witowski J,Jorres A, Aufricht C: Correlation between HSP-72 expression andIL-8 secretion in human mesothelial cells. Int J Artif Organs 2007,30:199203

32. Jorres A, Ludat K, Lang J, Sander K, Gahl GM, Frei U, DeJonge K,Williams JD, Topley N: Establishment and functional characterizationof human peritoneal fibroblasts in culture: regulation of interleukin-6production by proinflammatory cytokines. J Am Soc Nephrol 1996,7:21922201

33. Liu J, Farmer JD, Jr., Lane WS, Friedman J, Weissman I, SchreiberSL: Calcineurin is a common target of cyclophilin-cyclosporin A andFKBP-FK506 complexes. Cell 1991, 66:807815

34. Calandra T, Roger T: Macrophage migration inhibitory factor: a reg-ulator of innate immunity. Nat Rev Immunol 2003, 3:791800

35. Matzinger P: The danger model: a renewed sense of self. Science2002, 296:301305

36. Eigenbrod T, Park JH, Harder J, Iwakura Y, Nunez G: Cutting edge:critical role for mesothelial cells in necrosis-induced inflammationthrough the recognition of IL-1 alpha released from dying cells.J Immunol 2008, 181:81948198

37. van der Most RG, Currie AJ, Robinson BW, Lake RA: Decodingdangerous death: how cytotoxic chemotherapy invokes inflammation,immunity or nothing at all. Cell Death Differ 2008, 15:1320

38. Chen CJ, Kono H, Golenbock D, Reed G, Akira S, Rock KL: Identifi-cation of a key pathway required for the sterile inflammatory responsetriggered by dying cells. Nat Med 2007, 13:851856

39. Larsen CM, Faulenbach M, Vaag A, Volund A, Ehses JA, Seifert B,Mandrup-Poulsen T, Donath MY: Interleukin-1-receptor antagonist intype 2 diabetes mellitus. N Engl J Med 2007, 356:15171526

40. Witowski J, Tayama H, Ksiazek K, Wanic-Kossowska M, Bender TO,Jorres A: Human peritoneal fibroblasts are a potent source of neu-trophil-targeting cytokines: a key role of IL-1beta stimulation. LabInvest 2009, 89:414424

41. Chen Y, Voegeli TS, Liu PP, Noble EG, Currie RW: Heat shockparadox and a new role of heat shock proteins and their receptorsas anti-inflammation targets. Inflamm Allergy Drug Targets 2007,6:91100

42. DeMeester SL, Buchman TG, Cobb JP: The heat shock paradox:does NF-kappaB determine cell fate? FASEB J 2001, 15:270274

43. Ni M, Lee AS: ER chaperones in mammalian development and humandiseases. FEBS Lett 2007, 581:36413651

44. Aufricht C: HSP: helper, suppressor, protector. Kidney Int 2004,65:739740

45. Uekawa N, Nishikimi A, Isobe K, Iwakura Y, Maruyama M: Involve-ment of IL-1 family proteins in p38 linked cellular senescence ofmouse embryonic fibroblasts. FEBS Lett 2004, 575:3034

46. Wong HR, Odoms K, Sakthivel B: Divergence of canonical dangersignals: the genome-level expression patterns of human mononu-clear cells subjected to heat shock or lipopolysaccharide. BMCImmunol 2008, 9:24

Inflammation Impairs Heat Shock Response 1555AJP April 2011, Vol. 178, No. 4

Interleukin-1 Receptor-Mediated Inflammation Impairs the Heat Shock Response of Human Mesothelial CellsMaterial and MethodsCell CultureExperimental DesignFlow CytometryProtein Sample PreparationQuantification of Cytokine Release Into the SupernatantWestern BlottingStatistical TestsTwo-Dimensional Gel ElectrophoresisProtein Identification by Mass SpectrometryBioinformatics Procedures

ResultsResponses of MCs to PDF StressSterile Inflammation

DiscussionCell Damage and Repair Patterns in Response to PDF ExposurePotential Role of Inflammatory ProcessesCell Damage and PDF Exposure Induce Inflammatory Signaling via IL-1RCytokine Receptor Antagonists Might Be Used as Potential Cytoprotective Agents

References