Asbestos Induces Nitric Oxide Synthesis in Mesothelioma Cells via Rho Signaling Inhibition Chiara Riganti, Sara Orecchia, Francesca Silvagno, Gianpiero Pescarmona, Pier Giacomo Betta, Elena Gazzano, Elisabetta Aldieri, Dario Ghigo, and Amalia Bosia Department of Genetics, Biology and Biochemistry, and Interdepartmental Center “G. Scansetti” for Studies on Asbestos and Other Toxic Particulates, Universita di Torino; Research Center on Experimental Medicine (CeRMS), Torino; and Pathology Unit, ` Department of Oncology, Azienda Sanitaria Ospedaliera, Alessandria, Italy We have observed that in three human malignant mesothelioma cell lines, crocidolite asbestos induced the activation of the transcription CLINICAL RELEVANCE factor NF- B and the synthesis of nitric oxide (NO) by inhibiting the RhoA signaling pathway. The incubation with crocidolite decreased In human malignant mesothelioma cells, crocidolite asbes- the level of GTP-bound RhoA and the activity of Rho-dependent tos induces the activation of NF- B and the synthesis of kinase, and induced the activation of Akt/PKB and IkB kinase, nitric oxide by impairing the prenylation and subsequent leading to the nuclear translocation of NF- B. The effects of crocido- activation of RhoA. lite fibers on NF- B activation and NO synthesis were mimicked by Y27632 (an inhibitor of the Rho-dependent kinases) and toxin B (an inhibitor of RhoA GTPase activity), while they were reverted by mevalonic acid, the product of 3-hydroxy-3-methylglutaryl coen- eration, differentiation, and apoptosis (4). NO is synthesized by zyme A (HMGCoA) reductase. Furthermore, crocidolite, similarly to mevastatin, inhibited the synthesis of cholesterol and ubiquinone three NO synthase (NOS; EC 188.8.131.52) isoforms, which cata- and the prenylation of RhoA: these effects were prevented in the lyze the conversion of L-arginine to L-citrulline and NO with a presence of mevalonic acid. This suggests that crocidolite fibers 1:1 stoichiometry (5). Various stimuli, such as bacterial lipopoly- might inhibit the synthesis of isoprenoid molecules at the level saccharide, inﬂammatory cytokines, and oxidative stress, can of the HMGCoA reductase reaction or of an upstream step, thus stimulate the expression of the inducible NOS isoform (iNOS, impairing the prenylation and subsequent activation of RhoA. Akt NOS II) via the activation of NF- B (6). In addition, asbestos can stimulate NO synthesis via a double mechanism: it can activate ﬁbers have been shown to induce both NF- B activation and the inducible NO synthase via the NF- B pathway and the endothe- NO synthesis in alveolar macrophages, lung epithelial cells (7, lial NO synthase via a direct phosphorylation. Our results suggest 8), and, more recently, in mesothelial cells (9). that crocidolite increases the NO levels in mesothelioma cells by NF- B activity is controlled by members of the IkB family modulating both NO synthase isoforms. which bind directly to NF- B dimers in the cytoplasm, preventing the nuclear localization of the transcription factor, which is re- Keywords: crocidolite; mesothelioma; nitric oxide; RhoA; NF- B quired for DNA binding (3). Many agents activate NF- B via Human malignant mesothelioma (HMM) is a rare but aggressive serine phosphorylation, ubiquitination and proteasomal degra- tumor that originates from mesothelial cells and exhibits a strong dation of I B : in this way, NF- B is free to translocate to the correlation with exposure to asbestos ﬁbers, such as crocidolite, nucleus and modulate the expression of many genes, including amosite, and chrysotile (1, 2). Key changes in the development iNOS (6). of the disease are the loss of the normal restraints on prolifera- An increased iNOS activity has been induced in different tion and the acquisition of resistance to apoptosis: asbestos, cellular models by inhibitors of RhoA prenylation, such as statins which is a complete carcinogen for mesothelial cells, plays a (10–13), and by direct inhibitors of RhoA GTPase activity, such complex role in altering both proliferation and apoptosis (2). as the toxin B from Clostridium difﬁcile (10, 13). A link between NF- B is a redox-sensitive transcription factor comprised of inhibition of the small G proteins Rho and NF- B activation protein dimers, including the transcription-activating hetero- has been already suggested (13). dimer consisting of p50 and p65 subunits, that regulates expres- Starting from these observations, our study has been aimed sion of genes intrinsic to inﬂammation, cell proliferation, and to investigate in HMM cells the role played by the RhoA sig- apoptosis (3). One of the effects induced by NF- B activation naling pathway in crocidolite-induced NO synthesis and cell is the increased synthesis of nitric oxide (NO), a highly reactive proliferation. molecule involved in different cellular functions, including prolif- MATERIALS AND METHODS Materials (Received in original form January 9, 2006 and in final form November 29, 2006 ) Fetal bovine serum (FBS) and Ham’s F-12 nutrient mixture medium were supplied by BioWhittaker (Verviers, Belgium); plasticware for This work has been supported with grants from Fondazione Internazionale cell culture was from Falcon (Becton Dickinson, Bedford, MA); the Ricerche Medicina Sperimentale (FIRMS), Compagnia di SanPaolo, Regione Piemonte (Ricerca Sanitaria Finalizzata CIPE A201, 2004/2005), and Ministero cationic exchange resin Dowex AG50WX-8, N-(1-naphthylethylenedia- dell’Universita e della Ricerca. E.G. is a recipient of a post-doctoral fellowship ` mine) dihydrochloride and sulfanilamide were from Aldrich (Milan, funded by Regione Piemonte. Italy); L-[2,3,4,5-3H]arginine monohydrochloride (62 Ci/mmol) was ob- Correspondence and requests for reprints should be addressed to Dario Ghigo, tained from Amersham International (Bucks, UK); Y27632 was from Dipartimento di Genetica, Biologia e Biochimica (Sezione di Biochimica), Via Calbiochem (La Jolla, CA). Electrophoresis reagents were obtained Santena, 5/bis, 10126 Torino, Italy. E-mail: firstname.lastname@example.org from Bio-Rad Laboratories (Hercules, CA), and the protein content Am J Respir Cell Mol Biol Vol 36. pp 746–756, 2007 of cell monolayers and cell lysates was assessed with the BCA kit from Originally Published in Press as DOI: 10.1165/rcmb.2006-0011OC on February 22, 2007 Pierce (Rockford, IL). When not otherwise speciﬁed, other reagents Internet address: www.atsjournals.org were purchased from Sigma Chemical Co (St. Louis, MO). Riganti, Orecchia, Silvagno, et al.: Asbestos Inhibits Rho Signaling 747 Cells 2 g/ml leupeptin, and 0.1% NP-40, pH 7.6). This suspension was incubated for 10 min on ice with occasional vortexing, and centrifuged HMM cell lines MM98, OC99, and GF99 were obtained from the pleural for 30 s at 13,000 g to pellet nuclei, which were rinsed with 0.2 ml effusions of three patients with histologically conﬁrmed malignant me- of wash buffer B (2 M KCl, 25 mM Hepes, 0.1 mM EDTA, 1 mM sothelioma; the mesothelial origin of the isolated cells was conﬁrmed PMSF, 1 mM DTT, 10 g/ml aprotinin, and 2 g/ml leupeptin, pH 7.6) by positive immunostaining as previously described (14). Cells were and incubated at 4 C for 20 min. Then an equal volume of buffer C cultured in Ham’s F-12 medium supplemented with 10% FBS, 1% penicillin/streptomycin, and 1% L-glutamine, and were maintained in (25 mM Hepes, 0.1 mM EDTA, and 20% glycerol, pH 7.6) was added, a humidiﬁed atmosphere at 37 C and 5% CO2. The N11 mouse glial the mix was centrifuged for 15 min at 20,000 g, and the supernatant cell line was a gift from Dr. Marco Righi (CNR Institute of Neurosci- stored at 80 C until used for Western blotting or electrophoretic ence, Section of Cellular and Molecular Pharmacology, Milan, Italy). mobility shift assay (EMSA). At the end of each incubation period the experiments were performed on adherent cells, after having removed the incubation medium together EMSA with detached cells. When measurements were performed on extracellu- The probe containing the NF- B oligonucleotide consensus sequence lar medium, the medium was centrifuged previously at 12,000 g for was labeled with [ -32P]ATP (Amersham International) (3,000 Ci/mmol, 15 min to pellet cellular debris. During the 24-h incubation period 250 Ci), using T4 polynucleotide kinase (Roche, Basel, Switzerland). preceding experiments aimed to investigate cell protein phosphoryla- The sequence of oligonucleotide was (binding site underlined): 5 -AGT tion (I B , I , eNOS), cells were maintained with the above- TGAGGGGACTTTCCCAGG-3 (Promega Corporation, Madison, mentioned culture medium containing a low level of FBS (2% instead WI). Ten g of the nuclear extracts, obtained as described above, were of 10%), to minimize the known effects of serum on the baseline pattern incubated for 20 min with 20,000 cpm of 32P-labeled double-stranded of protein phosphorylation. MM98, OC99, and GF99 cells showed the oligonucleotide at 4 C in a reaction mixture containing: 2 l of 10 g/ml same behavior under each experimental condition: for sake of simplicity BSA, 2 l of buffer D (100 mM KCl, 20 mM Hepes, 0.5 mM EDTA, we chose to report in Results the data obtained in MM98 cells as 2 mM DTT, 0.1 mM PMSF, 20% glycerol, and 0.25% NP-40, pH 7.6), representative of the other two cell lines. 4 l of buffer E (300 mM KCl, 100 mM Hepes, 10 mM DTT, 100 M PMSF, and 20% Ficoll, pH 7.6) and 2 g of poly(dI-dC) (Roche). The Asbestos Fibers ﬁnal volume of the mix was brought to 25 l with water. In supershift UICC (Union International Contre le Cancer) crocidolite ﬁbers were assay, nuclear extracts were pre-incubated for 30 min at room tempera- ¨ sonicated (Labsonic sonicator, 100 W, 10 s; Labsonic, Gottingen, ture with 2 l of anti-p50 antibody (PC136; Calbiochem) or anti-p65 Germany) before incubation with cell cultures, in order to dissociate antibody (PC138; Calbiochem); then the reaction mixture containing ﬁber bundles and allow their better suspension and diffusion in the the 32P-labeled double-stranded oligonucleotide was added and samples culture medium. were treated as described previously. The DNA–protein complex was separated on a not-denaturating 4% polyacrilamide gel in TBE buffer Propidium Iodide Exclusion Assay (pH 8.0). After electrophoresis, the gel was dried and autoradiographed After incubation under different experimental conditions in 24-well by exposure to X-ray ﬁlm for 48 h. plates, cells were washed twice with fresh PBS and incubated for 10 min at room temperature in 1 ml of binding buffer (100 mM Hepes/ Western Blot Analysis NaOH, pH 7.5, 140 mM NaCl, 25 mM CaCl2) containing 2.5 M propid- Cells were directly solubilized in the lysis buffer (25 mM Hepes, ium iodide (PI). Then cells were washed three times with fresh PBS 135 mM NaCl, 1% NP40, 5 mM EDTA, 1 mM EGTA, 1 mM ZnCl2, and rinsed with 1 ml of binding buffer. An aliquot of cells suspension 50 mM NaF, 10% glycerol), supplemented with protease inhibitor cock- was sonicated and used to determine intracellular protein content. Fluo- tail set III (100 mM AEBSF, 80 M aprotinin, 5 mM bestatin, 1.5 mM rescence of each sample was recorded using a Perkin-Elmer LS-5 spec- E-64, 2 mM leupeptin, and 1 mM pepstatin; Calbiochem), 2 mM PMSF, troﬂuorimeter (Shelton, CT). Excitation and emission wavelengths were and 1 mM sodium orthovanadate. Whole cell extracts (or nuclear ex- 536 and 617 nm, respectively. A blank was prepared testing the cells tracts, for p50 and p65 detection) containing 30 g of proteins were in the absence of PI in each set of experiments, and its ﬂuorescence separated by SDS-PAGE, transferred to PVDF membrane sheets was subtracted from that measured in the samples. (Immobilon-P; Millipore, Bedford, MA), and probed with the following antibodies: anti-I B (from rabbit, diluted 1:500 in PBS-BSA 1%; Santa Lactate Dehydrogenase Activity Cruz Biotechnology, Santa Cruz, CA), anti-phospho(Ser 32)-I B After incubation under different experimental conditions, the extracel- (from mouse, diluted 1:250 in PBS-BSA 1%; Santa Cruz Biotechnol- lular medium was centrifuged at 12,000 g for 15 min to pellet cellular ogy), anti-I B kinase(IKK) (from rabbit, diluted 1:500 in PBS-BSA debris, whereas cells were washed with fresh medium, detached with 1%, Santa Cruz Biotechnology), anti-phospho(Ser 180)-IKK (from trypsin/EDTA, washed with PBS, re-suspended at 1 105 cells/ml in rabbit, diluted 1:250 in PBS-BSA 1%; Cell Signaling Technology Inc., 0.2 ml of 82.3 mM triethanolamine phosphate hydrochloride (TRAP, Beverly, MA), anti-Rho kinase (Rock) 1 (from rabbit, diluted 1:500 in pH 7.6), and sonicated on ice with two 10-s bursts. Lactate dehydroge- PBS-BSA 1%; Santa Cruz Biotechnology), anti-Rock2 (from rabbit, nase (LDH) activity was measured in the extracellular medium and in diluted 1:500 in PBS-BSA 1%; Santa Cruz Biotechnology), anti-eNOS the cell lysate, as previously described (15). One hundred microliters (from mouse, diluted 1:500 in PBS-BSA 1%; Transduction Labora- of supernatant from extracellular medium or 10 l of cell lysate were tories, Lexington, KY), anti-phospho-(Ser 1177) eNOS (from mouse, incubated at 37 C with 82.3 mM TRAP (pH 7.6) and 5 mM NADH diluted 1:500 in PBS-BSA 1%; Cell Signaling Technology Inc.), anti- (ﬁnal volume: 1 ml). The reaction was started by adding 20 mM pyruvic p50 (from mouse, diluted 1:250 in PBS-BSA 1%; Santa Cruz Biotech- acid and was followed for 10 min, measuring absorbance at 340 nm nology), anti-p65 (from rabbit, diluted 1:500 in PBS-BSA 1%; Santa with a Lambda 3 spectrophotometer (Perkin-Elmer). The reaction ki- Cruz Biotechnology), and anti-glyceraldehyde-3-phosphate dehydroge- netics were linear throughout the time of measurement. Both intracellu- nase (anti-GAPDH, from rabbit, diluted 1:500 in PBS-BSA 1%; Santa lar and extracellular enzyme activity were expressed as mol NADH Cruz Biotechnology). Expression of GAPDH, the product of an oxidized/min/dish, then extracellular LDH activity was calculated as housekeeping gene, was used as a control of equal loading. After a percentage of the total LDH activity in the dish. 1-h incubation, the membrane was washed with PBS-Tween 0.1% and subjected for 1 h to a peroxidase-conjugated anti-rabbit, anti-mouse, Preparation of Nuclear Extracts or anti-goat IgG (Amersham International, diluted 1:1,000 in PBS- Cells were plated in 100-mm-diameter dishes at conﬂuence, and all Tween with Blocker Not-Fat Dry Milk 5% [Bio-Rad]). The membrane procedures for nuclear protein extraction were performed at 4 C using was washed again with PBS-Tween and proteins were detected by ice-cold reagents, as described (16). Cells were mechanically scraped enhanced chemiluminescence (Immun-Star; Bio-Rad). To assess the in PBS, washed, and resuspended (1 107 cells/0.5 ml) in lysis buffer presence of the Proliferating Nuclear Cellular Antigen (PCNA), lysates A (15 mM KCl, 10 mM Hepes, 2 mM MgCl2, 0.1 mM EDTA, 1 mM were directly resolved on a 12% SDS-PAGE, transferred to PVDF phenylmethylsulfonylﬂuoride [PMSF], 1 mM DTT, 10 g/ml aprotinin, membrane sheets and probed overnight with an anti-PCNA antibody 748 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 36 2007 (from mouse, 1:500 in PBS-BSA 1%; Santa Cruz Biotechnology). The was incubated for 45 min at 4 C with agarose-glutathione beads coupled membrane was then washed and treated as described above. to a fusion protein containing glutathione S-transferase (GST) and the Rho-binding domain of the Rho effector protein rhotekin (Upstate, IKK Activity Assay Charlottesville, VA). The beads were then washed three times in MLB IKK activity was measured as previously described (17). Cells were buffer and harvested by addition of Laemmli buffer 2 . The RhoA washed with ice-cold PBS and solubilized in 0.5 ml of lysis buffer activity was analyzed, resolving the samples by 12% SDS-PAGE and (50 mM Hepes, 150 mM NaCl, 2 mM MgCl2, 1 mM EDTA, 0.1% NP40, Western blotting using anti-RhoA antibody to detect GTP-bound acti- 100 M NaF, 10 mM sodium orthovanadate, 10 g/ml leupeptin, vated RhoA. RhoA-GTP/total RhoA ratio was taken as an index of 10 g/ml pepstatin, 10 g/ml aprotinin, 1 mM PMSF, and 250 M the active fraction of RhoA. DTT). Samples were centrifuged at 13,000 g for 15 min and the supernatant was used for the assay and cell protein quantiﬁcation. To Rock Assay purify the IKK complex, equal amounts of the whole lysate (0.5 mg Rock activity was measured using the CycLex Rho-Kinase Assay Kit cell proteins/test) were immunoprecipitated with an anti-IKK antibody (CycLex Co.), a single site binding immunoassay. Cells were cultured (from rabbit, diluted 1:200 in PBS-BSA 1%; Santa Cruz Biotechnology) in 35-mm-diameter Petri dishes, washed with ice-cold PBS and lysed for 90 min at 4 C. Samples were centrifuged (13,000 g for 15 min) in 0.2 ml lysis buffer (50 mM Tris-HCl, pH 8.0, 0.1% Triton X-100, and washed three times with kinase buffer (20 mM Hepes, 20 mM 1mM EDTA, 1mM EGTA, 2 mM NaF, 2 mM sodium orthovanadate, -glycerolphosphate, 1 mM MnCl2, 5 mM MgCl2, 2 mM NaF, and 0.5 g/ml leupeptin, 1 g/ml pepstatin, 0.2 mM PMSF, and 10 mM 250 M DTT). A quantity of 0.1 g of the immunoprecipitated proteins -mercaptoethanol). Samples were sonicated on crushed ice with two was incubated with 1 mM ATP in the presence of the proteasome 10-s bursts and centrifuged at 13,000 g for 5 min at 4 C. Supernatants inhibitor MG132 (10 M); to provide the reaction mix with an excess were treated following manufacturer’s instructions, and protein content of I B protein (the substrate of IKK), 30 g of total cellular lysate was measured. Brieﬂy, samples were diluted 1:4 in the kinase buffer of SV40-positive HMM cells, obtained under not-denaturing conditions, provided with the kit, containing 20 mM ATP, and incubated fo were added. SV40-positive HMM cells were chosen because they were 60 min at 30 C in 96-well plates, pre-coated with the recombinant previously shown to exhibit a very high basal amount of I B (9). C-terminus of the myosin binding subunit (MBS) of myosin phospha- MG132 was added in the reaction mixture in order to avoid, during tase. Wells were washed ﬁve times with 2% Tween-20, and 100 l of the assay, the degradation of phospho-IkB protein by the proteasome the HRP-conjugated anti phospho(Thr 696)-MBS antibody were added. eventually still present in the cell lysate (used as a provider of the After a 60-min incubation at room temperature, samples were washed substrate). Reaction was carried over at 30 C for 30 min and stopped again, and 100 l of the chromogenic substrate tetra-methylbenzidine with 30 l of Laemmli buffer. Finally, samples were subjected to electro- were added. After a 15-min incubation at room temperature, the reac- phoresis in a 12% SDS-PAGE, transferred to PVDF membrane sheets, tion was stopped with 100 l of 0.5 N H2SO4 and absorbance was and probed with an anti-I B antibody (from rabbit, diluted 1:250 in read at 450 nm, using a Packard EL340 microplate reader (Bio-Tek PBS-BSA 1%; Santa Cruz Biotechnology) and an anti-phospho(Ser Instruments). For each set of experiments, a titration curve was pre- 32)-IkB antibody (from mouse, diluted 1:250 in PBS-BSA 1%; Santa pared, using serial dilution of recombinant Rho-Kinase II (MBL Inc., Cruz Biotechnology), respectively. Woburn, MA) in kinase buffer. Data were expressed as mU absorbance/ mg cell proteins. Akt Activity Assay Akt activity was measured using the CycLex Akt/PKB Kinase Assay/ RT-PCR Inhibitor Screening Kit (CycLex Co., Nagano, Japan). Cells were cul- Total RNA was obtained by the guanidinium thiocyanate-phenol- tured in 35-mm-diameter Petri dishes, washed with ice-cold PBS, and chloroform method (19). A quantity of 30 ng of total RNA was reversely lysed in 0.2 ml lysis buffer (50 mM Tris-HCl, pH 8.0, 0.1% Triton X-100, 1mM EDTA, 1mM EGTA, 10 mM NaF, 2 mM sodium orthovan- transcribed into cDNA with the Superscript II One-Step RT-PCR Sys- adate, 0.5 g/ml leupeptin, 1 g/ml pepstatin, 0.5 mM PMSF, and tem with Platinum Taq DNA Polymerase (cycling conditions: 1 cycle 10 mM -mercaptoethanol). Samples were sonicated on crushed ice 50 C for 30 min, 1 cycle 94 C for 2 min). cDNA products were deter- with two 10-s bursts and centrifuged at 13,000 g for 30 min at 4 C. mined by PCR ampliﬁcation, carried out in a total volume of 50 l, Supernatants were treated following manufacturer’s instructions, and according to the manufacturer’s recommendations. The RT-PCR efﬁ- protein content was measured. Brieﬂy, 15 l of samples were diluted ciency was controlled by amplifying a -actin fragment, used as an in 85 l of the kinase buffer provided with the kit, containing 125 M housekeeping gene. Primers for iNOS (0.3 M) were: 5 -TCCGAGG ATP, and incubated for 60 min at 30 C in 96-well plates, pre-coated CAAACAGCACATTCA-3 , 5 -GGGTTGGGGGTGTGGTGATGT-3 with the Akt substrate AkTide-2T, which is efﬁciently phosphorylated (462 bp); primers for -actin (0.5 M) were: 5 -GGTCATCTTCTCG by Akt at a serine residue. Wells were washed ﬁve times with 2% CGGTTGGCCTTGGGGT-3 , 5 -CCCCAGGCACCAGGGCGTGAT-3 Tween-20, and 100 l of horseradish peroxidase (HRP)-conjugated (230 bp). PCR ampliﬁcation for iNOS was: 1 cycle of denaturation at anti-phospho-AkTide-2T monoclonal antibody were added. After a 95 C for 2 min, 30 cycles of denaturation at 95 C for 30 s, annealing 60-min incubation at room temperature, samples were washed again, at 55 C for 1 min, elongation at 72 C for 30 s, and 1 cycle of extension and 100 l of the chromogenic substrate tetra-methylbenzidine were at 72 C for 10 min; for -actin: 1 cycle of denaturation at 94 C for added. After a 15-min incubation at room temperature, the reaction was 3 min, 35 cycles of denaturation at 94 C for 1 min, annealing at 58 C stopped with 100 l of 0.5 N H2SO4 and absorbance was read at 450 nm, for 1 min, elongation at 72 C for 1 min, and 1 cycle of extension at using a Packard EL340 microplate reader (Bio-Tek Instruments, Winooski, 72 C for 7 min. Samples were electrophoresed in 1.5% agarose gels VT). For each set of experiments, a titration curve was prepared, using containing ethidium bromide in Tris-acetate/EDTA buffer to visualize serial dilutions of recombinant Akt (CycLex Co.) in kinase buffer. Akt the PCR products. activity was expressed as mU absorbance/mg cell proteins. Nitrite Production RhoA-GTP Pull-Down Conﬂuent cell monolayers in 35-mm-diameter Petri dishes were incu- Biochemical assay for activity of RhoA was performed as described bated in fresh medium for 24 h under the experimental conditions (18). Cells were lysed in MLB buffer (125 mM Tris-HCl, pH 7.4, indicated in Results. Then nitrite production was measured by adding 750 mM NaCl, 1% NP40, 10% glycerol, 50 mM MgCl2, 5 mM EDTA, 0.15 ml of cell culture medium (centrifuged previously at 12,000 g 25 mM NaF, 1 mM sodium orthovanadate, 10 g/ml leupeptin, 10 g/ml for 15 min to pellet cellular debris) to 0.15 ml of Griess reagent (20) pepstatin, 10 g/ml aprotinin, and 1 mM PMSF) and centrifuged at in a 96-well plate, and, after a 10 min incubation at 37 C in the dark, 13,000 g for 10 min at 4 C. An aliquot of supernatant was taken out absorbance was measured at 540 nm with a Packard EL340 microplate for determination of protein content and analysis of total amount of reader (Bio-Tek Instruments). A blank was prepared for each experi- RhoA. Another aliquot of the same lysate was directly probed with an mental condition in the absence of cells, and its absorbance was sub- anti-RhoA antibody (1:250, in PBS-BSA 1%; Santa Cruz Biotechnol- tracted from that measured in the samples. Nitrite concentration was ogy), to measure total RhoA protein. A further 30 g of the supernatant expressed as nmol nitrite/mg cell proteins. Riganti, Orecchia, Silvagno, et al.: Asbestos Inhibits Rho Signaling 749 Measurement of NOS Activity ice-cold PBS, detached by trypsin/EDTA, and resuspended in 200 l of PBS. A 50- l aliquot was used for protein quantiﬁcation, while the Cells grown at conﬂuence on 35-mm-diameter Petri dishes, after incuba- remaining part was transferred in poliethylene vials and the radioactiv- tion under the experimental conditions described in Results, were ity was measured by liquid scintillation. [3H]thymidine incorporated in detached by trypsin/EDTA, washed with PBS, resuspended in 0.3 ml each sample was expressed as pmol/mg cell proteins. of Hepes/EDTA/dithiotreitol (DTT) buffer (20 mM Hepes, 0.5 mM EDTA, 1 mM DTT, pH 7.2) and then sonicated on crushed ice with Statistical Analysis two 10-s bursts. In each assay tube, the following reagents were added to 100 l of lysate at the following ﬁnal concentrations: 2 mM NADPH, All data in text and ﬁgures are provided as means SE. The results were analysed by a one-way ANOVA and Tukey’s test. P 0.05 was 2.5 Ci L-[3H]arginine ( 0.4 M), 100 M tetrahydrobiopterin, considered signiﬁcant. 1.5 mM CaCl2 (20). After a 15-min incubation at 37 C, the reaction was stopped by adding 2 ml Hepes-Na/EDTA buffer (20 mM Hepes sodium salt, 2 mM EDTA, pH 6); the whole reaction mixture was RESULTS applied to 2 ml columns of Dowex AG50WX-8 (Na form) and eluted Crocidolite Asbestos Elicits in HMM Cells a Dose- and Time- with 4 ml of water. The radioactivity corresponding to [3H]citrulline Dependent Increase of Both NO Synthesis and LDH Leakage, content in 6.1 ml eluate was measured by liquid scintillation counting. Citrulline synthesis was expressed as pmol citrulline/min/mg cell Which Are Inhibited by Coincubation with Mevalonic Acid proteins. We investigated the effects of different concentrations (1, 5, 25 g/cm2) of crocidolite asbestos ﬁbers on HMM cells viability Measurement of Isoprenoid Molecules Synthesis after a 24- to 48-h incubation (Figure 1). The incorporation of The synthesis of isoprenoid molecules was checked as intracellular PI, a marker of decreased cell viability (23), increased as a function accumulation of cholesterol and ubiquinone, and measured as pre- of time and concentration. We investigated further the effects viously described (21). Cells were incubated for 24 h with 10 Ci/ml of the incubation of HMM cells with different concentrations of [3H]acetate (3600 mCi/mmol; Amersham International) or 10 Ci/ml (1, 5, 25 g/cm2) of crocidolite asbestos ﬁbers, under different of [14C]mevalonic acid (67 mCi/mmol; Amersham International), then incubation times (6, 24, 48 h), on the extracellular accumulation washed twice with PBS and resuspended in 200 l of PBS. A 50- l of nitrite (a stable derivative of NO synthesis), and on the release aliquot was used for protein quantiﬁcation, while the remaining part of LDH in the extracellular medium (a sensitive index of cytotox- was transferred in glass microcentrifuge tubes. A quantity of 1.5 ml of icity) (24) (Figure 2A). Since after 6 h the NO synthesis and a 1:2 methanol/hexane solution was added, and cellular suspensions cytotoxic effect were negligible, whereas after 48 h the cytotoxic were vortexed for 1 h, then centrifuged at 2,000 g for 5 min. The effect was too marked, we decided to perform the subsequent upper phase was transferred in a new set of glass microcentrifuge tubes, while the lower phase was resuspended in 1 ml of hexane, vortexed experiments using a 24-h incubation time and a 25 g/cm2 con- overnight, and centrifuged at 2,000 g for 5 min: the new upper phase centration of ﬁbers, which appeared to be the minimal dose was added to the previously isolated phase. After a 24-h evaporation, among our conditions able to induce signiﬁcant accumulation samples were dissolved in 100 l of chloroform and resolved by thin of nitrite. Under these experimental conditions, we incubated layer chromatography on silica gel, using 1:1 diethyl ether/exane as the cells in the absence or presence of variable concentrations mobile phase. Standard solutions of 10 l of cholesterol (2 g/ml) and (50, 100, 200 M) of mevalonic acid (Figure 2B). A quantity of ubiquinone (2 g/ml) were employed. After the separation, the gel was 100 M mevalonic acid completely reverted both nitrite increase exposed to iodine gas for 2 h and the spots corresponding to cholesterol and LDH leakage induced by crocidolite. and ubiquinone were isolated. The radioactivity of each spot was mea- sured by liquid scintillation counting and expressed as cpm/mg cellular Crocidolite Asbestos and Inhibitors of RhoA and proteins. Rocks Activate NF- B in HMM Cells After a 24-h incubation, crocidolite, Y27632 (an inhibitor of the Assessment of RhoA Prenylation Rocks) and toxin B (an inhibitor of RhoA GTPase activity) After 24 h of treatment cells were lysed with 2% ice cold Triton X-114 in Tris buffered saline pH 7.4 and phase-separated as previously described (22) with some modiﬁcations. Cells were harvested in lysis buffer (25 mM Tris-HCl, pH 7.4, 150 mM NaCl, 2% Triton X-114, 5 mM MgCl2, 1 mM Na2HPO4, 1 mM sodium orthovanadate, 1 mM PMSF, protease inhibitor cocktail set III; Calbiochem) and incubated for 30 min at 4 C. After sonication, insoluble material was removed by centrifugation at 13,000 g for 10 min at 4 C and the supernatant was phase-separated. Brieﬂy, each sample was overlaid on a sucrose cushion and after warming at 37 C for 3 min the turbid solution was centrifuged at 300 g for 5 min at room temperature to separate the hydrophobic and aqueous phase. Both phases were collected and the separation was repeated. Finally, the aqueous phase was rinsed with 2% Triton X-114 without sucrose cushion and the detergent phase of this last condensa- tion was discarded. The protein content of total lysate, aqueous phase, and detergent phase was determined by Bradford test (Bio-Rad), and 6 g aliquots were analyzed by SDS-PAGE. An anti-RhoA antibody (Santa Cruz Biotechnology) was used to evaluate RhoA distribution between aqueous phase (soluble unprenylated form of RhoA) and detergent phase (hydrophobic prenylated form of RhoA), whereas an Figure 1. Effect of crocidolite on PI exclusion in HMM cells. HMM anti-GAPDH antibody (Santa Cruz Biotechnology) was used to verify (MM98) cells were incubated for 24 and 48 h in the absence (CTRL) the partitioning of soluble GAPDH in the aqueous phase. or presence of crocidolite (CRO, 1, 5, 25 g/cm2). Camptothecin (CAM, 10 nM for 24 h), a genotoxic and proapoptotic compound in mesotheli- [3H]thymidine Incorporation Assay oma cells, was used as a positive control. After these incubation times, Cells were grown in 35-mm-diameter Petri dishes and incubated for 24 h cells were checked for their content of PI, as described in MATERIALS AND under the experimental conditions described in Results, in a culture METHODS. Measurements were performed in duplicate and data are medium containing 1 Ci/ml [3H]thymidine (62 Ci/mmol; Amersham presented as means SE (n 6). *P 0.05, **P 0.001 versus CTRL; International). At the end of the incubation, cells were washed with P 0.05 versus CRO 25 g/cm2 24 h. 750 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 36 2007 Figure 2. Effect of crocidolite and mevalonic acid on NO syn- thesis and LDH leakage in HMM cells. HMM (MM98) cells were incubated for 6, 24, 48 h in the absence (CTRL) or presence of crocidolite (CRO, 1, 5, 25 g/cm2) (A ), and for 24 h in the absence (CTRL) or presence of crocidolite (CRO, 25 g/cm2) and/or mevalonic acid (MA, 50, 100, 200 M) (B ). After these incubation times, nitrite concentration in the extracellular medium and LDH activity in both extracellular medium and cell lysate were measured as described in MATERI- ALS AND METHODS. Each measurement was performed in dupli- cate. The experiments were repeated with the other two cell lines (OC99, GF99), giving superimposable results (data not shown). (A ) Data are presented as means SEM (n 6). *P 0.05, **P 0.001 versus CTRL. (B ) Data are presented as means SEM (n 5). *P 0.05, ** p 0.001 versus CTRL; P 0.05 versus CRO. strongly activated NF- B in HMM cells. The NF- B nuclear lysate) to phosphorylate its substrate I B : asbestos ﬁbers, as translocation was demonstrated by EMSA (Figure 3A) and by well as Y27632 and toxin B, induced IKK activity in HMM cells Western blotting performed on nuclear extracts using anti-p50 (Figure 4A). Mevalonic acid, per se devoid of any effect on IKK and anti-p65 subunits antibodies (Figure 3B). A similar nuclear activity, completely prevented the crocidolite-induced increase translocation was observed in mouse N11 glial cells (used as of IKK activity (Figure 4A). Activation of IKK implies its a positive control) incubated with bacterial lipopolysaccharide phosphorylation on serine and threonine (3): asbestos ﬁbers (LPS), a well-known NF- B activator and iNOS inducer (Figures signiﬁcantly increased the intracellular amount of phospho(Ser 3A and 3B). Cells were also incubated with mevalonic acid, the 180)-IKK , without changing the total levels of IKK (Figure product of 3-hydroxy-3-methylglutaryl coenzyme A (HMGCoA) 4B). Y27632 and toxin B also elicited an increment of phospho reductase: mevalonic acid alone did not induce nuclear transloca- (Ser 180)-IKK . Mevalonic acid, both alone and incubated together tion of the p50/p65 heterodimer, but it completely abolished with crocidolite, prevented IKK phosphorylation (Figure 4B). the effect of crocidolite (Figures 3A and 3B). Under the same LPS induced both IKK activity and phosphorylation in N11 cells experimental conditions, the NF- B inhibitor I B , which was (Figure 4). abundant in the cytosol of resting HMM cells, disappeared in Akt/PKB activates the IKK complex in some experimental cells incubated with crocidolite, Y27632, and toxin B (Figure 3B). models (25): after the incubation of HMM cells with crocidolite The appearance of phosphorylated I B followed a complemen- asbestos, the Akt activity in the cell lysate increased (Figure 5), tary pattern (Figure 3B). Again, mevalonic acid completely re- an effect which was completely prevented when cells were incu- verted the effect of asbestos ﬁbers on I B (Figure 3B). In N11 bated with crocidolite and mevalonic acid together (Figure 5). glial cells LPS induced a clear disappearance of the cytosolic A signiﬁcant increase of intracellular Akt activity was elicited also level of I B and the appearance of phospho-I B (Figure 3B). by Y27632 and toxin B in HMM cells and by LPS in N11 cells (Figure 5). Crocidolite Asbestos and Inhibitors of RhoA and of Rocks Elicit Since RhoA can bind GTP only when prenylated, GTP-bound IKK Phosphorylation, Akt Activation, and RhoA/Rock RhoA can be measured as an index of RhoA prenylation and Signaling Inhibition in HMM Cells activation (26). In HMM cells crocidolite ﬁbers markedly low- IKK is responsible for I B phosphorylation and degradation ered the level of GTP-bound RhoA, without changing the ex- (3). IKK activity in HMM cells was assayed as the ability of IKK pression of total RhoA: this effect was prevented when crocido- complex (immunoprecipitated from equal amounts of whole cell lite was incubated together with mevalonic acid (Figure 6A). As Riganti, Orecchia, Silvagno, et al.: Asbestos Inhibits Rho Signaling 751 Figure 3. Effect of crocidolite, mevalonic acid, and inhibitors of RhoA/Rock on the NF- B pathway. HMM (MM98) cells were incubated for 24 h without (CTRL) or with crocidolite fibers (25 g/cm2, CRO), in the absence or presence of mevalonic acid (MA, 100 M), Y27632 (Y276, 0.2 M), or toxin B (TOX, 0.1 ng/ml). N11 cells, incubated with bacterial lipopolysaccharide (LPS, 20 g/ml), were used as a positive control ( ); in each experiment one lane was loaded with bidistilled water ( ) in place of cellular extracts; in the absence of LPS the pattern was superimposable to that of HMM CTRL (not shown). (A ) EMSA detection of NF- B nuclear translocation, as described in MATERIALS AND METHODS. In the lane marked with Anti p50 and Anti p65, a supershift assay was performed on HMM cells incubated with 25 g/cm2 crocidolite fibers (see MATERIALS AND METHODS). EMSA was repeated also in the other two cell lines (OC99, GF99), giving superimposable results (data not shown). (B ) The nuclear translocation of the two subunits of NF- B, p50 and p65, was detected with Western blotting performed on nuclear extracts as described in MATERIALS AND METHODS. The expression of I B , phospho-I B , and GAPDH was detected with Western blotting performed on whole cellular extracts (see MATERIALS AND METHODS for details). Detection of GAPDH, evidence of housekeeping gene expression, was used as a control of equal loading in this and the subsequent experiments. The level of I B followed the same pattern in the other two cell lines (OC99, GF99). expected, Y27632 did not modify the amount of GTP-bound Crocidolite Increases the Expression of NOS II, the Activity RhoA or total RhoA protein, while toxin B reduced the level of NOS III, and the Synthesis of NO via NF- B and of GTP-bound RhoA (Figure 6A). After a 24-h incubation, Akt Activation in HMM Cells crocidolite asbestos, Y27632, and toxin B lowered the activity After a 24-h incubation, crocidolite, Y27632, and toxin B induced of Rock in HMM cells, and mevalonic acid reverted the effect the expression of iNOS mRNA in HMM cells: the effect of of crocidolite (Figure 6B). Neither crocidolite, Y27632, nor toxin crocidolite was blocked by the co-incubation with mevalonic B changed the expression of Rock1 and Rock2 proteins (data acid (Figure 8A). None of these agents modiﬁed the transcript not shown). of -actin, chosen as an housekeeping gene (Figure 8A). Crocidolite Inhibits the Synthesis of Isoprenoid Molecules and In a number of cellular models Akt phosphorylates the consti- the Prenylation of RhoA in HMM Cells tutive endothelial NOS (eNOS, NOS III) on serine 1177, causing a further enzyme activation (28). In untreated HMM cells most The ability of crocidolite to inhibit the synthesis of isoprenoid eNOS was in the nonphosphorylated form, and crocidolite in- molecules was checked by investigating the incorporation of 3H duced a marked increase of phospho-eNOS (Figure 8B). This and 14C in cholesterol and ubiquinone in HMM cells incubated effect was prevented by mevalonic acid. with [3H]acetate or [14C]mevalonic acid. In the presence of either These effects on NOS expression and phosphorylation were crocidolite or the HMGCoA reductase inhibitor mevastatin, the accompanied by changes of NOS activity. After a 24-h incuba- intracellular conversion of [3H]acetate into both cholesterol and tion, crocidolite, Y27632, and toxin B induced a signiﬁcant aug- ubiquinone was signiﬁcantly decreased (Figure 7A). Contrarily, mentation of extracellular nitrite (an NO stable derivative in the conversion of [14C]mevalonic acid to these two molecules oxygenated cell systems) and intracellular NOS activity in HMM was not affected, whereas in the same experimental conditions cultures (Figure 9A). Mevalonic acid signiﬁcantly prevented the squalene synthase inhibitor squalestatin (27) was fully effec- such effects (Figure 9A). Acting as NO scavengers (due to their tive in inhibiting cholesterol synthesis (favoring, as expected, a content of oxyhemoglobin), red blood cells lowered the nitrite greater conversion of mevalonic acid to the isoprenoid chain of concentration without modifying the NOS activity, as expected ubiquinone) (Figure 7B). We also observed that crocidolite, as (Figure 9A). well as mevastatin, inhibited the prenylation of RhoA, evaluated by investigating the relative amount of prenylated membrane- Crocidolite Induces Both Proliferation and Death of HMM associated and nonprenylated cytosolic RhoA in HMM cells Cells in an NO-Dependent Manner (Figure 7C). These effects were prevented by the presence of After a 24-h incubation, the expression of PCNA, an antigen mevalonic acid (Figure 7C). that accumulates in the cells during the early S phase (29), was 752 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 36 2007 Figure 5. Effect of crocidolite, mevalonic acid, and inhibitors of RhoA/ Rho kinase on Akt activity. HMM (MM98) cells were incubated for 24 h in fresh medium only (CTRL) or with crocidolite fibers (CRO, 25 g/ cm2), in absence or presence of mevalonic acid (MA, 100 M), Y27632 (Y276, 0.2 M), or toxin B (TOX, 0.1 ng/ml). Cell lysates were analysed for Akt activity as described in MATERIALS AND METHODS. N11 cells, incu- bated with bacterial lipopolysaccharide (LPS, 20 g/ml), were used as Figure 4. Effect of crocidolite, mevalonic acid, and inhibitors of RhoA/ a positive control; in the absence of LPS the pattern was superimposable Rho kinase on Ikk activation. HMM (MM98) cells were incubated for to that of HMM CTRL (not shown). Each measurement was performed 24 h in fresh medium only (CTRL) or with crocidolite fibers (CRO, in duplicate. The experiments were repeated with the other two cell 25 g/cm2), in the absence or presence of mevalonic acid (MA, 100 lines (OC99, GF99), giving superimposable results (data not shown). M), Y27632 (Y276, 0.2 M), or toxin B (TOX, 0.1 ng/ml). (A ) IKK Data are presented as means SEM (n 6). *P 0.05, **P 0.001 activity was assayed as the ability of the immunoprecipitated IKK com- versus CTRL; P 0.001 versus CRO. plex to phosphorylate I B , and its outcome was checked using an anti-phospho(Ser 32)-I B and an anti-I B antibody (as described in MATERIALS AND METHODS). MG132 (10 M) was added in the reaction mixture in order to avoid, during the assay, the degradation of phospho- also been implicated (30). The molecular basis of crocidolite- I B protein by the proteasome eventually still present in the SV40- positive HMM cell lysate (used as a provider of the substrate). associated carcinogenesis is still under investigation. Asbestos (B ) Detection of Ikk phosphorylation. Whole cell lysates were analysed can stimulate in a variety of cell types the expression of genes by Western blotting using the following antibodies: anti-phospho(- critical to cellular injury, proliferation, and inﬂammation, via Ser180)-I , anti-I , anti-GAPDH (as described in MATERIALS AND the activation of transcription factors, such as activated protein-1 METHODS). N11 cells, incubated with LPS ( , 20 g/ml), were used as and NF- B (31). The strong oxidative stress evoked by asbestos a positive control; in the absence of LPS the pattern was superimposable is likely to be responsible for NF- B activation (31), but other to that of HMM CTRL (not shown). Each panel is representative of two mechanisms may be involved. NF- B activation is known to be experiments with similar results. The experiments shown in B were promoted by different serine/threonine kinases, such as mitogen- repeated in duplicate with the other two cell lines (OC99, GF99), giving activated protein kinases and Akt (3). superimposable results (data not shown). We have previously observed that crocidolite can promote nuclear translocation of NF- B in HMM cells (9). The results of the present work show that the crocidolite-dependent nuclear translocation of NF- B is elicited through the inhibition of the clearly increased in HMM cells by crocidolite ﬁbers, Y27632, RhoA signaling pathway. RhoA is involved in different cellular and toxin B (Figure 9B). This effect was prevented by mevalonic crucial events, such as proliferation, tumorigenesis, and tumor acid and packed red blood cells, with a pattern similar to that invasion (32). It usually cycles between a nonprenylated, inactive observed in nitrite measurements (Figure 9A). Since PCNA form and a prenylated, GTP-bound active form (26): if active, may increase also as a consequence of DNA damage, we also RhoA can interact with several downstream effectors, such as the measured the cell incorporation of [3H]thymidine under the same two serine-threonine Rho-dependent kinases Rock1 and Rock2 experimental conditions: again, crocidolite, Y27632, and toxin B (33). The effect of RhoA modulation on NF- B activation may increased the DNA synthesis (Figure 10). In parallel, crocidolite, change depending on the cellular model. We observed that in Y27632, and toxin B increased also the leakage of LDH (a HMM cells crocidolite ﬁbers promoted NF- B translocation into sensitive marker of cell damage), thus suggesting that these the nucleus, elicited the phosphorylation of I B and lowered agents can exert both mitogenic and cytotoxic effects (Figure the intracellular content of I B protein. A similar response 10). Mevalonic acid and packed erythrocytes prevented these pattern was observed when cells were incubated with both toxin effects: this phenomenon was less evident after 48 h (not shown), B from Clostridium difﬁcile, which inhibits the GTPase activity probably because of mevalonic acid consumption and oxidation of RhoA and other small G proteins (34), and Y27632, a selective of red blood cell hemoglobin to methemoglobin (unable to scav- Rock inhibitor (35). Mevalonic acid, the product of HMGCoA enge NO). reductase, being a substrate for the synthesis of isoprenoid mole- cules may favor the prenylation and activation of RhoA. In our DISCUSSION experimental conditions mevalonic acid, which per se did not Crocidolite asbestos is considered to be the major causative factor change the nuclear content of NF- B, completely prevented in the development of HMM, although other co-carcinogens, such both the NF- B activation and the phosphorylation and decrease as simian virus 40 (SV40) infection and genetic factors, have of I B induced by crocidolite. Riganti, Orecchia, Silvagno, et al.: Asbestos Inhibits Rho Signaling 753 Figure 6. Effect of crocidolite and mevalonic acid on GTP binding of RhoA and on Rho kinase activity. HMM (MM98) cells were cultured for 24 h in the absence (CTRL) or presence of the following compounds, alone or in different combinations: crocidolite fibers (CRO, 25 g/cm2), mevalonic acid (MA, 100 M), Y27632 (Y276, 0.2 M), or toxin B from C. difficile (TOX, 0.1 ng/ml). Subsequently, the cells were lysed and checked for (A ) expression of RhoA-GTP and total RhoA (see MATERIALS AND METHODS; panel is representative of two experiments with similar results), and (B ) Rho kinase activity (see MATERIALS AND METHODS); data are presented as means SE (n 6). *P 0.001 versus CTRL; P 0.05 versus CRO. The experiments were repeated with the other two cell lines (OC99, GF99), giving superimposable results (data not shown). Figure 7. Effect of crocidolite and mevastatin on the synthesis of iso- prenoid molecules and on the prenylation and expression of RhoA. HMM (MM98) cells were cultured for 24 h with [3H]acetate (A ) or [14C]mevalonic acid (B ), in the absence (CTRL) or presence of the follow- The IKK complex phosphorylates I B favoring its proteaso- ing compounds, alone or in different combinations: crocidolite fibers mal degradation and allowing NF- B to translocate to the nu- (CRO, 25 g/cm2), mevastatin (MVS, 100 M), and squalestatin (SQ, cleus. IKK is fully active only when phosphorylated, mainly on 1 M). Subsequently, the cells were lysed and checked for incorporation serine 176 and 180 of IKK (3). In nonstimulated HMM cells, of 3H (A ) or 14C (B ) into cholesterol and ubiquinone (see MATERIALS AND IKK was weakly phosphorylated and its activity (measured as METHODS). Data are presented as means SE (n 6). *P 0.05, **P the ability of immunoprecipitated intracellular IKK complex to 0.001 versus CTRL. (C ) After the incubation, cells were lysed and phosphorylate I B ) was not detectable, while both crocidolite checked for the expression of total, prenylated, and nonprenylated and inhibitors of RhoA and Rock increased both IKK activity RhoA and of GAPDH (see MATERIALS AND METHODS). The figure is represen- and the amount of phospho-IKK . Mevalonic acid completely tative of two experiments with similar results. The experiments show blocked the effect of crocidolite. in A and B were repeated in duplicate with the other two cell lines (OC99, GF99), giving superimposable results (data not shown). The mechanisms of IKK phosphorylation have been exten- sively investigated and many serine/threonine kinases seem to be involved (3). For instance, in some cell types IKK is a target of the kinase activity of Akt/PKB, which promotes NF- B trans- In other experimental models, a relationship between IKK / location, cellular survival, and/or inhibition of apoptosis (36). I B /NF- B pathway and RhoA and Rock signaling has already Interestingly, crocidolite has been shown to activate Akt (by been proposed (12, 13). Exposure of HMM cells to crocidolite inducing its phosphorylation on serine 473) in a rat pleural meso- ﬁbers induced a signiﬁcant decrease of GTP-bound (active) thelial cell line (37). After a 24-h incubation of HMM cells with RhoA and of Rock activity, a result similar to that observed crocidolite, Akt activity was signiﬁcantly increased, an effect pre- after the incubation with toxin B, a speciﬁc RhoA inhibitor (34). vented by the presence of mevalonic acid. Moreover, Y27632 and Again, mevalonic acid prevented these crocidolite effects. toxin B, although at a lower extent, induced a signiﬁcant increase To our knowledge, this is the ﬁrst evidence that crocidolite of the Akt activity as well. So far, the crocidolite-dependent activa- inhibits the RhoA/Rock pathway. Furthermore, our data suggest tion of Akt is likely to be linked to the inhibition of RhoA and that the asbestos elicits the activation of Akt, IKK, and NF- B Rock. by inducing the blockage of RhoA signaling: indeed, crocidolite, 754 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 36 2007 Until now, no data are available about the effect of crocidolite asbestos on the activity of the other NOS isoforms in human mesothelioma. Since we found that crocidolite and inhibitors of RhoA pathway activate Akt, we investigated whether they could exert some effect on eNOS, which is a well-known target of Akt (28). Actually, crocidolite induced phosphorylation of eNOS in HMM cells. It is already known that Akt can stimulate NO synthesis via a double mechanism: it can activate the NF- B/ iNOS pathway by phosphorylating IKK (25), and it may phos- phorylate the eNOS isoform, leading to its activation (28). Our results suggest that crocidolite increases the NO levels in HMM cells by modulating both iNOS and eNOS. Prenylated RhoA seems to play an important role in inducing this redundant signal, because in the presence of mevalonic acid crocidolite was unable to induce iNOS, phosphorylate eNOS, and increase nitrite accu- mulation. The role of RhoA and Rock in crocidolite-induced expression of iNOS RNA is inferred indirectly by the similar effect of pharmacologic inhibitors of Rho signaling pathway: this hypothesis has to be conﬁrmed with the use of dominant- negative mutants of Rock. A number of experimental evidences suggests that asbestos may activate simultaneously both proliferation and apoptosis Figure 8. (A ) Effect of crocidolite, mevalonic acid and inhibitors of in mesothelial cells (2, 38) and lung epithelial cells (39). This RhoA/Rho kinase on iNOS induction. HMM (MM98) cells were incu- phenomenon could be observed also in our cell cultures. Crocido- bated in fresh medium only (CTRL) or with crocidolite fibers (CRO, 25 lite, as well as Y27632 and toxin B, increased the DNA synthesis, g/cm2), in the absence or presence of mevalonic acid (MA, 100 M), but also the leakage of LDH (a sensitive marker of cell damage), Y27632 (Y276, 0.2 M), and toxin B (TOX, 0.1 ng/ml). After 24 h, thus appearing able to exert both mitogenic and cytotoxic effects, total RNA was extracted and subjected to RT-PCR for iNOS mRNA, as which were prevented by mevalonic acid and packed erythrocytes. described in MATERIALS AND METHODS. One lane was loaded with bidistilled Interestingly, crocidolite, Y27632, and toxin B modiﬁed NOS activ- water ( ) in place of cellular extracts. The figure is representative of two experiments per cell line with similar results. (B ) Effect of crocidolite ity and nitrite levels according to DNA synthesis and LDH release. on eNOS phosphorylation. HMM (MM98, OC99, and GF99) cells were This suggests that NO could play a role both in the cytotoxic and incubated 24 h in the absence (CTRL) or presence on crocidolite fibers in the proliferative stimulus induced by crocidolite and by inhibitors (CRO, 25 g/cm2), in the absence or presence of mevalonic acid (MA, of RhoA signaling. Mevalonic acid, favoring the prenylation of 100 M), Y27632 (Y276, 0.2 M), and toxin B (TOX, 0.1 ng/ml). RhoA, is likely to inhibit cell toxicity and proliferation by decreas- Cellular extracts were analyzed by Western blotting with an anti-eNOS ing the synthesis of NO induced by crocidolite, Y27632, and toxin antibody and an anti-phospho(Ser1177)eNOS antibody, as indicated B. To conﬁrm this statement, the presence of packed erythrocytes, in MATERIALS AND METHODS. The figure is representative of two experi- used as scavengers of NO, clearly inhibited nitrite accumulation, ments with similar results. The experiments were repeated with the DNA synthesis, and LDH leakage under each experimental con- other two cell lines (OC99, GF99), giving superimposable results (data dition. This suggests that the prenylation status of RhoA may not shown). play a critical role in both mesothelioma apoptosis and prolifera- tion. Although NO is more known to act as a pro-apoptotic and antiproliferative agent, it has been also observed to inhibit apoptosis and stimulate cellular proliferation, depending on the like mevastatin, inhibits both the synthesis of isoprenoid mole- cell type and the NO concentration (4, 40); thus it may play a cules (cholesterol, ubiquinone) and the prenylation of RhoA. role as both pro- or antitumoral agent (41). These effects are prevented in the presence of mevalonic acid, During the preparation of this manuscript, a paper has been which allows cells to synthesize isoprenoid groups and then to published (42) showing that in primary mesothelial cells TNF- keep RhoA in the prenylated, active form. This observation induces NF- B activation and protects from crocidolite asbestos suggests that crocidolite ﬁbers might inhibit the synthesis of cytotoxicity. Since TNF- may induce Akt activation and NO isoprenoid molecules at the level of the HMGCoA reductase synthesis, we investigated its effect on HMM cells: after a 24-h reaction or of an upstream step, causing a depletion of mevalonic incubation with 10 ng/ml TNF- , we did not observe any signiﬁ- cant increase of Akt or NOS activity (data not shown). More- acid. over, TNF- did not revert the activation of Akt and NOS and We have previously shown that in HMM cells crocidolite the cytotoxic effect exerted by crocidolite in HMM (data not increases iNOS expression and NO synthesis via activation of shown). So far, our results would rule out a role for TNF- in NF- B (9). Several authors have reported an increased iNOS asbestos-induced NO synthesis, cytotoxicity, and Akt activation activity in cells incubated with different inhibitors of RhoA pro- in HMM cells. The transformed malignant phenotype of HMM teins, such as statins (10, 11) and inhibitors of geranylgeranyla- cells may account for their different response to TNF- in com- tion (13). Toxin B from C. difﬁcile (10) and Y27632 (13) also parison with normal mesothelial cells (42). elicited enhanced nitrite production in several cell lines. Our To verify whether mevalonic acid could inﬂuence the effects data show that, like crocidolite, inhibitors of RhoA and Rock of asbestos by modifying directly the ﬁber reactivity, we have may also increase iNOS mRNA and NO synthesis in HMM also incubated the crocidolite ﬁbers for 24 h in culture medium cells. More interestingly, overexpression and activation of iNOS without cells in the presence of mevalonic acid; the suspension induced by crocidolite are likely to be mediated by inhibition was then centrifuged, and ﬁbers were washed with fresh medium of the RhoA prenylation, since these effects were prevented by and ﬁnally incubated with cell cultures for 24 h. In the same the presence of mevalonic acid. experimental conditions previously shown in Results, such Riganti, Orecchia, Silvagno, et al.: Asbestos Inhibits Rho Signaling 755 Figure 9. Effects of crocidolite and signaling inhibitors on NO synthesis and on the level of PCNA. HMM (MM98) cells were grown for 24 h in the absence (CTRL) or presence of the following components, alone or in different combinations: crocidolite fibers (CRO, 25 g/cm2), mevalonic acid (MA, 100 M), Y27632 (Y276, 0.2 M), toxin B from C. difficile (TOX, 0.1 ng/ml), and packed red blood cells (RBC, 10 l/ml). (A ) NOS activity in cell lysates (open bars) and nitrite levels in the extracellular medium (hatched bars) was measured in tripli- cate, as described (see MATERIALS AND METHODS). Data are pre- sented as means SE (n 6). *P 0.05, **P 0.005 versus CTRL; P 0.05; P 0.001 versus CRO or Y276 or TOX, respectively. (B ) Whole extracts were subjected to Western blotting, using an anti-PCNA antibody, as described in MATERI- ALS AND METHODS; the figure is representative of two experi- ments with similar results. The experiments were repeated with the other two cell lines (OC99, GF99), giving superimposable results (data not shown). asbestos ﬁbers pretreated with mevalonic acid elicited in HMM In summary, our data suggest that crocidolite asbestos induces cells the same effects of nontreated ﬁbers, as far as nuclear NF- B activation and stimulates the synthesis of NO by inhib- translocation of NF- B, GTP binding of RhoA, and DNA syn- iting the RhoA signaling pathway. The increased production of thesis were concerned (data not shown). Furthermore, in order NO may be implicated in the proliferative response of HMM to check the putative antioxidant action of mevalonic acid, we cells that survive after exposure to asbestos ﬁbers. In our experi- measured the accumulation of thiobarbituric acid–reactive sub- mental conditions, mevalonic acid prevented any effect induced stances (TBARS), which are known markers of lipid peroxida- by crocidolite and by the HMGCoA reductase inhibitor mevas- tion. As expected, a 24-h incubation of HMM cells with crocido- tatin, under concentrations that per se did not exert toxic effects lite induced a signiﬁcant increase of TBARS accumulation versus and did not inhibit Rho kinase signaling. The reverting effect control (n 6, P 0.02; data not shown): the coincubation of of mevalonic acid suggests that crocidolite impairs the prenyla- cells with ﬁbers and mevalonic acid obtained the same effect of tion of RhoA, which is necessary to the activation of this small crocidolite alone, thus suggesting that MA does not exert per GTPAse. Although many steps of this pathway still remain un- se an antioxidant effect and that the correction elicited by MA on clear, we think that our results may open a new line of research crocidolite-dependent NF- B activation and RhoA GTP binding in the investigation of crocidolite effects on human mesothelial inhibition is not mediated by such a mechanism. cells and in the pathogenesis of malignant mesothelioma. Figure 10. Effects of crocidolite and signaling inhibi- tors on cell proliferation (measured as [3H]thymidine incorporation) and on the release of LDH in the extra- cellular medium. HMM (MM98) cells were grown for 24 h in the absence (CTRL) or presence of the follow- ing components, alone or in different combinations: crocidolite fibers (CRO, 25 g/cm2), mevalonic acid (MA, 100 M), Y27632 (Y276, 0.2 M), toxin B from C. difficile (TOX, 0.1 ng/ml), and packed red blood cells (RBC, 10 l/ml). Cell proliferation, extrapolated from the [3H]thymidine incorporation assay (open bars), and LDH leakage in the extracellular medium (hatched bars) were measured in triplicate, as de- scribed (see MATERIALS AND METHODS). Data are pre- sented as means SE (n 3). *P 0.05, **P 0.005 versus CTRL; P 0.05, P 0.005 versus CRO or Y276 or TOX, respectively. The experiments were re- peated with the other two cell lines (OC99, GF99), giving superimposable results (data not shown). 756 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL 36 2007 Conflict of Interest Statement : None of the authors has a financial relationship G, Bosia A. Chloroquine stimulates nitric oxide synthesis in murine, with a commercial entity that has an interest in the subject of this manuscript. porcine, and human endothelial cells. J Clin Invest 1998;102:595–605. 21. Seifert SC, Lucas JJ. Incorporation of mevalonate into dolichol and other References isoprenoids during estrogen-induced chick oviduct differentiation. Biochim Biophys Acta 1988;962:16–24. 1. Manning CB, Vallyathan V, Mossman BT. Diseases caused by asbestos: 22. Bordier C. Phase separation of integral membrane proteins in Triton mechanisms of injury and disease development. Int Immunopharmacol X-114 solution. J Biol Chem 1981;256:1604–1607. 2002;2:191–200. 23. Docherty NG, O’Sullivan OE, Healy DA, Murphy M, O’Neill AJ, 2. Leard LE, Broaddus VC. Mesothelial cell proliferation and apoptosis. Fitzpatrick JM, Watson RW. TGF-beta1-induced EMT can occur inde- Respirology 2004;9:292–299. pendently of its proapoptotic effects and is aided by EGF receptor 3. Viatour P, Merville MP, Bours V, Chariot A. Phosphorylation of NF- activation. Am J Physiol Renal Physiol 2006;290:F1202–F1212. kappaB and IkappaB proteins: implications in cancer and inﬂamma- 24. Kinnula VL, Aalto K, Raivio KO, Walles S, Linnainmaa K. Cytotoxicity tion. Trends Biochem Sci 2005;30:43–52. of oxidants and asbestos ﬁbers in cultured human mesothelial cells. 4. Murphy MP. Nitric oxide and cell death. Biochim Biophys Acta 1999; Free Radic Biol Med 1994;16:169–176. 1411:401–414. 25. Madrid LV, Wang CW, Guttridge DC, Schottelius AJG, Baldwin AS, 5. Nathan C, Xie Q-w. Regulation of biosynthesis of nitric oxide. J Biol Mayo MW. Akt suppresses apoptosis by stimulating the transactiva- Chem 1994;269:13725–13728. tion potential of the RelA/p65 subunit of NF-kB. Mol Cell Biol 2000; 6. Laroux FS, Pavlick KP, Hines IN, Kawachi S, Harada H, Bharwani S, 20:1626–1638. Hoffman JM, Grisham MB. Role of nitric oxide in inﬂammation. Acta 26. Laufs U, Liao JK. Targeting Rho in cardiovascular disease. Circ Res Physiol Scand 2001;173:113–118. 2000;87:526–528. 27. Baxter A, Fitzgerald BJ, Hutson JL, McCarthy AD, Motteram JM, Ross 7. Chao C-C, Park S-H, Aust AE. Participation of nitric oxide and iron in BC, Sapra M, Snowden MA, Watson NS, Williams RJ, et al. Squales- the oxidation of DNA in asbestos-treated human lung epithelial cells. tatin 1, a potent inhibitor of squalene synthase, which lowers serum Arch Biochem Biophys 1996;326:152–157. cholesterol in vivo. J Biol Chem 1992;267:11705–11708. 8. Quinlan TR, BeruBe KA, Hacker MP, Taatjes DJ, Timblin CR, Goldberg 28. Dimmeler S, Fleming I, Fisslthaler B, Hermann C, Busse R, Zeiher J, Kimberley P, O’Shaughnessy P, Hemenway D, Torino J, et al. Mech- AM. Activation of nitric oxide synthase in endothelial cells via Akt- anisms of asbestos-induced nitric oxide production by rat alveolar dependent phosphorylation. Nature 1999;399:601–605. macrophages in inhalation and in vitro models. Free Radic Biol Med 29. Tsurimoto T. PCNA, a multifunctional ring on DNA. Biochim Biophys 1998;24:778–788. Acta 1998;1443:23–29. 9. Aldieri E, Orecchia S, Ghigo D, Bergandi L, Riganti C, Fubini B, Betta 30. Pistolesi M, Rushtoven J. Malignant pleural mesothelioma: update, cur- PG, Bosia A. Simian virus 40 infection down-regulates the expression rent managment, and newer therapeutic strategies. Chest 2004;126: of nitric oxide synthase in human mesothelial cells. Cancer Res 2004; 1318–1329. 64:4082–4084. 31. Shukla A, Ramos-Nino M, Mossman B. Cell signaling and transcription 10. Hausding M, Witteck A, Rodriguez-Pascual F, von Eichel-Streiber C, factor activation by asbestos in lung injury and disease. Int J Biochem Forstermann U, Kleinert H. Inhibition of small G proteins of the Rho Cell Biol 2003;35:1198–1209. family by statins or Clostridium difﬁcile toxin B enhances cytokine- 32. Mackay DJG, Hall A. Rho GTPases. J Biol Chem 1998;273:20685–20688. mediated induction of NO synthase II. Br J Pharmacol 2000;131:553– 33. Kaibuchi K, Kuroda S, Amano M. Regulation of the cytoskeleton and 561. cell adhesion by the Rho family GTPases in mammalian cells. Annu 11. Kolyada AY, Fedtsov A, Madias NE. 3-hydroxy-3-methylglutaryl coen- Rev Biochem 1999;68:459–486. zyme A reductase inhibitors upregulate inducible NO synthase expres- 34. Sehr P, Joseph G, Genth H, Just I, Pick E, Aktories K. Glucosylation sion and activity in vascular smooth muscle cells. Hypertension 2001; and ADP ribosylation of Rho proteins: effects on nucleotide binding, 38:1024–1029. GTPase activity, and effector coupling. Biochemistry 1998;37:5296– 12. Kraynack NC, Corey DA, Elmer HL, Kelley TJ. Mechanism of NOS2 5304. 35. Ishizaki T, Uehata M, Tamechika I, Keel J, Nonomura K, Maekawa M, regulation by RhoGTPase signaling in airway epithelial cells. Am J Narumiya S. Pharmacological properties of Y-27632, a speciﬁc inhibi- Physiol Lung Cell Mol Physiol 2002;283:L604–L611. tor of Rho-associated kinase. Mol Pharmacol 2000;57:976–983. 13. Rattan R, Giri S, Singh AK, Singh I. RhoA negatively regulates cytokine- 36. Madrid LV, Mayo MW, Reuther JY, Baldwin AS. Akt stimulates the mediated inducible nitric oxide synthase expression in brain-derived transactivation potential of Rel/p65 subunit of NF-kB through utiliza- transformed cell lines: negative regulation of IKK . Free Radic Biol tion of the IkB kinase and activation of the mitogen-activated protein Med 2003;29:1037–1050. kinase p38. J Biol Chem 2001;276:18934–18940. 14. Orecchia S, Schillaci F, Salvio M, Libener R, Betta PG. Aberrant E- 37. Berken A, Abel J, Unfried K. 1-integrin mediates asbestos-induced cadherin and gamma-catenin expression in malignant mesothelioma phosphorylation of Akt and ERK1/2 in a rat pleural mesothelial cell and its diagnostic and biological relevance. Lung Cancer 2004;45:S37– line. Oncogene 2003;22:8524–8528. S43. 38. Faux SP, Houghton CE, Hubbard A, Patrick G. Increased expression of 15. Beutler E. Red cell metabolism: a manual of biochemical methods. New epidermal growth factor receptor in rat mesothelial cells correlates York: Grune & Stratton; 1971. with carcinogenicity of mineral ﬁbres. Carcinogenesis 2001;12:2275– 16. Aldieri E, Atragene D, Bergandi L, Riganti C, Costamagna C, Bosia A, 2280. Ghigo D. Artemisinin inhibits inducible nitric oxide synthase and 39. Manning CB, Cummins AB, Jung MW, Berlanger I, Timblin CR, Palmer nuclear factor NF-kB activation. FEBS Lett 2003;552:141–144. C, Taatjes DJ, Hemenway D, Vacek P, Mossman BT. A mutant epider- ` 17. Castrillo A, Diaz-Guerra MJM, Horetlano S, Martin-Sanz P, Bosca L. ` mal growth factor receptor targeted to lung epithelium inhibits Inhibition of IkB kinase and IkB phosphorylation by 15-deoxy- 12,14- asbestos-induced proliferation and proto-oncogene expression. Cancer prostaglandin J2 in activated murine macrophage. Mol Cell Biol Res 2002;62:4169–4175. 40. Dimmeler S, Zeiher AM. Nitric oxide and apoptosis: another paradigm 2000;20:1692–1698. for the double-edged role of nitric oxide. Nitric Oxide 1997;1:275–281. 18. Danen EH, Sonneveld P, Sonneberg A, Yamada KM. Dual stimulation 41. Jenkins DC, Charles IG, Thomsen LL, Moss DW, Holmes LS, Baylis of Ras/mitogen-activated protein kinase and RhoA by cell adhesion SA, Rhodes P, Westmore K, Emson PC, Moncada S. Roles of nitric to ﬁbronectin supports growth factor-stimulated cell cycle progression. oxide in tumor growth. Proc Natl Acad Sci USA 1995;92:4392–4396. J Cell Biol 2000;151:1413–1422. 42. Yang H, Bocchetta M, Kroczynska B, Elmishad AG, Chen Y, Liu Z, 19. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid Bubici C, Mossman BT, Pass HI, Testa JR, et al. TNF- inhibits guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem asbestos-induced cytotoxicity via NF-kB dependent pathway, a possi- 1987;162:156–159. ble mechanism for asbestos-induced oncogenesis. Proc Natl Acad Sci 20. Ghigo D, Aldieri E, Todde R, Costamagna C, Garbarino G, Pescarmona USA 2006;103:10397–10402.