Human Macrophages Constitute Targets for Immunotoxic Inorganic Arsenic Anthony Lemarie, Claudie Morzadec, Emilie Bourdonnay, Olivier Fardel and Laurent Vernhet This information is current as of September 28, 2021. J Immunol 2006; 177:3019-3027; ; doi: 10.4049/jimmunol.177.5.3019 http://www.jimmunol.org/content/177/5/3019 Downloaded from

References This article cites 44 articles, 14 of which you can access for free at: http://www.jimmunol.org/content/177/5/3019.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 28, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Human Macrophages Constitute Targets for Immunotoxic Inorganic Arsenic1

Anthony Lemarie,* Claudie Morzadec,* Emilie Bourdonnay,* Olivier Fardel,*† and Laurent Vernhet2*

Chronic exposure to inorganic arsenic, a widely distributed environmental contaminant, can lead to toxic effects, including immunosuppression. Owing to the established roles of human macrophages in immune defense, we determined, in the present study, whether inorganic arsenic can affect these major immune cells. Our results demonstrate that noncytotoxic concentrations

of arsenic trioxide (As2O3), an inorganic trivalent form, markedly impair differentiated features of human blood monocyte-derived ␮ macrophages. First, treatment of macrophages with 1 MAs2O3 induced a rapid cell rounding and a subsequent loss of adhesion. These morphologic alterations were associated with a marked reorganization of actin cytoskeleton, which includes retraction of Downloaded from peripheral actin extensions and formation of a cortical actin ring. In addition, As2O3 reduced expression of various macrophagic surface markers, enhanced that of the monocytic marker CD14, and altered both endocytosis and phagocytosis; unexpectedly, exposure of macrophages to the metalloid also strongly potentiated expression of TNF␣ and IL-8 induced by LPS. Finally, like

monocytes, As2O3-treated macrophages can be differentiated into dendritic-like cells. Impairment of macrophage function by

As2O3 mainly resulted from activation of a RhoA/Rho-associated kinase pathway; indeed, pretreatment of macrophages with the

Rho-associated kinase inhibitor Y-27632 prevented metalloid effects on cytoskeleton and phagocytosis. Moreover, As2O3 was found to increase level of the active GTP-bound form of RhoA and that of phosphorylated-Moesin, a major cytoskeleton adaptor http://www.jimmunol.org/ protein involved in RhoA regulation. Taken together, our results demonstrated that human macrophages constitute sensitive targets of inorganic arsenic, which may contribute to immunotoxicity of this environmental contaminant. The Journal of Im- munology, 2006, 177: 3019–3027.

norganic arsenic (iAs)3 is a potent environmental toxic to with reduced CD4 lymphocyte population, proliferative response which millions of people are exposed over the world, mainly to PHA, and IL-2 secretion. Moreover, low micromolar concen- I through contaminated drinking water (1). Epidemiological trations of arsenite, an inorganic trivalent arsenical compound, de- studies have demonstrated that long-term exposure to iAs can fa- lays in vitro proliferation of activated human T lymphocytes by by guest on September 28, 2021 vor development of cardiovascular diseases such as atherosclerosis reducing production and secretion of IL-2 (8). Alterations of lym- (2); it can also induce cancer of skin, bladder, lung, and, possibly, phocyte functions have also been observed in animal models. of other internal organs (3, 4). iAs particularly increases incidence Lymphocytes, from BALB/c mice exposed to 50 mg/L arsenic in of nonmelanoma skin cancers, a known complication of organ the drinking water for 4 wk, exhibited reduced proliferative re- transplantation along with the use of immunosuppression (5, 6). sponses following T cell mitogen stimulation (9); in addition, treat- Interestingly, it was recently reported that iAs can exert immuno- ment of B6C3F1 mice with gallium arsenide (200 mg/kg) de- suppressive effects that may contribute to its general toxicity. In- creased lymphocyte populations (10). deed, this metalloid was recently demonstrated to affect normal Besides lymphocytes, monocytes/macrophages may also consti- function of lymphocytes in chronically exposed humans (7). In tute major targets of iAs. Indeed, we and others have recently these individuals, increased urine arsenic levels were associated demonstrated that low micromolar concentrations of iAs inhibit in vitro macrophagic differentiation of human blood-derived mono- *Institut National de la Sante´et de la Recherche Me´dicale, Unite´620, De´toxication cytes (11, 12). Particularly, we showed that arsenic trioxide † et Re´paration Tissulaire, Universite´ de Rennes-1, and De´partement He´matologie, (As O ), an iAs trivalent form, triggers apoptosis of human mono- Immunologie et Therapie cellulaire, Centre Hospitalier Universitaire Pontchaillou, 2 3 Rennes, France cytes during their differentiation by down-regulating NF-␬B-de- Received for publication November 29, 2005. Accepted for publication June 8, 2006. pendent survival pathways (11). In addition, in vivo experiments The costs of publication of this article were defrayed in part by the payment of page have reported that iAs can alter macrophage functions. Thus, treat- charges. This article must therefore be hereby marked advertisement in accordance ment of Swiss albino mice with arsenite (0.5 mg/kg) for 15 days with 18 U.S.C. Section 1734 solely to indicate this fact. significantly increased bacterial load in blood and delayed bacte- 1 This work was supported by grants from Association pour la Recherche sur le rial clearance by spleen (13); these effects were associated with Cancer and Ligue Contre le Cancer (Comite´ d’Ille-et-Vilaine). A.L. and E.B. are recipients of a fellowship from Ligue Nationale Contre le Cancer and Ligue Contre marked reduction of adhesion property, chemotactic migration, le Cancer (Comite´d’Ille-et-Vilaine), respectively. and phagocytic activity of splenic macrophages. Mice exposure to 2 Address correspondence and reprint requests to Dr. Laurent Vernhet, Institut Na- gallium arsenide affected proteolytic cathepsin activities and Ag tional de la Sante´ et de la Recherche Me´dicale, Unite´ 620, Faculte´ des Sciences processing of splenic macrophages (14). Together, these results Pharmaceutiques et Biologiques, Institut Fe´de´ratif de Recherche 140, Universite´de Rennes-1, 2 avenue du Professeur Le´on Bernard, 35043 Rennes, France. E-mail ad- suggest that, in addition to lymphocytes, human macrophages may dress: [email protected] constitute by themselves sensitive targets of iAs. The present study 3 Abbreviations used in this paper: iAs, inorganic arsenic; ERM, Ezrin/Radixin/Moe- was designed to investigate this hypothesis. sin; RhoGDI, Rho guanine dissociation inhibitor; A5, annexin V; SG, sytox green; MFI, mean fluorescence intensity; RT-qPCR, reverse transcription-real-time quanti- In this work, we demonstrate that low concentrations of As2O3 tative PCR; ROCK, Rho-associated kinase. (0.25–1 ␮M), in the range of arsenic blood levels measured in

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00 3020 As2O3 IMPAIRS HUMAN PRIMARY MACROPHAGES

chronically exposed humans (10–60 ␮g/L) (15, 16), alter differ- Endocytosis assay entiation features of human primary macrophages such as mor- Cells were incubated with 1 mg/ml FITC-dextran for 30 min at 37°C. phology, surface marker expression, or phagocytic activity. These Cellular uptake of FITC-dextran was then monitored by flow cytometry at effects were initiated by a rapid reorganization of F-actin cytoskel- 525 nm. A negative control was performed in parallel by incubating cells eton, resulting from a sustained and persistent activation of a with FITC-dextran at 4°C instead of 37°C. Uptake of FITC-dextran was ⌬ RhoA-Rho-associated kinase (ROCK) pathway. Our results high- expressed as mean fluorescence intensity (MFI), i.e., MFI (uptake at 37°C) – MFI (uptake at 4°C). light human macrophages as marked sensitive targets of iAs, which may contribute to the immunosuppressive properties of this Phagocytosis of latex microspheres and FITC-labeled environmental contaminant. Escherichia coli Materials and Methods Cells were incubated with 15 ␮l of fluorescent latex microspheres for 30 min at 37°C. Cellular phagocytosis of latex beads was then monitored by Chemical reagents and Abs flow cytometry at 525 nm. A negative control was performed in parallel by incubating cells with latex beads at 4°C instead of 37°C. As2O3, FITC-conjugated dextran (molecular mass, 40,000 kDa), FITC- phalloidin, and LPS were purchased from Sigma-Aldrich. The WST-1 tet- For studying uptake of E. coli bacteria by macrophages, a Phagotest (BD razolium salt 4-[3–4(iodophenyl)-2-(4-nitrosophenyl)-2H-5-tetrazole]-1,3- Biosciences) was used according to the instructions of manufacturer. benzene disulfonate was purchased from Roche Diagnostics. Y-27632 was Briefly, 20 ␮l of precooled FITC-labeled E. coli bacteria were added to a from Calbiochem (VWR). GM-CSF (specific activity, 1.2 ϫ 108 UI/mg) 100-␮l suspension of macrophages. The resulting cell suspension was next was obtained from Shering Plough, whereas IL-4 (specific activity, 2 ϫ 108 incubated for3hat37°C or at 4°C (controls). Cells were then washed, and UI/ml) was from Promocell. Rabbit polyclonal Abs against phospho- and 100 ␮l of quenching solution was added. After washings, cells were ana-

total Ezrin/Radixin/Moesin (ERM) proteins were purchased from Cell Sig- lyzed on FACSCalibur. Phagocytosis of latex microspheres or FITC-la- Downloaded from naling Technology (Ozyme). Mouse mAb raised against Rho guanine dis- beled E. coli was expressed as ⌬ MFI, i.e., MFI (uptake at 37°C) – MFI sociation inhibitors (RhoGDI) was from Santa Cruz Biotechnology (Tebu- (uptake at 4°C). bio). FITC-conjugated mAbs against CD1a, CD11c, CD14, CD29, and CD71, and PE-conjugated mAb against CD11b were purchased from Im- Total RNA isolation and reverse transcription-real-time munotech. Fluorescent latex microspheres were provided by Polysciences. quantitative PCR (RT-qPCR) assay Cell cultures Total RNAs were extracted from macrophages using the TRIzol method (Invitrogen Life Technologies) and then subjected to RT-qPCR analyses as http://www.jimmunol.org/ PBMC were first isolated from bloody buffy coats of healthy donors described previously (22). RT-qPCR assays were performed using the flu- through Ficoll gradient centrifugation. Human monocytes were then pre- orescent dye SYBR Green methodology and an ABI Prism 7000 detector pared by a 2-h adhesion step, which routinely obtained Ͼ90% of adherent (Applied Biosystem). Specific primers were as follows: TNF␣, forward, CD14-positive cells as assessed by immunostaining. These monocytic cells AACCTCCTCTCTGCCATCAA; TNF␣, reverse, ATGTTCGTCCTCCT were next cultured for 6 days in RPMI 1640 medium supplemented with CACAGG; IL-8, forward, AAGAAACCACCGGAAGGAAC; IL-8, re- 10% FCS, 2 mM L-glutamine, 20 UI/ml penicillin, and 20 ␮g/ml strepto- verse, AAATTTGGGGTGGAAAGGTT. The specificity of each gene am- mycin, in the presence of 800 UI/ml GM-CSF to get macrophages as pre- plification was checked up at the end of qPCR through analysis of viously reported (17–19). To obtain dendritic cells, monocytic cells were dissociation curves of the PCR products. Amplification curves were read differentiated in the presence of 800 UI/ml GM-CSF and 500 UI/ml IL-4, with ABI Prism 7000 SDS software using the comparative cycle threshold as previously reported (20). Once differentiated, macrophages were cul-

method. Relative quantification of the steady-state target mRNA levels was by guest on September 28, 2021 tured in GM-CSF-free RPMI 1640 medium in the absence or presence of calculated after normalization of the total amount of cDNA to an 18S RNA As O for indicated time intervals. In some experiments, macrophages 2 3 endogenous reference. were first pretreated with 20 ␮M Y-27632 for 2 h, and then exposed to

As2O3. Y-27632 was dissolved in distilled water. For cell death assays and flow cytometric studies, both floating and adherent macrophages were an- Determination of cytokine levels alyzed. Adherent macrophages were collected after 15 min of incubation at Levels of TNF␣ and IL-8 in the supernatants of macrophage cultures were ␮ 37°C in PBS supplemented with 100 M EDTA. quantified using Duoset ELISA development system kits obtained from Cellular adhesion assay R&D Systems. Briefly, 96-well plates, initially coated with 1 ␮g/ml mouse anti-human TNF␣ or IL-8 Ab overnight at room temperature, were incu- Adherence of macrophages was analyzed using the WST-1 assay in 96- bated for 2 h with macrophage culture supernatants or recombinant human well microplates, as previously described (18). The yellow formazan prod- TNF␣ or IL-8 standard. After washing, plates were processed according to uct formed by viable adherent cells was quantified by detection of its ab- the manufacturer’s instructions. sorbance at 450 nm using a Titertek Multiskan spectrophotometer (Flow Laboratories). Analysis of F-actin expression Determination of apoptosis and necrosis Monocytes were first differentiated with GM-CSF into macrophages on glass coverslips for 6 days, and then treated with As O . After washing, To analyze alterations of the plasma membrane structure linked to apopto- 2 3 cells were fixed on coverslips with 3% paraformaldehyde in PBS for 30 sis, exposition of phosphatidylserine to the extracellular environment was min at 4°C and washed three times with PBS. Fixed cells were subse- studied. We determined binding of annexin V (A5), a calcium-dependent quently incubated with a blocking and permeabilizing solution (PBS, 2 protein with high affinity for phosphatidylserine, using A5 conjugated to mg/ml BSA, 0.2 mg/ml saponin) for1hatroom temperature. Cells were the fluorescent label Alexa 568 (Roche Diagnostics). Simultaneously, ne- then incubated with FITC-phalloidin (1.5 ␮M), to detect F-actin filaments, crotic cells, which have lost their plasma membrane integrity, were de- in blocking solution for2hatroom temperature and washed in blocking tected with the green fluorescent DNA dye sytox green (SG) (Interchim). solution. Thereafter, cells were costained with blocking solution containing After treatment, cells were collected, washed, and incubated with dyes as ϩ Ϫ Ϫ ϩ 1 ␮g/ml 4Ј,6Ј-diamidino-2-phenylindole, a fluorescent dye specific for previously described (21). Apoptotic (A5 SG ) and necrotic (A5 SG ϩ ϩ DNA, during 15 min. After washings, coverslips were mounted with PBS- and A5 SG ) cells were quantified using a fluorescence Olympus BX60 glycerol-Dabco. Fluorescent-labeled cells were captured with a DMRXA2 microscope, in comparison with living cells. At least 200 cells were Leica microscope and a COOLSNAP HQ CCD camera, using Metavue counted for each cell suspension. software (Molecular Devices). Flow cytometric immunolabeling assays RhoA-GTP pull-down assay After treatments, macrophages were collected and centrifuged. Then, phe- notypic analysis of cells was performed using flow cytometric direct im- RhoA-GTP levels were measured using the RhoA activation assay kit from munofluorescence assays (20); cells were first incubated for 30 min in PBS Cytoskeleton (Tebu-bio). Briefly, cells were rapidly lysed at 4°C and in- with 5% human AB serum at 4°C to avoid nonspecific mAb binding, and cubated with Rhotekin-RBD affinity beads to specifically pull-down RhoA- then incubated with appropriate FITC- or PE-conjugated mAbs for 20 min GTP. After washing, RhoA levels were quantified by running bead/protein at 4°C. Fluorescence related to immunolabeling was measured using a complexes in Laemmli buffer containing 0.1 M DTT and probing with the FACSCalibur flow cytometer (BD Biosciences). mouse monoclonal anti-RhoA Ab as recommended by manufacturers. The Journal of Immunology 3021

Western blot immunoassays phage (Fig. 1A); rounded and contracted morphology was first ob- After treatment, macrophages were harvested and lysed on ice with lysis served in some macrophages after 8 h; the amount of rounded buffer (60 mM ␤-glycerophosphate, 15 mM paranitrophenylphosphate, 25 macrophages was nearly maximal and stable after 72 h. Morpho-

mM MOPS (pH 7.2), 15 mM EGTA, MgCl2,1mMNaF,1mMphe- logical effects of arsenic were dose-dependent and detectable from nylphosphate, 2 mM DTT, 1 mM sodium orthovanadate, 1% protease in- ␮ 0.25 M (Fig. 1B). After a 6-day treatment with As2O3, macro- hibitor mixture (Roche)). Then, lysates were sonicated on ice. Protein con- phage rounding was associated with a loss of adhesion as assessed centration was quantified using the Bradford’s method (23). A total of 50 Ϯ Ϯ ␮g of each sample was heated for 5 min at 100°C, and then analyzed by by the WST-1 assay; indeed, only 63.2 6.2 and 31.3 4.5% of 12% SDS-PAGE, and electroblotted overnight onto nitrocellulose mem- macrophages, comparatively to untreated cells, were still adherent ␮ branes (Bio-Rad). After blocking, membranes were hybridized with pri- after treatment with 0.5 and 1 MAs2O3, respectively. Adhesion mary Abs overnight at 4°C and incubated with appropriate HRP-conju- and spreading of macrophages are mainly related to reorganization gated secondary Ab (DakoCytomation). Immunolabeled proteins were visualized by chemiluminescence. of the actin cytoskeleton. In our cellular model, untreated macro- phages displayed extended F-actins resembling filopodia, in their

Statistical analysis periphery, as previously reported (24). Interestingly, As2O3 in- The results are presented as means Ϯ SEM. Significant differences were duced a time-dependent reorganization of F-actin cytoskeleton in evaluated with the Student’s t test. Criterion of significance of the differ- macrophages treated for 72 h (Fig. 1C). Peripheral extensions of ence between means was p Ͻ 0.05. F-actin began to retract after 8 h and formed a cortical actin ring

after 24 h. As2O3-induced reorganization of F-actin was then sta- Results ble for at least 72 h.

As2O3 alters morphology, adhesion, and actin organization of Downloaded from human macrophages As2O3-induced morphological alterations do not result from cell Blood monocyte-derived macrophages are tightly adherent cells death displaying a “fried-egg”-like morphology, i.e., large spread cells Contribution of direct cytotoxicity to morphological effects of

with distinct nuclei (17) (Fig. 1A); these macrophages survived in As2O3 was then evaluated using the fluorescent A5 and SG probes, GM-CSF-free culture medium for several days without alteration which detected apoptosis and necrosis, respectively. As shown in

of their morphology. In contrast, a 6-day treatment of cultures with Fig. 2A, Ͼ85% of macrophages were viable after 6, 12, and 18 http://www.jimmunol.org/ ␮ ␮ 1 MAs2O3 induced a time-dependent rounding of the macro- days of cultures in the absence or presence of 1 MAs2O3.In by guest on September 28, 2021

FIGURE 1. As2O3 alters morphology and F-actin organization of human primary macrophages. Macro- phages were cultured at day 0 (d0) in the absence or ␮ presence of 1 MAs2O3 for the indicated time intervals (A and C) or for 6 days (d6) at the indicated concen- trations (B). A and B, Phase-contrast microscopy of macrophages from d0 to d6 (magnification, ϫ40). C, After fixation and permeabilization, immunostaining of cells was performed with FITC-phalloidin to specifi- cally detect F-actin. Cells were costained with 4Ј,6Ј- diamidino-2-phenylindole to detect nuclei and subse- quently viewed by fluorescence microscopy (magnification, ϫ400). Scale bar, 10 ␮m. Data are rep- resentative of four independent experiments. 3022 As2O3 IMPAIRS HUMAN PRIMARY MACROPHAGES

cytes, down-regulated by GM-CSF during macrophagic differen- ␮ tiation (26). However, a concentration of 1 MAs2O3 failed to modify CD11c levels, which indicates that metalloid effects were not due to a general alteration of surface markers. We then eval-

uated effects of As2O3 on endocytosis and phagocytosis, two major functions of macrophages. Exposure of macrophages to 1 ␮M

As2O3 for 6 days markedly reduced endocytosis of FITC-dextran (Fig. 3A) and phagocytosis of fluorescent microspheres (Fig. 3B), when compared with untreated cells. Effect of metalloid on mac- rophagic functions was also evaluated using Phagotest, a true func- tional assay investigating phagocytosis of bacteria. As expected, uptakes of FITC-labeled E. coli by macrophages incubated at 4°C ϭ ␮ were negligible (MFI 2). Fig. 3C demonstrates that 1 MAs2O3 significantly decreased the uptake of FITC-labeled E. coli after a 4- or 6-day treatment. We next determined whether metalloid could alter the ability of macrophages to secrete inflammatory cytokines in response to LPS, which constitutes another major function of APCs. LPS markedly enhanced secretion of both TNF␣ and IL-8

after 8 and 24 h (Fig. 4); concomitantly, treatment with LPS for 8 h Downloaded from strongly induced TNF␣ and IL-8 mRNA levels, as assessed using RT-qPCR assays (Table II). A treatment of 6 days with 1 ␮M

As2O3 had no obvious effect on cytokine secretion, whereas it marginally increased TNF␣ mRNA levels of untreated macro- phages. By contrast, it strongly potentiated secretion and mRNA

level of TNF␣ and IL-8 induced by LPS (Fig. 4, Table II). http://www.jimmunol.org/

As2O3 effects toward human macrophages are reversible FIGURE 2. As2O3-induced morphological alterations are not associated with cell death. Human macrophages were cultured in the absence or pres- To determine whether effects of metalloid were reversible, we first treated macrophages with 1 ␮MAsO for 6 days; cells were then ence of As2O3 for the indicated time intervals (A) and concentrations (B). 2 3 Then, cells were costained with annexin V-Alexa 568 (A5) and SG to washed and cultured with GM-CSF in arsenic-free medium for detect apoptotic (A5ϩSGϪ) and necrotic cells (A5ϪSGϩ and A5ϩSGϩ), respectively, and viewed by fluorescence microscopy. In A, only viable cells (A5ϪSGϪ) are represented in the graph. Values are means Ϯ SEM of

three (A) and eight (B) independent experiments. by guest on September 28, 2021

contrast, higher concentrations of metalloid (2–4 ␮M) decreased cell viability and specifically increased the percentage of apoptotic macrophages (A5ϩSGϪ) after a 6-day treatment (Fig. 2B).

As2O3 alters phenotype of human macrophages Besides morphology and adhesion, we next determined whether

As2O3 could affect other specific features of differentiated macro- phages. We first analyzed expression of some typical surface markers. Table I shows that treatment of macrophages for 6 days ␮ with 1 MAs2O3, markedly reduced expression of CD11b and CD29, two integrins involved in macrophage adhesion and spread-

ing (25). As2O3 also decreased expression of CD71, a well-known macrophagic differentiation marker (18). In contrast, it markedly increased that of CD14, a typical marker of human blood mono-

␮ a Table I. Phenotypic analysis of macrophages exposed to 1 MAs2O3

Surface Macrophages Macrophages As2O3-Treated Markers (day 0) (day 6) Macrophages (day 6)

FIGURE 3. As2O3 decreases endocytosis and phagocytosis by macro- CD14 4.8 Ϯ 1.6 3.2 Ϯ 0.6 16.2 Ϯ 2.7* phages. Differentiated macrophages were cultured in the absence or pres- Ϯ Ϯ Ϯ CD71 30.7 3 26.9 3.9 12.2 1.7* ence of 1 ␮MAsO for 4 (d4) (C) or 6 days (d6) (A–C). Macrophages CD29 38.3 Ϯ 4 34.8 Ϯ 3.5 16.9 Ϯ 2.4* 2 3 were incubated with FITC-dextran (A), fluorescent latex microbeads (B), or CD11b 158.4 Ϯ 12.4 140.6 Ϯ 14.7 58.7 Ϯ 10.4* CD11c 122.1 Ϯ 16.8 129.7 Ϯ 9 111.5 Ϯ 11 FITC-labeled E. coli (C) at 4°C (negative control) or 37°C to measure endocytosis (A) and phagocytosis (B and C). Cellular uptakes of FITC- a Macrophages (control, day 0) were cultured for 6 days in the absence or presence dextran, microbeads, or fluorescent E. coli were then determined by flow ␮ of 1 MAs2O3 (day 6). Phenotype analysis was then performed as described in cytometry. Uptakes, expressed as ⌬MFI (⌬MFI ϭ MFI – MFI ), are Materials and Methods. Ϯ 37°C 4°C Ͻ ء Results are expressed as MFI and are the means SEM of, Ϯ at least, five independent experiments. means SEM of, at least, three independent experiments. , p 0.05 Ͻ *, p 0.05, control day 6 vs As2O3 day 6. compared with untreated macrophages. The Journal of Immunology 3023

ing that fully differentiated macrophages failed to yield dendritic- like cells. In addition, like blood monocyte-derived dendritic cells, and in contrast to untreated macrophages, macrophages first ex-

posed to As2O3 and next cultured with GM-CSF and IL-4 in the absence of iAs were nonadherent (Fig. 6B).

Inhibition of ROCK prevents As2O3-induced morphological alterations The Rho-associated kinase ROCK is a major target of the small GTP-binding protein RhoA and a key signaling molecule involved in cytoskeleton regulation (27). Pretreatment of cultures with Y-27632, a specific inhibitor of ROCK (28), prevented both F- actin reorganization and cellular rounding in macrophages treated

with As2O3 for 24 h (data not shown) and 72 h (Fig. 7A). In addition, Fig. 7B indicates that the ROCK inhibitor significantly prevented, although partially, alteration of phagocytosis due to

As2O3 treatment. We next evaluated the effect of As2O3 on levels ␮ of the active GTP-bound form of RhoA and found that 1 M Downloaded from

As2O3 strongly increased its level after a treatment of 8 h (Fig. ␣ 8A). RhoA-GTP levels remained increased for at least 72 h. We FIGURE 4. As2O3 enhances TNF and IL-8 secretion induced by LPS. Human macrophages were cultured in the absence (control) or presence of finally analyzed expression of RhoGDI and ERM proteins, which ␮ are involved in the regulation of RhoA activity (29, 30). Fig. 8B 1 MAs2O3 for 6 days. Cells were then untreated or exposed to 200 ng/ml LPS for 8 or 24 h in the absence or presence of As2O3. Secretion of shows that As2O3 did not alter protein levels of RhoGDI; in con- p Ͻ 0.05. trast, it increased levels of phospho-Moesin without affecting those ,ء .cytokine in supernatants were then determined by ELISA http://www.jimmunol.org/ of total ERM proteins (Fig. 8C). In fact, ezrin and radixin are only slightly expressed in macrophages and their phosphorylated forms another 6 days. Fig. 5A shows that withdrawal of iAs, associated were barely detected, if at all, by Western blotting. Effects of met- with addition of GM-CSF, induced readhesion and spreading of alloid on phospho-Moesin levels were rapid and detectable as soon As O -treated macrophages. Moreover, reversion of cell rounding 2 3 as4h. was also associated with changes in surface marker expression (Fig. 5, B–E). Indeed, after withdrawal of iAs and addition of GM-CSF, expressions of CD11b, CD29, and CD71, decreased in Discussion As O -treated macrophages, returned to levels observed in un- The present results demonstrated that As O markedly alters dif- 2 3 2 3 by guest on September 28, 2021 treated macrophages; in addition, CD14 levels, up-regulated in ferentiation features of human macrophages, likely through a

As2O3-treated cells, were concomitantly reduced. mechanism involving a RhoA-ROCK pathway. Such effects result

in the acquisition by As2O3-treated macrophages of phenotypic As2O3-treated macrophages could be differentiated into properties usually exhibited by monocytic cells. dendritic-like cells First, our results show that treatment of macrophages with 0.5–1 ␮ Monocytes are not only able to differentiate into macrophages but MAs2O3, two noncytotoxic concentrations, induced rapid cel- also into immature dendritic cells, in response to appropriate stim- lular rounding and thereafter loss of adhesion. Cell rounding was uli such as treatment with GM-CSF and IL-4 (20). Because the tightly associated with a time-dependent reorganization of F-actin

above results suggest that As2O3 “de-differentiated” macrophages in macrophages. In fact, As2O3 induced retraction of peripheral into monocytic cells, we next analyzed whether As2O3-treated F-actin extensions and stable formation of a cortical actin ring; macrophages could be differentiated into dendritic cells as mono- interestingly, organization of F-actin cytoskeleton in arsenic- cytic cells. Fig. 6A shows that macrophages previously exposed to treated macrophages was very similar to that observed in freshly

As2O3 for 6 days and then cultured with GM-CSF and IL-4 in adherent monocytes (31). In addition, As2O3 decreased expression iAs-free medium for an additional 6 days, expressed high levels of of the CD11b and CD29 integrins, which play a major role in

the typical dendritic marker CD1a. This CD1a expression was sim- adhesion, spreading, and differentiation of monocytes (32). As2O3 ilar to that found in blood monocyte-derived dendritic cells (Fig. also decreased expression of the CD71 macrophagic marker, in- 6A); in contrast, treatment of control macrophages with GM-CSF creased that of the CD14 monocyte marker, and markedly reduced and IL-4 did not significantly increased CD1a expression, indicat- both endocytosis and phagocytosis. Finally, like blood monocytes,

␣ a Table II. As2O3 potentiates TNF and IL-8 mRNA levels induced by LPS

Fold Change of TNF␣ mRNA Level Fold Change of IL-8 mRNA Level

As2O3-treated As2O3-treated Treatment Macrophages macrophages (day 6) Macrophages macrophages (day 6)

Untreated 1 2.4 Ϯ 1.1 1 0.9 Ϯ 0.9 LPS, 8 h 689.7 Ϯ 191 3168 Ϯ 697* 181 Ϯ 19 1510 Ϯ 521* a ␮ Macrophages were cultured for 6 days in the absence or presence of 1 MAs2O3 and then treated with 200 ng/ml LPS for 8 h. mRNA levels were determined by RT-qPCR, normalized to those of untreated macrophages, and expressed as fold change. Data are the means Ϯ SEM of four independent experiments. *, p Ͻ 0.05 compared to LPS-treated macrophages. 3024 As2O3 IMPAIRS HUMAN PRIMARY MACROPHAGES

FIGURE 5. As2O3-induced alterations of macro- phages are reversible. Human macrophages were cul- tured in the absence (control) or presence of 1 ␮M

As2O3 for 6 days (d0 to d6). Then, macrophages were washed, centrifuged, and reincubated in fresh medium for another 6 days (d6 to d12) in the absence or pres- ence of GM-CSF. In A, phase-contrast microscopy was performed at d12. Experiments were repeated five times, with similar results. B–E, Cells were stained with FITC- or PE-conjugated mAbs directed against the sur- face markers CD14 (B), CD71 (C), CD11b (D), and CD29 (E), and analyzed by flow cytometry. Results are Downloaded from expressed as means Ϯ SEM of, at least, three indepen- .p Ͻ 0.05 ,ء .dent experiments http://www.jimmunol.org/ by guest on September 28, 2021

FIGURE 6. As2O3-treated cells can be differentiated into dendritic-like cells. Human blood monocytes and macrophages previously untreated (control d6) or ex- ␮ posed to 1 MAs2O3 for 6 days (d6) were cultured in the absence or presence of GM-CSF plus IL-4, in

As2O3-free RPMI 1640 medium for 6 days (d6 to d12). A, Cells were then stained with FITC-conjugated mAbs directed against the dendritic cell marker CD1a and an- alyzed by flow cytometry. Results are expressed as means Ϯ SEM of three independent experiments. B, Phase-contrast microscopy of cell cultures (magnifica- tion, ϫ40). Experiments were repeated three times, with similar results. The Journal of Immunology 3025

FIGURE 7. The ROCK inhibitor Y-27632 prevents

As2O3 effects on macrophage phenotype. A, Human mac- rophages were pretreated or not for 2 h with a selective inhibitor of ROCK, Y-27632 (20 ␮M), and then cultured ␮ in the absence or presence of 1 MAs2O3 for 72 h. F- actin immunolocalization by fluorescence microscopy (magnification, ϫ400; scale bar, 10 ␮m) and phase-con- trast microscopy were then achieved as previously de- scribed. B, Macrophages were incubated with FITC-la- beled E. coli at 4°C (negative control) or 37°C to measure phagocytosis. Cellular uptakes of fluorescent E. coli were then determined by flow cytometry. Uptakes, expressed as

⌬ ⌬ ϭ Ϯ Downloaded from MFI ( MFI MFI37°C – MFI4°C), are means SEM of p Ͻ 0.05 compared with ,ء .four independent experiments

As2O3-treated macrophages. http://www.jimmunol.org/

As2O3-treated macrophages could be differentiated into dendritic- expression were reversible, thus confirming that they did not re- like cells by GM-CSF and IL-4. Our study therefore clearly dem- flect a general unspecific toxicity toward macrophages. onstrated that metalloid effects on morphology and surface marker Macrophages are known to possess functional plasticity, which

allows them to adapt to changing microenvironment, notably after by guest on September 28, 2021 cytokine or LPS treatment (33). The present results suggest that human macrophages can also “de-differentiate” into monocyte-like cells when exposed to an appropriate chemical compound. This study and other recent reports highlight a marked influence of iAs on cell differentiation. Besides macrophages, iAs partially de-dif- ferentiates murine adipocytes (34); arsenite inhibits differentiation of preadipocytes and human epidermal cells likely by maintaining proliferative cell capacity during differentiation program (35, 36).

We have also recently demonstrated that As2O3 totally blocks dif- ferentiation of human blood monocytes into macrophages through down-regulation of survival pathways (11). On the other hand,

As2O3 was shown to promote differentiation of breast cancer cells (37) and that of acute promyelocytic leukemia cells (38). Together, these results demonstrate that iAs exerts complex effects on dif- ferentiation according to the cell type.

Different arguments support the idea that As2O3 impaired hu- man macrophage function, at least in part, through activation of a RhoA/ROCK signaling pathway. First, our results demonstrate that the ROCK inhibitor Y-27632 prevented F-actin reorganiza-

tion, cell rounding and phagocytosis inhibition in As2O3-exposed

macrophages. In addition, we report that As2O3 activates the small G protein RhoA, a potent ROCK inducer. Indeed, As O induces FIGURE 8. As2O3 increases RhoA-GTP and phospho-Moesin levels in 2 3 macrophages. Macrophages were cultured in the absence or presence of 1 a rapid and sustained increase of RhoA-GTP levels which pre- ␮ MAs2O3 for the indicated time intervals. A, The GTP-binding fraction of ceded its effects on morphology and cytoskeleton. Involvement of RhoA was pulled down as described in Materials and Methods. The bead/ RhoA/ROCK in As2O3 toxicity is further strengthened by the fact protein complexes and whole-cell lysates were then immunoblotted to de- that this pathway is known 1) to induce cell contraction through tect RhoA levels. Western blot analysis of Rho-GDI (B) and phospho-ERM reorganization of actin cytoskeleton, notably in macrophages (24) protein (C) expressions were performed with whole-cell lysates. Equal gel loading and transfer efficiency were checked by protein hybridization with and 2) to constitute a negative regulator of monocyte spreading anti-RhoA and anti-ERM Ab, respectively. Experiments in A–C were re- (39). In addition, the recent observation that RhoA-activated peated, at least three times, with similar results. ROCK negatively regulated phagocytosis of apoptotic cells (40) 3026 As2O3 IMPAIRS HUMAN PRIMARY MACROPHAGES argues for a role of ROCK in As2O3-induced down-regulation of Disclosures phagocytic activity. RhoA is regulated by several proteins and par- The authors have no financial conflict of interest. ticularly by RhoGDI proteins, which maintain RhoA in its inactive GDP-bound form (30). DNA microarrays have revealed that ar- References senite can down-regulate mRNA levels of RhoGDI proteins (41, 1. Rossman, T. G. 2003. Mechanism of arsenic carcinogenesis: an integrated ap- 42). However, our results show that As2O3 had no effect on proach. Mutat. Res. 533: 37–65. RhoGDI protein levels in macrophages; in contrast, it increased 2. Navas-Acien, A., A. R. Sharrett, E. K. Silbergeld, B. S. Schwartz, K. E. Nachman, T. A. Burke, and E. Guallar. 2005. Arsenic exposure and car- those of phospho-Moesin, a member of the ERM protein family. diovascular disease: a systematic review of the epidemiologic evidence. Phosphorylation of ERM proteins favors their activation by stabi- Am. J. Epidemiol. 162: 1037–1049. lizing the active open conformation (29). Once phosphorylated, 3. Brown, K. G., and G. L. Ross. 2002. Arsenic, drinking water, and health: a position paper of the American Council on Science and Health. Regul. Toxicol. these proteins can promote dissociation of RhoGDI from RhoA Pharmacol. 36: 162–174. and subsequent activation of RhoA (29, 30). The fact that phos- 4. Abernathy, C. O., D. J. Thomas, and R. L. Calderon. 2003. Health effects and risk phorylation of Moesin seems to precede increase of RhoA-GTP assessment of arsenic. J. Nutr. 133: 1536S–1538S. 5. Abel, E. A. 1989. Cutaneous manifestations of immunosuppression in organ levels suggests that this ERM protein was involved in RhoA ac- transplant recipients. J. Am. Acad. Dermatol. 21: 167–179. 6. Boukamp, P. 2005. Non-melanoma skin cancer: what drives tumor development tivation in As2O3-treated macrophages. Besides its effects on and progression? Carcinogenesis 26: 1657–1667. RhoA, iAs was also recently demonstrated to regulate actin orga- 7. Soto-Pena, G. A., A. L. Luna, L. Acosta-Saavedra, P. Conde, L. Lopez-Carrillo, nization through activation of the small G protein Cdc42 in murine M. E. Cebrian, M. Bastida, E. S. Calderon-Aranda, and L. Vega. 2006. Assess- vascular cells (43). These results thus indicate that such small G ment of lymphocyte subpopulations and cytokine secretion in children exposed to

arsenic. FASEB J. 20: 779–781. Downloaded from proteins constitute new molecular targets of iAs which merit fur- 8. Galicia, G., R. Leyva, E. P. Tenorio, P. Ostrosky-Wegman, and R. Saavedra. ther exploration. 2003. Sodium arsenite retards proliferation of PHA-activated T cells by delaying Together with our previous study (11), the present results sug- the production and secretion of IL-2. Int. Immunopharmacol. 3: 671–682. ␮ 9. Patterson, R., L. Vega, K. Trouba, C. Bortner, and D. Germolec. 2004. Arsenic- gest that low micromolar concentrations (0.25–1 M) of As2O3, induced alterations in the contact hypersensitivity response in BALB/c mice. which are relevant of arsenic blood levels measured in chronically Toxicol. Appl. Pharmacol. 198: 434–443. 10. Sikorski, E. E., L. A. Burns, M. L. Stern, M. I. Luster, and A. E. Munson. 1991. exposed humans (15), can target the monocyte/macrophage system Splenic cell targets in gallium arsenide-induced suppression of the primary an- http://www.jimmunol.org/ in humans. Indeed, our data likely indicate that iAs can reduce the tibody response. Toxicol. Appl. Pharmacol. 110: 129–142. pool of human functional macrophages by 1) inhibiting macroph- 11. Lemarie, A., C. Morzadec, D. Merino, O. Micheau, O. Fardel, and L. Vernhet. 2006. Arsenic trioxide induces apoptosis of human monocytes during macroph- agic differentiation of blood monocytes (11) and 2) impairing fea- agic differentiation through nuclear factor-␬B-related survival pathway down- tures of differentiated monocyte-derived macrophages, notably en- regulation. J. Pharmacol. Exp. Ther. 316: 304–314. docytosis and phagocytosis; at higher concentrations (4 ␮M), 12. Sakurai, T., T. Ohta, and K. Fujiwara. 2005. Inorganic arsenite alters macrophage generation from human peripheral blood monocytes. Toxicol. Appl. Pharmacol. As2O3 also induces apoptosis of differentiated macrophages. Rel- 203: 145–153. evance of these mechanisms is strengthened by in vivo studies 13. Bishayi, B., and M. Sengupta. 2003. Intracellular survival of Staphylococcus aureus due to alteration of cellular activity in arsenic and lead intoxicated mature reporting a decrease of phagocytic activity and antibacterial re- Swiss albino mice. Toxicology 184: 31–39. sponse in mice treated with arsenite (13). However, toxicity of iAs 14. Lewis, T. A., C. B. Hartmann, and K. L. McCoy. 1998. Gallium arsenide mod- by guest on September 28, 2021 ulates proteolytic cathepsin activities and antigen processing by macrophages. toward macrophages appears to be complex, because As2O3 J. Immunol. 161: 2151–2157. strongly increases the macrophage response to LPS. In fact, al- 15. Wu, M. M., H. Y. Chiou, I. C. Ho, C. J. Chen, and T. C. Lee. 2003. Gene though it had no significant effect on untreated macrophages, expression of inflammatory molecules in circulating lymphocytes from arsenic- exposed human subjects. Environ. Health Perspect. 111: 1429–1438. As2O3 appears to “prime” macrophages for LPS-induced secretion ␣ 16. Pi, J., Y. Kumagai, G. Sun, H. Yamauchi, T. Yoshida, H. Iso, A. Endo, L. Yu, of TNF and IL-8, two major proinflammatory cytokines. The K. Yuki, T. Miyauchi, and N. Shimojo. 2000. Decreased serum concentrations of superinduction of TNF␣ and IL-8 mRNA levels in macrophages nitric oxide metabolites among Chinese in an endemic area of chronic arsenic coexposed to LPS and iAs suggests that iAs can modulate LPS poisoning in inner Mongolia. Free Radic. Biol. Med. 28: 1137–1142. 17. Young, D. A., L. D. Lowe, and S. C. Clark. 1990. Comparison of the effects of intracellular signals. Such effects may contribute to the develop- IL-3, granulocyte-macrophage colony-stimulating factor, and macrophage colo- ment of iAs-related inflammation observed in chronically exposed ny-stimulating factor in supporting monocyte differentiation in culture: analysis of macrophage antibody-dependent cellular cytotoxicity. J. Immunol. 145: humans (15). 607–615. Interestingly, iAs, used under its trioxide form, is an effective 18. van Grevenynghe, J., S. Rion, E. Le Ferrec, M. Le Vee, L. Amiot, R. Fauchet, and antileukemic drug and its concentrations in blood patients are close O. Fardel. 2003. Polycyclic aromatic hydrocarbons inhibit differentiation of hu- man monocytes into macrophages. J. Immunol. 170: 2374–2381. to those acting in vitro on macrophages (44). This indicates that 19. Lecureur, V., E. L. Ferrec, M. NЈDiaye, M. L. Vee, C. Gardyn, D. Gilot, and macrophages, in addition to malignant cells, may also constitute O. Fardel. 2005. ERK-dependent induction of TNF␣ expression by the environ- targets in As O -treated patients, and therefore supports the idea mental contaminant benzo(a)pyrene in primary human macrophages. FEBS Lett. 2 3 579: 1904–1910. that a careful analysis of the monocyte/macrophage compartment 20. Laupeze, B., L. Amiot, L. Sparfel, E. Le Ferrec, R. Fauchet, and O. Fardel. 2002. may be of interest in such subjects. Polycyclic aromatic hydrocarbons affect functional differentiation and maturation of human monocyte-derived dendritic cells. J. Immunol. 168: 2652–2658. In conclusion, our results demonstrate that noncytotoxic con- 21. Lemarie, A., D. Lagadic-Gossmann, C. Morzadec, N. Allain, O. Fardel, and centrations of As2O3 down-regulate differentiation features of hu- L. Vernhet. 2004. Cadmium induces caspase-independent apoptosis in liver man primary macrophages likely through activation of a RhoA/ Hep3B cells: role for calcium in signaling oxidative stress-related impairment of mitochondria and relocation of endonuclease G and apoptosis-inducing factor. ROCK pathway. Moreover, they show that human macrophages Free Radic. Biol. Med. 36: 1517–1531. constitute by themselves sensitive targets of iAs, which may con- 22. Le Vee, M., E. Jigorel, D. Glaise, P. Gripon, C. Guguen-Guillouzo, and tribute to immunotoxicity of this major environmental O. Fardel. 2006. Functional expression of sinusoidal and canalicular hepatic drug transporters in the differentiated human hepatoma HepaRG cell line. Eur. contaminant. J. Pharm. Sci. 28: 109–117. 23. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248–254. Acknowledgments 24. Allen, W. E., G. E. Jones, J. W. Pollard, and A. J. Ridley. 1997. Rho, Rac and We are grateful to Dr. C. Leberre (Etablissement Franc¸ais du Sang, Cdc42 regulate actin organization and cell adhesion in macrophages. J. Cell Sci. 110: 707–720. Rennes, France) for providing us with blood buffy coats. Thanks to the 25. Berton, G., and C. A. Lowell. 1999. Integrin signalling in neutrophils and mac- Institut Fe´de´ratif de Recherche 140 microscopy platform in Rennes. rophages. Cell. Signal. 11: 621–635. The Journal of Immunology 3027

26. Kruger, M., J. G. Van de Winkel, T. P. De Wit, L. Coorevits, and J. L. Ceuppens. 36. Patterson, T. J., T. V. Reznikova, M. A. Phillips, and R. H. Rice. 2005. Arsenite 1996. Granulocyte-macrophage colony-stimulating factor down-regulates CD14 maintains germinative state in cultured human epidermal cells. Toxicol. Appl. expression on monocytes. Immunology 89: 89–95. Pharmacol. 207: 69–77. 27. Riento, K., and A. J. Ridley. 2003. Rocks: multifunctional kinases in cell behav- 37. Baj, G., A. Arnulfo, S. Deaglio, R. Mallone, A. Vigone, M. G. De Cesaris, iour. Nat. Rev. Mol. Cell Biol. 4: 446–456. N. Surico, F. Malavasi, and E. Ferrero. 2002. Arsenic trioxide and breast cancer: 28. Ishizaki, T., M. Uehata, I. Tamechika, J. Keel, K. Nonomura, M. Maekawa, and analysis of the apoptotic, differentiative and immunomodulatory effects. Breast S. Narumiya. 2000. Pharmacological properties of Y-27632, a specific inhibitor Cancer Res. Treat. 73: 61–73. of Rho-associated kinases. Mol. Pharmacol. 57: 976–983. 38. Miller, W. H., Jr., H. M. Schipper, J. S. Lee, J. Singer, and S. Waxman. 2002. 29. Ivetic, A., and A. J. Ridley. 2004. Ezrin/radixin/moesin proteins and Rho GTPase Mechanisms of action of arsenic trioxide. Cancer Res. 62: 3893–3903. signalling in leucocytes. Immunology 112: 165–176. 39. Aepfelbacher, M., M. Essler, E. Huber, A. Czech, and P. C. Weber. 1996. Rho is a negative regulator of human monocyte spreading. J. Immunol. 157: 5070–5075. 30. Olofsson, B. 1999. Rho guanine dissociation inhibitors: pivotal molecules in 40. Tosello-Trampont, A. C., K. Nakada-Tsukui, and K. S. Ravichandran. 2003. En- cellular signalling. Cell. Signal. 11: 545–554. gulfment of apoptotic cells is negatively regulated by Rho-mediated signaling. 31. Williams, L. M., and A. J. Ridley. 2000. Lipopolysaccharide induces actin reor- J. Biol. Chem. 278: 49911–49919. ganization and tyrosine phosphorylation of Pyk2 and paxillin in monocytes and 41. Andrew, A. S., A. J. Warren, A. Barchowsky, K. A. Temple, L. Klei, macrophages. J. Immunol. 164: 2028–2036. N. V. Soucy, K. A. O’Hara, and J. W. Hamilton. 2003. Genomic and proteomic 32. Shi, C., X. Zhang, Z. Chen, K. Sulaiman, M. W. Feinberg, C. M. Ballantyne, profiling of responses to toxic metals in human lung cells. Environ. Health Per- M. K. Jain, and D. E. Simon. 2004. Integrin engagment regulates monocyte spect. 111: 825–835. differentiation through the forkhead transcription factor Foxp1. J. Clin. Invest. 42. Rea, M. A., J. P. Gregg, Q. Qin, M. A. Phillips, and R. H. Rice. 2003. Global 114: 408–418. alteration of gene expression in human keratinocytes by inorganic arsenic. Car- 33. Stout, R. D., and J. Suttles. 2004. Functional plasticity of macrophages: reversible cinogenesis 24: 747–756. adaptation to changing microenvironments. J. Leukocyte Biol. 76: 509–513. 43. Qian, Y., K. J. Liu, Y. Chen, D. C. Flynn, V. Castranova, and X. Shi. 2005. 34. Wauson, E. M., A. S. Langan, and R. L. Vorce. 2002. Sodium arsenite inhibits Cdc42 regulates arsenic-induced NADPH oxidase activation and cell migration and reverses expression of adipogenic and fat cell-specific genes during in vitro through actin filament reorganization. J. Biol. Chem. 280: 3875–3884. adipogenesis. Toxicol. Sci. 65: 211–219. 44. Shen, Z. X., G. Q. Chen, J. H. Ni, X. S. Li, S. M. Xiong, Q. Y. Qiu, J. Zhu, Downloaded from 35. Trouba, K. J., E. M. Wauson, and R. L. Vorce. 2000. Sodium arsenite inhibits W. Tang, G. L. Sun, K. Q. Yang, et al. 1997. Use of arsenic trioxide (As2O3)in terminal differentiation of murine C3H 10T1/2 preadipocytes. Toxicol. Appl. the treatment of acute promyelocytic leukemia (APL). II. Clinical efficacy and Pharmacol. 168: 25–35. pharmacokinetics in relapsed patients. Blood 89: 3354–3360. http://www.jimmunol.org/ by guest on September 28, 2021