IMMUNOBIOLOGY

IgA Fc receptor I signals apoptosis through the FcR� ITAM and affectstumor growthYutaka Kanamaru,1,3 Houda Tamouza,1,3 Severine Pfirsch,1,3 Delphine El Mehdi,2,3 Claudine Guerin-Marchand,1,3

Marina Pretolani,2,3 Ulrich Blank,1,3 and Renato C. Monteiro1,3

1Institut National de la Sante et de la Recherche Medicale (INSERM) U699, Paris, France; 2Inserm U700, Paris, France; and 3Universite Paris 7-Denis Diderot,Faculte de Medecine, Site Xavier Bichat, Paris, France

The IgA Fc receptor (Fc�RI) has dualproinflammatory and anti-inflammatoryfunctions that are transmitted throughthe immunoreceptor tyrosine-based acti-vation motifs (ITAMs) of the associatedFcR� subunit. Whereas the involvementof Fc�RI in inflammation is well docu-mented, little is known of its anti-inflam-matory mechanisms. Here we show thatmonomeric targeting of Fc�RI by anti-Fc�RI Fab or serum IgA triggers apopto-sis in human monocytes, monocytic celllines, and Fc�RI� transfectants. However,the physiologic ligand IgA induced apo-

ptosis only when cells were cultured inlow serum conditions, indicating differ-ences with induction of anti-inflammatorysignaling. Apoptosis signaling requiredthe FcR� ITAM, as cells transfected withFc�RI or with a chimeric Fc�RI-FcR� re-sponded to death-activating signals,whereas cells expressing a mutatedFc�RIR209L unable to associate with FcR�,or an ITAM-mutated chimeric Fc�RI-FcR�,did not respond. Fc�RI-mediated apopto-sis signals were blocked by treatmentwith the pan-caspase inhibitor zVAD-fmk,involved proteolysis of procaspase-3, and

correlated negatively with SHP-1 concen-tration. Anti-Fc�RI Fab treatment of nudemice injected subcutaneously with Fc�RI�

mast-cell transfectants prevented tumordevelopment and halted the growth ofestablished tumors. These findings dem-onstrate that, on monomeric targeting,Fc�RI functions as an FcR� ITAM-depen-dent apoptotic module that may be funda-mental for controlling inflammation andtumor growth. (Blood. 2007;109:203-211)

© 2007 by The American Society of Hematology

Introduction

The receptor specific for the IgA Fc region (Fc�RI or CD89) isexpressed on blood myeloid cells, including monocyte/macro-phages, dendritic cells, Kupffer cells, neutrophils, and eosinophils.1

It transmits IgA-mediated effector functions via an IgA-bindingmodule (Ka about 106 M�1) that can be expressed with or withoutphysical association with the disulfide-linked signaling adaptorFcR�.2,3 The FcR�-associated receptor has signaling functions thatinvolve an immunoreceptor tyrosine-based activation motif (ITAM),consisting of a conserved sequence that is present in the cytoplas-mic domain of FcR� and is shared by different signaling subunitsassociated with receptors such as TCR and BCR and other FcRs.Although the ITAM was initially reported to mediate activatingresponses, recent data suggest that, in certain conditions, this motifcan also trigger inhibitory responses mediated by a variety ofreceptors.4-6 The latter include Fc�RI, a dual-function receptor thatcan mediate both activating and inhibitory responses depending onthe type of interaction with its ligand. Sustained multimericaggregation mediates activation of target-cell functions, such assuperoxide production, cytokine release, and antigen presenta-tion.7-9 Monomeric targeting with IgA or with a variety ofanti-Fc�RI (A77, A59, or A62, but not A3) Fab fragments triggersan inhibitory response to heterologous immunoreceptors such asFc�RI and Fc�R.5 Like the activating functions, the inhibitorycross-talk is dependent on the FcR� ITAM. However, in contrast toreceptors bearing an immunoreceptor tyrosine-based inhibitorymotif (ITIM), such as FcR�IIB,10 the inhibitory signal does not

require coaggregation. Monomeric targeting of Fc�RI also haspowerful anti-inflammatory activity in vivo because administrationof anti-Fc�RI Fab suppresses manifestations of allergic asthma inFc�RI transgenic mice.5

Activation of certain FcRs also promotes cell death. Thus,Fc�RIIB expressed on B cells not only terminates the activationsignals of coengaged receptors, via an ITIM-dependent mecha-nism, but, in case of isolated aggregation, is also able to triggeran apoptotic response.11 The latter involves ITIM and Srchomology 2 domain–containing inositide phosphatase (SHIP)–independent and c-abl-family kinase-dependent pathways.12-14

Recruitment of the Src homology 2 domain-containing phospha-tase (SHP)–1 to the B- and T-cell antigen receptors (BCR andTCR) was also shown to negatively correlate with the inductionof apoptosis.12,15 Indeed, SHP-1–deficient mice are unusuallysusceptible to clonal deletion of their B- and T-cell compart-ment.15-17 Although the precise mechanisms whereby SHP-1negatively regulates apoptosis remains unknown, a possiblehypothesis is that SHP-1 raises the threshold required forantigen receptor signaling spread thereby linking it to otherbiologic outcomes such as cell death.

Incubation of neutrophils with immobilized IgA or secretoryIgA also induces apopotosis.18 This likely involves Fc�RI, the onlyIgA receptor so far detected on neutrophils. Previous reports alsosuggest that ITAMs are essential for transmitting growth-arrest andapoptotic signals, and these functions of ITAMs are positively

Submitted June 2, 2006; accepted August 12, 2006. Prepublished online asBlood First Edition Paper, September 25, 2006; DOI 10.1182/blood-2006-06-025882.

An Inside Blood analysis of this article appears at the front of this issue.

The publication costs of this article were defrayed in part by page chargepayment. Therefore, and solely to indicate this fact, this article is herebymarked ‘‘advertisement’’ in accordance with 18 USC section 1734.

© 2007 by The American Society of Hematology

203BLOOD, 1 JANUARY 2007 � VOLUME 109, NUMBER 1

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom

regulated by tyrosine phosphorylation.19 Because we have previ-ously shown that monomeric targeting of Fc�RI can induce cellularsignaling, we investigated whether such signaling might generatedeath-activating signals. We found that monomeric occupancy ofFc�RI by anti-Fc�RI Fab or IgA directly induced apoptosis onmonocytes and monocytic cell lines. This required the Fc�RI-associated FcR� subunit and the ITAM, but, contrary to inhibitorysignaling, was also induced by A3 Fab. Apoptosis induction by thenatural ligand, IgA, required an additional signal, because it wasobserved only in low serum conditions. Finally, Fc�RI targeting invivo prevented tumor development and growth, with possibleimplications for cancer therapy.

Materials and methods

Animals

Nude mice were purchased from Janvier (Le Genest-St-Isle, France). Micewere bred and maintained at the mouse facilities of IFR02 at Bichat MedicalSchool. Transgenic mice expressing human Fc�RI (CD89, line 83) wereback-crossed for more than 12 generations with C57BL/6 mice.20 Allexperiments were done in accordance with national ethical guidelines andwere approved by a local committee (animal facility of IFR02 of theUniversity of Paris 7; authorizations A75-18-01 and 7413).

Immunoglobulins and antibodies

The mouse monoclonal antibodies (mAbs) A3, A59, A62, and A77 specificfor Fc�RI, and an irrelevant IgG1� control mAb (clones 320.1)21 were usedas either F(ab�)2 or Fab fragments.22 Rabbit anti–mouse IgG antibody(Southern Biotechnology, Birmingham, AL) was used as F(ab�)2. F(ab�)2

fragments were obtained by pepsin digestion as described.23 Fab wereprepared by pepsin digestion followed by reduction with 0.01 M cysteineand alkylation with 0.15 M iodoacetamide at pH 7.5. Complete digestionand purity were controlled by sodium dodecyl sulfate–polyacrylamide gelelectrophoresis (SDS-PAGE) and Coomassie blue staining. The anti–caspase-3 antibody was purchased from Cell Signaling Technology (Bev-erly, MA). The antiscramblase antibody has been described previously.5

Purified serum IgA (Jackson ImmunoResearch, West Grove, PA), IgG(Southern Biotechnology), goat anti–human IgA (Southern Biotechnology)and rabbit anti-SHP1 (Santa Cruz Biotechnology, La Jolla, CA) werealso used.

Cell culture and stimulation

Mononuclear cells (MNCs) were isolated from whole blood by Ficoll-Hypaque (Pharmacia, Piscataway, NJ) gradient centrifugation. Enrichedmonocyte populations (60%-80% pure) were obtained by subjecting MNCsto plastic adherence as described24 and were cultured in either 24- or 6-wellplates at a density of 2 � 106 cells/mL of RPMI-Glutamax mediumsupplemented with 10% fetal calf serum (FCS) and penicillin-streptomycin(Invitrogen, Paisley, United Kingdom) before use. RBL-2H3 cells express-ing wild-type (WT) human Fc�RI, mutant Fc�RIR209L, chimericFc�RIR209L/�, or chimeric Fc�RIR209L/� mutated in the FcR�ITAM (Y268/Y278F) were maintained in DMEM supplemented with 10% FCS, 2 mMglutamine, and 1 mg/mL G418 (Invitrogen) as described.5 Cells weretrypsinized, centrifuged, and resuspended in fresh media in 24-well plates ata density of 106/mL for 24 hours before use. Mouse peritoneal macrophageswere obtained following thioglycollate broth (4 days) before cell harvest asdescribed.25 The cells were then either plated on glass coverslips or kept insuspension in polypropylene tubes on ice. Purity was greater than 98%, asdetermined by CD11b staining.

Measurement of apoptosis by annexin V and PI staining

Apoptotic cells were identified by staining with annexin V and propidiumiodide (PI).26 Briefly, cells (106) were treated with apoptosis-inducing

agents for the times indicated, then washed and suspended in a bindingsolution (10 mM HEPES/NaOH, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2)containing annexin V–biotin, at the dilutions recommended by the manufac-turer (Beckman Coulter, Villepinte, France). After 30 minutes the cells werewashed and streptavidin-allophycocyanin (APC; 5 �g/mL) and PI (1�g/mL) were added for 10 minutes before washing and analysis with aFACSCalibur flow cytometer (Becton Dickinson ImmunoCytometry Sys-tem, San Jose, CA). At least 10 000 cells were analyzed per sample. Earlyapoptosis and late apoptosis, respectively, were determined as the percent-ages of annexin V/PI� and annexin V/PI cells.

Measurement of apoptosis by counting apoptotic nucleiwith hypoploid DNA

The proportion of cells with hypoploid DNA was measured by using amodification of the Nicoletti method.27 Briefly, cells (106) were incubated at37°C with anti-Fc�RI (A77) Fab or an irrelevant control (320) Fab. Themedium was removed and cells were fixed in 70% ethanol at �20°C for atleast 30 minutes. After ethanol removal, phosphate buffer consisting of0.2 M Na2HPO4, 0.1 M citric acid (pH 7.8) was added for 30 minutes atroom temperature. The cells were centrifuged and stained for 30 minutes atroom temperature with 1 mL PI staining buffer (0.1% Triton X-100,10 �g/mL RNAse A, and 10 �g/mL PI in PBS) before analysis by flowcytometry. Apoptotic nuclei containing degraded hypoploid DNA (lowerfluorescence intensity) were easily distinguished from nonapoptotic nuclei(sharp peaks), and their percentage was determined.

Counting of apoptotic cells by DAPI staining

DAPI staining was performed as previously described.28 Briefly, cells wereincubated at 37°C with anti-Fc�RI (A77) Fab or an irrelevant control (320)Fab. The medium was removed and the cells were washed twice with PBSand fixed by incubation with 70% ethanol for 30 minutes. Followingwashing in PBS, cells were incubated in 1 �g/mL DAPI solution for 30minutes in the dark. Apoptotic cells appear as cells with highly condensedfluorescent nuclei. Fluorescent cells were counted with a Nikon TE2000-Umicroscope coupled to a DXM1200F camera (Nikon, Champigny surMarne, France), using a 1� auxiliary lens and a direct 1� C-mount.Objective magnifications are indicated.

Measurement of mitochondrial apoptosis by stainingwith DiOC6

Changes in mitochondrial membrane potential (�m) were assessed bymeasuring uptake of the lipophilic cationic dye 3,3-dihexyloxacarbocya-nine iodide (DiOC6

3; Molecular Probes, Eugene, OR), which accumulatesin polarized mitochondria under the influence of �m.29 Cells werewashed and incubated at 37°C for 30 minutes with 0.1 �M DiOC6

3, and 2.5�M PI was added immediately before analysis by flow cytometry toexclude necrotic cells. In negative control experiments, cells were preincu-bated for 10 minutes at 37°C with 200 �M mitochondrial uncouplercarbonyl cyanide m-chlorophenylhydrazone (CCCP, Sigma, St Louis, MO),which collapses �m.30 A total of 5000 events were analyzed by flowcytometry (DiOC6

3 emission in FL-1 and PI in FL-2). The percentage ofDiOC6

3-high cells was determined.

Measurement of caspase-3 activation

After induction of apoptosis, cells were washed in ice-cold PBS andsolubilized in lysis buffer (50 mM HEPES, pH 7.4, 0.3% Triton X-100, 50mM NaF, 50 mM NaCl, 1 mM Na3VO4, 30 mM Na4P2O7, 50 U/mLaprotinin, 10 �g/mL leupeptin), and then postnuclear supernatants wereprepared. Proteins were resolved by SDS-PAGE and transferred to PVDFmembranes. After blocking in 4% BSA, membranes were incubated withanticaspase mAb for 1 hour at room temperature, followed by goatanti–mouse immunoglobulin coupled to HRP (Ig-HRP). After decapping,membranes were reprobed with antiscramblase antibody followed by goatanti–mouse Ig-HRP. Filters were developed by enhanced chemolumines-cence (ECL; Amersham-Pharmacia Biotech, Uppsala, Sweden).

204 KANAMARU et al BLOOD, 1 JANUARY 2007 � VOLUME 109, NUMBER 1

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom

Study of the effect of caspase inhibition by zVAD-fmk

Cells (106/mL) were incubated for 2 hours at 37°C with 20 �M pan-caspaseinhibitor z-VAD-fmk (N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl-ketone; R&D Systems, Minneapolis, MN). The test agents were added andincubation was continued for the times indicated at 37°C before apopto-sis assay.

SHP-1 knock-down experiments using siRNA

Rat SHP-1 in RBL cells was targeted with a cocktail of 3 siRNAs(Eurogentec, Louvain, Belgium): siRNA1 (sense strand) CAGGCACCAU-CAUUGUCAU; siRNA2: GAACCGCUACAAGAACAUU; siRNA3: CA-GAGCUGGUGGAGUACUA. A universal scramble negative siRNA GC-CCCGUCUACAUACAUGU was used as a control. Fc�RIR209L/� RBLcells were transfected with siRNA by 2 successive electroporations 24hours apart. For each transfection, cell density was adjusted to 1 � 107/mLin electropermeabilization buffer (120 mM KCl, 10 mM NaCl, 1 mMKH2PO4, 10 mM glucose, 20 mM HEPES, pH 7.0) before electroporationwith annealed siRNAs (0.05 �M) using an Easyject electroporationapparatus (Eurogentec) at 250 V and 2100 �F. Between transfections,floating dead cells were discarded. Two successive transfections werepreferred over a single one to improve the effective knock-down of SHP-1.After each electroporation, cells were cultured in complete medium. Afterthe second electroporation, apoptosis experiments were started by incubat-ing cells with anti-Fc�RI (A77) or control antibody (320). The effective-ness of siRNA treatment was tested by SHP-1 immunoblotting of extractsof untreated cells grown in parallel.

Skin grafting of Fc�RI RBL transfectantsand treatment protocols

Adherent Fc�RI RBL transfectants grown as monolayers were detached bytrypsinization, washed once with complete medium and once with serum-

free DMEM, and were then resuspended in the same medium. ChimericFc�RIR209L/� transfectants (1 � 107) were injected subcutaneously into6-week-old female nude mice to induce the development of subcutaneoustumors.31 Preventive and curative treatments were tested. In the former, 50�g A77 or 320 control Fab was injected subcutaneously, starting on day 0,as indicated in the legends; curative treatment was started on day 12, asindicated in the figure legends. Four groups of 6 mice were used. Mortalityand tumor size were recorded.

Detection of apoptotic cells by TUNEL

Tumors were fixed in 20% formalin and embedded in paraffin or immedi-ately frozen in Tissue-Tek OCT compound (Sakura Finetek, Zoeterwoude,the Netherlands) for immunohistochemical analysis. Formalin-fixed sec-tions 5 �m thick were analyzed after PAS staining. Apoptotic cells weredetected on 5-�m cross-sections by the TdT-mediated dUTP nick-endlabeling (TUNEL) method implemented with the in situ apoptosis detectionkit (Chemicon, Temeculo, CA) following the manufacturer’s protocol.

Results

Monomeric targeting of Fc�RI with anti-Fc�RI Fab in humanmonocytes and Fc�RI transfectants induces apoptosis

Because monomeric targeting of Fc�RI transmits inhibitory sig-nals,5 we examined whether, like Fc�IIB,11 it could also promoteapoptosis. Monocytes isolated from human blood and treated withirrelevant 320 Fab showed a background level of apoptosis similarto that of solvent (PBS)–treated cells (Figure 1A, left and rightpanels). Treatment with A77 anti-Fc�RI Fab, however, caused ashift to an annexin V/PI population within 24 hours, indicating

Figure 1. Induction of apoptosis by Fc�RI targeting in human monocytes and Fc�RI RBL-2H3 transfectants. (A-B) Analysis of annexin V/PI double-positive populationsin human blood monocytes (A) and RBL-2H3 cells expressing WT Fc�RI (B) after 24 hours of treatment with 10 �g/mL anti-Fc�RI (A77) Fab or control (320) Fab. Cells werestained with annexin V and PI and analyzed by flow cytometry as described in “Materials and methods.” The mean � SD percentage of annexin V/PI double-positive cellstreated with PBS, 320, or A77 Fab in 5 experiments is shown. *P .05. (C) Kinetic and dose-response analysis of annexin V/PI double-positive populations of WT Fc�RIRBL-2H3 transfectants after treatment with 10 �g/mL anti-Fc�RI (A77) Fab or control (320) Fab for 1 to 24 hours or for 24 hours with the indicated concentrations of A77 or 320Fab. The percentage of annexin V/PI double-positive populations was determined as described in panel A. (D) Analysis of hypoploid nuclei in blood monocytes and WT Fc�RIRBL-2H3 transfectants after treatment with 10 �g/mL anti-Fc�RI (A77) or control (320) Fab for 24 hours. Cells were stained with PI and analyzed for the appearance ofhypoploid nuclei as described in “Materials and methods.” Numbers indicate the percentage of cells with hypoploid nuclei (a sign of apoptosis). (E) Analysis of condensedapoptotic nuclei in WT Fc�RI RBL-2H3 transfectants after 24 hours of treatment with 10 �g/mL anti-Fc�RI (A77) Fab or control (320) Fab. Following treatment, cells werestained with DAPI as described in “Materials and methods,” before analysis by fluorescence microscopy (top panels; 40�/1.3 NA objective lens; insets represent a zoom of oneof the nuclei.). Apoptotic cells have condensed nuclei and crescent shape (inset). Treated cells were also examined by light microscopy (lower panels, 40�/1.3 NA objectivelens). Apoptotic cells contained large vacuoles (inset).

IgA Fc RECEPTOR SIGNALS APOPTOSIS 205BLOOD, 1 JANUARY 2007 � VOLUME 109, NUMBER 1

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom

apoptosis induction in a significant fraction of cells (Figure 1A,middle and right panels). Similar data were obtained withperitoneal macrophages from Fc�RI humanized transgenic mice(not shown).

To study Fc�RI-mediated apoptosis, we took advantage ofpreviously established rat RBL mast-cell transfectants expressingWT human Fc�RI that are able to associate with endogenous ratFcR�.3 As shown in Figure 1B, monomeric targeting of Fc�RI withA77 Fab in these transfectants induced apoptosis (between 30%and 40% of cells). Kinetic analysis showed that double-positive(annexin V/PI) cells could be detected as early as 3 hours and thattheir numbers were maximal after 24 hours (Figure 1C, left panel).When annexin V single staining was used to detect early signs ofapoptosis, a positive population was seen after only 1 hour (notshown). Dose-response studies at 24 hours revealed that apoptosiswas induced at Fab concentrations as low as 0.01 �g/mL and wasmaximum between 5 and 10 �g/mL (Figure 1C, right panel).Apoptosis was also confirmed by determining the appearance ofcells with hypoploid nuclei. An increased proportion was seen inA77-treated cells both in blood monocytes and Fc�RI transfec-tants as compared to 320 Fab-treated cells (Figure 1D). Similarly,A77 Fab-treated cells show an increased proportion of cells withhighly condensed nuclei as revealed by staining with DAPI andwith a highly vacuolated phenotype (Figure 1E). Apoptosis was notinduced in non-transfected cells (not shown).

Induction of apoptosis requires the FcR� ITAM

To determine which type of Fc�RI—associated or not associatedwith FcR�3—accounts for the apoptotic response, we comparedcells transfected with WT Fc�RI to cells transfected with atransmembrane mutant receptor (Fc�RIR209L) unable to associatewith the FcR� signaling adaptor. This transfectant expressedsimilar levels of receptor as WT Fc�RI transfected cells5 (and notshown). Cells were treated for 24 hours with A77 or control 320Fab and were then examined for signs of cell death by annexin V

and PI staining (Figure 2A, left panel). Only a very smallpopulation (�3%) of double-positive cells was detected amongcells transfected with the mutant receptor Fc�RIR209L and among320 Fab-treated cells. Similar data were obtained by countingapoptotic nuclei (Figure 2A, right panel).

These results suggested an important role of FcR� in thedeath-activating signal generated through Fc�RI. A transfectantbearing a chimeric receptor, Fc�RIR209L/�, in which the intracellu-lar portion of Fc�RIR209L was replaced by FcR�, was thus tested forits capacity to mediate apoptosis after targeting with A77 Fab. Asshown in Figure 2B (left panel), the addition of the intracytoplas-mic portion of FcR� to the mutant receptor restored the capacity ofFc�RIR209L to induce apoptosis after monomeric targeting of A77Fab-treated but not 320 Fab-treated cells. The proportion ofannexin V/PI double-positive cells was higher among Fc�RIR209L/�RBL transfectants (�60%) than among control 320 Fab-treatedcells (8.2%). This was confirmed by counting hypoploid apoptoticnuclei (Figure 2B, right panel). The role of FcR� ITAM in theinhibitory effect was then examined. RBL cells transfected with achimeric receptor, Fc�RIR209L-�, in which 2 Y-to-F point mutationsare introduced in the ITAM, were tested for apoptosis followingtreatment with A77 Fab. As shown in Figure 2C (left panel), noapoptosis was seen in these mutant transfectants, indicating theimportance of the FcR� ITAM in this mechanism. This observationwas confirmed by cell-cycle analysis (Figure 2C, right panel).

Apoptosis signaling by Fc�RI differs from inhibitory signaling

Previous studies have shown that Fab fragments A59, A62, andA77 derived from anti-Fc�RI mAbs can induce inhibitory signal-ing, whereas fragment A3 cannot.5 We therefore examined thecapacity of these fragments to trigger an apoptotic response. Asshown in Figure 3A, the apoptotic response was induced by all 4Fabs, including A3. We also tested the effect of the physiologicligand, IgA. As shown in Figure 3B (left panel), as for control IgG,no apoptosis was observed after incubation with monomeric IgA,

Figure 2. Fc�RI-induced apoptosis is mediated by FcR�

ITAM. Induction of apoptosis in RBL-2H3 cells transfected withtransmembrane mutated Fc�IRR209L (A), chimeric Fc�RIR209L/�(B), or chimeric ITAM-mutated Fc�RIR209L/�Y268/278F (C). Cellswere treated with 10 �g/mL A77 or irrelevant control (320) Fab for24 hours before annexin V/PI staining (left panel) or analysis ofhypoploid nuclei (middle panel) as described in “Materials andmethods.” The percentage of apoptotic (annexin V/PI double-positive) cells is shown (right panel). Values are the percentagesof double-positive cells and are means � SD of at least 3 separateexperiments performed in duplicate. *P .01.

206 KANAMARU et al BLOOD, 1 JANUARY 2007 � VOLUME 109, NUMBER 1

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom

even at high concentrations, thus contrasting with data that haveshown powerful anti-inflammatory properties of IgA.5,32 However,incubation with IgA at low serum concentrations (0.5%) resulted inan apoptotic response in transfectants and monocytes, contrary toIgG (Figure 3B right panel), indicating that IgA requires additionalsignals to induce apoptosis. As shown in Figure 3C, susceptibilityto apoptosis induction by IgA began at a serum concentration of 1%and was maximal in the complete absence of serum. The requiredIgA concentrations were relatively high, with a maximal effect atabout 100 �g/mL. In the absence of serum, incubation with A77induced a much higher level of apoptosis (Figure 3D).

We then examined whether multimeric cross-linking could alsoinduce death-activating signals. We compared cells incubated withA77 Fab and cells incubated with preformed immune complexes ofA77 Fab plus rabbit anti–mouse IgG F(ab�)2. As shown in Figure4A, both types of interaction induced an apoptotic response inFc�RI-transfected RBL cells, but the response was much stronger(�80%) after monomeric targeting than after multimeric cross-linking. No apoptosis was induced by control 320 Fab or RAM (notshown) alone. We also tested the effect of IgA cross-linking byanti-IgA in serum-free medium. As shown in Figure 4B, whereasmonomeric IgA induced apoptosis of about 20% to 25% of

Fc�RI-transfected RBL cells (as before), multimeric cross-linkingof IgA induced a much lower level of apoptosis.

Fc�RI death signals involve caspase activation and areblocked by SHP-1

Caspase activation is an early step in cell death and is one of themature intracellular signaling pathways that promotes apoptosis.33

To investigate whether Fc�RI-mediated death-activating signalsinvolve caspase activation, we evaluated apoptosis induction in thepresence of zVAD-fmk, a general caspase inhibitor. ChimericFc�RIR209L/� and WT Fc�RI RBL transfectants were incubatedwith 20 �M zVAD-fmk for 2 hours before treatment with A77 or320 Fab. After 24 hours the cells were stained with annexin V/PIand examined for numbers of cells with hypoploid apoptotic nuclei.As shown in Figure 5A, pretreatment with 20 �M zVAD-fmkstrongly reduced the percentage of apoptotic cells elicited bytreatment with 10 �g/mL Fab A77 in WT Fc�RI (left panel) and inchimeric Fc�RIR209L/� (right panel) transfectants. In contrast, A77Fab had no effect on DiOC6 staining, which reveals changes inmitochondrial membrane potential. Control experiments usingCCCP, an inducer of the mitochondrial apoptosis pathway, mark-edly modified the mitochondrial membrane potential (Figure 5B).To directly evaluate the caspase activation status, we measured thecleavage of the proform of caspase-3. As shown in Figure 5C,monomeric targeting of chimeric Fc�RIR209L/� transfectants andmonocytes with A77 Fab rapidly led to the appearance of the17-kDa cleaved (active) form of caspase-3. This was also con-firmed by intracellular immunostaining using anti–cleavedcaspase-3 antibody analyzed by flow cytometry (not shown).The appearance of cleaved caspase-3 suggested that apoptosiswas directly induced by Fc�RI signaling. Because the inhibitoryfunction of Fc�RI depends on SHP-1 recruitment,5 we examinedwhether the Fc�RI-mediated apoptotic signal was also depen-dent on this phosphatase by using an siRNA approach to knockdown the expression of SHP-1. As shown in Figure 5D (leftpanel), transfection with SHP-1–specific siRNAs strongly re-duced (� 60%) the expression of SHP-1 in chimeric Fc�RIR209L/�transfectants. In the corresponding experiment to determineapoptosis, no differences in the background levels (in thepresence of 320 Fab) were found between cells transfected witha scrambled siRNA control or with the SHP-1–targeting siRNA(Figure 5D, right panel). By contrast, after treatment with A77,

Figure 3. Induction of apoptosis in Fc�RI RBL transfectants and bloodmonocytes by purified human IgA. (A) Comparison of apoptosis induction inchimeric Fc�RIR209L/� RBL transfectants treated with different anti-Fc�RI Fabfragments. Cells were incubated for 24 hours with 10 �g/mL A77, A62, A59, A3, orcontrol (320) Fab fragments and the percentage of apoptotic cells was determined byanalysis of the annexin V/PI double-positive population. One representative experi-ment of 3 is shown. (B) Induction of apoptosis in chimeric Fc�RIR209L/� RBLtransfectants and blood monocytes by IgA was measured in the presence of 10% or0.5% serum. Cells were treated with 50 �g/mL human IgA or control human IgG for 10hours before annexin V/PI staining. Values indicate the percentage of double-positivecells and are means � SD of at least 3 separate experiments performed in duplicate.*P .01. (C) Comparison of apoptosis induction in chimeric Fc�RIR209L/� RBLtransfectants after incubation with 50 �g/mL IgA and 10 �g/mL A77 Fab in thepresence of 0.5% serum. After 10 hours cells were analyzed for annexin V/PIdouble-positivity as described in “Materials and methods.” Values indicate thepercentage of double-positive cells and are means � SD of at least 3 separateexperiments performed in duplicate. (D)Analysis of apoptosis in chimeric Fc�RIR209L/�RBL transfectants after treatment with 50 �g/mL human IgA (�) or control IgG (f) inthe presence of various percentages of serum (left panel) and after treatment withvarious concentrations of immunoglobulin in the presence of 0.5% serum (rightpanel). After 10 hours cells were analyzed for annexin V/PI double-positivity asdescribed in “Materials and methods.” Values indicate the percentage of double-positive cells and are means � SD of at least 3 separate experiments performed induplicate. *P .01.

Figure 4. Fc�RI monovalent targeting induces apoptosis more strongly thanmultimeric targeting. (A) Fc�RIR209L/� RBL transfectants were incubated with 10�g/mL anti-Fc�RI (A77) Fab or preformed complexes of anti-Fc�RI (A77) F(ab�)2 andF(ab�)2 rabbit anti–mouse IgG. After 24 hours cells were analyzed for annexin V/PIdouble-positivity as described in “Materials and methods.” Values indicate thepercentage of double-positive cells and are means � SD of at least 3 separateexperiments performed in duplicate. *P .01. (B) Fc�RIR209L/� RBL transfectantswere incubated with 50 �g/mL IgA or preformed complexes of 50 �g/mL IgA and 100�g/mL goat anti–human IgA. After 10 hours cells were analyzed for annexin V/PIdouble-positive population as described in “Materials and methods.” Values indicatethe percentage of double-positive cells and are means � SD of at least 3 separateexperiments performed in duplicate. *P .01.

IgA Fc RECEPTOR SIGNALS APOPTOSIS 207BLOOD, 1 JANUARY 2007 � VOLUME 109, NUMBER 1

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom

the apoptotic effect in chimeric Fc�RIR209L/� transfectants wasincreased in cells that had been targeted with SHP-1 siRNAcompared with those treated with scrambled siRNA (Figure 5Dright panel). In 4 independent experiments in which SHP-1levels fell by more than 60%, the increase in A77-inducedapoptosis in SHP-1 siRNA-transfected cells was 1.53 0.12-fold (mean � SD; P .05) compared with cells transfectedwith scrambled siRNA. These observations suggested that highlevels of SHP-1 inhibited the Fc�RI-mediated apoptotic signal.

Monomeric targeting of Fc�RI with anti-Fc�RI Fab inducesapoptosis in human monocytic cell lines

To analyze whether monomeric targeting of Fc�RI in human tumorcell lines induces apoptosis, we took advantage of human myelo-blastic monoMac6 and U937 cells which express Fc�RI. As shownin Figure 6A, monomeric targeting of Fc�RI with anti-Fc�RI A77Fab in MonoMac6 and U937 cell lines induced apoptosis (between10% and 35% of cells) as revealed by cell-cycle analysis, whereasno apoptosis was seen with irrelevant 320 Fab. This variability inapoptosis induction was linked to the level of Fc�RI expression, asthe highest induction of apoptosis was observed in MonoMac6cells, which express high levels of Fc�RI protein than U937 cells(not shown). We also tested the effect of the physiologic ligand,IgA. Incubation with IgA, but not IgG, at low serum concentrationsof FCS (0.5%) resulted in an apoptotic response in these cell lines(not shown). To directly evaluate the caspase activation status, weexamined the cleavage of the proform of caspase-3 in these celllines. As shown in Figure 6B, monomeric targeting with A77 Fabrapidly led to the appearance of the 17-kDa cleaved (active) formof caspase-3.

Antitumoral effect of anti-Fc�RI Fab treatment of nude mice

The tumoricidal activity of A77 Fab was tested in nude miceinjected subcutaneously with chimeric Fc�RI RBL transfectants.We used 2 different protocols to measure the effect of A77 Fab. Inthe first protocol (prevention model), A77 was administered shortlyafter the injection of tumor cells. As shown in Figure 7A, tumorsize was clearly diminished compared with PBS- and 320 Fab-treated mice, and the mice survived significantly longer thancontrols (Figure 7C). In the second protocol (treatment model),A77 Fab administration was started on day 10, once the subcutane-ous tumors had already developed. Tumor development wasimmediately halted, whereas it continued in mice treated with PBSor 320 Fab (Figure 7B). Again the A77 Fab-treated mice survivedsignificantly longer than controls (Figure 7D). To confirm that theantitumoral effect was due to apoptosis, we applied the TUNELmethod to tumor tissue sections from each group. The number ofpositive cells was far higher in A77 Fab-treated animals than incontrols (Figure 7E).

Discussion

We report that monomeric targeting of Fc�RI by different anti-Fc�RI Fab promotes apoptosis of blood human monocytes, humanmonocytic cell lines, and RBL mast cells transfected with Fc�RI.Apoptosis was not induced in non-transfected RBL cells or in cellstransfected with a mutant Fc�RIR209L receptor unable to associatewith the FcR� subunit. Expression of a chimeric mutant receptor inwhich the intracytoplasmic domain was replaced by FcR� restoredthe apoptotic signal, whereas mutating the 2 tyrosines contained inthe FcR� ITAM abolished the apoptotic signal. These results show

Figure 5. Analysis of the apoptosis pathways induced by Fc�RI targeting. (A) WT Fc�RI and Fc�RIR209L/� RBL transfectants were treated with 20 �M zVAD-fmk or solventfor 40 minutes before adding 10 �g/mL anti-Fc�RI (A77) Fab. After 24 hours cells were analyzed for annexin V/PI double-positivity as described in “Materials and methods.”Values indicate the percentage of double-positive cells and are means � SD of at least 3 separate experiments performed in duplicate. *P .01. (B) Analysis of themitochondrial pathway of apoptosis. Fc�RIR209L/� RBL transfectants were incubated with the mitochondrial apoptosis pathway-inducing compound CCCP (200 �M) or with 10�g/mL anti-Fc�RI (A77) Fab or control (320) Fab. After 24 hours cells were stained with the mitochondrial membrane potential–sensitive dye DiOC6 and analyzed by flowcytometry as described in “Materials and methods.” Induction of the mitochondrial pathway resulted in a strong loss of DiOC6 staining. (C) Analysis of the “cleaved” (activated)form of caspase-3. Fc�RIR209L/� cells were treated with 10 �g/mL A77 or control (320) Fab for the times indicated. Cells were then lysed, separated by SDS-PAGE, andtransferred to PVDF membranes before immunoblotting with an antibody recognizing the cleaved form of caspase-3. To control for equal loading, blots were stripped andreprobed with antiscramblase (PLSCR) antibody plus goat anti–rabbit Ig-HRP. One representative experiment of 3 is shown. (D) Effect of SHP-1 siRNA treatment onFc�RI-mediated apoptosis in Fc�RIR209L/� RBL transfectants. Analysis of relative SHP-1 expression levels by anti–SHP-1 immunoblotting in lysates from cells transfected withscrambled or SHP-1–specific siRNA (left panel). Analysis of the percentage of Fc�RI-mediated apoptosis in Fc�RIR209L/� cells transfected with scrambled siRNA orSHP-1–targeting specific siRNA (right panel). Fc�RIR209L/� cells were transfected twice, 24 hours apart, with scrambled siRNA or SHP-1–targeting specific siRNA. After thesecond transfection, cells were treated with 10 �g/mLA77 or irrelevant control (320) Fab (right panel). Cells were then analyzed for hypoploid nuclei as described in “Materialsand methods.” Numbers are the percentage of cells with hypoploid nuclei at 18 hours in a representative experiment.

208 KANAMARU et al BLOOD, 1 JANUARY 2007 � VOLUME 109, NUMBER 1

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom

that Fc�RI can not only mediate activating and inhibitory signalsinvolved in proinflammatory and anti-inflammatory functions butcan also control cell survival in a manner that depends on afunctional ITAM. These data provide a mechanistic explanation forthe involvement of IgA in apoptosis18,34 and identify Fc�RI as thetarget receptor.

Although Fc�RI-mediated apoptosis required intact signalingthrough the ITAM of the associated FcR� adaptor, some substantialdifferences to previously described proinflammatory and anti-inflammatory functions were noted. The mAb A3, which wasincapable of anti-inflammatory signaling after monomeric targetingof Fc�RI, readily induced apoptosis. The A3 mAb recognizes theinterdomain between the first and second immunoglobulin-likedomains of Fc�RI, whereas the other 3 functional antibodies usedhere and in previous studies recognize the second immunoglobulin-like domain.5,35 Furthermore, whereas monomeric IgA inducedinhibitory signaling, it failed to promote apoptosis in high serumconditions. IgA, however, induced apoptosis in low serum condi-tions, suggesting that it required a second signal. Given that IgA ispresent at high concentrations in human blood, the relevantapoptotic signals must be suppressed to avoid apoptosis ofFc�RI-bearing blood cells. IgA can, however, enhance pro-grammed cell death in the presence of low serum concentrationindicating requirements for other signals, and this phenomenonmight come into play in inflammatory sites where stress andischemic conditions predominate and thus can provide additionalsignals for IgA-mediated apoptosis.

The data obtained with Fc�RI are reminiscent of those obtainedwith other ITAM-bearing receptors in the immune system, includ-ing antigen receptors on B and T cells (BCR and TCR), respec-tively. Apoptosis triggered by BCR and TCR is crucial forestablishing the lymphocyte repertoire.36,37 It is generally agreedthat strong engagement promotes negative selection through apopto-sis, whereas engagement with lower-affinity ligands leads to clonalexpansion. Both responses crucially depend on phosphorylation ofITAM tyrosines,19,38,39 but qualitative differences may occur amongthe different ITAMs of the associated subunits.40-42 Our data onFc�RI also point to a crucial role of the FcR� ITAM in cellactivation, inhibition, and apoptosis. However, what determines thedifferent signals and outcomes is unclear. Our previous studies ofinhibitory function suggested that weak phosphorylation of theFcR� ITAM after monomeric targeting led to preferential recruit-ment of SHP-1 phosphatase rather than Syk, which recruited to aheavily phosphorylated ITAM following sustained multimericaggregation.5 Our present data suggest that monomeric targetingprovides also a strong apoptotic signal. However, multimericsignals also induced apoptosis (albeit at somewhat lower levels),

Figure 6. Induction of apoptosis by Fc�RI targeting in human monocytic celllines. (A) Induction of apoptosis in MonoMac6 or U937 cells after 24 hours oftreatment with 20 �g/mL anti-Fc�RI (A77) Fab or control (320) Fab. Cells werestained with PI and analyzed for the appearance of hypoploid nuclei as described in“Materials and methods.” Numbers indicate the percentage of cells with hypoploidnuclei. Values indicate the percentage of double-positive cells and are means � SDof at least 3 separate experiments performed in duplicate. *P .01. (B) Analysis ofthe “cleaved” (activated) form of caspase-3 following anti-Fc�RI Fab treatment. TheMonoMac6 or U937 cells were treated with 20 �g/mL A77 or control (320) Fab forindicated times. Cells were then lysed, separated by SDS-PAGE, and transferred toPVDF membranes before immunoblotting with an antibody recognizing the cleavedform of caspase-3. To control for equal loading, blots were stripped and reprobed withantiscramblase (PLSCR) antibody plus goat anti–rabbit Ig-HRP. One representativeexperiment of 3 is shown.

Figure 7. Monovalent targeting of FcRI prevents andarrests subcutaneous tumor development by Fc�RIRBL transfectants in nude mice. Nude mice (6/group)were injected subcutaneously with WT Fc�RI RBL trans-fectants. In the prevention model (A,C), animals weretreated intraperitoneally with 50 �g anti-Fc�RI (A77) Fab,control (320) Fab, or PBS on days 0, 3, 6, 9, 12, 15, 18,and 21. In treatment model (B,D), mice were treatedintraperitoneally with 50 �g anti-Fc�RI (A77) Fab, control(320) Fab, or PBS on day 12 when tumors had started todevelop followed by injections on days, 15, 18, and 21.(A-B) Tumor size was evaluated on day 25 and isrepresented as the mean � SD (n � 6). (C-D) Survival ofmice treated with PBS, 320 Fab, or A77 Fab. (E) TUNELanalysis of RBL tumor cells after treatment with control(320) Fab or anti-Fc�RI (A77) Fab. Images show PASstaining of formalin-fixed tumor sections. Apoptotic cellswere visualized by TUNEL analysis of cryosections asdescribed in “Materials and methods (10�/0.30 NA objec-tive lens).

IgA Fc RECEPTOR SIGNALS APOPTOSIS 209BLOOD, 1 JANUARY 2007 � VOLUME 109, NUMBER 1

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom

suggesting a complex relationship and the need for specificadditional signals. This is supported by our data showing that thenatural ligand, IgA, induces apoptosis only in medium containinglittle or no serum. This putative requirement for additional signalsmay be overcome in very strong conditions of Fc�RI aggregation,as recently reported for human neutrophils.18 Independently ofthese considerations, our data suggest that Fc�RI signaling acti-vates an intrinsic caspase-dependent pathway, because apoptosisinduction by Fc�RI was essentially blocked by the pan-caspaseinhibitor zVAD-fmk. One of the downstream targets is caspase-3,the proform of which is rapidly cleaved into the active form inRBL-cell transfectants, blood monocytes, and monocytic cell lines.

Our previous studies on the inhibitory function of IgA sug-gested that the ITAM-dependent anti-inflammatory action criticallyinvolves SHP-1, similarly to ITIM-mediated inhibition.5 Interest-ingly, SHP-1–mediated inhibitory signals block apoptosis of B andT cells,12,15 whereas SHIP recruitment attenuates a proapoptoticsignal initiated by Fc�RIIB.13 Here, decrease of SHP-1 expressionin Fc�RI transfectants induced by siRNA promoted an increasedsusceptibility to the apoptotic effect of anti-Fc�RI Fab suggesting acrucial role for the physiologic levels of SHP-1 to regulate thebalance between Fc�RI-mediated apoptosis and previously de-scribed inhibitory signaling.5 SHP-1 may thus represent an impor-tant target in the switch mechanisms between these functions.

Finally, we show that Fc�RI is expressed in some leukemic celllines and thus Fc�RI targeting might be useful for treating tumorsthat are susceptible to Fc�RI-mediated apoptosis. Indeed, tumorformation and progression after subcutaneous grafting of humanFc�RI-bearing RBL cells in nude mice were prevented or arrestedby an Fc�RI-targeting Fab but not by an irrelevant control Fab.

Taken together, our data show that Fc�RI is a new apoptoticmodule that, in addition to its anti-inflammatory, inhibitory signal-

ing activity, may be important for controlling tumor growth. Fc�Ris thus a highly versatile receptor that can trigger multipleresponses ranging from activation and inhibition to the triggeringof apoptosis, depending on the physiologic context. This versatilitymight have therapeutic implications, both in inflammation5 and, asshown here, in cancer.

Acknowledgments

This work was supported by INSERM and a grant from theAssociation pour la Recherche sur le Cancer (no. 3613). We thankM. Arcos-Fajardo, H. Cohen, and J. Bex for technical support andhelp in animal care.

Authorship

Contribution: Y.K. designed and performed experiments and wrote themanuscript; H.T. performed apoptotic nuclei analysis; S.P. performedtransfectant analysis; D.E. performed apoptosis inhibitor experiments;C.G.-M. performed siRNA experiments; M.P. designed inhibitor andbiochemical experiments; U.B. interpreted and designed experimentsand wrote the manuscript; R.C.M. conceived the idea, designed theexperiments, and wrote the manuscript.

Conflict-of-interest disclosure: The authors declare no compet-ing financial interests.

Correspondence: Renato Monteiro, Inserm U699, Faculte deMedecine Paris 7, Site Xavier Bichat, 16 rue Henri Huchard, 75780Paris Cedex 18; e-mail: renato.monteiro@bichat.inserm.fr.

References

1. Monteiro RC, van de Winkel JGJ. IgA Fc recep-tors. Annu Rev Immunol. 2003;21:177-204.

2. Pfefferkorn LC, Yeaman GR. Association ofIgA-Fc receptors (Fc�R) with Fc�RI� 2 subunitsin U937 cells: aggregation induces the tyrosinephosphorylation of �2. J Immunol. 1994;153:3228-3236.

3. Launay P, Patry C, Lehuen A, Pasquier B, BlankU, Monteiro RC. Alternative endocytic pathwayfor immunoglobulin A Fc receptors (CD89) de-pends on the lack of FcR� association and pro-tects against degradation of bound ligand. J BiolChem. 1999;274:7216-7225.

4. Hamerman JA, Tchao NK, Lowell CA, Lanier LL.Enhanced Toll-like receptor responses in the ab-sence of signaling adaptor DAP12. Nat Immunol.2005;6:579-586.

5. Pasquier B, Launay P, Kanamaru Y, et al. Identifi-cation of Fc�RI as an inhibitory receptor that con-trols inflammation: dual role of FcR� ITAM. Immu-nity. 2005;22:31-42.

6. Hamerman JA, Lanier LL. Inhibition of immuneresponses by ITAM-bearing receptors. Sci STKE.2006;2006:re1.

7. Shen L. A monoclonal antibody specific for immu-noglobulin A receptor triggers polymorphonuclearneutrophil superoxide release. J Leukoc Biol.1992;51:373-378.

8. Patry C, Herbelin A, Lehuen A, Bach JF, MonteiroRC. Fc� receptors mediate release of tumour ne-crosis factor-� and interleukin-6 by human mono-cytes following receptor aggregation. Immunol-ogy. 1995;86:1-5.

9. Geissmann F, Launay P, Pasquier B, et al. A sub-set of human dendritic cells expresses IgA Fc re-ceptor (CD89), which mediates internalization

and activation upon cross-linking by IgA com-plexes. J Immunol. 2001;166:346-352.

10. Nimmerjahn F, Ravetch JV. Fc� receptors: oldfriends and new family members. Immunity. 2006;24:19-28.

11. Ashman RF, Peckham D, Stunz LL. Fc receptoroff-signal in the B cell involves apoptosis. J Im-munol. 1996;157:5-11.

12. Ono M, Okada H, Bolland S, Yanagi S, KurosakiT, Ravetch JV. Deletion of SHIP or SHP-1 revealstwo distinct pathways for inhibitory signaling. Cell.1997;90:293-301.

13. Pearse RN, Kawabe T, Bolland S, Guinamard R,Kurosaki T, Ravetch JV. SHIP recruitment attenu-ates Fc�RIIB-induced B cell apoptosis. Immunity.1999;10:753-760.

14. Tzeng SJ, Bolland S, Inabe K, Kurosaki T, PierceSK. The B cell inhibitory Fc receptor triggers apo-ptosis by a novel c-Abl family kinase-dependentpathway. J Biol Chem. 2005;280:35247-35254.

15. Zhang J, Somani AK, Watt S, Mills GB, Simino-vitch KA. The Src-homology domain 2-bearingprotein tyrosine phosphatase-1 inhibits antigenreceptor-induced apoptosis of activated periph-eral T cells. J Immunol. 1999;162:6359-6367.

16. Cyster JG, Goodnow CC. Protein tyrosine phos-phatase 1C negatively regulates antigen receptorsignaling in B lymphocytes and determinesthresholds for negative selection. Immunity. 1995;2:13-24.

17. Cassady-Cain RL, Kaushik AK. Increased nega-tive selection impairs neonatal B cell repertoirebut does not directly lead to generation of dis-ease-associated IgM auto-antibodies. Int Immu-nol. 2006;18:661-669.

18. Schettini J, Salamone G, Trevani A, et al. Stimu-

lation of neutrophil apoptosis by immobilized IgA.J Leukoc Biol. 2002;72:685-691.

19. Yao XR, Flaswinkel H, Reth M, Scott DW. Immu-noreceptor tyrosine-based activation motif is re-quired to signal pathways of receptor-mediatedgrowth arrest and apoptosis in murine B lym-phoma cells. J Immunol. 1995;155:652-661.

20. Launay P, Grossetete B, Arcos-Fajardo M, et al.Fc� receptor (CD89) mediates the developmentof immunoglobulin A (IgA) nephropathy (Berger’sdisease): evidence for pathogenic soluble recep-tor-IgA complexes in patients and CD89 trans-genic mice. J Exp Med. 2000;191:1999-2009.

21. Pastorelli C, Veiga J, Charles N, et al. IgE recep-tor type I-dependent tyrosine phosphorylation ofphospholipid scramblase. J Biol Chem. 2001;276:20407-20412.

22. Monteiro RC, Cooper MD, Kubagawa H. Molecu-lar heterogeneity of Fc� receptors detected byreceptor-specific monoclonal antibodies. J Immu-nol. 1992;148:1764-1770.

23. Silvain C, Patry C, Launay P, Lehuen A, MonteiroRC. Altered expression of monocyte IgA Fc re-ceptors is associated with defective endocytosisin patients with alcoholic cirrhosis: potential rolefor IFN-�. J Immunol. 1995;155:1606-1618.

24. Monteiro RC, Kubagawa H, Cooper MD. Cellulardistribution, regulation, and biochemical nature ofan Fc� receptor in humans. J Exp Med. 1990;171:597-613.

25. Sung SS, Nelson RS, Silverstein SC. Yeast man-nans inhibit binding and phagocytosis of zymosanby mouse peritoneal macrophages. J Cell Biol.1983;96:160-166.

26. Vermes I, Haanen C, Steffens-Nakken H, Reutel-ingsperger C. A novel assay for apoptosis: flow

210 KANAMARU et al BLOOD, 1 JANUARY 2007 � VOLUME 109, NUMBER 1

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom

cytometric detection of phosphatidylserine ex-pression on early apoptotic cells using fluoresceinlabelled annexin V. J Immunol Methods. 1995;184:39-51.

27. Nicoletti I, Migliorati G, Pagliacci MC, Grignani F,Riccardi C. A rapid and simple method for mea-suring thymocyte apoptosis by propidium iodidestaining and flow cytometry. J Immunol Methods.1991;139:271-279.

28. Miller T, Beausang LA, Meneghini M, Lidgard G.Death-induced changes to the nuclear matrix: theuse of anti-nuclear matrix antibodies to studyagents of apoptosis. Biotechniques. 1993;15:1042-1047.

29. Macouillard-Poulletier de G, Belaud-RotureauMA, Voisin P, et al. Flow cytometric analysis ofmitochondrial activity in situ: application to acetyl-ceramide-induced mitochondrial swelling and ap-optosis. Cytometry. 1998;33:333-339.

30. Zamzami N, Marchetti P, Castedo M, et al. Re-duction in mitochondrial potential constitutes anearly irreversible step of programmed lymphocytedeath in vivo. J Exp Med. 1995;181:1661-1672.

31. Hamawy MM, Siraganian RP. Large scale isola-

tion of intact rat basophilic leukemia (RBL-2H3)cells. J Immunol Methods. 1997;201:11-14.

32. Olas K, Butterweck H, Teschner W, Schwarz HP,Reipert B. Immunomodulatory properties of hu-man serum immunoglobulin A: anti-inflammatoryand pro-inflammatory activities in human mono-cytes and peripheral blood mononuclear cells.Clin Exp Immunol. 2005;140:478-490.

33. Rathmell JC, Thompson CB. The central effectorsof cell death in the immune system. Annu RevImmunol. 1999;17:781-828.

34. Reljic R, Crawford C, Challacombe S, Ivanyi J.Mouse IgA inhibits cell growth by stimulating tu-mor necrosis factor-� production and apoptosis ofmacrophage cell lines. Int Immunol. 2004;16:607-614.

35. Morton HC, van Zandbergen G, van Kooten C,Howard CJ, van de Winkel JG, Brandtzaeg P. Im-munoglobulin-binding sites of human Fc�RI(CD89) and bovine Fc�2R are located in theirmembrane-distal extracellular domains. J ExpMed. 1999;189:1715-1722.

36. Strasser A, Bouillet P. The control of apoptosis inlymphocyte selection. Immunol Rev. 2003;193:82-92.

37. Palmer E. Negative selection—clearing out thebad apples from the T-cell repertoire. Nat RevImmunol. 2003;3:383-391.

38. Liu Y, He Y, Zhang J, Jiang L, Liu S, Zheng D. Ty-rosine mutation in CD3�-ITAM blocked T lympho-cyte apoptosis mediated by CD3�. Immunol In-vest. 2003;32:59-70.

39. Tseng J, Eisfelder BJ, Clark MR. B-cell antigenreceptor-induced apoptosis requires both Ig� andIg�. Blood. 1997;89:1513-1520.

40. Combadiere B, Freedman M, Chen L, ShoresEW, Love P, Lenardo MJ. Qualitative and quanti-tative contributions of the T cell receptor zetachain to mature T cell apoptosis. J Exp Med.1996;183:2109-2117.

41. Chae WJ, Lee HK, Han JH, et al. Qualitatively differ-ential regulation of T cell activation and apoptosis byT cell receptor � chain ITAMs and their tyrosine resi-dues. Int Immunol. 2004;16:1225-1236.

42. Reichlin A, Hu Y, Meffre E, et al. B cell develop-ment is arrested at the immature B cell stage inmice carrying a mutation in the cytoplasmic do-main of immunoglobulin �. J Exp Med. 2001;193:13-23.

IgA Fc RECEPTOR SIGNALS APOPTOSIS 211BLOOD, 1 JANUARY 2007 � VOLUME 109, NUMBER 1

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom

online September 21, 2006 originally publisheddoi:10.1182/blood-2006-06-025882

2007 109: 203-211  

Guérin-Marchand, Marina Pretolani, Ulrich Blank and Renato C. MonteiroYutaka Kanamaru, Houda Tamouza, Séverine Pfirsch, Delphine El Mehdi, Claudine tumor growth

ITAM and affectsγIgA Fc receptor I signals apoptosis through the FcR 

http://www.bloodjournal.org/content/109/1/203.full.htmlUpdated information and services can be found at:

(1930 articles)Signal Transduction    (5388 articles)Immunobiology   

(746 articles)Apoptosis    Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

  Copyright 2011 by The American Society of Hematology; all rights reserved.of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society

For personal use only.on May 31, 2016. by guest www.bloodjournal.orgFrom