sign in sign up
<<
Home , Phagocytosis

Published March 4, 2016, doi:10.4049/jimmunol.1502198 The Journal of

CD44 Inhibition of Phagocytosis Targets Fcg Receptor– and 3–Dependent Mechanisms

Alaa Amash,*,† Lin Wang,‡ Yawen Wang,‡ Varsha Bhakta,x Gregory D. Fairn,† Ming Hou,‡ Jun Peng,‡ William P. Sheffield,*,x and Alan H. Lazarus*,†,{

Targeting CD44, a major leukocyte adhesion molecule, using specific Abs has been shown beneficial in several models of autoim- mune and inflammatory diseases. The mechanisms contributing to the anti-inflammatory effects of CD44 Abs, however, remain poorly understood. Phagocytosis is a key component of function and can play a pivotal role in autoimmune states where CD44 Abs have shown to be effective. In this study, we show that the well-known anti-inflammatory CD44 Ab IM7 can inhibit murine macrophage phagocytosis of RBCs. We assessed three selected macrophage phagocytic receptor systems: Fcg receptors (FcgRs), complement receptor 3 (CR3), and dectin-1. Treatment of with IM7 resulted in significant inhibition of FcgR-mediated phagocytosis of IgG-opsonized RBCs. The inhibition of FcgR-mediated phagocytosis was at an early stage in the phagocytic process involving both inhibition of the binding of the target RBC to the macrophages and postbinding events. This CD44 Ab also inhibited CR3-mediated phagocytosis of C3bi-opsonized RBCs, but it did not affect the phagocytosis of zymosan particles, known to be mediated by the C-type lectin dectin-1. Other CD44 Abs known to have less broad anti-inflammatory activity, including KM114, KM81, and KM201, did not inhibit FcgR-mediated phagocytosis of RBCs. Taken together, these findings demonstrate selective inhibition of FcgR and CR3-mediated phagocytosis by IM7 and suggest that this broadly anti- inflammatory CD44 Ab inhibits these selected macrophage phagocytic pathways. The understanding of the immune-regulatory effects of CD44 Abs is important in the development and optimization of therapeutic strategies for the potential treatment of autoimmune conditions. The Journal of Immunology, 2016, 196: 000–000.

he CD44 transmembrane glycoprotein is one of the most glycoproteins (80–200 kDa in size) is encoded by a single, highly structurally and functionally variable surface molecules conserved gene (2). The wide molecular diversity is in part gen- T present on most nucleated cells (1). The CD44 group of erated by alternative splicing, which affects predominantly the extracellular domains of the protein, and by posttranslational modifications, which differ depending on the type and growth *Canadian Services Centre for Innovation, Toronto, Ontario K1G 4J5, Canada; conditions (3). The standard or hematopoietic isoform (CD44s) is †Keenan Research Centre for Biomedical Science, Li Ka Shing Knowledge Institute, present on the of most vertebrate cells (4). In addition ‡ St. Michael’s Hospital, Toronto, Ontario M5B 1T8, Canada; Department of Hema- to its role in hyaluronic acid (HA) binding, numerous studies tology, Qilu Hospital, Shandong University, Jinan, Shandong 250012, China; xDepartment of Pathology and Molecular Medicine, McMaster University, Hamilton, suggest that CD44 may modulate immune responses and inflam- { Ontario L8N 3Z5, Canada; and Department of Laboratory Medicine and Patho- mation through at least two major mechanisms. First, CD44 was , Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 1A1, Canada shown to play an important role in controlling leukocyte extrav- asation and infiltration into inflammatory sites (5–7). Second, ORCIDs: 0000-0001-8922-8826 (L.W.); 0000-0003-4876-1207 (Y.W.); 0000-0001- 6508-168X (G.D.F.); 0000-0002-5870-8189 (W.P.S.). CD44 can induce signal transduction pathways and cell activation. Received for publication October 13, 2015. Accepted for publication February 8, Engagement of CD44 was shown to trigger cytotoxic function in 2016. T cells, and CD44 expression correlates positively with regulatory This work was supported by peer-reviewed operating grants from the Canadian Blood T cell potency and function (8–10). Additionally, association be- Services/Canadian Institutes of Health Research Partnership to A.H.L. (Grant tween CD44 deficiency and enhanced proinflammatory cytokine FRN6897) and from the Canadian Blood Services Intramural Operating Grant pro- gram to A.H.L. (Grant 2013-IG-AL00545). Resources for both programs are pro- production by peritoneal macrophages in response to Escherichia vided by Health Canada, a division of the federal government of Canada. A.A. is the coli was recently reported (11). Recent studies suggest a possible recipient of a Canadian Blood Services postdoctoral fellowship. The views expressed role for CD44 in negative regulation of TLR-mediated inflam- herein do not necessarily represent the views of the federal government of Canada. matory responses (12, 13). Address correspondence and reprint requests to Dr. Alan H. Lazarus, Keenan Re- search Centre for Biomedical Science, St. Michael’s Hospital, 209 Victoria Street, Phagocytosis is a critical component of both innate and adaptive Toronto, ON M5B 1T8, Canada. E-mail address: [email protected] and can play key roles in the first line of defense against The online version of this article contains supplemental material. invading as well as in tissue homeostasis and Abbreviations used in this article: BMDM, marrow–derived macrophage; C3bi- remodeling (14). Dysregulation of phagocytosis can lead not only SRBC, C3bi-opsonized SRBC; CMFDA, 5-chloromethylfluorescein diacetate; com- to impaired host immune responses to , but may also plete RPMI, RPMI 1640, 10% FBS, 4 mM L-glutamine, penicillin (100 U/ml), and streptomycin (100 mg/ml); CR3, complement receptor 3; FcgR, Fcg receptor; HA, lead to alterations in immune tolerance and to the development of hyaluronic acid; IgG-SRBC, IgG-opsonized SRBC; ITP, immune thrombocytopenia; chronic inflammation and autoimmunity (15). The phagocytic PerM, peritoneal macrophage; PFA, paraformaldehyde; PNGase, peptide-N-glycosi- process can be initiated by a wide range of surface receptors dase; RAW, RAW264.7. which often involve two main groups: nonopsonic receptors that Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 are capable of direct recognition and binding of target particles,

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1502198 2 REGULATION OF MACROPHAGE PHAGOCYTOSIS BY ANTI-CD44 and opsonic receptors that recognize , such as Abs and marrow–derived macrophages (BMDM) from C57BL/6- and CD44- complement. Of these opsonic receptors, Fcg receptors (FcgRs), deficient mice were generated as previously described by Zhang et al. which bind the Fc portion of IgG, and complement receptor 3 (36). Peritoneal macrophages (PerM) were prepared by injection of cold RPMI 1640 into the peritoneum of euthanized CD1 female mice, followed (CR3), which binds the C3bi component of complement, are the by rapid removal of all fluid. The peritoneal cells obtained were washed best described (16). For the nonopsonic receptors, dectin-1 is the with cold PBS and resuspended in complete RPMI. Cells were cultured in most studied receptor and is thought to be the major receptor for Nunclon surface cell culture dishes (Nunc, Roskilde, Denmark) at 37˚C the fungal component b-glucan in macrophages (17). and 5% CO2 for 4 h, followed by extensive washing with cold PBS to remove nonadherent cells. Adherent cells were removed with a scraper and Abs to CD44 have been successfully used in the treatment of transferred to complete RPMI, counted, and cultured in flat-bottom 12-well several animal models of autoimmune and inflammatory diseases, tissue culture plates (Corning, Corning, NY) at 5 3 105 cells/well at 37˚C including experimental autoimmune encephalomyelitis (18, 19), and 5% CO2 overnight. Cells were then washed and incubated in complete collagen- or Ab-induced arthritis (7, 20–23), autoimmune diabetes RPMI for the analysis of phagocytosis. The purity of the PerM population . (24, 25), experimental autoimmune uveoretinitis (26, 27), exper- following this procedure was 90% as determined by flow cytometry using fluorescently labeled anti-CD45 and anti-F4/80 Abs (data not imental colitis (28), experimental pulmonary eosinophilia (29), shown). and cutaneous delayed-type hypersensitivity (30). The mAb rat For use in FcgR- or CR3-mediated phagocytosis assays, RAW cells, anti-human/mouse CD44, IM7, which recognizes an extracellular BMDM, or PerM were cultured at 5 3 105/well overnight in 12-well tissue epitope on the CD44s, has been extensively used in many of these culture plates. For zymosan phagocytosis, RAW macrophages were cul- tured at 105/well overnight in 24-well plates. studies, and robust anti-inflammatory effects have been reported with the administration of the Ab in these disease models (7, 18, FcgR-mediated phagocytosis 19, 21, 22, 24–26). In comparison, the mAbs KM81, KM201, and KM114 have only been shown to have limited anti- SRBCs (MP Biomedicals) were washed with PBS, resuspended at 5 3 108/ml inflammatory activity in arthritis (20–23). In addition to the in PBS, and opsonized with a subagglutinating concentration (50 mg/ml) of polyclonal rabbit anti-SRBC IgG for 1 h at room temperature with broad anti-inflammatory effects of IM7, it is also been demon- gentle agitation, followed by washing with PBS to remove the unbound strated to inhibit hematopoiesis as well as homing of bone Ab. Subsequent to their sensitization, SRBCs were adjusted to a concen- marrow cells (31–33). Although in our study we have not found tration of 2 3 107/ml and held in complete RPMI. Sensitization of SRBCs evidence for significant direct cytotoxicity of IM7, it has been was confirmed by flow cytometry using FITC-labeled F(ab9)2 of goat anti- rabbit IgG (Life Technologies; data not shown). shown to induce programmed cell death in some cells (34, 35). Phagocytosis was assessed as previously described with minor modifi- Despite that several mechanisms of action have been attributed cations (37). Briefly, RAW macrophages or BMDM were cultured at 5 3 to this CD44 mAb such as reduction in the HA-binding function 105/well overnight in 12-well tissue culture plates and washed with RPMI of CD44 (5), potentially due to its proteolytic removal from the 1640 prior to the addition of CD44 mAbs or inhibitors (Fc Block, cyto- cell surface (7), the underlining mechanisms of these anti- chalasin D [Life Technologies]) for 30 min at 37˚C. Subsequently, cells were washed again to remove free Ab before the addition of IgG- inflammatory effects are yet to be defined. opsonized SRBCs (IgG-SRBCs) at a 20:1 ratio (RBC/macrophage). In this study, we provide evidence that the anti-inflammatory Macrophages were incubated with IgG-SRBCs for 30 min at 37˚C fol- CD44 mAb IM7, but not the other CD44 mAbs studied, can se- lowed by washing with cold PBS to remove unbound SRBCs. Bound lectively inhibit macrophage FcgR and CR3-mediated phagocy- (noninternalized) SRBCs were lysed and removed with distilled water for 90 s followed by fixation with 4% paraformaldehyde (PFA). Phagocytosis tosis of opsonized RBCs, but not dectin-1–mediated phagocytosis was visualized using the Nikon Eclipse TS100 inverted microscope system of zymosan particles. Furthermore, we show that this inhibitory (Nikon Instruments, Melville, NY), and images of five different fields from effect is at an early stage of the phagocytic process. each well were collected. Quantification of phagocytosis was done by counting the number of internalized RBCs in 300–500 macrophages from each sample and the phagocytic index (PI) was calculated according to the Materials and Methods following formula: PI = (total number of internalized RBCs/total number Mice and Abs of counted macrophages) 3 100. For PerM, cells were cultured at 5 3 105/well overnight in 12-well tissue C57BL/6 mice and CD-1 mice were purchased from Charles River Lab- culture plates and washed with RPMI 1640 prior to the addition of CD44 oratories (Montreal, QC, Canada or Kingston, NY). CD44-deficient mice tm1Hbg mAb for 30 min at 37˚C. SRBCs were labeled with the green fluorescent (B6.129(Cg)-Cd44 /J) were from The Jackson Laboratory (Bar Har- 8 dye 5-chloromethylfluorescein diacetate (CMFDA; 100 mM per 5 3 10 bor, ME). All animal studies were approved by the St. Michael’s Hospital SRBCs in 1 ml; Life Technologies) prior to sensitization with IgG. PerM Animal Care Committee. were washed to remove free Abs, followed by the addition of the CMFDA- Polyclonal rabbit anti–SRBC IgG was purchased from MP Biomedicals labeled IgG-SRBCs at a 20:1 ratio (RBC/macrophage) for 30 min at 37˚C. (Solon, OH). Normal rabbit IgG and rat IgG were from Jackson PerM were then washed to remove unbound SRBCs and treated with water ImmunoResearch Laboratories (West Grove, PA). The mAb rat anti-CD44 for 90 s to lyse and remove bound SRBCs. Cells were incubated with the (clone IM7), rat IgG2b, rat IgG1, and mouse IgG1 isotype control were fixable viability dye eFluor 450 (eBioscience; dilution 1:1000) for 10 min from Bio X Cell (West Lebanon, NH). Rat anti-CD44 (clone KM114), rat at 4˚C, followed by washing and fixation in 4% PFA. Cells were then anti-FcgRII/III (Fc Block; clone 2.4G2), and PE-conjugated mouse anti- harvested by scraping and transferred to polystyrene staining tubes (VWR FcgRI (clone X54-5/7.1) and PE–rat anti-FcgRII/III as well as the isotype International, Mississauga, ON, Canada) on ice and stained with control PE–mouse IgG1k, PE–rat IgG2b, and PE–rat IgG2a were from BD allophycocyanin-conjugated anti-F4/80 (1 mg/ml). Phagocytosis was Biosciences (San Jose, CA). PE-conjugated rat anti-FcgRIII (clone evaluated by flow cytometry using the MACSQuant analyzer (Milteny 275003) and the rat IgG2a isotype control Abs were from R&D Systems Biotec, San Diego, CA), and data were analyzed using FlowJo software (Minneapolis, MN). Rat anti-CD44 (clone KM81), rat anti-F4/80 (clone + (FlowJo, Ashland, OR) by gating on the F4/80 population. To facilitate CI:A3-1), as well as rabbit anti-SRBC IgM were all from Cedarlane comparison, the experimental data were normalized to the mean fluores- Laboratories (Burlington, ON, Canada). Rat anti-CD44 (clone KM201) cence values of the control (untreated macrophages plus IgG-SRBCs) was from SouthernBiotech (Birmingham, AL), and PE-conjugated rat within the same experiment, which was designated as 100%. anti-F4/80 (clone BM8) was from eBioscience (San Diego, CA). Phagocytosis of CMFDA-labeled IgG-SRBCs by PerM was also eval- Allophycocyanin-conjugated rat anti-F4/80 was purchased from Bio- uated by confocal microscopy using the same conditions as for flow 9 Legend (San Diego, CA). Alexa Fluor 633–conjugated F(ab )2 goat anti- cytometry except that cells were cultured on 18-mm cover glasses. Fol- rabbit IgG was from Life Technologies (Carlsbad, CA). lowing incubation of IgG-SRBCs with PerM at 37˚C, cells were washed Cell culture with PBS to remove unbound RBCs and fixed with 4% PFA. Cells were then stained with Alexa Fluor 633–conjugated F(ab9)2 goat anti-rabbit IgG RAW264.7 (RAW) cells (American Type Culture Collection, Manassas, to label bound RBCs, and with Alexa Fluor 555–conjugated wheat germ VA) were maintained in RPMI 1640, 10% FBS, 4 mM L-glutamine, pen- agglutinin (Life Technologies) to label cell . Cover glasses icillin (100 U/ml), and streptomycin (100 mg/ml) (complete RPMI). Bone were then mounted on microscope slides using the ProLong Gold antifade The Journal of Immunology 3 reagent (Life Technologies). Fluorescent images were collected using the Zeiss CR3-mediated phagocytosis LSM 700 confocal microscope system (Carl Zeiss Canada, Toronto, ON, 3 8 Canada), equipped with 363/1.4 oil immersion objective lens, and standard For CR3-mediated phagocytosis, SRBCs (2 10 /ml) were opsonized first laser lines and filters. Zen Black software (Carl Zeiss) was used for image with polyclonal rabbit anti-SRBC IgM (32 ml IgM/ml SRBCs) in gelatin acquisition, and ImageJ2 software (38) was used for digital image processing. veronal buffer (Sigma-Aldrich, Saint Louis, MO) for 1 h at 37˚C with The ability of the macrophage to bind the opsonized SRBC target cell gentle agitation. SRBCs were then washed and resuspended in gelatin was assessed using RAW macrophages and IgG-SRBCs, which were veronal buffer followed by the addition of human C5-deficient serum cultured and prepared as described for FcgR-mediated phagocytosis. RAW (Sigma-Aldrich) at a final serum dilution of 1:12.5 and incubated for 20 macrophages were pretreated with CD44 mAbs and other inhibitors for 30 min at 37˚C to opsonize the SRBCs with C3bi. SRBCs were then washed, min at 37˚C, followed by washing and addition of IgG-SRBCs at 10:1 counted, and resuspended in RPMI 1640 at 108/ml. The opsonization with ratio (RBC/macrophage) for 30 min on ice. Subsequently, cells were C3bi was confirmed by flow cytometry using FITC-labeled mouse anti- washed at 4˚C to remove unbound RBCs and fixed in 4% PFA. Images of human C3/C3bi (Cedarlane Laboratories; data not shown). five fields from each sample were collected using the Nikon Eclipse Complete RPMI was removed from the adherent RAW macrophages inverted microscope, and quantification of the binding was done by (5 3 105/well) followed by the addition of 0.5 ml/well of 100 ng/ml PMA counting bound RBCs in 300–500 macrophages. The binding index (BI) (Sigma-Aldrich) for 30 min at 37˚C to activate CR3. The CD44 mAb IM7 was calculated using the following formula: BI = (total number of bound was then added (2 mg; 0.5 ml/well) into the PMA-containing wells RBCs/total number of counted macrophages) 3 100. for an additional incubation of 30 min at 37˚C. Subsequently, 100 ml

FIGURE 1. Inhibition of macrophage FcgR-mediated phagocytosis of RBCs by IM7. (A) Phagocytosis of IgG-SRBCs by RAW macrophages in the presence of rat IgG2b isotype control or IM7 (1 mg/ml). The arrows point to two engulfed RBCs in a macrophage. Representative images of .20 fields from four independent experiments are shown. Original magnification 340. (B and C) Dose response curve of IM7-mediated inhibition of phagocytosis of IgG-SRBCs by (B) RAW macrophages or (C) BMDM as analyzed by bright field microscopy (n = 5–10 per group from three to four independent ex- periments). (D) Dose response curve of IM7-mediated inhibition of phagocytosis of CMFDA-labeled IgG-SRBCs by PerM as analyzed by flow cytometry (n = 3 per group from three independent experiments). (E) Confocal microscopy images comparing phagocytosis of CMFDA-labeled IgG-SRBCs by PerM in presence of the rat IgG2b isotype control (1 mg/ml) or IM7 (1 mg/ml). Bound but external RBCs (red) were distinguished from internalized SRBCs

(green) using Alexa Fluor 633–conjugated F(ab9)2 goat anti-rabbit IgG as described in Materials and Methods. Cell membranes were labeled with Alexa Fluor 555–conjugated wheat germ agglutinin (yellow). Representative images of 20 fields from two independent experiments are shown. Scale bar, 10 mm. (F) Dose response curve of the inhibition of IgG-SRBC phagocytosis mediated by anti-F4/80 Ab and the rat IgG2b isotype control by RAW macrophages as analyzed by bright field microscopy (n = 8 per group from four independent experiments). Data are presented as mean 6 SEM. The p values indicate results of a two-way ANOVA. *p , 0.05, **p , 0.01, ***p , 0.001. PI, phagocytic index. 4 REGULATION OF MACROPHAGE PHAGOCYTOSIS BY ANTI-CD44

Table I. Properties of CD44 mAbs and their effects on FcgR-mediated phagocytosis

Ab Isotype Specificitya Binding to RAW Macrophagesd Inhibition of FcgR-Mediated Phagocytosise IM7 Rat IgG2b All isoformsb +++ Yes (p , 0.0001) KM114 Rat IgG1 All isoformsc +++ No (p = 0.55) KM201 Rat IgG1 All isoformsc +++ No (p = 0.6) KM81 Rat IgG2a All isoformsc ++ No (p = 0.38) aSpecificity is as described by Zheng et al. (41). bRecognizes a different epitope than KM114, KM201, and KM81. cAll recognize epitopes in the hyaluronan binding region. dFor binding to RAW macrophages (at 10 mg/ml Ab), see Supplemental Fig. 2A. eInhibition of FcgR-mediated phagocytosis was assessed in dose response experiments ranging from 0.01 to 10 mg/ml of the CD44 Ab being evaluated. The p values are shown for the 10 mg/ml dosage and are the results of a two-way ANOVA. ++, medium binding; +++, high binding.

C3bi-opsonized SRBCs (C3bi-SRBCs) were added to the RAW mac- cells as well as for contaminating undigested IgG by flow cytometry rophages at a 20:1 ratio (RBC/macrophage) for 30 min at 37˚C. Un- using PE-conjugated F(ab9)2 of goat anti-rat IgG [F(ab9)2 specific; bound SRBCs were then washed away, and bound RBCs were lysed Abnova, Cedarlane Laboratories] and FITC-conjugated F(ab9)2 of goat and removed with water for 90 s followed by fixation with 4% PFA. anti-rat IgG (Fc specific; Jackson ImmunoResearch Laboratories), Phagocytosis was assessed using the Nikon Eclipse inverted micro- respectively. scope, and quantification of phagocytosis was done as described for FcgR-mediated phagocytosis in RAW macrophages. CR3-dependent Detection of macrophage FcgRs phagocytosis was blocked with anti-CR3, but not blocked with anti- RAW macrophages cultured at 2 3 106/well in six-well plates were FcgRII/III Abs (Fc Block), confirming the validity and specificity of pretreated with CD44 mAbs or inhibitors for 30 min at 37˚C, followed by the assay (Supplemental Fig. 1A). washing with PBS and incubation with the viability dye eFluor 450 for 10 min at 4˚C. RAW cells were then fixed with 4% PFA, harvested by Dectin-1–mediated phagocytosis scraping, and transferred into staining tubes. FcgRs were detected Zymosan particles conjugated with the pH-sensitive fluorophore pHrodo using specific PE-labeled mAbs against FcgRI (clone X54-5/7.1), FcgRII/ Green (Life Technologies) were resuspended in RPMI 1640 at 0.5 mg/ml, III (clone 2.4G2), and FcgRIII (clone 275003) and the macrophage sonicated twice for 1 min at 30% amplitude, and used as phagocytic cell surface markers F4/80 and CD11b (controls). Cells were then ana- targets. RAW macrophages were cultured at 105/well in 24-well plates lyzed by flow cytometry and the effects of the different treatments on and pretreated with CD44 mAbs or inhibitors prior to the addition of FcgR expression were compared. 200 ml/well of 0.5 mg/ml zymosan particles (final ratio 10:1 [particle/ macrophage]). Phagocytosis was allowed to proceed for 2 h at 37˚C, Data analysis followed by washing, harvesting by scraping, and analysis of phago- Data are presented as mean 6 SEM unless otherwise stated. Statistical by flow cytometry. analysis of results was performed by ANOVA followed by Sidak or Phagocytosis of zymosan particles was also evaluated by confocal Tukey multiple comparisons tests using GraphPad Prism (version 6.02; microscopy. RAW macrophages were cultured on 18-mm cover glasses GraphPad Software, La Jolla, CA). A p value ,0.05 was considered (Fisher Scientific), and zymosan particles were prepared as described significant. above. Following 2 h of incubation of the zymosan particles with the RAW macrophages at 37˚C, the cells were washed and the cover glasses immediately transferred into a 37˚C and 5% CO2 incubation chamber and analyzed for fluorescence (live cell imaging; 20 min/sample) using the Zeiss LSM 700 system, and images were processed using the ImageJ2 software.

Deglycosylation of CD44 mAb Deglycosylation of mAb IM7 was performed under nondenaturing conditions by incubating the Ab with peptide-N-glycosidase (PNGase) F (10 U/1 mg Ab; New England Biolabs, Beverly, MA) for 24 h at 37˚C, followed by an additional 0.5 U PNGase F/1 mg Ab for an additional 8 h to ensure complete . PNGase F was removed by protein G– agarose chromatography (Life Technologies). The digested Ab was eluted from the resin with 0.1 M glycine (pH 2.5), neutralized with 1 M unbuffered Tris, concentrated, and applied onto an EndoTrap red column (Hyglos, Bernried, Germany) for removal of potential endotoxins. The effectiveness of PNGase F in deglycosylating the IM7 Ab was assessed by detecting a shift in the molecular mass of the H chain band, but not the L chain, on a Coomassie blue–stained gel as expected. Moreover, the loss of the full structure was verified using Sambucus nigra agglutinin lectin blot, which showed reactivity with undigested, but not the PNGase-digested, Ab (data not shown).

Preparation of F(ab9)2 fragments of anti-CD44 mAb

F(ab9)2 fragments of the IM7 mAb and control rat IgG2b were prepared as previously described (39) with minor modifications. Abs were digested with strain V8 endoproteinase Glu-C FIGURE 2. IM7 inhibits the binding of IgG-SRBCs to macrophages. (Sigma-Aldrich) at a 30:1 ratio of Ab to protease in reaction buffer (0.1 Binding of unopsonized or IgG-SRBCs to RAW macrophages in the M sodium phosphate, 2 mM EDTA [pH 7.8]) and overnight incubation presence versus absence of Ab 2.4G2 (Fc Block, 10 mg/ml), rat IgG2b (1 at 37˚C. Digestion was stopped by the addition of PMSF (final con- centration 25 mM; Sigma-Aldrich) prior to 5-fold dilution with PBS. mg/ml), or IM7 (1 mg/ml) as evaluated by bright field microscopy (n = 4–8 The digested material was passed over a protein G–agarose column per group from four independent experiments) is shown. Data are pre- (Life Technologies), dialyzed against PBS (pH 7.4), filtered, and stored sented as mean 6 SEM. The p value indicates results of a one-way , at 280˚C. The F(ab9)2 preparations were tested for binding to RAW ANOVA. ***p 0.001. The Journal of Immunology 5

FIGURE 3. Inhibition of phagocytosis by IM7 and FcgR expression. (A) Effect of IM7 versus KM114 on the expression of FcgRI, FcgRIII, FcgRII+III, or F4/80 by RAW macrophages. Macrophages were pretreated with 1 mg/ml of the indicated Ab (on the left of the panel), and the expression of FcgRs and F4/80 was assessed using specific PE-conjugated anti-FcgRI, -FcgRIIII, -FcgRII+III, and -F4/80 versus a corresponding matched PE-conjugated isotype control (shown in the bottom histogram, [Untreated+]). Representative histograms from three independent experiments are shown. (B) Effect of degly- cosylated IM7 (dotted lines) versus control rat IgG2b (solid lines) on the expression of FcgRI, FcgRIII, or FcgRII+III by RAW macrophages. Filled histograms show the matched PE-conjugated isotype control Abs. Representative histograms from three independent experiments are shown. (C) Dose response curve of deglycosylated IM7 (Deg-IM7) inhibiting the phagocytosis of IgG-SRBCs by RAW macrophages. Phagocytosis was evaluated as in Fig. 1B (n = 3–4 per group from three independent experiments). The p values indicate the results of a two-way ANOVA. (D) Binding of IgG-SRBCs to RAW macrophages pretreated with control rat IgG2b versus Deg-IM7 (1 mg/ml). The binding of SRBCs to macrophages was evaluated as described in

Fig. 2 (n = 6 per group from three independent experiments). (E) Dose response curve of the effect of F(ab9)2 of IM7 versus (Figure legend continues) 6 REGULATION OF MACROPHAGE PHAGOCYTOSIS BY ANTI-CD44

FIGURE 4. IM7 inhibits FcgR-mediated phagocytosis of prebound SRBCs. (A) Rep- resentative bright field images and (B) quantitative analysis of phagocytosis of prebound IgG-SRBCs by RAW macro- phages in the absence (untreated) or pres- ence of cytochalasin D (10 mg/ml), rat IgG2b (1 mg/ml), or IM7 (1 mg/ml). Origi- nal magnification 340 (n = 4–6 per group from three independent experiments). Data are presented as mean 6 SEM. The p values indicate results of a one-way ANOVA. ***p , 0.001.

Results absence of inhibition by these other Abs may be due to lower The CD44 mAb IM7, but not other CD44 mAbs, directly binding of macrophage CD44, we compared the binding of all the inhibits FcgR-mediated RBC phagocytosis CD44 mAbs to RAW macrophages. Although the relative binding of IM7 to RAW cells was better at lower doses, IM7, KM201, and The broad anti-inflammatory properties of the CD44 mAb IM7 KM114 all bound well at 10 mg/ml (Supplemental Fig. 2A). Be- prompted us to investigate whether this Ab inhibits macrophage cause no inhibitory effects were seen with KM201 and KM114 at phagocytosis. To assess the effect of IM7 on FcgR-mediated this 10 mg/ml dose, we speculate that the lack of inhibition by phagocytosis, we evaluated phagocytosis of IgG-SRBCs by these other CD44 Abs was not likely due to differences in Ab RAW macrophages by microscopy. IM7 pretreatment of mac- binding. rophages inhibited FcgR-mediated phagocytosis of RBCs by RAW cells in a dose-dependent manner, reaching complete in- IM7 can inhibit macrophage-target cell binding hibition at 1 mg/ml Ab (Fig. 1A, 1B). Similarly, pretreatment of To investigate the level of IM7 inhibition of the phagocytic process, BMDM with IM7 resulted in substantial inhibition of FcgR- we first examined whether IM7 inhibits the binding of IgG-SRBCs mediated phagocytosis, also at 1 mg/ml Ab (Fig. 1C). to macrophages. The FcgRII/III blocking Ab (2.4G2; Fc Block) Primary PerM treated with IM7 also underwent dose-dependent caused 80% inhibition of binding, whereas incubation of macro- phagocytic inhibition of CMFDA-labeled IgG-SRBCs, reaching phages with IM7 resulted in 92% inhibition of IgG-SRBC–mac- maximum inhibition (60%) by 1 mg/ml Ab (Fig. 1D). These rophage binding (Fig. 2), potentially suggesting that the inhibition findings with PerM were also confirmed by confocal microscopy of FcgR-mediated phagocytosis by this Ab could be explained (Fig. 1E), and the specificity of inhibition was confirmed using by inhibition of the binding of the phagocytic target to the anti-F4/80 and an isotype control Ab (Fig. 1F). The lesser inhi- macrophage. bition of phagocytosis in PerM was not due to differences in CD44 expression within the PerM population as assessed by flow Inhibition of phagocytosis by IM7 and FcgR expression cytometry (data not shown), but we speculate that it may be due to The inhibition of macrophage–target cell binding could occur at an the cellular heterogeneity of the PerM population (40), with IM7 early stage involving selective IM7-induced changes in FcgR affecting a distinct subset of PerM. FcgR-mediated phagocytosis expression. IM7 in fact partially decreased the expression of using CD44-deficient (CD442/2) BMDM was not affected FcgRI and FcgRIII on RAW macrophages (Fig. 3A). Although by IM7, suggesting that engagement of CD44 is essential for this decrease in FcgR expression could potentially explain some its inhibition (Supplemental Fig. 1B). The level of FcgR- level of inhibition of phagocytosis, the same effect was seen mediated phagocytosis in CD442/2 BMDM was slightly with KM114 (Fig. 3A), which did not inhibit RBC phagocytosis lower but comparable to wild-type BMDM (Fig. 1C versus (Table I). Interestingly, these effects on FcgRI and FcgRIII me- Supplemental Fig. 1B). diated by IM7 and KM114 were also observed in fixed cells (not The IM7 Ab has superior anti-inflammatory effects in arthritis as shown), suggesting that this biophysical phenomenon may not be compared with KM201 (20) and KM81 (21). It also successfully sufficient to explain the inhibition of phagocytosis. To help ad- inhibited the development of diabetes in NOD mice in comparison dress this issue and examine whether IM7 with impaired FcgR with another CD44 mAb (the IRAWB14.4), which did not (24). To binding properties could inhibit phagocytosis, we deglycosylated explore whether other selected CD44 mAbs can also inhibit FcgR- IM7 to impair its FcgR binding properties (42). The binding of the mediated phagocytosis of SRBCs, we examined the effects of the deglycosylated IM7 to RAW macrophages was comparable to that CD44 mAbs KM114, KM201, and KM81 on the phagocytosis of of IM7 at higher doses (10 mg/ml; Supplemental Fig. 2B), and the IgG-SRBCs by RAW macrophages. None of these Abs signifi- Ab did not decrease FcgR expression (Fig. 3B); however, it was cantly inhibited FcgR-dependent phagocytosis of SRBCs by RAW still able to significantly inhibit FcgR-mediated phagocytosis macrophages at any dose tested (Table I). To examine whether the (Fig. 3C) and binding (Fig. 3D). The deglycosylated IM7 did not

equimolar doses of F(ab9)2 of control rat IgG2b on phagocytosis of IgG-SRBCs by RAW macrophages (n = 3–6 per group from three independent ex- periments). Data are presented as mean 6 SEM. The p values indicate results of a one-way ANOVA. *p , 0.05, ***p , 0.001. N.T., not tested. The Journal of Immunology 7

FIGURE 5. IM7 inhibits CR3-mediated phago- cytosis. (A) Bright field microscopy images of CR3- mediated phagocytosis of C3bi-SRBCs by RAW macrophages in the presence of rat IgG2b versus IM7 (1 mg/ml). Representative images of .16 fields from four independent experiments are shown. Original magnification 340. (B) Dose response curve of IM7 inhibiting the phagocytosis of C3bi- SRBCs by RAW macrophages (n = 4 per group from four independent experiments). Data are pre- sented as mean 6 SEM. The p values indicate re- sults of a two-way ANOVA. **p , 0.01, ***p , 0.001.

completely inhibit phagocytosis; whether this effect is due to (Fig. 5A). This inhibition was also dose-dependent, reaching FcgR-dependent events, receptor–dependent events, or maximal inhibition (87%) at 1 mg/ml Ab (Fig. 5B). These data other interactions of the Fc is unknown. To further explore suggest that IM7 can inhibit phagocytosis mediated by this dif- the role of the Fc domain of IM7 in its inhibitory effect, we tested ferent opsonic macrophage phagocytic receptor. the capacity of F(ab9)2 fragments of IM7 to inhibit phagocytosis. The binding of F(ab9)2 of IM7 to RAW macrophages was com- IM7 does not affect zymosan phagocytosis parable to that of intact IM7 (Supplemental Fig. 2C), yet it did not Dectin-1 is a major nonopsonic pattern recognition receptor that affect FcgR-mediated phagocytosis in these cells even at higher can induce the phagocytosis of b-glucan–expressing yeast particles doses (66.67 nM; corresponding to 10 mg/ml intact IM7) of Ab (e.g., zymosan) by macrophages. To examine whether dectin-1– (Fig. 3E). Taken together, these results suggest that the Fc domain mediated phagocytosis is affected by the IM7 Ab, we assessed the of IM7 is necessary for inhibition of FcgR-mediated phagocytosis. uptake of pHrodo-labeled zymosan particles by RAW macro- IM7 retains the ability to inhibit FcgR-dependent phagocytosis phages incubated with IM7 or a control rat IgG2b Ab. The uptake even when the initial IgG–FcgR interaction step is bypassed of zymosan particles by RAW macrophages was affected by cy- tochalasin D (as expected) but was not affected by IM7, as To explore whether inhibition of macrophage–target cell binding is assessed by confocal microscopy (Fig. 6A). These findings were the only step affected by IM7, we examined the ability of IM7 to also confirmed by flow cytometric analysis (Fig. 6B, 6C). We bypass the initial engagement between the sensitizing IgG and conclude that IM7 does not interfere with this dectin-1–mediated macrophage FcgR. RAW macrophages were first incubated with phagocytic pathway in RAW macrophages. IgG-SRBCs for 30 min at 4˚C to allow FcgR-dependent target cell binding (37), followed by washing away the unbound RBCs. IM7 (1 mg/ml) was then added to RAW cells for 30 min at 4˚C, and the Discussion cells were washed and adjusted to 37˚C to allow phagocytosis to The CD44 mAb IM7 has been reported to exhibit robust anti- proceed. Bound but not internalized RBCs were then lysed with inflammatory effects in numerous experimental autoimmune and water as described in Materials and Methods. IM7 completely inflammatory diseases. Several mechanisms have been suggested inhibited phagocytosis of these already-bound IgG-SRBCs for its broad anti-inflammatory effects, including the induction of (Fig. 4). Collectively, these results suggest that IM7 interferes CD44 shedding, regulation of HA–CD44 interactions, and with FcgR-mediated phagocytosis at a multistep level. blocking of leukocyte infiltration to sites of inflammation, al- though the particular mechanisms underlying these effects are yet IM7 inhibits CR3-mediated phagocytosis to be fully understood (18–25). Our data demonstrate that IM7, To assess whether IM7 can affect other major phagocytic pathways, but not other tested CD44 mAbs, can directly inhibit macrophage we investigated whether IM7 can also affect CR3-mediated FcgR-mediated RBC phagocytosis, as well as binding of IgG- phagocytosis by RAW macrophages using C3bi-SRBCs as target RBCs to the macrophage. Moreover, IM7 also directly inhibits cells. IM7 dramatically inhibited CR3-mediated phagocytosis CR3-mediated RBC phagocytosis but did not affect dectin-1– 8 REGULATION OF MACROPHAGE PHAGOCYTOSIS BY ANTI-CD44

FIGURE 6. IM7 does not affect the classical dectin-1–mediated phagocytosis of zymosan particles. (A) Confocal fluorescent images comparing phagocytosis of pHrodo-labeled zymosan particles by RAW macrophages in the absence (untreated) or presence of cytochalasin D (10 mg/ml), rat IgG2b (1 mg/ml), or IM7 (1 mg/ml). Representative images of 10 fields from two independent experiments are shown. Scale bar, 10 mm. (B) Representative flow cytometric histograms for the phagocytosis of zymosan particles by RAW macrophages (untreated [ far left], cytochalasin D [middle left], rat IgG2b [middle right], and IM7 [ far right]). The numbers show the percentage of green pHrodo+ cells. (C) Quantitative analysis for the phagocytosis of zymosan particles by RAW macrophages as evaluated by flow cytometry (n = 9 per group from three independent experiments). Data are presented as mean 6 SEM. The p values indicate results of a one-way ANOVA. ***p , 0.001. mediated phagocytosis of zymosan. These findings suggest that in expression alone is not sufficient to explain the inhibitory effect addition to its previously reported inhibitory effects on phagocytic of IM7. Interestingly, significant yet partial inhibition of phago- cell infiltration into inflammatory sites (27), IM7 is able to se- cytosis and binding of IgG-SRBCs was also achieved with the lectively inhibit two different opsonic phagocytic mechanisms in deglycosylated IM7, wheras no inhibition was observed with F macrophages. (ab9)2 of IM7. These data suggest that the inhibitory effect of IM7 Abs against macrophage surface markers have been suggested on FcgR-mediated phagocytosis has at least some requirement for to inhibit FcgR-mediated phagocytosis by blockade of FcgRs the Fc portion of the Ab. Whether a specific interaction with one presumably via the formation of Ag–mAb–FcgR complexes on of the macrophage FcgRs or other receptors that can interact with the cell surface (43, 44). This could also occur with CD44 mAbs the Fc region may be involved in this inhibition is unknown, but it on the macrophage cell surface. In accordance with this, our data is worthy of further investigation. Interestingly, however, IM7 show that IM7 induced a partial decrease in FcgR expression by inhibited phagocytosis of IgG-SRBCs even when the initial RAW macrophages. However, because a similar decrease in FcgR binding step was bypassed. expression was also induced by the KM114 mAb, which did not The IM7-induced direct inhibition of macrophage phagocytosis inhibit RBC phagocytosis, this suggests that decreased FcgR mediated by FcgRs and CR3, but not by dectin-1, suggests that The Journal of Immunology 9

IM7 could be selectively interfering with signaling components opsonized RBCs mediated through FcgRs and CR3 and that this that are required for macrophage FcgR- and CR3-mediated inhibition is dependent on the Fc domain of the Ab and can occur phagocytosis but not for the dectin-1 pathway. Despite that at more than one stage of the phagocytic process. The ability to phagocytosis mediated by these three phagocytic receptors is selectively inhibit FcgR- and CR3-mediated phagocytosis by IM7 thought to share many signaling and adaptor proteins, the has important implications in the development of mAb-based signaling elicited by these receptors is different. During FcgR- therapeutics against immune-mediated diseases. mediated phagocytosis, clustered FcgRs trigger activation of tyrosine kinases, including Src family kinases, which phosphor- Disclosures ylate the ITAM on the FcgR or on the associated FcR g subunit. A.H.L. has patents on the use of mAbs to treat ITP. The other authors have The phosphorylation of ITAM leads to recruitment of the spleen no financial conflicts of interest. tyrosine kinase Syk as well as additional adaptor proteins, and these resulting signaling platforms activate multiple lipid modifi- cation as well as guanine nucleotide exchange factors for References 1. Gallatin, W. M., I. L. Weissman, and E. C. Butcher. 1983. A cell-surface mol- small GTPases of the Rho family. The activation of the Rho ecule involved in organ-specific homing of lymphocytes. Nature 304: 30–34. GTPases Cdc42, Rac 1 and Rac 2, and RhoG activates the 2. Screaton, G. R., M. V. Bell, D. G. Jackson, F. B. Cornelis, U. Gerth, and nucleation complex Arp2/3, which in turn elicits actin polymeri- J. I. Bell. 1992. Genomic structure of DNA encoding the lymphocyte homing receptor CD44 reveals at least 12 alternatively spliced exons. Proc. Natl. Acad. zation that drives pseudopod extension during the phagocytic Sci. USA 89: 12160–12164. process (16). Growing evidence also indicates interactions be- 3. Zo¨ller, M. 2011. CD44: can a cancer-initiating cell profit from an abundantly tween the FcgRs and CR3 phagocytic systems. The presence of expressed molecule? Nat. Rev. Cancer 11: 254–267. 4. Naor, D., S. B. Wallach-Dayan, M. A. Zahalka, and R. V. Sionov. 2008. In- complement component 3 was shown to significantly reduce the volvement of CD44, a molecule with a thousand faces, in cancer dissemination. amount of bound IgG required to induce FcgR-mediated phago- Semin. Cancer Biol. 18: 260–267. 5. DeGrendele, H. C., P. Estess, and M. H. Siegelman. 1997. Requirement for CD44 cytosis by (45), and blocking CR3 function using anti- in activated T cell extravasation into an inflammatory site. Science 278: 672–675. CR3 Abs reduces FcgR-mediated phagocytosis without affecting 6. Wang, Q., P. Teder, N. P. Judd, P. W. Noble, and C. M. Doerschuk. 2002. CD44 binding of FcgRs to IgG (46). Binding of IgG to FcgRs was found deficiency leads to enhanced migration and injury in pneumonia in mice. Am. J. Pathol. 161: 2219–2228. to induce the lateral movement and accumulation of CR3 to FcgR- 7. Mikecz, K., F. R. Brennan, J. H. Kim, and T. T. Glant. 1995. Anti-CD44 treat- containing phagocytic cups and enhance C3bi binding to CR3 ment abrogates tissue oedema and leukocyte infiltration in murine arthritis. Nat. (47). Moreover, stimulation of individual FcgRs can differently Med. 1: 558–563. 8. Galandrini, R., N. Albi, G. Tripodi, D. Zarcone, A. Terenzi, A. Moretta, modulate CR3-mediated phagocytosis (48). Despite that the cy- C. E. Grossi, and A. Velardi. 1993. to CD44 trigger effector functions tosolic domain of dectin-1 bears an ITAM-like motif, it was shown of human T cell clones. J. Immunol. 150: 4225–4235. that at least in macrophages, dectin-1 seems to use different cel- 9. Firan, M., S. Dhillon, P. Estess, and M. H. Siegelman. 2006. Suppressor activity and potency among regulatory T cells is discriminated by functionally active lular mechanisms for phagocytosis of zymosan particles, in which CD44. Blood 107: 619–627. Syk, PI3K, and the Rho GTPases Cdc42 and Rac-1 do not play an 10. Liu, T., L. Soong, G. Liu, R. Ko¨nig, and A. K. Chopra. 2009. CD44 expression positively correlates with Foxp3 expression and suppressive function of CD4+ essential role (49). Treg cells. Biol. Direct 4: 40. The absence of inhibition of phagocytosis by the KM114, 11. van der Windt, G. J., C. van ’t Veer, S. Florquin, and T. van der Poll. 2010. CD44 KM201, and KM81 mAbs was not due to lower binding of deficiency is associated with enhanced Escherichia coli-induced proin- flammatory cytokine and chemokine release by peritoneal macrophages. Infect. macrophage CD44 and is not likely to be related to the isotype of Immun. 78: 115–124. the Ab because the rat IgG2b anti-F4/80 Ab did not inhibit 12. Kawana, H., H. Karaki, M. Higashi, M. Miyazaki, F. Hilberg, M. Kitagawa, and phagocytosis. However, the four CD44 mAbs tested recognize K. Harigaya. 2008. CD44 suppresses TLR-mediated inflammation. J. Immunol. 180: 4235–4245. different epitopes on the CD44s molecule. Although KM114, 13. Liang, J., D. Jiang, J. Griffith, S. Yu, J. Fan, X. Zhao, R. Bucala, and P. W. Noble. KM201, and KM81 are known to bind to epitopes in the HA 2007. CD44 is a negative regulator of acute pulmonary inflammation and -TLR signaling in mouse macrophages. J. Immunol. 178: binding site and to interfere with HA binding by CD44, the epi- 2469–2475. tope recognized by IM7 is located outside of the HA binding site, 14. Stuart, L. M., and R. A. Ezekowitz. 2005. Phagocytosis: elegant complexity. and this Ab does not directly block HA binding (41). However, that Immunity 22: 539–550. 15. Elliott, M. R., and K. S. Ravichandran. 2010. Clearance of apoptotic cells: F(ab9)2 of IM7 did not inhibit phagocytosis suggests that the in- implications in health and disease. J. Cell Biol. 189: 1059–1070. hibitory effect of IM7 is not likely to be related to its specific 16. Flannagan, R. S., V. Jaumouille´, and S. Grinstein. 2012. The cell biology of epitope on the CD44 Ag. phagocytosis. Annu. Rev. Pathol. 7: 61–98. 17. Brown, G. D., P. R. Taylor, D. M. Reid, J. A. Willment, D. L. Williams, Being such a potent inhibitor of FcgR-mediated phagocytosis L. Martinez-Pomares, S. Y. Wong, and S. Gordon. 2002. Dectin-1 is a major in vitro, it would be assumed that IM7 will also be able to block b-glucan receptor on macrophages. J. Exp. Med. 196: 407–412. this process in macrophages in vivo. However, we have previously 18. Brocke, S., C. Piercy, L. Steinman, I. L. Weissman, and T. Veromaa. 1999. Antibodies to CD44 and integrin a4, but not L-selectin, prevent central nervous shown that IM7 did not improve counts in a passive model system inflammation and experimental encephalomyelitis by blocking secondary of immune thrombocytopenia (ITP) (50). These seemingly con- leukocyte recruitment. Proc. Natl. Acad. Sci. USA 96: 6896–6901. 19. Brennan, F. R., J. K. O’Neill, S. J. Allen, C. Butter, G. Nuki, and D. Baker. 1999. tradictory findings are not due to a lack of anti-inflammatory ac- CD44 is involved in selective leucocyte extravasation during inflammatory tivity of IM7 in ITP. Deglycosylated IM7 (which more minimally central nervous system disease. Immunology 98: 427–435. depletes ) does have anti-inflammatory effects in ITP (A. 20. Mikecz, K., K. Dennis, M. Shi, and J. H. Kim. 1999. Modulation of hyaluronan receptor(CD44)functioninvivoinamurine model of rheumatoid arthritis. Arthritis Amash and A.H. Lazarus, unpublished data). More interestingly, Rheum. 42: 659–668. we have also shown that other CD44 mAbs such as KM114, 21. Nedvetzki, S., M. Walmsley, E. Alpert, R. O. Williams, M. Feldmann, and KM81, and KM201 are able to ameliorate ITP (50). The lack of D. Naor. 1999. CD44 involvement in experimental collagen-induced arthritis (CIA). J. Autoimmun. 13: 39–47. direct inhibition of phagocytosis with these other Abs that ame- 22. Mott, P. J., and A. H. Lazarus. 2013. CD44 antibodies and immune thrombocy- liorate ITP demonstrates that these Abs may work by an indirect topenia in the amelioration of murine inflammatory arthritis. PLoS One 8: e65805. 23. Huta´s, G., E. Bajnok, I. Ga´l, A. Finnegan, T. T. Glant, and K. Mikecz. 2008. mechanism in ITP, or that platelet phagocytosis is distinct from CD44-specific antibody treatment and CD44 deficiency exert distinct effects on RBC phagocytosis. leukocyte recruitment in experimental arthritis. Blood 112: 4999–5006. In summary, we have provided evidence suggesting that the anti- 24. Weiss, L., S. Slavin, S. Reich, P. Cohen, S. Shuster, R. Stern, E. Kaganovsky, E. Okon, A. M. Rubinstein, and D. Naor. 2000. Induction of resistance to dia- inflammatory CD44 mAb IM7, but not other tested CD44 mAbs, betes in non-obese diabetic mice by targeting CD44 with a specific monoclonal can selectively inhibit the macrophage phagocytic function of antibody. Proc. Natl. Acad. Sci. USA 97: 285–290. 10 REGULATION OF MACROPHAGE PHAGOCYTOSIS BY ANTI-CD44

25. Li, C. R., E. E. Mueller, and L. M. Bradley. 2015. Targeting CD44 augments the 39. Rousseaux, J., R. Rousseaux-Pre´vost, H. Bazin, and G. Biserte. 1983. efficacy of Tregs in autoimmune diabetes. Immunol. Lett. 163: 199–205. of rat IgG subclasses by Staphylococcus aureus V8 proteinase. Biochim. Bio- 26. Xu, H., A. Manivannan, J. Liversidge, P. F. Sharp, J. V. Forrester, and I. J. Crane. phys. Acta 748: 205–212. 2002. Involvement of CD44 in leukocyte trafficking at the blood-retinal barrier. 40. Ghosn, E. E., A. A. Cassado, G. R. Govoni, T. Fukuhara, Y. Yang, J. Leukoc. Biol. 72: 1133–1141. D. M. Monack, K. R. Bortoluci, S. R. Almeida, L. A. Herzenberg, and 27. Xu, H., A. Manivannan, I. Crane, R. Dawson, and J. Liversidge. 2008. Critical L. A. Herzenberg. 2010. Two physically, functionally, and developmentally but divergent roles for CD62L and CD44 in directing blood trafficking distinct peritoneal macrophage subsets. Proc. Natl. Acad. Sci. USA 107: 2568– in vivo during inflammation. Blood 112: 1166–1174. 2573. 28. Wittig, B., C. Schwa¨rzler, N. Fo¨hr, U. Gunthert,€ and M. Zo¨ller. 1998. Curative 41. Zheng, Z., S. Katoh, Q. He, K. Oritani, K. Miyake, J. Lesley, R. Hyman, treatment of an experimentally induced colitis by a CD44 variant V7-specific A. Hamik, R. M. Parkhouse, A. G. Farr, and P. W. Kincade. 1995. Monoclonal antibody. J. Immunol. 161: 1069–1073. antibodies to CD44 and their influence on hyaluronan recognition. J. Cell Biol. 29. Katoh, S., N. Matsumoto, K. Kawakita, A. Tominaga, P. W. Kincade, and 130: 485–495. S. Matsukura. 2003. A role for CD44 in an -induced murine model of 42. Goulabchand, R., T. Vincent, F. Batteux, J. F. Eliaou, and P. Guilpain. 2014. pulmonary eosinophilia. J. Clin. Invest. 111: 1563–1570. Impact of autoantibody glycosylation in autoimmune diseases. Autoimmun. Rev. 30. Camp, R. L., A. Scheynius, C. Johansson, and E. Pure´. 1993. CD44 is necessary 13: 742–750. for optimal contact allergic responses but is not required for normal leukocyte 43. Kurlander, R. J. 1983. Blockade of -mediated binding to U-937 cells extravasation. J. Exp. Med. 178: 497–507. by murine monoclonal antibodies directed against a variety of surface . 31. Khaldoyanidi, S., S. Karakhanova, J. Sleeman, P. Herrlich, and H. Ponta. 2002. J. Immunol. 131: 140–147. CD44 variant-specific antibodies trigger hemopoiesis by selective release of 44. Shepard, S. L., and A. G. Hadley. 1997. Monocyte-bound monoclonal antibodies cytokines from bone marrow macrophages. Blood 99: 3955–3961. inhibit the FcgRI-mediated phagocytosis of sensitized red cells: the efficiency 32. Miyake, K., K. L. Medina, S. Hayashi, S. Ono, T. Hamaoka, and P. W. Kincade. and mechanism of inhibition are determined by the nature of the antigen. Im- 1990. Monoclonal antibodies to Pgp-1/CD44 block lympho-hemopoiesis in munology 90: 314–322. long-term bone marrow cultures. J. Exp. Med. 171: 477–488. 45. Schreiber, A. D., J. Parsons, P. McDermott, and R. A. Cooper. 1975. Effect of 33. Khaldoyanidi, S., A. Denzel, and M. Zo¨ller. 1996. Requirement for CD44 in corticosteroids on the human monocyte IgG and complement receptors. J. Clin. proliferation and homing of hematopoietic precursor cells. J. Leukoc. Biol. 60: Invest. 56: 1189–1197. 579–592. 46. Brown, E. J., J. F. Bohnsack, and H. D. Gresham. 1988. Mechanism of inhibition 34. Fedorchenko, O., M. Stiefelhagen, A. A. Peer-Zada, R. Barthel, P. Mayer, of -mediated phagocytosis by monoclonal antibodies that L. Eckei, A. Breuer, G. Crispatzu, N. Rosen, T. Landwehr, et al. 2013. CD44 recognize the Mac-1 antigen. J. Clin. Invest. 81: 365–375. regulates the apoptotic response and promotes disease development in chronic 47. Jongstra-Bilen, J., R. Harrison, and S. Grinstein. 2003. Fcg-receptors induce lymphocytic leukemia. Blood 121: 4126–4136. Mac-1 (CD11b/CD18) mobilization and accumulation in the phagocytic cup for 35. Rajasagi, M., A. von Au, R. Singh, N. Hartmann, M. Zo¨ller, and R. Marhaba. optimal phagocytosis. J. Biol. Chem. 278: 45720–45729. 2010. Anti-CD44 induces in T lymphoma via mitochondrial depo- 48. Huang, Z. Y., S. Hunter, P. Chien, M. K. Kim, T. H. Han-Kim, Z. K. Indik, and larization. J. Cell. Mol. Med. 14(6B): 1453–1467. A. D. Schreiber. 2011. Interaction of two phagocytic host defense systems: Fcg 36. Zhang, X., R. Goncalves, and D. M. Mosser. 2008. The isolation and charac- receptors and complement receptor 3. J. Biol. Chem. 286: 160–168. terization of murine macrophages. Curr. Protoc. Immunol. Chapter 14: Unit 14.1. 49. Herre, J., A. S. Marshall, E. Caron, A. D. Edwards, D. L. Williams, doi:10.1002/0471142735.im1401s83. E. Schweighoffer, V. Tybulewicz, C. Reis e Sousa, S. Gordon, and G. D. Brown. 37. Chow, C. W., G. P. Downey, and S. Grinstein. 2004. Measurements of phago- 2004. Dectin-1 uses novel mechanisms for yeast phagocytosis in macrophages. cytosis and phagosomal maturation. Curr. Protoc. Cell Biol. Chapter 15: Unit Blood 104: 4038–4045. 15.7. doi:10.1002/0471143030.cb1507s22. 50. Crow, A. R., S. Song, S. J. Suppa, S. Ma, M. P. Reilly, P. Andre, S. E. McKenzie, 38. Schneider, C. A., W. S. Rasband, and K. W. Eliceiri. 2012. NIH Image to and A. H. Lazarus. 2011. Amelioration of murine immune thrombocytopenia by ImageJ: 25 years of image analysis. Nat. Methods 9: 671–675. CD44 antibodies: a potential therapy for ITP? Blood 117: 971–974.