M. tuberculosis-initiated human mannose receptor signaling temporally regulates macrophage recognition and vesicle trafficking by FcRγ-chain, Grb2 and SHP-1

Summary

Despite its prominent role as C-type lectin (CTL) pattern recognition receptor, mannose receptor (MR, CD206)-specific signaling molecules and pathways are unknown. The MR is highly expressed on human macrophages, regulating endocytosis, phagocytosis and immune responses, and mediating Mycobacterium tuberculosis (M.tb) phagocytosis by human macrophages thereby limiting phagosome-lysosome (P-L) fusion. We identified human MR-associated proteins using phosphorylated and non-phosphorylated MR cytoplasmic tail peptides. We found that MR binds FcRγ-chain, which is required for MR plasma membrane localization and M.tb cell association. Additionally, we discovered that MR-mediated M.tb association triggers immediate MR tyrosine residue phosphorylation and Grb2 recruitment, activating the Rac/Pak/Cdc-42 signaling cascade important for M.tb uptake. MR activation subsequently recruits SHP-1 to the M.tb-containing phagosome, where its activity limits PI(3)P generation at the phagosome and M.tb P-L fusion, and promotes M.tb growth. In sum, we identify human MR signaling pathways that temporally regulate phagocytosis and P-L fusion during M.tb infection.

eTOC Blurb

The human mannose receptor (MR) mediates macrophage phagocytosis and immune regulation. MR-specific signaling remains a major gap in the field. Rajaram et al. identify the importance of FcRγ-chain and Grb2 during MR-mediated phagocytosis and subsequent MR-dependent recruitment of SHP-1 to the M.tb phagosome, thereby limiting PI(3)P generation and phagosome-lysosome fusion.

Introduction

The macrophage mannose receptor (CD206; MRC1) was discovered in 1978 (Stahl et al., 1978; Stahl and Gordon, 1982) and cloned in 1990 (Ezekowitz et al., 1990). It is part of a large family of C-type lectin (CTL) receptors, which can cross-talk with other pattern recognition receptor (PRRs) (Goyal et al., 2016). CD206 is a calcium-dependent type I transmembrane glycoprotein that contains an extracellular N-terminal cysteine-rich (CR) domain, a fibronectin II (FNII) domain, eight carbohydrate recognition domains (CRDs), a transmembrane domain, and a cytoplasmic tail (Stahl et al., 1980). The MR cytoplasmic tail is 49 amino acids long and does not contain an ITAM or ITIM motif, but does have a tyrosine residue at the 18^th position that is critical for mediating endocytosis (Schweizer et al., 2000). Early work indicated that this receptor is recycled (Stahl et al., 1980) and focused on its endocytic and phagocytic functions following engagement of specific carbohydrate motifs, i.e. mannose, fucose and N-acetyl-glucosamine (Shepherd et al., 1982). The MR is currently recognized as a prototypic PRR expressed on macrophages, some dendritic cells (Sallusto et al., 1995) and other cell types (Stahl and Ezekowitz, 1998), orchestrating innate and adaptive immune communication (He et al., 2007), and is emerging as a critical regulator of health and disease (Lee et al., 2002; Zhang et al., 2012; Hattori et al., 2010; Hattori et al., 2009). Despite its well-known central role in host recognition of a variety of microbes and ability to modulate inflammatory responses, its receptor-specific role in transducing signals to regulate phagocytosis and inflammatory pathways remains in question (Le Cabec et al., 2005; Gazi and Martinez-Pomares, 2009; Goyal et al., 2016; Hoving et al., 2014; Martinez-Pomares, 2012; Sancho and Reis E Sousa, 2012, Taylor et al., 2005; Gordon, 2016).

The majority of CTL receptors signal through associated adaptor proteins. Recent studies have indicated that MR binds heat shock proteins (HSPs) in unstimulated cells (Yang et al., 2013), but it is unknown how this changes following MR activation, and what role HSPs and other binding partners have on MR-mediated signalling. Of the identified CTL adaptor proteins, FcRγ-chain is the most common and its association is based on amino acid charge (Kanazawa et al., 2003; Yamasaki et al., 2008; Miyake et al., 2013). The adaptor molecule Grb2 plays an essential role in orchestrating actin remodeling during phagocytosis. Phagocytosis is an actin-dependent process mediated by receptor clustering, receptor tyrosine phosphorylation (Lowell, 2011), recruitment of various adaptor proteins and downstream activation of the Rho family of small GTPases which lead to actin cytoskeleton remodeling (May and Machesky, 2001) and cellular immune functions. MR-dependent phagocytosis is actin-dependent (Kang and Schlesinger, 1998; Zhang et al., 2005), although MR does not co-localize with actin in unstimulated cells (Kato et al., 2007). The interactions of FcRγ-chain and Grb2 with the MR and their functional relevance to pathogen phagocytosis and trafficking are not known.

MR plays an important role in the phagocytosis of virulent strains of Mycobacterium tuberculosis (M.tb) by human macrophages (Schlesinger, 1993), a pathway that is particularly important in lung alveoli since alveolar macrophages express abundant MR (Stephenson and Shepherd, 1987). MR engagement by M.tb and its cell wall component mannose-capped lipoarabinomannan (ManLAM) limits phagosome-lysosome (P-L) fusion (Kang et al., 2005), up-regulates the expression and activity of the ligand-activated nuclear receptor peroxisome proliferator-activated receptor gamma (PPARγ) (Rajaram et al., 2010), and regulates immune responses, including those to mycobacteria (Pathak et al., 2005; Nigou et al., 2001). Importantly, MR blockade or knockdown significantly reduces the phagocytosis of M.tb (Kang and Schlesinger, 1998) and M. avium by human macrophages (Sweet et al., 2010) and increases P-L fusion (Kang et al., 2005; Sweet et al., 2010). MR activation by its ligand is also linked to the production of anti-inflammatory cytokines (Chieppa et al., 2003), and inhibition of IL-12 (Nigou et al., 2001), TNF-α, IL-1β, IL-6, and IL-23 (Zenaro et al., 2009), and reactive oxygen species (ROS) production (Astarie-Dequeker et al., 1999). Elucidation of MR-mediated signaling pathways will fill an essential gap in our knowledge for tuberculosis (TB) pathogenesis and macrophage responses in general.

In this study we used a biochemical approach to identify human MR-interacting proteins that could define specific signalling cascades for phagocytosis, subsequent vesicle trafficking and immune regulation. We identified MR-associated proteins by mass spectrometry and Western blot using phosphorylated and non-phosphorylated MR cytoplasmic tail peptides to capture MR interacting proteins from macrophage lysates. The proteins identified (FcRγ-chain, Grb2, SHP-1) were confirmed by functional assays and demonstrate spatial and temporal biochemical signaling events that accompany MR activation. These pathways explain at a molecular level how MR interaction with its ligands leads to specific roles in phagocytosis (FcRγ-chain and Grb2) and immunoregulatory events (SHP-1) in human macrophages.

Results

Human macrophage MR expression and tyrosine phosphorylation by M.tb or ManLAM

Macrophages from different mammalian species express several types of CTLs (Robinson et al., 2006). The M.tb outer surface contains abundant mannosylated molecules, especially mannose-capped LAM which serves as a ligand for the MR on human macrophages and for Dendritic Cell-Specific Intercellular adhesion molecule-3-Grabbing Non-integrin (DC-SIGN) on human dendritic cells (Tailleux et al., 2003; Schlesinger et al., 1994; Geijtenbeek et al., 2003). Thus we determined the baseline expression of CTLs on human monocyte-derived macrophages (hMDMs) by flow cytometry. Results show that 85.13 ± 5.936% hMDMs are positive for MR (CD206) and 19.76 ±1.694 % are positive for Dectin-1 (mean ± SEM, N = 3), whereas DC-SIGN expression is more limited, and Dectin-2 and Mincle are not expressed (Figures 1A-F), in contrast to mouse macrophages that express Dectin-2, several isoforms of DC-SIGN and Mincle (McGreal et al., 2004; Yamasaki et al., 2008; Yonekawa et al., 2014). In support of these data we found that hMDMs express high levels of MR and lesser levels of Dectin-1 by confocal microscopy (Figure 1G). In addition, mRNA levels of CD206 are higher than those of the other C-type lectins (Figure 1H).

Figure 1. CTL expression by human macrophages.

Day 5 PBMCs were immunostained for (A) MR/CD206, (B) Dectin-1, (C) Mincle, (D) Dectin-2, or (E) DC-SIGN/CD209. hMDMs were gated by forward and side scatter and surface expression on non-permeabilized hMDMs was determined by flow cytometry. Flow cytometry data were analyzed using FlowJo software. The specific MFI (F) (mean ± SEM, N=3) was calculated by subtracting the MFI for the respective isotype control Abs (MFI for isotype Abs are: CD206=363; Dectin-1=190; Mincle=314; Dectin-2=72.1; DC-SIGN=190). CTL expression in non-permeabilized MDMs was examined by confocal microscopy (G) and qRT-PCR (H). The data shown in H is a representative experiment (mean ± SD) of 3 independent experiments using 3 different donors. Scale bar = 10 µm.

Phosphorylation of a tyrosine residue in the MR cytoplasmic tail is essential for its endocytic function (Schweizer et al., 2000). Incubation of hMDMs with M.tb or ManLAM led to phosphorylation of the tyrosine residue in the MR (Figures 2A-B), this was confirmed by immunoprecipitating (IP’ing) lysates with phosphor-tyrosine Abs, and detecting MR (Fig 2C). The MR does not contain an ITAM-like motif but does contain a tyrosine residue in the cytoplasmic tail that is important for endocytosis. Auto- or Src kinase-mediated phosphorylation of this tyrosine residue may occur with MR clustering similar to other CTLs like CLEC2 and CLEC9A (Mocsai et al., 2010). We found that Syk, a non-receptor tyrosine kinase, is phosphorylated early upon MR activation and co-localizes with the MR (Figure S1). Thus is it possible that Syk is activated by a recruited Src kinase or alternatively by binding to FcRγ-chain which contains an ITAM motif.

Figure 2. MR-interacting proteins in human macrophages upon activation.

(A-B) hMDMs were incubated with M.tb (MOI 20:1) by synchronized phagocytosis (A) or ManLAM (5µg/mL) (B). Cells were lysed at the indicated times and cell lysates were analyzed by WB using a phosphor-tyrosine Ab (4G10) and re-probed with anti-MR Ab. WBs shown are representative experiments (N=3, using 3 different donors). The graphs shown in lower panels are cumulative data of band intensities of phosphorylated MR (normalized to total MR) from the 3 donors (mean ± SEM). (C) hMDMs were incubated with M.tb ( MOI 20:1) and after the indicated times (min), cells were lysed and IP’d with phosphor-tyrosine Ab (4G10) followed by immunoblot with anti-MR Ab. Arrow shows the presence of MR in the immunoprecipitate. N=2 (D) Synthetic peptides corresponding to 28 amino acids of the MR cytoplasmic tail with (MRCTP, peptide 2) or without (MRCT, peptide 1) phosphorylation of the tyrosine residue at the 18^th position were N-terminal tagged with biotin. (E) hMDM lysates were mixed with MR peptides, pulled down with streptavidin beads and separated by gradient SDS-PAGE. The protein bands were visualized with Gel code blue. The boxes shown depict the locations where the lanes were cut and analyzed by mass spectrometry. (F) The identified proteins were compared between MRCT and MRCTP groups. (G) Selected proteins associated with human MR cytoplasmic peptides. (H) hMDM lysates were mixed with MR peptides, pulled down with streptavidin beads and binding of Grb2 (MW = 25.3 kDa), SHP-1 (MW = 67.9 kDa) and FcRγ-chain (MW = 9.7 kDa) was determined by WB. N=3. See also Figure S1.

Identification of MR-interacting proteins

Understanding of MR-mediated signaling pathways and associated signaling molecules following phosphorylation is limited, likely due to the difficulties in immunoprecipitating the MR. We therefore undertook a biochemical approach to identify MR-interacting proteins. We synthesized biotin-conjugated peptides of the MR cytoplasmic tail, which include the tyrosine residue at the 18^th intracellular position, with (MRCTP) or without phosphorylation (MRCT) (Figure 2D). Incubation of the MRCT or MRCTP peptides with hMDM cell lysates and recovery of the peptides by streptavidin bead affinity chromatography reveals a different pattern and intensity of peptide-associated proteins as observed by SDS-PAGE (Figure 2E). Protein bands in segments of lanes were digested with trypsin and analyzed by LC-MS/MS Orbitrap mass spectrometry (depicted as boxes around segments of the lanes in Figure 2E). The amino acid sequences were analyzed using the MASCOT data base, and the identified proteins were compared (Tables S1–S3). This procedure identified 103 proteins associated with MRCTP and 76 proteins associated with MRCT (after removal of keratin peptides); 29/76 and 56/103 were unique to the MRCT and MRCTP groups, respectively, and 47 proteins were common to both groups (Figure 2F). This approach identified heat shock proteins binding to the MR, including Grp 75, similar to previously reported work indicating that HSPs bind unstimulated MR (Yang et al., 2013). We selected three proteins for further validation and characterization based on their potential roles in endocytosis and cell signaling (Figure 2G): Grb2 and SHP-1 that bound specifically to MRCTP and FcRγ-chain which bound to both MRCTP and MRCT peptides. Interaction of these proteins with MRCT and/or MRCTP was confirmed by Western blot (WB) (Figure 2H). None of these proteins were pulled down using streptavidin beads without peptides (data not shown). Thus, our biochemical approach identified several potential MR-interacting proteins, either constitutively bound (FcRγ-chain) or recruited following tyrosine phosphorylation (SHP-1 and Grb2), which may be involved in downstream transmission of signals for functions such as phagocytosis and immune responses to pathogens.

FcRγ-chain is required for MR cell surface expression on human macrophages

Since the ITAM – containing adaptor molecule FcRγ-chain was constitutively bound to the MR cytoplasmic peptides, we examined whether FcRγ-chain interacts with the MR in human macrophages by performing co-immunoprecipitation (IP) of FcRγ-chain. Consistent with the peptide pull down assay, MR co-precipitated with FcRγ-chain constitutively (Figure 3A). This binding likely occurs at the two arginine (RR) residues in the interface of the transmembrane and cytoplasmic tail regions of the MR, since FcRγ-chain binds other CTLs through interactions with positively charged amino acids located in the transmembrane domain (Robinson et al., 2006; Yamasaki et al., 2008).

Figure 3. FcRγ-chain is required for MR cell surface expression on human macrophages.

(A) hMDM lysates were incubated with anti FcRγ-chain Ab or isotype control IgG Ab, immunoprecipitated, proteins resolved by gradient SDS-PAGE, and then immunoblotted with human MR Ab followed by anti FcRγ-chain Ab. The data shown are representative of 3 experiments. (B) Scramble control or FcRγ-chain siRNA transfected hMDM monolayers were fixed with/without permeabilization and stained with anti-MR Ab or isotype control Ab. Representative images (n=2) are shown (upper panels). Scale bar = 10 µm. MR surface expression and total MR levels were determined by MFI (lower panels). (C) Transfected hMDMs were incubated with mCherry-M.tb (MOI 10:1), fixed and M.tb association was examined by fluorescence microscopy. Approximately 100 macrophages were counted for each experiment. (D) Control or MR siRNA transfected hMDMs were incubated with mCherry-M.tb (MOI 10:1) for 2h or (E) with mannose BSA beads (MOI 20:1) for 1h. Cells were fixed and M.tb association or percent bead uptake (uptake index) was determined by counting 100 and 150 cells, respectively. Graphs shown are cumulative data (mean ± SEM) from two independent donors. * p< 0.05, ** p< 0.005. MR knockdown was confirmed in each experiment by WB using MR Ab followed by actin Ab (Fig S2C & G).

Since FcRγ-chain mediates cell surface expression of other CTLs (Miyake et al. 2013; Yamasaki et al., 2008), we asked whether FcRγ-chain is essential for MR surface expression. We knocked down the FcRγ-chain in hMDMs using siRNA (Figure S2A & E) and examined MR localization in non-permeabilized and permeabilized cells by confocal microscopy. FcRγ-chain knockdown significantly reduced surface expression of MR on hMDMs without affecting total cellular MR (Figure 3B). FcRγ-chain knockdown specifically affected MR surface expression since surface expression of CD11b was unaffected (Figure S2A, E, & I). Since M.tb binds to the MR on hMDMs (Schlesinger, 1993), we examined M.tb cell association with FcRγ-chain knockdown macrophages. FcRγ-chain knockdown significantly reduces M.tb cell association at 30 and 60 min (Figure 3C) similar to MR blockade by Ab (Schlesinger, 1993) and knockdown (Figures 3D and S2C & G). In like fashion, MR knockdown reduced the uptake of mannose BSA beads (Figure 3E). Thus, herein we demonstrate that FcRγ-chain is required for MR cell surface expression and it is also critical for M.tb association with human macrophages.

Grb2 interacts with activated MR but not Dectin-1 during M.tb infection

Grb2 is an adapter protein recruited to surface receptors upon tyrosine phosphorylation through the specific binding motif YxNx (Kessels et al., 2002) and a similar motif is contained in the MR cytoplasmic region (Figure 4A). Since Grb2 associated with MRCTP (Figure 2H), we examined Grb2 interaction with the MR during M.tb infection. We incubated hMDMs with M.tb and then Grb2-MR association was determined by co-IP of MR in cell lysates with anti-Grb2 Abs. Consistent with our hypothesis, Grb2 bound to the MR upon M.tb addition (Figure 4B). Grb2 binding increased over time and appeared to reach maximum within 10 min (Figure 4B). Additionally, we examined whether Grb2 associates with Dectin-1, which is expressed to limited extent on human macrophages (Figure 1). Our results indicate that Grb2 does not interact with human Dectin-1 (Figure S3A & B) consistent with the absence of a Grb2 binding motif in human Dectin-1. These results provide evidence that Grb2 associates with phosphorylated MR, serving as a potential intermediate in transmitting signals downstream in human macrophages.

Figure 4. Grb2 interacts with the MR and initiates Rac-1/Cdc42 activation in hMDMs.

(A) Grb2 binding motif in the human MR. (B) hMDMs were infected with M.tb (MOI 20:1) for the indicated times or left uninfected (R). The cells were lysed and cell lyates were IP’d with Grb2 or isotype control IgG Ab; proteins were visualized by WB using anti-MR Ab and re-probed with Grb2 Ab. Arrows indicate MR (upper figure) and Grb2 (lower figure). (C, D) hMDMs were incubated with mannose-BSA beads or BSA beads (MOI 20:1) for the indicated times. The amount of Rac-1 or Cdc42 bound to PAK (indicative of active GTP-Rac-1/Cdc42) was determined by WB (N=2). (E) Grb2 siRNA or control transfected hMDMs were incubated with mannose BSA beads (MOI 20:1) for the indicated times and Rac-1 activity was determined as described above. The WB shown is representative of 2 independent experiments using 2 different donors. The graph shown (lower panel) is cumulative data from N=2 (mean ± SEM; * p< 0.05; ** p<0.005). (F) Bead uptake in transfected cells was assessed at 2h by dividing the number of cells that took up beads by total cells counted in each test group (250 total cells). Cumulative data from 2 experiments (mean ± SEM). **p<0.005 (G) Transfected hMDMs were incubated with M.tb-Lux (MOI 5:1) and luciferase activity determined. RLUs = Relative Light Units. Cumulative data from 2 experiments (mean ± SEM). **p<0.005. See also Figures S2 and S3.

Grb2 plays a key role in MR-mediated signaling cascades in human macrophages

Grb2 recruits guanine exchange factor Sos to activate Rac, a critical molecule for activating various signaling pathways including phagocytosis by regulating actin cytoskeleton rearrangement (Belov and Mohammadi, 2012; Nimnual and Bar-Sagi, 2002). Since Grb2 associates with the MR (Figure 4B), we determined whether Grb2 is involved in the MR-mediated phagocytic signaling pathway by examining activation of the Rac/Cdc42/PAK complex. We incubated hMDMs with mannose BSA beads [an MR ligand (Wileman et al., 1986)] and determined the amount of Rac-1 and Cdc42 bound to PAK-1 by a pull down assay. There were increased GTP-bound Rac-1 and Cdc42 within 5 min following mannose BSA bead addition (Figure 4C). Requirement for MR engagement was assessed using control BSA beads. In contrast to mannose BSA beads, BSA beads did not mediate Rac-1 activation at the early time point (5 min), but did so at later time points (similar to mannose BSA beads), likely due to background lower level “nonspecific” internalization of BSA beads (Figure 4D). To assess the involvement of Grb2 in Rac-1 activation, we knocked down Grb2 (Figure S2B & F) prior to mannose BSA bead addition. Rac-1-PAK association was significantly reduced following Grb2 knockdown (Figure 4E). Additionally, we examined the role of Grb2 in MR-mediated phagocytosis by assaying for the uptake of mannose-BSA beads by hMDMs (Kang and Schlesinger, 1998). Grb2 knockdown significantly reduced bead uptake (Figures S3C and 4F). MR-mediated phagocytosis of M.tb limits P-L fusion as a bacterial survival mechanism (Kang et al., 2005), thus we determined bacterial survival after Grb2 knockdown. We observed significantly reduced M.tb growth in Grb2 siRNA-transfected hMDMs at 72 h (Figure 4G) providing further evidence for the importance of Grb2 in mediating signaling events critical for MR-mediated phagocytosis and M.tb growth.

M.tb infection induces SHP-1 phosphorylation and co-localization with the M.tb-containing phagosome

Protein tyrosine phosphatases (PTPs) play a central role in regulating immune functions in various cell types (Zhang et al., 2000). SHP-1 is one such PTP that down-regulates IFN-γ mediated nitric oxide production and IFN-γ inducible gene expression (Blanchette et al., 2009; Olivier et al., 1998). SHP-1 interacts to a greater extent with MRCTP than MRCT (Figure 2H). SHP-1 must be phosphorylated to exert phosphatase activity (Xiao et al., 2010). Thus we examined whether SHP-1 is phosphorylated following M.tb infection and found that SHP-1 phosphorylation is increased after 15–30 min of M.tb infection (Figure 5A). Similar findings were observed in hMDMs incubated with mannose BSA beads, but not control BSA beads (Figure S4). This timing (15–30 min) suggested that SHP-1 activation is delayed until M.tb enters the phagosome. To confirm this result we examined SHP-1 co-localization with M.tb-containing phagosomes. Our results show that SHP-1 co-localizes with 97.67 ± 0.88 % and 91.00 ± 1.73 % of M.tb-containing phagosomes at 30 and 120 min, respectively (Figures 5B & C). Thus our data indicate that M.tb phagocytosis leads to SHP-1 activation and recruitment to the M.tb phagosome.

Figure 5. MR regulates the recruitment and activation of SHP-1 in hMDMs.

(A) hMDMs underwent synchronized phagocytosis with mCherry-M.tb (MOI 20:1) for the indicated times and SHP-1 phosphorylation was determined using phospho-specific SHP-1 Ab followed by total SHP-1 Ab. Shown is a representative blot (upper panel) and band intensities of cumulative data from N=3 (lower panel, mean ± SEM). (B) hMDMs were infected with mCherry-M.tb (MOI 10:1) and stained with SHP-1 Ab (recognizes total SHP-1) or rabbit isotype control Ab. Nuclei were stained with DAPI. (C) Twenty M.tb-containing phagosomes were counted for SHP-1 co-localization and calculated as % SHP-1 colocalization (mean ± SD). (D) Control or MR siRNA transfected hMDMs were infected with mCherry-M.tb (MOI 10:1) and stained with SHP-1 Ab. (E) Percent SHP-1 co-localization with M.tb was determined for the experiment in D (mean ± SD). (F) Phosphatase activity of SHP-1 in M.tb-infected hMDMs was determined following synchronized phagocytosis with mCherry-M.tb (MOI 20:1) (mean ± SD). (G) Transfected hMDMs were incubated with mCherry-M.tb (MOI 10:1). The cells were lysed and SHP-1 phosphatase activity determined (mean ± SD). (B-G) are representative of 3 independent experiments. * p< 0.05; ** p<0.005; *** p<0.0005. Scale bars = 10 µm (isotype) and 20 µm (SHP-1 labeling). See also Figures S2 and S4.

Recruitment and activation of SHP-1 on the M.tb phagosome is MR-dependent

To determine whether SHP-1 recruitment to the M.tb phagosome depends on the MR, we transfected MR siRNA into hMDMs before infecting them with M.tb and then assessed SHP-1 recruitment to the phagosome. As shown in Figures 5D & E, MR knockdown (Figure S2) significantly reduces co-localization of SHP-1 with M.tb-containing phagosomes by 57.29 ± 5.81%. Next we determined whether SHP-1 is functionally active by measuring SHP-1 phosphatase activity in M.tb-infected hMDMs. SHP-1 phosphatase activity was significantly increased within 30 min and further increased over time (Figure 5F). Furthermore, this phosphatase activity was markedly reduced if the MR was knocked down (Figure 5G).

Since SHP-1 recruitment to the M.tb phagosome depends on the MR, we next considered whether SHP-1 interacts directly with the MR or through MR-bound Grb2. The MR does not contain a SHP-1 binding motif [(I/V)xYxx(L/V)], whereas both MR and SHP-1 contain the Grb2 binding motif (Figures 4A and S5A), SHP-1 contains multiple such motifs (Minoo et al., 2004; Kon-Kozlowski et al., 1996). We incubated hMDMs with M.tb and then SHP-1-Grb2 association was determined in cell lysates by co-IP of Grb2 with anti-SHP-1 Ab. The interaction of Grb2 with SHP-1 was confirmed (Figure S5B). Together these data provide evidence that recruitment of enzymatically active SHP-1 to M.tb phagosomes is via Grb2 bound to the MR. We next determined if this interaction involves FcRγ-chain. However, FcRγ-chain does not contain a Grb2 binding motif and IP with anti-SHP-1 Ab does not pull down FcRγ-chain during M.tb infection (in control and Grb2 siRNA-transfected hMDMs; Figure S5C), indicating that SHP-1 and Grb2 interact with the MR independently of FcRγ-chain, a finding consistent with the fact that FcRγ-chain binds constitutively to MR whereas Grb2 and SHP-1 are recruited upon MR phosphorylation.

Inhibition of SHP-1 activity enhances P-L fusion and decreases bacterial growth in human macrophages

Since SHP-1 is activated upon M.tb infection and co-localizes with the M.tb phagosome, we directly examined the role of SHP-1 in regulating P-L fusion in M.tb-infected hMDMs. We pretreated hMDMs with sodium-stibogluconate (SSG 10 µg/mL), a selective inhibitor of SHP-1 (Pathak and Yi, 2001; Choi et al., 2017), then infected with M.tb, and assessed co-localization of bacteria with the late endosomal-lysosomal marker LAMP-1. SSG pre-treatment led to a significant increase in LAMP-1 co-localization with M.tb-containing phagosomes (Figures 6A & B). Similar findings were observed in hMDMs that were transfected with SHP-1 siRNA (Figures 6C & D and Figure S2D & H). Given the increase in P-L fusion with SHP-1 inhibition, we next assessed whether SHP-1 inhibition limits M.tb growth. M.tb growth was significantly reduced in hMDMs that were treated with SSG or underwent SHP-1 knockdown (Figure 6E & F). Taken together, these data define how MR limits P-L fusion, i.e., MR activation leads to SHP-1 recruitment to the M.tb phagosome, thereby limiting P-L fusion and facilitating M.tb growth in human macrophages.

Figure 6. MR regulates P-L fusion through SHP-1 in hMDMs.

(A) hMDMs were pre-treated with the SHP-1 inhibitor SSG (10 µg/mL) or DMSO for 1 h and infected with mCherry-M.tb (MOI 10:1). Scale bars = 15 µm (left panels) and 20 µm (right panel). (B) Percent P-L fusion was determined by localization of LAMP-1 with mCherry-M.tb. (C) Day 5 PBMCs were transfected with control siRNA or SHP-1 siRNA and after 48h of incubation the transfected hMDMs were infected with mCherry-M.tb (MOI 5:1), then fixed, permeabilized and stained with LAMP-1 Ab. Scale bar = 25 µm. (D) Percent LAMP-1 co-localization with mCherry M.tb. White arrows in A and C show bacteria not co-localized with LAMP-1. (E) Vehicle control or SHP-1 inhibitor-treated hMDMs, or (F) control or SHP-1 siRNA transfected hMDMs were infected with M.tb-Lux (MOI 5:1) and M.tb growth was determined by luciferase assay. RLUs = Relative Light Units. Shown are representative images and graphs of 3 independent experiments. B, D, E and F are mean ± SD. * p<0.05; ** p<0.005. See also Figure S2.

SHP-1 reduces class III PI3K activity by interacting with the Vps15 and Vps34 complex to reduce the amount of PtdIns (3) P on M.tb phagosomes

Generally, phagosome maturation is regulated by the activity of Class III PI3K (Vps34) and its product PtdIns (3) P [PI(3)P] which serves as a docking site for EEA1, recruitment of Rab7 and P-L fusion. The activity of class III PI3K is complex and partially regulated by Vps15 phosphorylation (Stack et al., 1993). Recruitment of EEA1 to the M.tb-containing phagosome is restricted and PI(3)P is not generated by the Class III PI3K Vps34 (Fratti et al., 2001) but the reason for this is not clear. Since we show that SHP-1 is localized on M.tb phagosomes, SHP-1 deficiency increases P-L fusion, and there is a consensus SHP-1 binding motif in both Vps15 and Vps34 (Figure 7A), we next investigated the impact of SHP-1 on Class III PI3K activity during M.tb infection. We first determined whether SHP-1 binds Vps15 and Vps34 during M.tb infection. Vps15 and Vps34 co-precipitated with SHP-1 within 30 min post infection and at subsequent time points (Figure 7B). Since Vps15 is a protein kinase required for Vps34 lipid kinase activity (Backer, 2008), we hypothesized that the interaction of SHP-1 with Vps15 and Vps34 limits Vps34 activity and subsequent production of PI(3)P on phagosomal membranes to limit P-L fusion. To test this we determined the level of PI(3)P in M.tb-infected hMDMs. M.tb infection increased the production of PI(3)P at 120 min post infection (Figure 7C). However, PI(3)P production was significantly increased after MR or SHP-1 knockdown by 46.78 ± 1.75% and 38.27 ± 1.12%, respectively (Figure 7D) confirming that MR and SHP-1 limit P13K activity. In order to prove that SHP-1 activation decreases the accumulation of PI(3)P on M.tb-containing phagosomes, we treated hMDMs that express p40PX-EGFP [p40PX binds to PI(3)P] (Kanai et al., 2001) with SHP-1 inhibitor (SSG 10µg/mL) and subsequently incubated the cells with mannose BSA beads or mCherry M.tb. Interaction of the p40PX domain with phagosomes was examined by confocal microscopy. We found that inhibition of SHP-1 enhances localization of p40PX-EGFP with phagosomes (Figure 7E-H) providing evidence that SHP-1 inhibits the production of PI(3)P on M.tb phagosomes. Together these data provide strong evidence that M.tb activation of MR leads to SHP-1 activity and association with Vps15 and Vps34, leading to inhibition of Class III P13K Vps34 activity, thereby limiting PI(3)P production on the M.tb phagosome and subsequent P-L fusion in human macrophages.

Figure 7. SHP-1 regulates class III PI3K activity in human macrophages.

(A) SHP-1 binding motif on hVps15 and hVps34. (B) hMDMs underwent synchronized phagocytosis with mCherry-M.tb (MOI 20:1) for the indicated times and cell lysates were IP’d with SHP-1 or control IgG Ab followed by WB with Vps15 (top panel), Vps34 (middle panel), and SHP-1 Abs (lower panel). The WBs shown are representative of 2 independent experiments. (C) hMDMs underwent synchronized phagocytosis with mCherry-M.tb (MOI 20:1) for the indicated times. Cell lysates were IP’d with anti-hVps34 or isotype IgG control Ab and the IPs were used to determine PI3K activity. (D) Transfected hMDMs were infected with M.tb (MOI 20:1) and PI3K activity determined. (E, F) hMDMs were transfected with p40PX-EGFP and after 16h cells were pre-treated with the SHP-1 inhibitor SSG (10 µg/mL) or DMSO for 1 h and then incubated with mannose BSA beads (E) or mCherry-M.tb (MOI 10:1) (F). (G, H) Percent co-localization of p40PX-EGFP with beads or M.tb was determined from the images acquired using confocal microscopy (~20 images, representing ~150 cells). Red arrows signify co-localization. White arrows signify no co-localization. Scale bars = 25 µm. The graphs shown (C&D, G&H) are representative of 2 independent experiments (mean ± SD). * p<0.05; ** p<0.005. See also Figure S2.

Human MR activates SHP-1 in murine macrophages

Some prior studies in mice showed that loss of MR function does not have a phenotype for M.tb pathogenesis (Court et al., 2010) but MR is critical in human macrophages (Kang et al., 2005) suggesting differences between the murine and human contexts. Consistent with this, there are amino acid sequence differences and consequent charge differences in the cytoplasmic tail between human and mouse MR, suggesting that the MR may recruit different proteins upon activation in humans compared to mice and thus signal differently (Figure S6A). To address this possibility, we used a human MR transgenic (hMR) mouse (He et al., 2007) to assess the impact of the human MR expressed in murine macrophages on MR signaling Expression of the human MR transgene in the spleen and lungs of hMR mice was confirmed by WB and IHC (Figure S6B & C).As a read-out of MR signaling, we assessed SHP-1 phosphorylation following M.tb infection of bone marrow derived macrophages (BMDMs) and alveolar macrophages (AMs) from WT and hMR mice. There is increased SHP-1 phosphorylation following M.tb infection in both BMDMs and AMs from hMR mice compared to wild type (WT) mice (Figure S7 A-F). We next determined whether increased SHP-1 phosphorylation in BMDMs from hMR mice enhances M.tb intracellular growth. Consistent with our results using hMDMs, there was a significant increase in M.tb growth in BMDMs from hMR mice compared to WT mice (Figure S7G), despite no significant difference in initial bacterial association (Figure S7H). These results indicate a specific function of human MR in regulating SHP-1 activation, which contributes to M.tb survival in human cells, providing another difference between mouse and man (Seok et al., 2013, PNAS; Rigamonti et al., 2008).

Discussion

This study presents evidence for human MR-specific signaling molecules that help explain the biological responses of human macrophages to M.tb. The MR is a prototypic PRR that regulates inflammatory responses, pathogen recognition, entry and trafficking, and antigen presentation (Stahl and Ezekowitz, 1998; Schlesinger, 1993; Kang et al., 2005); M2 macrophage differentiation (Martinez et al., 2008); tumor associated macrophage biology (Mantovani et al., 2002); and therapeutic targeting (Azad et al., 2014). The MR is a member of the CTL family, which includes Dectin-1, Dectin-2, DC-SIGN, Mincle, and Endo 180 (East and Isacke, 2002) and is the major CTL expressed on hMDMs (Figure 1). Despite this, how the receptor initiates signaling events has remained unclear.

To define mechanisms by which MR functions in macrophages, we identified proteins interacting with the MR cytoplasmic tail, which included 47 proteins present in both MRCT and MRCTP peptide pulldown groups. Of these proteins, we demonstrate that FcRγ-chain interacts with MR and this association is required for surface localization of MR. Other CTLs like Dectin-2, Mincle and MCL also couple with the FcRγ-chain for signal transduction, and furthermore, interaction of FcRγ-chain with FcγRs, Dectin-2 and Mincle facilitates surface expression on macrophages (Miyake et al., 2013; Yamasaki et al., 2008; Yonekawa et al., 2014; van Vugt et al., 1996). Constitutive FcRγ-chain MR interaction may be due to the local positive charge in the interface of the MR transmembrane (TM) region and cytoplasmic tail, since interaction of FcRγ-chain with other CTLs occurs through a positively charged arginine residue (Yamasaki et al., 2008). In contrast, the murine MR TM region and cytoplasmic tail are neutrally charged, thus we speculate that this difference reduces FcRγ-chain recruitment and murine MR surface exposure, thereby accounting in part for differences between human and mouse MR functions.

In order to transmit activating signals, the recruitment of second messengers/adaptor molecules to transmembrane receptors like the MR is essential. To define these, we identified 56 proteins specifically associated with phosphorylated MR. Grb2 is a signaling adaptor molecule that recognizes a specific binding motif (YxNx) (Kessels et al., 2002) contained in the MR. We found that Grb2 is recruited to phosphorylated MR during M.tb infection and knockdown of Grb2 significantly reduces macrophage uptake of mannose-BSA beads. Furthermore, we found that MR activation leads to activation of Rac-1, Cdc42 and PAK-1 which depend on the adaptor molecule Grb2 and are important for F-actin polymerization. These data are supported by an earlier finding that Cdc42 and RhoB activation is required for MR-mediated phagocytosis by human alveolar macrophages (Zhang et al., 2005). Importantly, our kinetic analyses indicate that MR activation quickly (within 5 min) leads to activation of a signaling cascade required for phagocytosis (i.e., Grb2 and the Rac/Cdc42/PAK complex).

We demonstrate that phosphorylated MR recruits SHP-1 to the M.tb phagosome following phagocytosis. SHP-1 is a tyrosine phosphatase that is activated by several microbes to subvert host innate immune functions (Blanchette et al., 2009; Massa and Wu, 1996; Olivier et al., 1998). This identifies a phosphatase localized to the M.tb phagosome which has important implications for the unique nature of this vacuole. Our data support a mechanism whereby SHP-1 recruitment occurs via Grb2 that binds the MR early (within 5 min). In contrast, SHP-1 phosphorylation upon M.tb infection occurs after 15 min. Thus we postulate a temporal and spatial interplay of MR-associated signaling molecules such that early Grb2-MR association for initiation of phagocytosis is followed by Grb2-mediated recruitment of SHP-1 to the MR on the M.tb phagosome, which then reduces P-L fusion. In support of this, SHP-1 is recruited to latex bead- or opsonized zymosan bead-containing phagosomes (Gomez et al., 2012) and M.tb phagosomes (Figure 5B). MR knockdown in hMDMs significantly reduced SHP-1 recruitment to phagosomal membranes demonstrating the dependence on MR signaling. Our data do not support a direct interaction between SHP-1 and FcRγ-chain as recently reported by Iborra et al. (Iborra et al., 2016) which studied Mincle-mediated inhibitory signaling events in murine GMCSF-treated bone marrow-derived cells suggesting receptor and/or mammalian cell differences.

Phagosome maturation involves Rab5 recruitment to the early endosome, leading to recruitment and phosphorylation of the regulatory subunit Vps15, activation of the Class III PI3K Vps34, generation of PI(3)P and recruitment of Rab5 effector protein early endosome autoantigen-1(EEA1) which contributes to membrane tethering and fusion with lysosomes (Backer, 2008; Stack et al., 1993; Stack et al., 1995; Fratti et al., 2001). It has been determined that although Rab5 and Vps34 are recruited to M.tb-containing phagosomes, EEA-1 is not recruited and PI(3)P availability is limited suggesting that Vps34 is not functional (Fratti et al., 2001). However, the mechanism for inhibition of Vps34 and the role of Vps15 are unknown. Our data show that SHP-1 knockdown or inhibition in human macrophages increases LAMP-1 localization with M.tb phagosomes, which suggests that SHP-1 activation and recruitment inhibits phagosome maturation. In support of this, Vps15 and Vps34 contain the consensus SHP-1 binding motif [(I/V)xYxxL] and interact with SHP-1 upon M.tb infection (Figure 7A&B). The interaction of Vps15 and Vps34 with SHP-1 may facilitate a conformational change in Vps34, thereby inhibiting its ability to produce PI(3)P. Indeed, our data clearly show that MR or SHP-1 knockdown increases the production of PI(3)P, an indicator of Vps34 activation (Class III PI3K activity), in human macrophages. Thus, we propose that MR-dependent recruitment of SHP-1, which limits PI(3)P production, contributes to the ability of M.tb to limit P-L fusion and enhance its growth in human macrophages. In contrast to human macrophages, SHP-1 activation is not robust in murine cells following M.tb interaction. Through creation of transgenic mice expressing hMR, we demonstrate that SHP-1 activation by human MR, but not murine MR, provides a potential explanation for the absence of a phenotype in M.tb growth in MR-KO mice compared to WT mice (Court et al., 2010).

Herein, we identify signaling molecules that interact specifically with human MR and provide a mechanistic explanation for MR cell surface expression and for MR-dependent reduction in P-L fusion following M.tb infection. Our results define the temporal and spatial orchestration of MR-interacting proteins which explains both the induction of phagocytosis of M.tb and the maturation inhibition of M.tb-containing phagosomes by this PRR. Importantly, this mechanism of phagosome maturation inhibition induced by M.tb-MR interaction facilitates M.tb survival in human cells. In addition, the demonstration that human MR but not murine MR activates SHP-1, which is important for inhibition of phagosome maturation, provides an explanation for why MR is critical for M.tb pathogenesis in human cells but not murine cells, further suggesting that M.tb’s evolution as a human pathogen exploits unique properties of the human MR.

Experimental procedures

Bacterial strains and culture

Lyophilized M.tb H₃₇R_v (ATCC #25618) was reconstituted and used as described (Schlesinger et al., 1990). The mCherry-M.tb strain was kindly provided by Dr. Sarah Fortune, Harvard University. A luciferase-expressing reporter strain of M.tb (M.tb-Lux) was used in which the plasmid construct pMV306hsp+Lux (which contains the entire bacterial Lux operon cloned in a mycobacterial integrative expression vector) was introduced into M.tb H₃₇R_v (Andreu et al., 2010; Salunke et al., 2015; Guirado et al., 2015). The concentration of bacteria (1–2 × 10⁸ bacteria /ml) and the degree of clumping (≤10%) were determined by counting in a Petroff-Hausser chamber. Bacteria prepared in this fashion are ≥ 90% viable by colony-forming unit (CFU) assay.

Isolation and culture of human monocyte-derived macrophages (hMDMs)

hMDM monolayers were prepared from healthy, purified protein derivative (PPD)-negative human volunteers using an approved OSU IRB protocol as described (Schlesinger, 1993). Briefly, PBMCs were isolated from heparinized blood on a Ficoll cushion and then cultured in Teflon wells (Savillex, Minnetonka, MN, USA) for 5 days in 20% autologous serum. hMDMs in the cultured PBMCs were adhered to tissue culture plates (Falcon, Becton Dickinson labware, Franklin Lakes, NJ, USA) for 2–3h at 37°C/5% CO₂ in 10% autologous serum. Lymphocytes were then washed away, and hMDM monolayers were repleted with RPMI containing 10% autologous serum and incubated overnight.

Peptide pulldown and LC-MS/MS analysis

MR cytoplasmic tail peptides MRCT (biotin-NH-AYFFYKKRRVHLPQEGAFENTLYFNQS-COOH) and MRCTP (biotin-NH-AYFFYKKRRVHLPQEGAFENTL-[Y(PO3H2)]FNQS-COOH) were synthesized and purchased from ThermoFisher Scientific (Waltham, MA, USA). The HPLC grade peptides were dissolved in 50% (v/v) DMSO/endotoxin-free water. hMDM cell lysates (500 µg) were mixed with 5 µg of peptides and incubated overnight at 4°C on a nutator. Subsequently, the peptides were pulled down by addition of 25 µL of streptavidin agarose beads (Invitrogen) and incubated at 4°C for an additional 2 h. Beads were collected by centrifugation, washed seven times with TN-1 lysis buffer, boiled with SDS-PAGE sample buffer and proteins were separated by gradient SDS-PAGE (4–16%). The protein bands were visualized with Gel code blue stain (ThermoFisher Scientific). Bands in the stained gels in the 2 groups (whole lanes) were cut in segments, digested with proteases and analyzed by LC-MS/MS Orbitrap mass spectrometry. The amino acid sequences were analyzed using the MASCOT data base and the identified proteins were compared between the MRCT and MRCTP groups. The presence or absence of bands in the 2 groups was compared as well as significant differences in the amounts of protein in the 2 groups.

Rac1/Cdc42 assay

Following Grb2 knockdown hMDMs were incubated with mannose-BSA beads or BSA-beads (MOI 20:1) for 5, 10 or 30 min and cells were lysed with TN-1 lysis buffer. The Rac-1/Cdc42 assay was performed according to the manufacturer’s instruction (EMD Millipore, Cat#17-441). Briefly, equal amounts (300µg) of protein lysates were incubated with 10µg of the Rac/cdc42 assay reagent containing PAK-1 PBD (protein binding domain)-coated agarose beads (serve as substrate for activated Rac-1 and Cdc42) and incubated at 4°C for 60 min on a rocking platform. The beads were collected by centrifugation, washed 3 times with 0.5 mL lysis buffer, re-suspended in 40µL of 2X Laemmli reducing sample buffer and boiled for 5 min, then subjected to SDS-PAGE and WB using anti Rac-1 or anti cdc42 Ab.

SHP-1 phosphatase assay

The SHP-1 phosphatase assay was performed according to the manufacturer’s instructions using the PTP assay kit (EMD Millipore; Cat#17-125). Briefly, SHP-1 protein was IP’d from uninfected or M.tb-infected (30, 60 or 120 min) hMDM lysates using anti-SHP-1 Ab. IP with control Abs was done in cell lysates of hMDMs infected for 120 min. Ag-Ab complexes were pulled down with protein G agarose beads. The IPs were washed 7 times with wash buffer (10mM Tris-HCL, pH 7.4) and subsequently incubated with tyrosine phosphorylatedpeptide as substrate (RRLIEDAEpYAARG) (Upstate Biotechnology) in 10 mM Tris-HCL, pH 7.4, for 30 min. The reaction was stopped by adding 100µL of malachite green solution, incubated for an additional 15 min, and the absorbance was measured at 630 nm. The amount of phosphate released from these samples was calculated based on phosphate standards used in every experiment.

Class III PI3K activity assay

hMDMs were washed with ice cold PBS, and lysed with lysis buffer (20mM Tris, pH 7.5, 137 M NaCl, 1mM MgCl₂, 1mM CaCl₂, 100mM NaF, 10 mM Na Pyrophosphate, 100µM Na3VO4, 1% Nonidet P-40, 10% glycerol and 0.35mg/mL PMSF). hVps34 protein was IP’d from uninfected or M.tb-infected cells using anti hVPS34 Ab (Echelon BioSciences Inc. Salt lake City, UT,USA) or IgG control Ab. Ag-Ab complexes were pulled down with protein A Sepharose beads and washed with PBS/1% NP-40 (3x), 100mM Tris-HCl/500mMLiCl (3x), and 10mM Tris-HCl/100mM NaCl/1mM EDTA (2x). hVps 34 enzyme activity was determined according to a Class III PI3-Kinase assay kit (K-3000) (Echelon BioSciences).

Statistics

When using hMDMs, the magnitude of the response from each independent experiment varied among the donors; however, the pattern of experimental results was the same from donor to donor. To account for this variability, we normalized the data to an internal control in each experiment. A ratio of experimental results to control was obtained, and the mean ratio was then tested for a significant difference using t statistics. Statistical significance was defined as *p < 0.05. Unless indicated, results were obtained from 3 independent experiments using 3 different donors. N= experimental replication. Additional methods can be found in the supplemental information.

Supplementary Material

2. Supplemental Table 1 (related to Figure 2). Identified Proteins from MRCT peptide pulldown assay.

a The following abbreviations and their meaning are used in the protein description: OS = Organism species, GN = gene name, PE = evidence for protein existence (1 = protein level evidence and 2 = transcript level evidence), SV = sequence version. b The protein score is derived from Mascot and provides an indication of how well the peptides matched the indicated protein sequence. The actual score is calculated by the following equation: protein score = −10*Log(P), where P is the probability that the protein match is a random event. Scores above 100 indicate that p < 0.05. c The number is peptide=spectral matches between the acquired MS data and the identified protein in the database searched against. d The protein match score indicates the number of unique peptides that matched the sequence of the identified protein. Two unique peptide matches to a protein sequence confirms the identity of a protein.

3. Supplemental Table 2 (related to Figure 2). Identified Proteins from MRCTP peptide pulldown assay.

a The following abbreviations and their meaning are used in the protein description: OS = Organism species, GN = gene name, PE = evidence for protein existence (1 = protein level evidence and 2 = transcript level evidence), SV = sequence version. b The protein score is derived from Mascot and provides an indication of how well the peptides matched the indicated protein sequence. The actual score is calculated by the following equation: protein score = −10*Log(P), where P is the probability that the protein match is a random event. Scores above 100 indicate that p < 0.05. c The number is peptide=spectral matches between the acquired MS data and the identified protein in the database searched against. d The protein match score indicates the number of unique peptides that matched the sequence of the identified protein. Two unique peptide matches to a protein sequence confirms the identity of a protein.

4. Supplemental Table 3 (related to Figure 2). Common Proteins Identified from MRCTP and MRCT peptide pulldown assays.

a The following abbreviations and their meaning are used in the protein description: OS = Organism species, GN = gene name, PE = evidence for protein existence (1 = protein level evidence and 2 = transcript level evidence), SV = sequence version. b The protein score is derived from Mascot and provides an indication of how well the peptides matched the indicated protein sequence. The actual score is calculated by the following equation: protein score = −10*Log(P), where P is the probability that the protein match is a random event. Scores above 100 indicate that p < 0.05. c The number is peptide=spectral matches between the acquired MS data and the identified protein in the database searched against. d The protein match score indicates the number of unique peptides that matched the sequence of the identified protein. Two unique peptide matches to a protein sequence confirms the identity of a protein.

Highlights.

• MR interaction with FcRγ-chain is essential for MR macrophage surface exposure

• MR activation mediates Grb2 recruitment to initiate phagocytosis signaling

• MR-dependent SHP-1 recruitment limits PI3K activity and phagosome-lysosome fusion