|Home | About | Journals | Submit | Contact Us | Français|
The neonatal Fc receptor (FcRn) for IgG, an MHC class I-related molecule, functions to transport IgG across polarized epithelial cells and protect IgG from degradation. However, little is known about whether FcRn is functionally expressed in immune cells. We show here that FcRn mRNA was identifiable in human monocytes, macrophages, and dendritic cells. FcRn heavy chain was detectable as a 45-kDa protein in monocytic U937 and THP-1 cells and in purified human intestinal macrophages, peripheral blood monocytes, and dendritic cells by Western blot analysis. FcRn colocalized in vivo with macrosialin (CD68) and Ncl-Macro, two macrophage markers, in the lamina propria of human small intestine. The heavy chain of FcRn was associated with the β2-microglobulin (β2m) light chain in U937 and THP-1 cells. FcRn bound human IgG at pH 6.0, but not at pH 7.5. This binding could be inhibited by human IgG Fc, but not Fab. FcRn could be detected on the cell surface of activated, but not resting, THP-1 cells. Furthermore, FcRn was uniformly present intracellularly in all blood monocytes and intestinal macrophages. FcRn was detectable on the cell surface of a significant fraction of monocytes at lower levels and on a small subset of tissue macrophages that expressed high levels of FcRn on the cell surface. These data show that FcRn is functionally expressed and its cellular distribution is regulated in monocytes, macrophages, and dendritic cells, suggesting that it may confer novel IgG binding functions upon these cell types relative to typical FcγRs: FcγRI, FcγRII, and FcγRIII.
The neonatal Fc receptor (FcRn)4 is structurally related to the MHC class I family (1–3) and consists of a membrane-bound heavy chain (45 kDa for human, 51 kDa for rodents) in nonconvalent association with β2-microglobulin (β2m; 12 kDa). FcRn was originally characterized as a transport receptor involved in the uptake of maternal IgG by an intestinal route in rodents (4–8) and probably via syncytiotrophoblastic cells within human placenta, respectively (9–13). Additionally, FcRn has been considered to function in the protection of IgG from degradation. This idea was first proposed by Brambell (14) and is supported by recent observations that mice deficient in β2m exhibit significant reduction in the serum half-life of IgG (15–17). Recent evidence for FcRn expression by endothelial cells suggested that this may be the cell type most prominently involved in IgG protection (18).
A hallmark of FcRn interaction with its ligand is its strictly pH-dependent IgG binding in both epithelial and endothelial cells. FcRn preferentially binds IgG at acidic pH (6–6.5), but is unable to bind IgG at neutral pH (7–7.4) (19–21). FcRn is expressed in a variety of cell types and tissues, including intestinal epithelial cells (IECs) of neonatal rodents, syncytiotrophoblasts of humans, endothelial cells of adult rodents and humans, adult rat hepatocytes, and adult epithelial cells of bovine mammary gland, human intestine, and human kidney (22–27).
Immune cells, such as B lymphocytes, macrophages, dendritic cells, NK cells, mast cells, and granulocytes, typically express single or multiple receptors for the Fc portion of IgG, including FcγRI (CD64), FcγRII (CD32), FcγRIII (CD16), and their splice variants. These FcγRs play a pivotal role in linking the cellular and humoral arms of the immune response. Specifically, these receptors are involved in internalization of immune complexes, Ag presentation, Ab-dependent cellular cytotoxicity, negative regulation of effector functions of FcγR-bearing cells, regulation of the inflammatory cascade, and autoimmunity (28–31). However, FcRn expression has not been characterized in immune cells, especially in FcγR+ cells. Therefore, we tested the hypothesis in this study that FcRn is functionally expressed in human immune cells. We found by several criteria that FcRn was expressed in human monocytes, macrophages, and dendritic cells and in human monocytic cell lines and exhibits pH-dependent binding of IgG in these cells. Moreover, the cellular distribution of FcRn expression between intracellular and cell surface locations appears to be differentially regulated. These studies indicate that FcRn is the fourth FcR for IgG to be defined on macrophages and dendritic cells and significantly extend the potential function of FcRn and the cell types involved in the known functions of this novel MHC class I-like molecule.
HeLa (cervical epithelial cell line), Jurkat (thymoma cell line), U937 (monocyte cell line), Raji (B cell line), and 721.721 (HLA-A-, -B-, and -C-negative B cell line) were purchased from American Type Culture Collection (Manassas, VA). THP-1 (monocytic cell line), NK3.3 (NK cell line), and NKL (NK cell line) were gifts from Dr. Mark Birkenbach (University of Chicago, Chicago, IL), Dr. Paul Anderson (Harvard Medical School, Boston, MA), and Dr. Marco Colonna (Basel Institute for Immunology, Basel, Switzerland), respectively. The U937 (promonocytic cell line), Raji, and 721.721 cell lines were cultivated in suspension in RPMI 1640 medium (Life Technologies, Gaithersburg, MD) supplemented with 10% FCS, 1% l-glutamine, and 1% penicillin/streptomycin. A CD1d-transfected cell line, 721.721CD1d, generated by transfecting the 721.221 cell line with the full-length CD1day cDNA in the PSRαneo expression vector, was cultivated in the same medium supplemented with 500 µg/ml G418 (Life Technologies). The THP-1 cell line was cultivated in the same medium with 5 × 10−5 M 2-ME (Sigma, St. Louis, MO). NK3.3 and NKL were cultivated in 10% RPMI 1640 medium with 10% human serum. HeLa cells were cultivated with 10% FCS in DMEM (Life Technologies). Cell viability was assessed by trypan blue dye exclusion.
The human FcRn codons (11) corresponding to the α1 (1–87), the α2 (88–177), and the α3 (178–274) domains were amplified by PCR and subcloned into the EcoRI sites of the pGEX4T-1 (Amersham Pharmacia Biotech, Piscataway, NJ) expression vector. The primer pairs for α1 (5′-CCGGAATTCGCAGAAAGCCACCTCTCCCT-3′, 5′-GGCGAATTCTCAACCTTTTCCCCCCAA-3′), α2 (5′-GGCGAATTCTACACTCTGCAGGGCCTGCT-3′, 5′-CGCGAATTCTCACTTCCACTCCAGGTTT-3′), and α3 (5′-CCGGAATTCGAGCCCCCCTCCAT, 5′-GGCGAATTCGGAGGACTTGGCTGGAGATT-3′) were used for amplification by Pfu polymerase (Stratagene, La Jolla, CA). The EcoRI site in the primers are underlined, and human FcRn sequences are italicized. The plasmid encoding the full-length human FcRn, provided by Dr. Neil Simister (Brandeis University, Waltham, MA), was used as a template. All subclones were verified by sequencing. The production of recombinant proteins was performed by a method modified from that previously described (32) and analyzed by SDS-PAGE electrophoresis. Five micrograms of the purified GST-α1, GST-α2, and GST-α3 proteins were respectively emulsified in CFA and injected s.c. into each BALB/c mouse. Mice were boosted twice at 3-wk intervals with fusion protein emulsified with IFA. Sera were sampled 2 wk following the final dose. Furthermore, the immunization of rabbits with purified fusion protein was performed by Charles River Breeding Laboratories (Wilmington, MA).
Lamina propria macrophages were isolated from surgical human normal tissue by neutral protease digestion of intestinal tissue sections with counterflow centrifugal elutriation as previously described (33, 34). Briefly, sections of normal human jejunum were incubated in 0.2 M EDTA (Fisher Scientific, Norcross, GA) plus 10 mM 2-ME (Sigma) to remove the epithelium, minced, and then treated twice (45 min, 200 rpm, 37°C) in RPMI (Mediatech, Washington, DC) containing 100 µg/ml DNase and 75 µg/ml of the neutral protease dispase (grade I; Roche, Indianapolis, IN) to release the lamina propria mononuclear cells (33). After straining to remove debris and gradient sedimentation to remove residual nonmononuclear cells, the cells were separated into highly purified populations of lamina propria macrophages and lymphocytes by counterflow centrifugal elutriation using a J-6 M elutriation centrifuge (Beckman, Palo Alto, CA) (33, 34). The cells isolated by this procedure contained <1% CD3+ lymphocytes and displayed the size distribution, morphological features, ultrastructure, and phagocytic activity of macrophages (33).
Peripheral blood monocytes were isolated from leukopaks from healthy donors by elutriation. Both cell populations were rested for 2 days in DMEM (Quality Biologicals, Gaithersburg, MD) plus 50 mg/ml gentamicin and 10% human AB serum (Atlanta Biologicals, Atlanta, GA) before study. Cell purity was assessed by flow cytometry as previously described (3–5) using mAbs against the following cell markers: HLA-DR, CD3, CD13, CD14, CD20, and CD80 (Becton Dickinson, San Jose, CA) and CD103 and CD83 (Immunotech, Westbrook, ME). Isotype-matched irrelevant mAbs were used as controls.
The monocyte-derived dendritic cells were obtained by a previously described method (35). Briefly, monocytes were isolated from PBMCs by adherence to plastic for 2 h and were cultured for 8 days in RPMI 1640 (Life Technologies) supplemented with 10% FCS, 10 mM HEPES, 2 mM l-glutamine, 5 × 10−5 2-mecaptoethanol, penicillin (100 U/ml), streptomycin (100 mg/ml), recombinant human GM-CSF (100 U/ml), and recombinant human IL-4 (1000 U/ml). The medium was replaced every 3–4 days. After 8 days, cells displaying dendritic morphology and predominantly expressing CD1a and HLA-DR, but that had lost most of the expression of the monocyte marker CD14, were obtained. Immature dendritic cells were obtained by culturing the adherent fraction of normal human PBMCs in the presence of GM-CSF and IL-4 for 3 days.
Cells were pelleted and resuspended at 106 cells/ml in Tri-Reagent (Molecular Research Center, Cincinnati, OH). Total RNA was extracted according to the method recommended by the manufacturer. First-strand cDNA was synthesized from 1 µg of total RNA using Moloney murine leukemia virus reverse transcriptase (Promega, Madison, WI) and an oligo(dT) primer (Promega) as recommended by manufacturer. The human FcRn gene was amplified from cDNA by a primer pair (5′-CCGGAATTCGCAGAAAGCCACCTCTCCCT, 5′-CGGAATTCTTAGCAGTCGGAATGGCGGA-3′) that contained EcoRI sites in the 5′ extension to facilitate cloning. Amplification was performed by hot start PCR using 35 cycles each consisting of 95°C for 1 min, 55°C for 1 min, and 72°C for 1 min. At the end of the 35 cycles, samples were run for an additional 10 min at 72°C and then maintained at 4°C until analyzed by agarose gel electrophoresis. The mRNA was also amplified by GAPDH-specific primers as an internal control to monitor the quality of the RNA purification and cDNA synthesis.
The FcRn codon (1–343) was amplified from an FcRn-containing plasmid (11) with the primer pair 5′-ATAAGAATGCGGCCGCGGCAGAAAGCCACCTCTCCCT-3′ and 5′-TGCTCTAGATTAGGCGGTGGCTGGAATCA-3′. The upstream primer introduced a NotI site, and the downstream primer introduced an XbaI site to facilitate cloning (underline). Amplification was performed using Pfu DNA polymerase with initial heating to 95°C for 5 min, followed by 35 cycles each consisting of 95°C for 1 min, 58°C for 1 min, and 74°C for 1.5 min, and was terminated by a final extension step at 72°C for 10 min. The PCR product was purified by agarose gel using a GeneClean II kit (Bio 101, Vista, CA). The DNA fragment was digested with NotI and XbaI and ligated into the plasmid pFlagCMV-1 (Sigma) to generate the plasmid, pFlagCMVhFcRn. In this plasmid a Flag epitope (DYKDDDDK, single-letter amino acid code) was fused into the N terminus of the FcRn gene. The plasmid pCDNAhβ2m was constructed as previously described (36). The open reading frames of plasmids pCDNAhβ2m and pFlagCMVhFcRn were verified by sequencing both strands to confirm the fidelity of amplification and cloning.
Transfection of HeLa was performed by electroporation (Electroporator II; Invitrogen, San Diego, CA) using 20 µg of pFlagCMVhFcRn and 2 µg of pCDNAhβ2m to ensure that β2m concentrations were not substrate limiting for FcRn expression. Transfected cells were grown under selection with 1 mg/ml of G418 (Life Technologies). Single colonies of transfected HeLa were expanded under 500 µg/ml of G418. Positive colonies were confirmed by Western blotting using the FcRn anti-α2-specific serum as described. The chosen positive transfectant was designated HeLaFcRn + β2m.
Cell lysates were prepared in PBS with 0.5% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% SDS by adding a protease inhibitor cocktail (Sigma). A postnuclear supernatant was analyzed for total protein concentrations by the Bradford method with BSA as a standard (Bio-Rad, Hercules, CA). The proteins were separated on 12% SDS-PAGE gels under reducing conditions and transferred onto nitrocellulose (Schleicher & Schuell, Keene, NH). The membranes were blocked with 5% nonfat milk and probed with mouse anti-human FcRn α2 Ab (1/500) for 1 h, then with HRP-conjugated goat anti-mouse IgG Fc Ab (1/10,000). All blocking, incubation, and washing steps were performed in PBS containing 0.05% Tween 20 and 5% milk. The final product was visualized by ECL (Pierce, Rockford, IL).
Immunoprecipitations were performed as previously described (22). Briefly, 5 × 105 log-phase-grown THP-1 and U937 cells were metabolically labeled with 0.5 mCi of trans-35S-labeled methionine and cysteine (ICN Biomedicals, Costa Mesa, CA) in methionine- and cysteine-free RPMI 1640 medium (ICN Biomedicals) supplemented with 10% dialyzed FCS and incubated at 37°C for 5 h. After washing with PBS, cells were lysed in buffer (0.15 M NaCl, 1 mM EDTA, 50 mM Tris (pH 8), and 10 mM iodoacetamide) with protease inhibitors and detergent as described above. Cells were lysed and subsequently centrifuged at 14,000 × g for 30 min. Radioimmunoprecipitations were performed using mouse anti-α2-specific serum coupled to protein G-Sepharose beads (Pierce). A β2m mAb (Sigma) was used to deplete β2m from cell lysates.
For immunofluorescence assays, HeLaFcRn + β2m was grown on glass coverslips overnight. A total of 1 × 105 U937 and THP-1 were mounted onto adhesive microscope slides, air-dried, and fixed in 3.7% paraformaldehyde. After washes, cells were permeabilized with 0.1% digitonin in PBS for 10 min at room temperature, washed, and blocked for 30 min at room temperature with 10% heat-inactivated goat serum (Sigma) in PBS. Cells were then incubated with a mouse anti-α2-specific serum in PBS (1/250) containing 10% goat serum (Sigma) for 1 h at room temperature. Primary Ab was detected with an FITC-conjugated F(ab)2 goat anti-mouse Ab (1/100) for 1 h at room temperature. As a negative control, cells were incubated with normal mouse serum. Nuclei were stained with 0.1 µg/ml 4′6′-diamidino-2-phenylindole (Molecular Probes, Eugene, OR) in PBS for 5 min. After final washes, cells were mounted. Images were captured using a fluorescence microscope (Microphot FXA; Nikon, Tokyo, Japan) and processed with Adobe Photoshop 5.0. Positive samples and negative controls were viewed using the same contrast and brightness settings.
IgG Fc binding assays were performed as previously described (1, 18, 22) with the following modifications. Cells were lysed by shaking in sodium phosphate buffer (pH 6.0 or 7.5) with 0.5% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS; Sigma) and protease inhibitor cocktail on ice for 1 h. Postnuclear supernatants containing 0.5–1 mg of soluble proteins was diluted with an equal volume of sodium phosphate buffer containing 0.1% CHAPS and incubated with human IgG-Sepharose (Amersham Pharmacia Biotech) at 4°C for 4 h or overnight. The unbound proteins were removed with sodium phosphate buffer (pH 6.0 or 7.5) containing 0.1% CHAPS. The adsorbed proteins were eluted with sodium phosphate buffer (pH 8) or boiled with electrophoresis sample buffer at 100°C for 5 min. The eluted fractions were subjected to 12% reducing SDS-PAGE analysis. Proteins were visualized by Western blot using anti-α2-specific serum. For the blocking experiments, 250–500 µg/ml of human Fc or F(ab)2 (ICN Pharmaceuticals, Aurora, OH) were added to IgG-Sepharose beads before adding FcRn cell lysates. For the removal of CD64, CD32, and CD16 molecules, cell lysates (pH 7.5) were incubated with protein G that was previously incubated with mAbs specific for CD64, CD32, and CD16 (Caltag, Burlingame, CA) at 4°C overnight with shaking.
Cell surface biotinylation was performed as previously described (18). THP-1 and U937 cells (5 × 107) were suspended in 5 ml of PBS, pH 7.5, to which 2.5 ml of sulfo-NHS-biotin in PBS (1 mg/ml; Pierce) was added. The mixture was incubated at room temperature with rotation for 30 min. After washing with sodium phosphate buffer (pH 6.0) containing 0.1% CHAPS, the pellet was resuspended in 5 ml of sodium phosphate buffer (pH 6.0) with 0.5% CHAPS. A postnuclear supernatant was diluted 2-fold by sodium phosphate buffer (pH 6.0) with 0.1% CHAPS, then incubated with IgG-Sepharose. Following washings at pH 6.0, the bound protein was eluted in loading buffer at 100°C or with sodium phosphate buffer, pH 7.5. The eluted proteins were resolved by SDS-PAGE followed by blotting with streptavidin-HRP (Pierce). To confirm the specificity, the proteins eluted with sodium phosphate buffer were immunoprecipitated by mouse anti-α2-specific serum bound to protein G-Sepharose beads. Following incubation at 4°C on ice, the beads were washed, resuspended in loading buffer, resolved by SDS-PAGE, transferred onto nitrocellulose, and blotted with streptavidin-HRP. The final product was visualized using ECL (Pierce).
Surface and intracellular expressions of FcRn were examined in either fixed or permeabilized monocytes, macrophages, or THP-1 cells by flow cytometry. For staining, 1 × 106 cells were washed with PharMingen stain buffer (FBS; PharMingen, San Diego, CA), followed by blocking with PBS containing 10% normal goat sera (Jackson ImmunoResearch, West Grove, PA) on ice for 20 min. For surface Ag staining, 10 µl of diluted anti-α2-specific serum was added to each tube and incubated for 20 min at room temperature. Surface staining was also conducted at 4°C to minimize internalization, and the results were identical with those observed at room temperature (data not shown). For intracellular staining, the cells were first permeabilized with Cytofix/Cytoperm (PharMingen) on ice for 20 min and then washed with 1× perm/wash buffer. Anti-α2-specific serum was added as described. After washing, 20 µl of 1/50 diluted goat anti-mouse IgG-FITC Ab (Jackson ImmunoResearch) was added to each tube and incubated at room temperature for 15 min. After washing, cells were fixed with Cytofix and analyzed using a FACScan flow cytometer and CellQuest software (Becton Dickinson). The mouse IgG (0.5 µg/million cells) was used as a negative control.
Normal adult human small intestine was obtained from patients undergoing gastric bypass surgery under a protocol that was approved by the human studies committee of the Brigham and Women’s Hospital. Tissue was embedded in Tissue-Tek OCT compound (Sakura-Finite, Torrance, CA). Samples were sectioned on a Leach CM3050 cryomicrotome (Leica, Nussloch, Germany). A frozen section (5 µm) was air-dried at room temperature, fixed in 4% paraformaldehyde in PBS, washed in PBS, and blocked in 10% nonimmune goat serum (Zymed, South San Francisco, CA). Sections were stained with an affinity-purified FcRn-specific anti-peptide Ab (aa 174–188; provided by Dr. Neil Simister, Brandeis University) or against Ncl-Macro (Novocastra, Newcastle upon Tyne, U.K.) diluted in PBS containing 10% nonimmune goat serum and 0.02% Tween 20. Primary Abs were detected with appropriate fluorophore-conjugated secondary Abs for epifluorescence microscopy. All staining reactions were accompanied by a negative control that consisted of an affinity-purified, isotype-matched irrelevant Ab. Sections were mounted in ProLong antifade reagent (Molecular Probes, Eugene, OR) and viewed with a Zeiss Axiophot microscope (Zeiss, New York, NY) equipped with a Spot digital camera (Diagnostic Instruments, Sterling Height, CA). Electronic images were captured and edited using Adobe Photoshop. The sections were stained with either mouse anti-α2-specific serum or an FcRn-specific anti-peptide Ab and anti-CD68 (Santa Cruz Biotechnology, Santa Cruz, CA) using an avidin-biotin complex method (37).
To generate FcRn-specific serum Abs, we fused the codons corresponding to the α1, α2, or α3 domains of FcRn in-frame to the GST gene. The anti-GST Abs in the mouse sera were removed by incubation with GST bound to glutathione-Sepharose beads. Sera contained only Abs specific for the FcRn as shown by Western blot. We selected the GST-α2 for further immunization of rabbits. To show specificity of the mouse and rabbit anti-human α2-specific serum Abs, we probed cell lysates from 721.221, 721.221CD1d, Jurkat, and HeLa cells by Western blotting. The α2 domain-specific serum did not react with classical MHC or nonclassical MHC class I-like CD1d molecules (data not shown). Despite the 22–29% similarity between the α2 domains of MHC class I and FcRn (2), the anti-FcRn Abs recognized only a 45-kDa protein from HeLa cells transfected with plasmids encoding both FcRn heavy chain and human β2m as defined by Western blotting (Fig. 1) and immunoprecipitation of metabolically labeled protein (data not shown). Mock-transfected HeLa cells were negative.
Expression of FcRn heavy chains in immune cells was examined by RT-PCR from a variety of cell lines and from isolated monocytes, macrophages, and dendritic cells with FcRn-specific primers. The purity of the isolated cells is shown in Fig. 2. The results of RT-PCR screening are shown in Fig. 3. The amplified PCR products had a size similar to that of a product amplified from an FcRn-encoding plasmid or T84 cells, a polarized IEC line expressing functional FcRn (25). Moreover, the FcRn heavy chain mRNA detected in the human macrophage and dendritic cells had a DNA sequence identical with that previously described from human placenta as defined by sequencing five independent bacterial colonies (11) (data not shown). These results showed that human FcRn transcripts were expressed in monocytes, macrophages, and dendritic cells, but not in NK, T, and B cell lines (Fig. 3).
Western blotting studies confirmed these results. Fig. 4A shows that a band of 45 kDa was detected by the FcRn anti-α2-specific serum Abs in the promonocytic U937 and monocytic THP-1 cell lines as well as in freshly isolated monocytes, macrophages, and dendritic cells. A band was also observed in transfected, but not in untransfected, HeLa cells. It should be noted that the size of the band in the transfected HeLa cells was slightly larger than that of the band detected in the monocytic cell lines, monocytes, macrophages, and dendritic cells due to a Flag epitope that was inserted into the N terminus of the FcRn gene. Untransfected and transfected HeLa cells also exhibited two minor nonspecific bands with Western blotting that migrated above the 45-kDa specific band (Fig. 4A). However, these were not observed when immunoprecipitation of radiolabeled HeLahFcRn + β2m cells was performed (data not shown). Furthermore, Jurkat and untransfected HeLa cells lacked this 45-kDa band. In Jurkat, a band smaller than 45 kDa was detected, similar to a minor weak band present in the monocytic cell lines as well as macrophages and dendritic cells. This band probably represents a nonspecific immunoreactive protein given the absence of FcRn-specific mRNA in Jurkat cells (Fig. 3). In addition, immunoprecipitation of radiolabeled proteins from the Jurkat and Raji cell lines failed to show expression of FcRn (data not shown).
To further document the expression of FcRn in mononuclear cells, we performed immunofluorescence studies on the monocytic cell lines. Using the anti-α2-specific serum, we observed that THP-1 (Fig. 4B, panel C) and U937 (Fig. 4B, panel E) cell lines stained positively. For comparison, staining of the HeLaFcRn + β2m transfectants is shown in panel A. The negative control, normal mouse serum, failed to stain HeLa (data not shown), THP-1 (data not shown), or U937 cells (Fig. 4B, panel G), and the anti-α2-specific serum failed to stain the untransfected HeLa cell line (data not shown). Nuclear staining of all four cell lines is provided for reference ( panels B, D, F, and H).
FcRn is expressed as a 45-kDa membrane-bound heavy chain in nonconvalent association with 12-kDa β2m. Immunoprecipitation of [35S]methionine- and [35S]cysteine-labeled THP-1 and U937 cell lysates with an anti-α2-specific serum produced two bands, 45- and 12-kDa proteins, that were coimmunoprecipitated by the presence of anti-α2-specific serum, but not by preimmune serum (Fig. 4C). This is consistent with an association between the FcRn heavy chain and β2m in other cells. Further confirmation that the 12-kDa band was β2m was obtained when lysates from metabolically labeled THP-1 cells were subjected to three rounds of depletion with an anti-β2m mAb that removed the 12-kDa band from the autoradiogram. The molecular size of FcRn immunoprecipitated from U937 lysates was slightly larger than that of FcRn from THP-1. The explanation for this result is not clear. However, it may reflect different post-translational modifications of FcRn in the two cell lines, which are immortalized at different stages of monocyte maturation. We also observed that the associated β2m band did not appear to be stoichiometric with FcRn. Because β2m contains three methionine and cysteine residues compared with the nine such residues contained in human FcRn, we believe that this is an artifact of the metabolic labeling technique. Another possibility is that since mutations of the β2m molecule have been described in transformed cell lines (36), it may be that a β2m mutation has occurred in the monocytic cell lines, resulting in a low affinity of association with FcRn. Therefore, we sequenced the cDNA of β2m derived from the U937 cell line. The sequence aligned perfectly with the sequence of β2m deposited in GenBank (accession no. GI4757825), thus ruling out this possibility.
There is a large population of macrophages in the normal human intestinal mucosa (33). To determine whether FcRn is expressed in macrophages in vivo, we stained intestinal macrophages for FcRn and Ncl-Macro, a marker for human macrophages. Crypt and villus enterocytes exhibited a punctate apical membranous staining pattern for FcRn visible at the apical plasma membrane and in the apical cytoplasm (Fig. 5, a, e, and f, arrowheads) as previously described (25). Resident lamina propria macrophages also expressed FcRn (Fig. 5a, arrow). FcRn staining was absent from both enterocytes and macrophages in the presence of an irrelevant antiserum (Fig. 5b). Abs against Ncl-Macro specifically stained lamina propria macrophages (Fig. 5c, arrows), and this staining was not observed in the presence of an irrelevant isotype-matched mAb (Fig. 5d). Double labeling with both anti-FcRn and anti-NCL-Macro Abs revealed colocalization of FcRn and Ncl-Macro in lamina propria macrophages of the villi (Fig. 5e, arrows) and crypts (Fig. 5f, arrows). We also colocalized FcRn and macrosialin (CD68), an activated macrophage marker, in intestinal macrophages by an avidin-biotin complex method (37, 38). FcRn- and CD68 positively stained cells were clearly detectable in normal lamina propria of intestine with either the mouse anti-α2-specific serum or with an FcRn-specific anti-peptide Ab and an anti-CD68 mAb. Additionally, double-color staining revealed that some FcRn- and CD68-positive cells colocalized (data not shown). Therefore, the same result was obtained when two different Abs specific for FcRn in macrophages were used, thus confirming the expression of FcRn in tissue macrophages.
IgG binding assays were performed at both pH 6.0 and 7.5. Because macrophages and dendritic cells express conventional Fc receptors for IgG, which could confound the interpretation of functional IgG binding assays, we assessed pH-dependent binding by biochemical methods. FcRn was specifically immunoprecipitated from U937 and THP-1 cell lysates using human IgG bound to Sepharose 4B as the ligand at pH 6.0, but not at pH 7.5 (Fig. 6A). An ~32-kDa band was also detected in binding assays using the U937 cells at both pH 7.5 and 6.0. Because this band was detectable in FcRn-negative cells (data not shown), it is presumed that this represents a nonspecific precipitated protein. Isolated intestinal macrophages and monocyte-derived, peripheral blood dendritic cells also displayed the same pattern of pH-dependent binding (Fig. 6B). Because it is possible that FcRn failed to bind IgG at pH 7.5 due to competition from other FcγRs, especially high affinity FcγRI (CD64), we removed CD64, CD32, and CD16 molecules by incubating THP-1 cell lysates with excess amounts of anti-CD64, CD32, and CD16 mAbs immunoadsorbed to protein G at pH 7.5. Despite preclearing the THP-1 cell lysates of these IgG binding proteins, the 45-kDa protein binding of IgG could still not be detected at pH 7.5 (data not shown).
To further demonstrate that the Fc portion of IgG is responsible for the FcRn interaction, we performed an IgG binding assay at pH 6.0 in the presence of soluble human IgG Fc and F(ab)2. The results are shown in Fig. 7 and reveal that the binding of FcRn to IgG-Sepharose was inhibited by the presence of excess human IgG Fc fragments, but not by the presence of excess human IgG F(ab)2 (Fig. 7). Furthermore, this inhibition of IgG binding to FcRn by Fc fragments was concentration dependent, indicating that the binding of IgG Fc to FcRn in macrophage was specific and saturable.
Because FcRn binds IgG in a pH-dependent manner, it is important to know whether FcRn is expressed on the cell surface and/or intracellularly. First, cell surface biotinylation experiments were performed. Following cell surface biotinylation, FcRn could not be detected on the cell surface of monocytic THP-1 cells (Fig. 8A). The failure to detect FcRn on the cell surface may be associated with either the activation state or the degree of cellular differentiation, because THP-1 is a monocyte-like cell without complete maturation. PMA treatment can activate THP-1 cells with morphological changes consistent with differentiation. When THP-1 cells were labeled with biotin after PMA treatment, FcRn was readily detectable on the cell surface (Fig. 8A). Similar results were obtained by flow cytometry (Fig. 8B). Whereas resting THP-1 cells expressed FcRn solely intracellularly, THP-1 cells activated by PMA expressed FcRn both on the cell surface and intracellularly. This appearance of FcRn on the cell surface was detectable within 6 h of PMA activation and was sustained for up to 48 h. During this time period, intracellular levels of FcRn expression were maintained or even increased, suggesting that redistribution of FcRn to the cell surface was associated with increases in total cellular FcRn levels. Because PMA is reported to induce apoptosis in the HL-60 cell line (39), it is possible that apoptosis could result in leakage of cell membranes in THP-1 cells. However, we found that PMA-activated THP-1 cells did not stain with trypan blue (data not shown). These data suggest that the cellular distribution of FcRn may be regulated by either cellular maturation and/or activation in cells of the monocyte lineage.
To assess the in vivo relevance of these observations with the THP-1 cell line, we performed flow cytometry for FcRn expression on monocytes and macrophages whose purities were described in Fig. 2. This type of analysis provided the following results (Fig. 8C). Virtually all peripheral blood monocytes expressed FcRn intracellularly, and the majority (72.8%) exhibited detectable FcRn on the cell surface, albeit at lower levels (mean fluorescence intensity (MFI): surface, 9.02; intracellular, 22.56). These data indicate that monocytes uniformly express FcRn with the majority of the FcRn contained within intracellular compartments, and a lesser, but still substantive, proportion displayed on the cell surface on a majority of cells. Although the macrophages purified from the small intestine were also uniformly positive for FcRn expression (95.7% positive), only a small subset (23.2%) of these cells displayed FcRn on the cell surface. Interestingly, the MFI of these FcRn surface-positive macrophages was equivalent to that observed intracellularly (MFI: surface, 40.59; intracellular, 42.51). Thus, with differentiation to a macrophage, FcRn expression persists, but is redistributed intracellularly, except for a minor subset of cells that exhibits extremely high levels of surface FcRn expression. Taken together with the observations generated with the THP-1 cell line, these results suggest that the cellular distribution of FcRn is regulated in monocytes and macrophages. Moreover, they suggest that FcRn may function intracellularly and extracellularly in monocytes and predominantly intracellularly in the majority of tissue macrophages.
FcRn is highly expressed in mouse and rat during the first 3 wk after birth. In IECs it plays a major function in the passive acquisition of neonatal immunity. Following weaning, the expression of FcRn in the IECs is rapidly and profoundly diminished (1, 4). However, it is also known that FcRn expression persists into adult life in human IECs and in a limited range of other cell types in mammalian, including hepatocytes and endothelial cells (22–24). This expression beyond neonatal life is potentially relevant to other postnatal functions, including, importantly, the protection of IgG from catabolism.
This study examined the hypothesis that FcRn, an MHC class I-related Fc receptor for IgG, is functionally expressed in monocytes, tissue macrophages, and dendritic cells that are already well known to abundantly express other conventional FcRs for IgG. Our study for the first time has demonstrated that FcRn is expressed by monocytes, macrophages, and dendritic cells. The presence of FcRn heavy chain in macrophages from small intestine and dendritic cells was specifically demonstrated by RT-PCR amplification with FcRn-specific primer pairs (Fig. 3), Western blotting (Fig. 4A), and immunofluorescence staining with FcRn specific serum Abs (Fig. 4B) in vitro, and immunohistochemical colocalization of FcRn heavy chain with the macrophage-specific marker Ncl-Macro (Fig. 5) and CD68 (data not shown) in the lamina propria of human small intestine. We reason that macrophages in other tissues would also express FcRn, because monocytes express FcRn, although this should be further confirmed. Additional evidence to support this conclusion was that we were able to detect murine FcRn, a homologue of human FcRn, in a macrophage cell line, RAW264.7 (data not shown). The association between FcRn and β2m was also demonstrated in monocyte-like cell lines (Fig. 4C), proving that FcRn is structurally intact in this cell type. Therefore, our results support the previous finding that FcRn is expressed beyond neonatal life. In our examination of FcRn expression, we found that established cell lines derived from B lymphocyte, T lymphocyte, and NK cell lineages failed to express FcRn heavy chain (Fig. 3). However, we cannot exclude the possibility that FcRn is expressed in freshly isolated or activated T lymphocytes, B lymphocytes, and NK cells. We also do not know whether FcRn is expressed in other myeloid-derived lineages, such granulocytes and platelets. These issues will need further investigation.
FcRn binds IgG at acidic pH in macrophages and dendritic cells. As described in the intestine of neonatal rodent, FcRn binds IgG in the slightly acidic pH of gut lumen and releases IgG into the bloodstream of newborn animals at the neutral pH of the interstitium, pH 7.4 (1, 40). The amino acid residues isoleucine 254 and histidine 310 within the CH2 domain and the sequence -H-N-H-Y (aa 433–436) of the CH3 domain in mouse and human IgG1 appear to be of particular functional significance in this pH-dependent binding (41–43). Our results show that FcRn displays complete pH-dependent binding of IgG binding in monocyte-like cell lines and in vivo isolated macrophage and dendritic cells (Fig. 6). This pH-dependent IgG binding can be inhibited by Fc fragments that contain the IgG binding motifs, but not by Fab that do not contain these motifs (Fig. 7), supporting specificity for the Fc portion of IgG.
Studies on transcytosis of IgG through yolk sac (44) and human placenta (12, 13) have suggested that FcRn resides primarily within acidified vesicles where ligand binding is likely to occur after fluid phase uptake. Also, several in vitro studies that have modeled transcytosis of IgG in polarized epithelial cells support this idea (25, 45, 46). For example, pH gradient disruption in intracellular vesicles with bafilomycin A1 and monensin completely inhibited IgG transcytosis in a model human intestinal or rat kidney epithelial cell line (25, 46). We also reason that FcRn is likely to reside primarily within acidic vesicular compartments of cells of monocyte lineage. Our data support this conclusion, because FcRn was barely detectable on the cell surface of a resting monocyte-like cell line (Fig. 8A), and the majority of FcRn expressed by monocytes and tissue macrophages was intracellular (Fig. 8B).
Interestingly, FcRn could also be expressed on the cell surface. When the THP-1 cell line was treated with the phorbol ester, PMA, which also drives THP-1 differentiation toward a macrophage-like phenotype (47), FcRn was readily detectable on the cell surface (Fig. 8, A and B). Similarly, a significant fraction of peripheral blood monocytes and a subset of tissue macrophages were observed to express FcRn on the cell surface, albeit at lower levels than intracellularly, except in the case of the tissue macrophage subset that expressed extremely high levels. This suggests that the cellular distribution of FcRn may be related to the activation and/or differentiation state of the cell, which has not been previously appreciated in other cell types. Because FcRn binds IgG strongly in a pH-dependent manner, the appearance of FcRn on the cell surface would suggest that FcRn may be nonfunctional on the cell surface in terms of IgG binding under physiological conditions. However, it is possible that FcRn tethered on the cell surface of monocytes, macrophages, and dendritic cells might be functional in pathological conditions such as tissue inflammation (48, 49) and tumor infiltration (50, 51), where acidic conditions are created by alterations in tissue metabolism. The interstitial pH within solid tumors has been observed to be below physiological levels, ranging from 5.6 –7.7, which includes the pH optimum of FcRn binding (50). Macrophages are recruited in the earliest phases of inflammation such as inflammatory bowel disease (52), and they are widely infiltrated in solid tumor tissues (53).
Alternatively, the expression of FcRn on the cell surface may reflect other significant functions of FcRn on these cell types under physiological conditions: a role in shuttling IgG from the intracellular to extracellular milieu in protecting IgG from catabolism. With regard to IgG protection, there is a significant body of evidence that suggests that FcRn is directly involved in the control of serum IgG levels (14–17, 42). The proposed model is that pinocytotic vacuole formation by cells expressing FcRn results in uptake of IgG from surrounding fluids, and following a lowering of pH in early endosomes, some IgG molecules bind to FcRn. Enzymes present in organelles downstream of endosomes, such as lysosomes, digest the unbound IgG, but the IgG bound to FcRn is protected and recycled into the surrounding tissue fluid. Data to support this model are the decrease in serum half-life of IgG in β2m−/− mice (16), because loss of β2m presumably disables the function of FcRn, and the fact that mutated Fc fragments that exhibit a higher affinity for FcRn have a longer serum half-life than wild-type Fc fragments (42). Currently, the cell type responsible for this protection of IgG has not been clearly defined, although endothelial cells have been proposed. The monocytic U937 cell line was shown to be capable of recycling monomeric IgG by an unknown mechanism (54). Therefore, we reason that the expression of FcRn by virtually all monocytes in peripheral blood and the significant levels of FcRn expression detectable on the cell surface of this cell type may reflect a role of monocytic FcRn in the protection of IgG from catabolism and the maintenance of IgG levels in peripheral blood. Therefore, the prominent expression of FcRn on the cell surface of monocytes may reflect highly active sorting of IgG by FcRn from the endocytic pathway to the cell surface.
However, the relative distribution of FcRn on tissue macrophages was distinct from monocytes, with most FcRn in the former cell type intracellularly except for a small subset of cells that resembled the distribution of FcRn in monocytes, i.e. intracellular and cell surface (Fig. 8C). This suggests that the function of FcRn in most macrophages may be distinct and skewed toward protecting IgG from degradation intracellularly and thus prolonging the intracellular half-life of IgG. For a macrophage involved in Ag presentation, such a property may be advantageous, and this suggests that FcRn may influence Ag presentation. In macrophages and dendritic cells, FcγRs can promote the internalization of immune complexes into the endosomes, lysosomes, and MHC class II compartment (MIIC) to increase the efficiency of MHC class II presentation to CD4+ T lymphocytes (28, 55, 56). FcRn, in contrast, may influence Ag presentation pathways by protecting these immune complexes once inside cells in acidic compartments such as early endosomes (pH 6.0–6.5), late endosomes (pH 5.0–6.0), lysosomes (pH 4.5–5.0), and MIIC (57). Generally, antigenic peptides, which are ultimately associated with MHC class II molecules, are generated from internalized exogenous Ags by the movement of MHC class II sequentially through early endosomes, late endosomes, lysosomes, and MIICs (58). Several lines of evidence support this probability. First, because FcRn is able to bind immune complexes (59), it may be able to maintain high levels of these immune complexes at the sites of Ag processing. Second, FcRn binds IgG in the pH range of endosomes and lysosomes (pH 4.5–6.5; data not shown). Third, the appearance of a dileucine-based motif in the cytoplasmic tails of FcRn and the MHC class II-associated invariant chain suggests that FcRn and MHC class II molecules might be colocalized primarily in acidic compartments. The invariant chain has been shown to target MHC class II to acidic compartments (60). Therefore, the role of FcRn in protecting IgG may have an influence on Ag presentation in APCs such as macrophages and dendritic cells.
In summary, FcRn, the only known Fc receptor for IgG with MHC class I-like structure, is functionally expressed by monocytes, macrophages, and dendritic cells. Furthermore, the cellular distribution of FcRn expression on these cell types is regulated between intracellular and extracellular sites. These features of FcRn expression may confer upon monocytes, macrophages, and dendritic cells novel functions involving protection of IgG from catabolism that may relate to prolonging the IgG half-life in the extracellular (monocytes) and intracellular (macrophages and dendritic cells) milieu, which may impact the Ag presentation functions of these cells. Future studies must be aimed at testing these hypotheses.
We thank Dr. Sheldon Randall (Department of Surgery, Brigham and Women’s Hospital) for human intestinal tissue. We gratefully acknowledge the FcRn-containing plasmid and peptide Ab from Dr. Neil Simister. We thank Drs. Victor M. Morales and Neil Simister for critically reviewing the manuscript. We thank Dr. Kamran Badizadegan and Atul Bhan for advice about immunohistochemistry. We thank Dr. Tom Kupper for advice in purification of dendritic cells. We thank Drs. Mark Birkenbach, Paul Anderson, and Marco Colonna for the cell lines THP-1, NK3.3, and NKL. We also thank Dr. Jianhua Xu for human cDNA from PBMC, and Steven M. Claypool for technical help.
1This work was supported by research grants from the National Institutes of Health (DK/AI-53056 (to R.S.B. and W.I.L.), DK44319 and DK51362 (to R.S.B.), DK48107 (to W.I.L.), and DK-47322, AI-41530, DK-54495, and DE-72621 (to P.D.S.)), a Department of Veterans Affairs Merit Review Award (to P.D.S.), and the Harvard Digestive Disease Center. X.Z. was supported by a Career Development Award from the Crohn’s and Colitis Foundation of American.
4Abbreviations used in this paper: FcRn, neonatal Fc receptor; β2m, β2-microglobulin; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; MFI, mean fluorescence intensity; IEC, intestinal epithelial cell; MIIC, MHC class II compartment.