Cellular and Molecular Mechanisms of Inflammation: Receptors of Inflammatory Cells: Structure—Function Relationships

Receptors of Inflammatory Cells: Structure-Function Relationships is the first in a new serial on Cellular and Molecular Mechanisms of Inflammation. The purpose of this serial is to bring together the latest knowledge in various areas of research in this actively developing field around a topical focus. These volumes are not intended to present comprehensive reviews. Rather, each contribution is meant to be a status report from laboratories actively working in an area. This volume presents an analysis of the structure-function relationships of receptors. It is clear that the structure of receptors provides the initial guidance for numerous functions of each cell in the organism. Through an analysis of the submolecular features of the receptors that are responsible for the initiation of activity of diverse biochemical pathways within the cells, a molecular understanding of the all important initial, guiding events of cell functions will emerge. In the broad sense of cells involved in inflammation, this includes mitogenesis, gene transcription, generation of lipid metabolites and oxidants, clearance of molecules from the surrounding medium, and release of granular constituents from cytoplasmic vesicles into the external medium, among others. The contents of this first volume will serve as a foundation for the subject of the second volume, which is signal transduction. Four additional volumes are in preparation, including Endothelial Leukocyte-Adhesion Molecules, Leukocyte Adhesive Mechanisms in Inflammation and Immunity, a second volume on Signal Transduction, and Stimulation of Inflammatory Cells.

• Language: English
• Publisher: Academic Press
• Release date: Oct 22, 2013
• ISBN: 9781483191515

Book preview

1

Preface

This volume is the first in a new serial on Cellular and Molecular Mechanisms of Inflammation. The purpose of this serial is to bring together the latest knowledge in various areas of research in this actively developing field around a topical focus. These volumes are not intended to present comprehensive reviews. Rather, each contribution is meant to be a status report from laboratories actively working in an area. The editors accept the responsibility for bringing together a spectrum of contributions to provide the reader with knowledge in a given topic area of sufficient breadth to serve as a basis for further research initiatives. By avoiding any requirement for comprehensive review, by encouraging contributors to provide their expert viewpoint, and by reducing publication time to a minimum through the use of computer typesetting, each volume of this serial will prove to be a timely and useful contribution to the research community’s efforts in this area.

In the current issue, an analysis of the structure–function relationships of receptors is presented. It is clear that the structure of receptors provides the initial guidance for numerous functions of each cell in the organism. Through an analysis of the submolecular features of the receptors that are responsible for the initiation of activity of diverse biochemical pathways within the cells, a molecular understanding of the all important initial, guiding events of cell functions will emerge. In the broad sense of cells involved in inflammation, this includes mitogenesis, gene transcription, generation of lipid metabolites and oxidants, clearance of molecules from the surrounding medium, and release of granular constituents from cytoplasmic vesicles into the external medium, among others.

Needless to say, the contents of this first volume will serve as a foundation for the subject of the second volume, which is signal transduction. Four additional volumes are in preparation, including Endothelial Leukocyte-Adhesion Molecules, Leukocyte Adhesive Mechanisms in Inflammation and Immunity, a second volume on Signal Transduction, and Stimulation of Inflammatory Cells.

The editors are particularly grateful to the contributors for the promptness of their efforts. This is reflected in the fresh quality of each contribution. We also wish to thank Monica Bartlett for her precise clerical attention to each facet of the project.

Charles G. Cochrane

Michael A. Gimbrone, Jr.

CHAPTER 1

Fcγ Receptors: A Diverse and Multifunctional Gene Family

Joseph A. Odin, Catherine J. Painter and Jay C. Unkeless,     Department of Biochemistry, Mount Sinai School of Medicine, New York, New York 10029

Publisher Summary

Fc receptors for IgG (FcγRs) and soluble Ig-binding factors play important roles in immunity. These roles can be grouped into three areas: (1) cellular immune defense and lymphocyte regulation, (2) immunoglobulin transcytosis, and (3) autoimmune pathology. In cellular immune defense, upon cross-linking by IgG aggregates, FcγRs activate a variety of leukocyte responses—such as phagocytosis, antibody-dependent cell cytotoxicity (ADCC) and release of lysosomal hydrolases, reactive oxygen metabolites, arachidonate metabolites, and other mediators of inflammation. Soluble immunoglobulin-binding factors (IBFs) and cross-linking of cell-surface antibodies to membrane-bound FcγRs inhibit β cell differentiation. All FcRs, except CD23 (FcγRII), are members of the Ig supergene family and are homologous to each other. Most FcγRs are type I membrane glycoproteins with one transmembrane domain and a cytoplasmic domain. However, huFcγRIII-l is anchored in the neutrophil plasma membrane by a glycan phosphatidylinositol moiety. Low-avidity forms of membrane-bound FcγRs contain two extracellular Ig-like regions, whereas high-avidity forms contain three Ig-like regions. Assigning functions to individual FcγRs has been a challenging task because many share immunologically indistinguishable extracellular domains, within a subclass, and their cellular distributions overlap considerably. This chapter reviews the current state of knowledge of the structure, function, and signaling mechanisms of mouse and human Fcγ receptors.

Introduction

At a FASEB-sponsored conference in June of 1987, Fc receptors for IgG (FcγRs) were defined as a family of receptors that specifically bind IgG via the Fc domain and that mediate physiologic functions. Today, it is clear that these receptors and soluble Ig-binding factors play important roles in immunity. These roles can be grouped into three areas: cellular immune defense and lymphocyte regulation, immunoglobulin transcytosis, and autoimmune pathology. In cellular immune defense, upon cross-linking by IgG aggregates, FcγRs activate a variety of leukocyte responses, such as phagocytosis, antibody-dependent cell cytotoxicity (ADCC), and release of lysosomal hydrolases, reactive oxygen metabolites, arachidonate metabolites, and other mediators of inflammation. Soluble immunoglobulin-binding factors (IBFs) and cross-linking of cell surface antibodies to membrane-bound FcγRs inhibit B cell differentiation (Teillaud et al., 1987; Phillips and Parker, 1985). Though not yet tested, it seems likely that FcγRs on neonatal rat gut epithelium (Simister and Mostov, 1989) and syncytiotrophoblasts (Stuart et al., 1989) are involved in transcytosis of immunoglobulin. Dysfunction of macrophage FcγRs (Hoffman et al., 1989; Clarkson et al., 1986b) as well as the presence of high titers of anti-FcγR immunoglobulin (Sipos et al., 1988; Boros et al., 1990a) have been reported in both human and mouse autoimmune disease.

All FcRs, except CD23 (Fc∈RII), are members of the Ig supergene family and are homologous to each other. Most FcγRs are type I membrane glycoproteins with one transmembrane domain and a cytoplasmic domain. However, huFcγRIII-1 is anchored in the neutrophil plasma membrane by a glycan phosphatidylinositol (GPI) moiety (Selvaraj et al., 1988; Huizinga et al., 1988; Ravetch and Perussia, 1989; Scallon et al., 1989; Ueda et al., 1989). Low-avidity forms of membrane-bound FcγRs contain two extracellular Ig-like regions, whereas high-avidity forms contain three Ig-like regions. Assigning functions to individual FcγRs has been a challenging task because many share immunologically indistinguishable extracellular domains, within a subclass, and their cellular distributions overlap considerably. This review summarizes the current state of knowledge of the structure, function, and signaling mechanisms of mouse and human Fcγ receptors. Prior reviews cover older literature in greater depth (Unkeless et al., 1988; Anderson, 1989).

Nomenclature

A nomenclature for the family of Fc receptors was agreed upon in June, 1987. However, since then, numerous new FcγR genes have been isolated. The nomenclature we have adopted for this review is consistent with that originally agreed upon and also incorporates additional information. To save space, prior designations (pre-1987) (Unkeless et al., 1988) have not been included. The species of origin is designated by two lower case letters (e.g., mo for mouse and hu for human). The subscript Greek letter refers to the major class of immunoglobulin bound by the receptor. The Roman numeral refers to the distinct subclass of the designated receptor class. The subclasses are based on structural similarity and reactivity with specific monoclonal antibodies (mAbs). Any symbol following the Roman numeral refers to a distinct gene within that subclass. Subscript Arabic numerals designate a specific splice form of that gene. Because no consistent nomenclature exists for the newly discovered huFCγRII genes, we selected appropriate designations.

huFcγRI

FcγRs on human monocytes and macrophages were demonstrated by their ability to rosette IgG-sensitized erythrocytes in the absence of serum (Jandl and Tomlinson, 1958; Archer, 1965; LoBuglio et al., 1967). The rosettes were specifically inhibited by IgG or its Fc fragment. Monocytes phagocytosed erythrocytes coated with either anti-Rh0 IgG or nonimmune IgG, but did not phagocytose those coated with IgM or Fab fragments of IgG (Huber and Fudenberg, 1968). IgG subclass inhibition studies indicated that IgG1 and IgG3 were 1000-fold more effective inhibitors of phagocytosis than IgG2 or IgG4, and direct binding assays using radiolabeled IgG demonstrated the higher avidity of IgG1 and IgG3 (Alexander et al., 1978; Hay et al., 1972; Huber et al., 1971; Fries et al., 1982).

The histiocytic cell line U937 has about 18,000 high-affinity (Ka ∼10⁸ M−1), trypsin-insensitive IgG1-binding sites per cell (Anderson and Abraham, 1980) and is a good model for human monocytes, which have 20,200 sites with a Ka of ∼8.6 × 10⁸ M−1 for human IgG1 (Kurlander and Batker, 1982). High-affinity binding (10⁸–10⁹ M−1) of murine IgG subclasses IgG2a and IgG3 to both monocytes and U937 cells has been shown as well (Lubeck et al., 1985). The receptor, now termed huFcγRI (CD64), was purified by Sepharose–IgG affinity chromatography from the U937 cell line as well as from monocytes (Anderson, 1982; Cohen et al., 1983). It has an Mr as determined by sodium dodecyl sulfate and polyacrylamide gel electrophoresis (SDS–PAGE) of ∼72K. The broad band seen after SDS–PAGE was not affected by neuraminidase treatment, although the charge heterogeneity demonstrated by isoelectric focusing (IEF) and SDS–PAGE was significantly reduced (Anderson, 1982). Treatment of the receptor with endo-β-N-acetylglucosaminidase F or N-glycanase yielded a core protein of 40 or 50 kDa, respectively (Frey and Engelhardt, 1987; Peltz et al., 1988).

Several anti-huFcγRI mAbs have been characterized. mAbs 32 (Anderson et al., 1986) and 62 bind to an epitope distinct from that to which mAbs 22 and 44 bind (Guyre et al., 1989). None of these antibodies, however, is capable of blocking ligand binding. mAb 197, an IgG2a anti-FCγRI, blocks ligand binding, perhaps through binding of its Fc region to huFcγRI (Guyre et al., 1989). mAb 10.1 may bind near the ligand- binding site, as it can inhibit binding of immune complexes but not monomeric IgG (Dougherty et al., 1987). mAb FR51 or its F(ab’)2 inhibits the binding of both monomeric and aggregated IgG to U937 cells and the myeloblast cell line HL-60 (Frey and Engelhardt, 1987).

FcγRI is univalent for human IgG1 (O’Grady et al., 1986). Recent cloning of cDNAs for the huFcγRI showed that the extracellular domain contains six potential N-linked glycosylation sites and six cysteine residues, presumably disulfide linked to form three C2-set (Williams and Barclay, 1988) Ig-like regions. In contrast, huFcγRII and huFcγRIII encode only two Ig-like regions. The transmembrane domain is 21 residues and the cytoplasmic domain is short and highly charged (Allen and Seed, 1989). The mouse FcγRI is similar in structure (Sears et al., 1990). Homology also exists between the first two N-terminal external Ig-like regions of each FcγRI and the analogous domains of mouse and human FcγRII and huFcγRIII (Allen and Seed, 1989; Sears et al., 1990). The uniqueness of the third domain and its conservation between human and mouse, as well as preliminary mutational analysis (Allen and Seed, 1989), suggest that this third membrane-proximal domain of FcγRI endows high-affinity ligand-binding capacity.

The IgG site that interacts with FcγRI has been examined by a combination of techniques: aglycosylated ligand binding, mAb inhibition of ligand binding, and site-directed mutagenesis (Leatherbarrow et al., 1985; Burton et al., 1988; Duncan et al., 1988). IgG glycosylation is important for FcγR function. Aglycosylated murine IgG2b immune complexes bound poorly to the murine M1 macrophage cell line, were cleared slowly, and inefficiently mediated ADCC (Kurlander and Gartrell, 1983). Aglycosylated murine IgG2a did not bind to human monocytes, though complement fixation and activation were only slightly decreased (Leatherbarrow et al., 1985). Inhibition of monomeric human IgG binding to huFcγRI cells by mAbs directed against various IgG epitopes suggested that the binding site is low in the IgG hinge region (Burton et al., 1988). In support of this conclusion, site-directed mutagenesis of a single residue in the hinge region (IgG2b Glu²³⁵ → Leu²³⁵) converted a low-affinity IgG2b ligand to one with high avidity (Ka = 3.1 × 10⁸ M−1) for huFcγRI (Duncan et al., 1988).

The huFcγRI binding site may contain a readily oxidizable residue. Porphyrin photosensitization in vitro of monocytes and U937 cells selectively reduced murine IgG2a binding to huFcγRI (Krutmann et al., 1989). This was not due to loss of the receptor from the cell surface, as mAb surface staining was still possible with some anti-huFcγRI mAbs, which do not affect binding of ligand. Scavenger experiments suggest that generation of O2−. is responsible for the reduced binding.

It is not clear why there are multiple FcγRs, as many of the functions are subtended by more than one receptor. Indeed, several members of a Belgian family have a complete absence of huFcγRI expression on their peripheral blood monocytes (Ceuppens et al., 1985a,b, 1988). Two explanations of their apparent good health are as follows: (1) these individuals may possess a developmental defect such that huFcγRI is not expressed on monocytes but is expressed on tissue macrophages, where it functions normally; (2) an absence of huFcγRI may be of little consequence due to the redundancy of functions among leukocyte FcγRs (Unkeless, 1989a; Anderson, 1989).

Several investigators have shown that the expression of huFcγRI on monocytes and various related cell lines in increased from 2- to 10-fold upon incubation with interferon-γ (IFN-γ) (10–1000 U/ml) (Guyre et al., 1983; Perussia et al., 1983; Akiyama et al., 1984), and this effect is blocked by cycloheximide or actinomycin D (Perussia et al., 1983). Furthermore, IFN-γ treatment (at 50 ng/ml) of neutrophils, which normally do not express huFcγRI, resulted in the expression of ∼13,600 monomeric IgG-binding sites (Perussia et al., 1983). In a clinical study (Maluish et al., 1988), IFN-γ doses of 0.1 mg/m² were effective in elevating FcγR expression, as analyzed by binding of fluoroscein isothiocyanate (FITC)-labeled IgG (Guyre et al., 1983). In this study, however, objective toxicity included leukopenia and granulocytopenia. At a higher dose (0.25 mg/m²), the percentage of monocytes bearing FcγRs dropped from 48 to 11% during the 2-week course of daily treatment and dropped further to 2% by the second day posttreatment.

The IFN-γ-induced increased expression of high-affinity huFcγRI sites on monocytes and macrophages can be further augmented with dexamethasone (200 nM) (Crabtree et al., 1979; Girard et al., 1987), whereas this IFN-γ effect is abrogated by dexamethasone for HL-60 (Crabtree et al., 1979) and U937 (Shen et al., 1984) cell lines, as well as for neutrophils (Petroni et al., 1988). The positive effect of dexamethasone on IFN-γ-treated monocytes may be explainable by a dexamethasone-mediated increase in IFN-γ receptors (Strickland et al., 1986). Glucocorticoid treatment alone has been reported to decrease huFcγR expression in HL-60 and U937 cell lines (Crabtree et al., 1979; Shen et al., 1984), as well as on monocytes obtained following glucocorticoid therapy (Fries et al., 1983). Monocytes treated in vitro with glucocorticoid have an unaltered level of huFcγRI expression (Girard et al., 1987).

The effects of IFN-γ and dexamethasone on neutrophils shed light on the functional role of huFcγRI. Dexamethasone inhibited both the increased expression of FcγRI as well as the increased capacity for ADCC and phagocytosis demonstrated by IFN-γ-treated neutrophils. Dexa methasone reduction of IFN-γ-induced phagocytosis was more marked, however, than was the decrease in huFcγRI expression (Petroni et al., 1988).

Cross-linking of huFcγRI (by either ligand complexes or anti-huFcγRI mAbs) results in a number of functional responses; monomeric interactions have not been conclusively shown to generate any of these responses. In contrast to immune complexes, which are endocytosed rapidly (Kurlander and Gartrell, 1983; Segal et al., 1983; Jones et al., 1985b), monomeric ligand is not internalized or degraded through huFcγRI (Jones et al., 1985b). These results imply that huFcγRI, when occupied by monomeric ligand, does not recycle (Jones et al., 1985b). Cross-linking of huFcγRI using mAb 32 with a secondary anti-IgG reagent results in O2−. production (Anderson et al., 1986). Continuous O2−. production via huFcγRI is dependent on continuous de novo formation of cross-linked huFcγRI (Pfefferkorn and Fanger, 1989). HuFcγRI on monocytes, macrophages, and IFN-γ-treated neutrophils mediates phagocytosis of erythrocytes coated with heteroantibodies composed of Fab fragments of anti-huFcγRI mAb and Fab fragments of antierythrocyte antibody (Anderson and Shen, 1985).

Recent work has shown that huFcγRI is capable of ADCC toward target cells. The ADCC response is dependent on the effector cell type and maturity, as well as on the target cell type (Shen et al., 1989; Fanger et al., 1989; Graziano et al., 1989b). Inflammatory mediators may also be important, because exogenous C1q reconstitutes FcγR-mediated ADCC and phagocytosis in mouse peritoneal macrophages (Leu et al., 1989). HuFcγRI on monocytes and macrophages effects ADCC of both hybridoma and erythrocyte targets. IFN-γ treatment augments huFcγRI-mediated ADCC of monocytes and induces that of neutrophils (Shen et al., 1987, 1989; Fanger et al., 1989; Akiyama et al., 1984). Studies of ADCC with putative effector cell lines are largely unsatisfactory. Myeloid cell lines HL-60, U937, and THP-1 are unable to kill either erythroid or hybridoma targets, though IFN-γ treatment of these lines resulted in cytotoxicity against erythroid targets. When further differentiated (by 2-day culture), THP-1 cells exhibited slight huFcγRI-mediated cytotoxicity (Fanger et al., 1989).

huFcγRII

A second subclass of human FcγRs, huFcγRII (CD32), was initially identified by affinity chromatography of U937 lysates on IgG–Sepharose (Anderson, 1982). The anti-huFcγRII mAb IV.3 (Rosenfeld et al., 1985) immunoprecipitates an antigen of about 40 kDa. huFcγRII is found on monocytes, neutrophils, platelets, B cells, eosinophils (Kulczycki, 1984; Looney et al, 1986b), basophils (Anselmino et al, 1989), and trophoblasts (Stuart et al, 1989). The receptor binds aggregated IgG with low avidity (Ka = 1–3 × 10⁶ M−1) and does not bind monomeric IgG (Jones et al, 1985a). The affinity with which huFcγRII on platelets binds IgG subclasses is as follows: IgG1 = IgG3 > > IgG2 = IgG4 (Karas et al, 1982). The huFcγRII-positive cell lines Daudi and K562, which do not express other FcγR subclasses, will not form erythrocyte and IgG (EIgG) rosettes using erythrocytes coated with highly purified preparations of either human IgG2 or IgG4 (Walker et al, 1989). Similarly, neutrophils do not bind dimeric IgG complexes containing human IgG2 or human IgG4 (Huizinga et al, 1989a). In addition, only murine IgG1-sensitized erythrocytes were bound by U937, Daudi, or K562 cells, whereas previous work suggested that both murine IgG1 and IgG2b bound to huFcγRII (Looney et al, 1986a).

huFcγRII has two allotypes that differ in their affinity for murine IgG1. Homozygosity in the allotypic form that binds with lower affinity can be detected in screening assays by the failure of human T cells to respond to murine IgG1 anti-CD3 mAbs, such as mAb Leu 4, in the presence of FcγRII-bearing accessory cells (Unkeless, 1989a). Murine IgG2b can also be used to identify nonresponders in the T cell proliferation induction assay. The allotype that binds murine IgG1 has Arg¹³³ substituted for His¹³³ (Clark et al, 1989) and has a somewhat different isoelectric focusing pattern (Looney et al, 1988). The anti-huFcγRII mAb 41H. 16 detects the mAb Leu 4-responsive form of huFcγRII (Micklem et al, 1990).

The observation that mAb IV.3 did not react with Daudi cells, although a 40-kDa huFcγR could be immunoprecipitated with a polyclonal anti-huFcγRII serum (Looney et al, 1986a), suggested the possibility of isotypic variation. mAb IV.3 reacts with the 40-kDa receptor on neutrophils, macrophages, and platelets, and mAbs 41H.16, KuFc79, and KB61 recognize another 40-kDa molecule on B cells, neutrophils, and macrophages (Vaughn et al, 1985; Antoun et al., 1989). Recent studies have shown that some cross-reactivity for huFcγRII forms is evident among these and other huFcγRII-specific mAbs (Micklem et al, 1990; Gosselin et al, 1990). Initial cDNA clones of huFcγRII appeared to be nearly identical products of a single gene with no differential splicing (Hibbs et al, 1988a; Stuart et al., 1987). Subsequently, additional cDNA clones were isolated, showing that at least three genes encode huFcγRII proteins (Seki, 1989; Stuart et al., 1989; Brooks et al., 1989; Stengelin et al, 1988). All of the huFcγRIIs have homologous extracytoplasmic domains and are most homologous to moFcγRIIβ, especially huFcγRIIb. No common nomenclature exists for these three genes, so we have simply called them huFcγRIIa (the original huFcγRII gene cloned), hufcγRIIb [called huFcγRIIb in Brooks et al. (1989) and huFcγRIIC in Stuart et al. (1989)], and huFcγRIIc [called huFcγRIIa’ in Brooks et al. (1989) and huFcγRIIB in Stuart et al. (1989)].

The cellular distribution of each huFcγRII mRNA transcript was analyzed. huFcγRIIa was found in neutrophils, cultured adherent monocytes, chronic myelogenous leukemia cells, various monocyte-like cell lines, dimethyl sulfoxide (DMSO)-differentiated HL-60, and the erythroleukemic cell line K562 (Brooks et al, 1989). Of five lymphoid cell lines (Daudi, Raji, AW Ramos, IM-9, and MOLT-4), only the Burkitt lymphoma cell line Daudi expresses huFcγRIIa. huFcγRIIa transcripts alone were expressed on K562 and DMSO-treated HL-60. The cellular distribution of huFcγRIIb includes B cells, neutrophils, and cultured adherent monocytes (Brooks et al., 1989). Both huFcγRIIb and muFcγRIIβ undergo differential splicing in their cytoplasmic domains (Brooks et al., 1989). The splice form of huFcγRIIb detected in each cell type was not determined. huFcγRIIc is nearly identical to huFcγRIIb in its signal sequence and extracytoplasmic domains, but in its cytoplasmic domain and 3′ untranslated region, huFcγRIIc has high homology to huFcγRIIa, which made Northern analysis complex. The distribution of huFcγRIIc includes B cells, cultured adherent monocytes, neutrophils, and U937 cells, but not T lymphoid cell lines (Brooks et al., 1989). In situ hybridization studies showed that huFcγRIIb is present in syncytiotrophoblast cells of placenta (Stuart et al, 1989).

Fluorescence-activated cell-sorting (FACS) analysis demonstrated that huFcγRII on B cells is strongly recognized by mAbs 41H.16 and KB56 and weakly recognized by mAbs 2E1 (Micklem et al, 1990) or IV.3 (Gosselin et al, 1990). mAbs 41H.16 and KB56 apparently specifically immunoprecipitate a single antigen of 41 kDa from B cells (Micklem et al, 1990). This most likely is huFcγRIIc, because huFcγRIIa is not expressed in most B cells and huFcγRIIb mRNA levels are barely detectable (Brooks et al., 1989). mAbs 2E1 and IV.3 do, however, strongly stain monocytes (Micklem et al, 1990), probably primarily due to recognition of huFcγRIIa.

On platelets, neutrophils, and monocytes, all CD32-specific mAbs immunoprecipitate antigens between 40 and 42 kDa (Micklem et al., 1990). However, using a monocyte-like cell line, it is possible to detect slight differences in the Mr among proteins immunoprecipitated by different CD32-specific mAbs. mAbs KB61 and 41H.16 both immunopre-cipitate antigens of about 37 and 41 kDa from U937 cell lysates, whereas mAbs 2E1 and IV.3 immunoprecipitate an antigen of 42 kDa (Micklem et al., 1990). Unexpectedly though, a mAb KB61 affinity column precleared all three antigens, including the 42-kDa form, from U937 cell lysates. Thus KB61 must have some low affinity for the 42-kDa antigen as well. Competition experiments did support this assumption and showed that mAb IV.3 has some reactivity for the lower Mr antigens on U937 cells.

Sequence analysis of peptides of two proteins (37 and 41 kDa) immunoprecipitated from hairy cell spleen extracts by mAb KB61 suggested they were alleles of the same gene with different levels of glycosylation (Micklem et al., 1990). Five peptide sequences were obtained. Four peptide sequences, common to both proteins, align with sequences of huFcγRIIb1 or huFcγRIIb3, which are splice forms of huFcγRIIb (Brooks et al., 1989) that only differ in their signal sequences. However, a fifth peptide only found in the 37-kDa protein had a slightly different sequence, not seen in any huFcγRII cDNA, which possibly eliminated a potential N-linked glycosylation site.

On the erythroleukemic cell line K562, mAbs KB61 and 41H.16 were noted to immunoprecipitate a 41-kDa protein, whereas mAb IV.3 immunoprecipitated a 42-kDa protein (Antoun et al., 1989). In contrast to results obtained using U937 cells (Micklem et al., 1990), preclearing studies of K562 cell lysates using affinity chromatography did not reveal any cross-reactivity between mAbs 41H.16 and KB61 and mAb IV.3. If K562 lysates were extensively precleared (three passes) using a mAb 41H.16 affinity column, mAb IV.3 still precipitated a 42-kDa protein and vice versa, albeit the amount of material immunoprecipitated after preclearing was less than without preclearing.

Numerous functional studies have been done involving huFcγRII, though most were done before the existence of multiple isotypes was known. huFcγRII plays a role in ADCC by neutrophils and monocytes, as shown by the killing of IV.3 mAb-bearing hybridomas (Graziano and Fanger, 1987) and murine IgG1-coated erythrocytes (Boot et al., 1989), respectively. Upon short-term IFN-γ or granulocyte–monocyte colony-stimulating factor (GM-CSF) treatment (6 hr) of neutrophils or GM-CSF treatment (6 hr) of eosinophils, the FcγR-mediated cytotoxic potential of these cells is activated, although neither lymphokine increases receptor number (Graziano et al., 1989a). Stimulation by GM-CSF of both cell types occurred earlier than IFN-γ activation of neutrophils. Killing of specific anti-FcγR hybridoma cell lines by neutrophils incubated with IFN-γ or GM-CSF for 6 hr was solely mediated by huFcγRII (Graziano et al., 1989a), although after longer IFN-γ incubation (18 hr), a huFcγRI- mediated component of ADCC was observed (Graziano and Fanger, 1987). GM-CSF-treated (6 hr) eosinophils also mediate ADCC solely through huFcγRII (Graziano et al., 1989a). IFN-γ treatment (6 hr) of eosinophils did not activate ADCC, but no studies have been done with prolonged IFN-γ exposure (Graziano et al.,