Specific binding of bovine, ovine, caprine and equine IgG subclasses to defined types of immunoglobulin receptors in gram-positive cocci

One hundred and twenty bacterial strains were tested for non-immune binding of radiolabelled bovine, ovine, caprine and equine immunoglobulins. Bacteria possessing previously defined IgG receptors interacted in a well defined manner with purified IgG subclass immunoglobulins. Human group C and G streptococci carrying IgG receptors type III were capable of binding all IgG subclasses in the four mammalian species studied. Protein A-containing staphylococci demonstrated a restricted specificity with binding of bovine IgG1, ovine IgG1, caprine IgG1 and IgG2 as well as equine IgG(ab). Group A streptococci which can bind human IgG did not show specific reactivity. A new type of binding unrelated to the regular Fc-mediated binding was observed with equine IgG(T).The differences in specificity for IgG subclasses suggest that structures with binding capacity to streptococcal type III Fc receptors are different from staphylococcal protein A reactive sites. Inhibition experiments performed with purified immunoglobulins showed that individual IgG subclasses differed greatly in their inhibiting capacity reflecting differences in avidity.The high avidity and the broad, unrestricted immunoglobulin G reactivity of streptococcal IgG receptor type III indicate that human group C and G streptococci may provide a valuable tool for solid phase absorption of immunoglobulins from several mammalian species.     

Mononuclear phagocyte receptors in rhesus monkeys (Macaca mulatta) and their role in hemolytic disease of the newborn

Hemolytic disease of the newborn does not develop in rhesus monkeys because placentally-transferred maternal antibodies do not induce immune clearance of the newborn's erythrocytes. In an in vitro RBC adherence assay, rhesus peripheral blood monocytes did not bind newborn's RBC which had been coated in utero or in vitro with maternal antibodies. Nevertheless, rhesus phagocytes possess receptors that are specific for the Fc portion of IgC and for the C3b. Using purified human IgG subclasses as inhibitors of RBC adherence, rhesus Fc receptors preferentially bind IgG1 and IgG3. Thus, it may be that maternal antibodies are non-opsonic because they belong to IgG subclasses that do not bind effectively to rhesus Fc receptors. Also, RBC adherence appears to be controlled by the level of antibody coating which in turn is determined by avidity of the antibodies and by the number of RBC membrane determinants. The failure of maternal antibodies to opsonize the newborn's RBC and thus cause hemolytic disease is very likely due to the low avidity of antibodies and to the weak expression of blood group determinants on the membranes of these RBC.     

Heterogeneity of human lymphocyte Fc receptors: Studies using heat-aggregated and antigen-complexed IgG from human, rabbit, guinea pig, horse and goat

We have studied the ability of human peripheral blood lymphocytes (HuPBL)⁴ to interact with IgG from several animal species. Three functions or activities that are reported to depend on an interaction between complexed IgG and HuPBL receptors (R) for the Fc piece of IgG (FcγR) were compared: (1) antibody-dependent cell-mediated cytotoxicity (ADCC); (2) binding of heat-aggregated IgG (aggG); and (3) rosette formation with IgG-sensitized erythrocytes [RBC-A(γ)]. IgG (and IgM) antibodies to chicken erythrocytes (CRBC) were purified from the sera of the following species after injection with CRBC stroma: (1) horse (Ho); (2) goat (Go); (3) rabbit (Ra); and (4) guinea pig (Gp). Good IgG-agglutinating antibody titers were obtained from each injected species.

Using ⁵¹Cr-labeled CRBC targets and HuPBL effector cells, only Ra anti-CRBC IgG gave good ADCC at high dilutions. Ho and Go anti-CRBC (IgG) failed to give A C , and p anti-CRBC (IgG) gave approx. 30% of the level of kill as Ra. Ra Fab'₂ fragments of IgG antibody failed to produce ADCC.

Treatment of HuPBL with Ra anti-lymphocyte serum (ALS) almost totally ablated ADCC, whereas HoALS failed to alter ADCC. Pretreatment of HuPBL with aggG showed that Ra or Hu aggG gave essentially equal inhibition of ADCC, Gp gave approx. 30% of the degree of inhibition as Hu and Ra, and Ho or Go aggG had essentially no effect of ADCC. These results confirmed the following order of ability of IgG to interact with HuPBL ADCC killer (K) cells: (Hu )Ra > Gp Ho, Go. The data suggest that Gp IgG interacts with only a subpopulation (≈ 30%) of HuPBL K cells.

The binding of aggG to total HuPBL failed to strictly correlate with the ADCC results or with the results of rosette formation between total HuPBL and CRBC-A(γ). The observations suggest that there is a heterogeneity of FcγR between K and non-K cell subpopulations of HuPBL both in terms of the type of complexed IgG they are able to bind, and in terms of the species of origin of the IgG. The data also support contentions that FcγR that bind RBCA(γ) complexes differ from those that bind aggG.     

The subclass heterogeneity of pig IgG

The recognition and distinction of pig IgG subclasses is a difficult task because no pathological monoclonal immunoglobulins of pig origin are available. Attempts at elucidating the subclass composition of pig IgG based on subfractionation of pooled IgG and preparation of antisera has led to only partial success. Chemical evidence for heterogeneity of the constant portion of pig gamma-chains pointed to the existence of more than one subclass; a final answer to the question of the number of subclasses could not be obtained. Functionally different antibodies of the IgG class may be isolated from immune sera collected at various stages of the immune response. Precipitating and nonprecipitating anti-dinitro-phenyl (DNP) antibodies differ chiefly in their capacity for forming insoluble complexes with DNP-carrier proteins. Examination of these antibodies by physical methods showed that the precipitating antibody is less compact, more flexible and that the distance between the two binding sites on the same molecule are greater than for the nonprecipitating anti-DNP antibody. The precipitating antibody is very resistant to cleavage by pepsin compared to the non-precipitating antibody. This resistance, together with the differences in the molecular parameters described above provides evidence that precipitating and nonprecipitating anti-DNP antibodies in swine are representatives of separate IgG subclasses. Mouse monoclonal antibodies to pig IgG show a degree of selectivity for certain IgG subpopulations when tested against pooled pig IgG. The limited number of monoclonal antibodies currently available does not allow final conclusions to be drawn on the number of pig IgG subclasses.     

Immunoglobulin G subclass distribution in canine leishmaniosis: a review and analysis of pitfalls in interpretation

Infection with Leishmania may have different outcomes in genetically distinct individuals and the course of infection is determined by the nature of the host innate and adaptive immune response. Thus in experimentally infected mice, and in naturally infected dogs or humans, the protective (self-healing or asymptomatic) phenotype is associated with the induction of Th1-regulated cell-mediated immunity. By contrast, a Th2-regulated humoral immune response is associated with severe symptomatic disease. In the murine model system there is strong correlation between clinicopathological phenotype and the nature of the antigen-specific humoral immune response. Symptomatic infection and Th2-regulation is associated with elevation in antigen-specific IgG1 and IgE, whereas asymptomatic infection with Th1-regulation is linked with IgG2a production. IgG subclass restriction is less clear in human disease with only some clinical forms being correlated to a specific serological profile. Although numerous studies have questioned whether infected dogs develop skewed IgG subclass usage, the results of these have been conflicting-suggesting bias towards IgG1 or IgG2 or neither subclass in different investigations. This confusion could relate to the specificity of the commercially available polyclonal antisera used to detect the canine IgG1 and IgG2 subclasses. More meaningful results might be obtained by the use of the panel of monoclonal antibodies with well-validated specificity for all four canine IgG subclasses.     

Serum containing ovine IgG2 antibody specific for maedi visna virus envelope glycoprotein mediates antibody dependent cellular cytotoxicity

Antibody-dependent cell-mediated cytotoxicity (ADCC) specific for maedi visna virus (MVV) has never been described. The IgG antibody response to MVV is restricted to an IgG1 response whilst MVV specific IgG2 is never seen in persistently infected sheep. To determine whether the isotypic restriction of the antibody response is responsible for the lack of ADCC, an ADCC assay was developed using polyclonal serum raised to recombinant MVV ENV protein. Sheep immunised with a recombinant GST:SUenv fusion protein in complete Freund's adjuvant produced an antibody response which contained IgG1 and IgG2 antibodies. The activity of this serum in an ADCC assay was compared to serum from persistently infected sheep. Serum from immunised sheep mediated ADCC reactions whilst no activity was ever seen in persistently infected sheep serum. IgG2 may therefore be the possible effector isotype for ADCC reactions against MVV. Failure of the IgG2 dependent ADCC system in vivo may contribute to the persistence of MVV-infected macrophages in vivo.     

IgG subclass responses in a longitudinal study of canine visceral leishmaniasis

Visceral leishmaniasis (VL), caused by Leishmania infantum, is an important disease of domestic dogs. Here, we present data on the IgG subclass antibody response to crude L. infantum antigen in a cohort of naturally infected Brazilian dogs. Specific IgG1-IgG4 responses could be detected in 98, 58, 70 and 82%, respectively of 57 dogs that were seropositive for specific IgG. Levels of all IgG subclasses were strongly inter-correlated. Levels of all IgG subclasses increased at the time of seroconversion, before reaching a plateau after several months. Levels of all IgG subclasses were higher in sick dogs than healthy dogs, and levels of all except IgG2 were higher in parasite-positive (by PCR) than parasite-negative dogs. However, levels of IgG2 relative to IgG1 were lower in sick or parasite-positive dogs compared to healthy or parasite-negative infected dogs. In contrast to previous studies, the results suggest that canine VL is associated with upregulation of specific antibody of all IgG subclasses, particularly IgG1, IgG3 and IgG4.     

Canine heterophilic antibodies as a source of false-positive B-type natriuretic peptide sandwich ELISA results

BACKGROUND: Interference by heterophilic antibodies is a well-known cause of false-positive sandwich ELISA results in human medicine. They are considered rarely in veterinary species and have not been characterized but could become important as newer, highly sensitive sandwich immunoassay technologies are developed. OBJECTIVES: The goals of this study were to use a B-type natriuretic peptide (BNP-32) sandwich ELISA to determine the effect of heterophilic antibodies on test performance; to characterize canine heterophilic antibodies; and to develop and test a method for heterophilic antibody removal. METHODS: A sandwich ELISA was developed using a mouse IgG(1)K monoclonal and a rabbit polyclonal antibody to two synthetic peptides of canine BNP-32. The effects on false-positive results of heterophilic antibody depletion and blocking by various techniques were compared. The titers of canine heterophilic antibodies were compared with various blood antigens from other species and the relative amount of canine IgG was compared with that of IgM heterophilic antibody. RESULTS: Heterophilic antibodies in dog plasma were shown to be capable of causing false-positive ELISA results. They reacted with blood proteins from a variety of animal species at relatively low titers and consisted of both IgG and IgM. Protein A agarose antibody precipitation, in conjunction with mouse IgG(1)K blocking antibody, was effective in eliminating false-positive sandwich ELISA results while retaining adequate test performance. CONCLUSIONS: Canine heterophilic antibodies can interfere with sandwich ELISA assays and cause false-positive test results. An effective technique for their removal that has a potentially broad application was developed, and allows measurement of canine blood constituents at low picomolar concentrations.     

Investigations on the nature and pathogenicity of circulating antikeratinocyte antibodies in dogs with pemphigus foliaceus

In humans with pemphigus foliaceus (PF), pathogenic autoantibodies are principally of IgG4 subclass and they cause superficial vesiculation when injected into neonatal mice. The objectives of this study were to determine the isotypes of circulating antikeratinocyte antibodies in dogs with PF, to assess whether serum antikeratinocyte antibody titres decreased during successful treatment, and to study whether such antibodies were pathogenic in passive transfers. Using indirect immunofluorescence with neonatal mouse skin substrates, circulating antikeratinocyte IgG antibodies were detected in 36 of 44 dogs with PF (82%). Serum autoantibodies belonged predominantly to IgG4 (three of 44; 80%) and IgG1 (30 of 44; 68%) subclasses. Antikeratinocyte IgG antibodies were detected in 16 of 20 normal dogs (80%), and these antibodies were IgG1 (16 of 20, 80%) but rarely IgG4 (two of 20; 10%) isotypes. In four dogs, IgG4 antikeratinocyte antibody titres decreased concomitantly to lesions nearing or reaching complete remission. In contrast, IgG or IgG1 titres remained stable or increased when lesions abated. Antikeratinocyte antibodies targeted mainly intercellular autoantigen(s) in the stratum granulosum, while in fewer dogs, such antibodies bound to cytoplasmic basal antigen(s). Intradermal injections of PF or pemphigus vulgaris (PV) IgG into neonatal mice caused subgranular or suprabasal acantholytic vesiculation without granulocyte infiltration, respectively. Similar transfers of normal dog IgG did not cause vesiculation. These observations suggest that antikeratinocyte IgG4 antibodies could be relevant to disease pathogenesis. Importantly, canine PF or PV IgG appear to be pathogenic when transferred passively into mice, causing vesiculation at epidermal levels similar to those of the natural disease.     

DNA binding proteins in canine sera. A method for removal of nonspecific DNA binding in the Farr assay

A panel of canine sera, the majority of which were collected from clinically healthy dogs, were investigated for antibodies against double stranded (dsDNA) by the Farr radioimmunoassay technique. Non-specific DNA binding agents interfering with the Farr assay were detected in all sera. Heat inactivation at 60 degrees C or treatment with dextran sulphate was shown to eliminate this kind of unspecific DNA binding while not affecting true antibodies to dsDNA. Canine sera positive in the Farr assay after inactivation at 60 degrees C were positive also in immunofluorescence for anti-nuclear antibody on rat liver sections and for dsDNA with Chrithidia luciliae as antigen preparation. IgG or glycoprotein nature of the non-specific DNA binding could be excluded by means of affinity chromatography on protein A and the lectin lentil.     

Cloning and characterization of ovine immunoglobulin G Fc receptor III (FcγRIII)

Receptors for the Fc regions of immunoglobin G (IgG) play a critical role in immunoregulation and immune defenses against pathogens. In this study, we describe the cloning, eukaryotic expression and IgG subclass specificity of ovine Fc gamma receptor III (FcγRIII). The newly cloned ovine FcγRIII cDNA contains a 940 bp open-reading frame (ORF), and is predicted to encode a 250 amino acid transmembrane glycoprotein composed of two immunoglobulin-like extracellular domains, a transmembrane region and a short cytoplasmic tail. The overall identity of the ovine FcγRIII amino acid sequence to its cattle, pig and human counterparts was 83.2%, 62.0%, 60.7%, respectively. Overlapping PCR was performed with the extracellular domain of ovine FcγRIII and the transmembrane and intracellular region of ovine Fc gamma chain to construct a chimeric receptor. Rosetting analysis showed that transfected COS-7 cells required Fc receptor gamma chain for the expression of FcγRIII on the surface. COS-7 cells expressing FcγRIII were able to bind chicken erythrocytes sensitized with ovine IgG1, but not IgG2. Identification of ovine FcγRIII will further our understanding of the ovine immune system.     

Evaluation of enzyme-linked immunosorbent assay for quantitation by subclass for bovine antibodies.

The enzyme-linked immunosorbent assay (ELISA) was evaluated for use in the quantitative measurement of bovine immunoglobulin IgG1 and IgG2 antibodies. A method for standardization was devised in which IgG1 or IgG2 was directly adsorbed to polystyrene tubes and the actual degree of binding was calculated by using different input amounts of 125I-labeled IgG1 or IgG2. Values for quantity of IgG1 antibodies to human serum albumin were only slightly higher when measured by the ELISA than when measured by quantitative precipitation although the value measured by the ELISA for IgG2 antibodies was twice that determined by quantitative precipitation. This discrepancy could result from conjugate cross reactivity, differences in affinity between antibodies of the 2 subclasses, or the occurrence of IgG2 nonprecipitating antibodies. The danger of overlooking subclass anti-globulin cross reactivity because of the failure to detect it by immunoprecipitation, also is illustrated. In addition, only enzyme-antibody conjugates prepared with specifically purified antibodies were effective, and reproducibility of individual data points required that 4 replicate determinations be performed. Advantages, pitfalls, and limitations of the ELISA are discussed.     

Immunochemical studies to purify rabbit and chicken immunoglobulin G antibody by protein A-Sepharose chromatography

The immunoglobulin (Ig) G binding properties of protein A made affinity chromatography with protein A an acceptable method for isolation of IgG in mammals. Rabbit IgG was isolated by fractionation of serum on a column containing protein A coupled to Sepharose. The IgG molecules of subclasses 1, 2, and 4 and their fragments containing the Fc region had this affinity, but not IgG-3. The second peak, after washing with 1M acetic acid eluate (pH 7.0) as shown in the Ouchterlony agar immunodiffusion test, was the IgG fraction. Similar studies on normal and hyperimmune chicken sera indicated that chicken IgG does not bind to protein A molecules. Further studies indicated that chicken plasminogen did not bind covalently to protein A molecules and subsequently did not interfere in the affinity of protein A to bind to IgG molecules. The inhibition of binding of chicken IgG to protein A molecule is still unknown.     

Temporal IgG subclasses response in dogs following vaccination against Leishmania with Leishmune®

Canine Visceral Leishmaniasis (CVL) is a widespread zoonotic disease with mandatory euthanasia of infected dogs determined by the Brazilian Ministry of Health. Development of vaccines against CVL may provide a prophylactic barrier, but transitory peak of antibody response detected by standard diagnostic techniques in vaccinated dogs may be interpreted as natural infection. Accordingly, the aim of the present study was to sequentially evaluate total and IgG subclasses response between naturally Leishmania-infected and dogs vaccinated with Leishmune(?). A total of 172 mongrel dogs were divided in four groups: Group 1 (G1) with 45 clinically healthy dogs, Group 2 (G2) and Group 3 (G3) with 45 dogs naturally infected by Leishmania sp. each, symptomatic and asymptomatic respectively, and G4 (G4) with 37 healthy dogs submitted to a complete protocol of a commercially available vaccine against CVL, monitored and evaluated in 5 different chronological moments (M0-M4) up to 180 days after M0. Total IgG, IgG1 and IgG2 were unable to differentiate between infected (G2 and G3) and vaccinated (G4) dogs, demonstrating that polyclonal commercial antibodies do not distinguish these groups apart. Total and IgG subclasses antibodies were not detected until 21 days of the second vaccination dose; however, seroconversion was observed on 21 days and sustained positivity up to 6 months after the vaccination start. A peak of antibodies response was observed on 90 days (M3), when results for total IgG, IgG1, IgG2, IgG3 and IgG4 where highly significant when compared to M0 (P<0.0001). Neither total IgG nor IgG1 effectively differentiated between infected (G2 and G3) and vaccinated (G4) dogs. In conclusion, despite dogs may test serologically negative immediately after vaccination against CVL with Leishmune(?), subsequent seroconversion, antibody peak and positivity up to six months may lead vaccinated dogs to be mistakenly identified as naturally infected dogs during this period.     

Detection of reaginic antibodies against Faenia rectivirgula from the serum of horses affected with Recurrent Airway Obstruction by an in vitro bioassay

Reaginic antibodies, mainly of the IgE and some IgG subclasses, play an important role in the induction of type I immediate hypersensitivity reactions. These antibodies bind through their Fc fragment to high affinity receptors (FcεRI) present in the membrane of mast cells and basophils. Previously, several studies have investigated the role of reaginic antibodies in the pathogenesis of RAO. However, whereas immunological aspects of RAO have been extensively studied, the precise sequence of events is still not well understood and role of IgE in this disease still remains controversial. Therefore, in this study a bioassay was developed for reaginic antibody determination in serum from RAO-affected horses in order to determine the etiology of disease. The technique involves measuring in vitro calcium mobilization in RBL-2H3 cells following incubation with horse serum from RAO-affected or unaffected horses and one of the RAO antigens (Faenia rectivirgula). Results demonstrated that 15% of samples from the RAO-affected horses reacted positively in this in vitro bioassay, whereas the samples from unaffected horses did not. This bioassay indicates that reaginic antibodies could be involved in the immunological mechanism leading to RAO; and this technique may facilitate future research in other allergic diseases in horses.     

Production and characterization of monoclonal antibodies specific for sheep IgG subclasses IgG1 or IgG2

Monoclonal antibodies specific for sheep IgG subclasses IgG1 or IgG2 were produced using conventional cell fusion techniques. Monoclonal antibodies detected by preliminary screening assays were further characterized and their specificity verified by titration of ascites in radioimmunoassay or passive haemagglutination using pure sheep IgG1 or IgG2. Further evidence for the subclass specificity of the antibodies was obtained from immunoelectrophoretograms of sheep serum or purified proteins developed with monoclonal antibodies. Reaction of monoclonal antibodies with various IgG fragments showed that the determinants recognised were located on the pFc' portion of the heavy chain.     

Development and in vitro characterization of canine CD40-Ig

We recently reported that blockade of the CD40-CD154 ligand interaction with the cross-reacting mouse anti-human CD154 antibody, 5c8, together with donor-specific transfusion led to enhanced but not completely successful engraftment in a canine model of DLA-identical marrow transplantation after 100cGy total body irradiation (TBI). In order to improve the transplantation outcomes, we sought to develop a canine-specific reagent. To that end, we fused the extracellular domain of the canine CD40 with a mouse IgG2a Fc tail and tested the immunosuppressive effectiveness of the fusion protein in mixed leukocyte reactions. The extracellular domain of canine CD40 was fused with the Fc portion of mouse IgG2a in a pcDNA3.1+vector. Dhfr-deficient CHO cells were co-transfected with the CD40-Ig vector and a dhfr-containing vector. Stable, high producing clones were selected under increasing methotrexate concentrations. The fusion protein was purified, tested in mixed leukocyte reactions, and its immunosuppressive effect compared to that of the anti-CD154 antibody 5c8. The transfected cell line produced a CD40-Ig dimer whose identity was confirmed by mass spectroscopy. The purified canine CD40-Ig blocked mixed leukocyte reactions at a concentration of 1nM, which was 10 times more effective than the anti-CD154 antibody. Canine CD40-Ig is more immunosuppressive than the anti-human CD154 antibody 5c8 in canine mixed leukocyte reactions and may be more effective in vivo in a model of marrow transplantation.     

The immune response to disialoganglioside GD3 vaccination in normal dogs: a melanoma surface antigen vaccine

As a result of its metastatic potential, canine malignant melanoma like its human counterpart like its human counter part, has a poor response to conventional treatment protocols. This prompted us to investigate the possibility of enhancing the immune response against the melanoma cell surface antigen, disialoganglioside GD3. Initially a flow cytometric study was designed in which the incidence of GD3 on the cell surface, recognized by the monoclonal antibody Mel-1 (R24), was established in canine melanoma cell lines. Results from the flow cytometry found GD3 to be highly expressed (94.2%) in six out of seven canine melanoma cell lines. Since it was thus potentially a good target, a study in which normal dogs were vaccinated intradermally with a vaccine containing GD3 plus adjuvants was designed. The adjuvant included CpG oligodeoxynucleotide (CpG-ODN) sequences and RIBI-adjuvant, which are known to target toll-like receptors (TLR) of the innate immune system. From a cohort of 10 dogs, 4 were vaccinated 3 times, at 4 weekly intervals with GD3 plus adjuvant, and 4 received only RIBI-adjuvant, and 2 phosphate buffered saline. Caliper measurements were collected to assess skin reaction at the vaccination site and sera assayed for IgM and IgG antibodies against GD3 and cell-mediated cytotoxicity against a melanoma cell line. Results from the study found significant differences (P<0.05) in the vaccine site reactions, IgM/IgG levels and cell-mediated cytotoxicity in the vaccinated versus unvaccinated dogs. The addition of CpG-ODN sequences and increasing GD3 concentration in the vaccine increased the inflammation response at the injection site. GD3 IgG and IgM antibodies in vaccinated dogs showed increasing titers over time and achieved significance at weeks 9 and 12, respectively. Cell-mediated cytotoxicity was only detected in peripheral blood mononuclear cells from vaccinated dogs. In conclusion, by combining the tumor antigen GD3 (a known weak self-antigen) and an adjuvant, tolerance was overcome by an innate and adaptive immune response in this population of normal dogs.     

Identification of protein A from Staphylococcus intermedius isolated from canine skin

Protein A was identified in cell wall-bound and secreted forms from Staphylococcus intermedius isolated from canine skin. A direct binding radioimmunoassay for the detection of bacterial surface Fc receptors identified 48 of 50 S intermedius isolates that contained cell wall-bound protein A. Using a competitive binding radioimmunoassay for the detection of Fc-reactive proteins in bacterial culture supernatants, we identified 9 of 50 clinical isolates of S intermedius that secreted measurable quantities of an Fc receptor into the culture medium. Concentrated culture supernatants from these isolates were analyzed by western blotting techniques and probed with either a radiolabeled human IgG Fc-specific probe or a radiolabeled affinity-purified chicken antibody against protein A. The studies reported here confirmed that Fc receptors are secreted by S intermedius isolates from dogs and are antigenically and functionally similar or are identical to staphylococcal protein A. Analysis of Fc receptor secretion by S intermedius strains, isolated from dogs with a variety of dermatologic conditions, suggested a trend between severity of skin disease and the extent of Fc receptor secretion.