Up-regulation of CD64-expressing monocytes with impaired FcγR function reflects disease activity in polyarticular psoriatic arthritis.

OBJECTIVES: The aim of this study was to assess monocyte Fc receptor (FcR) status and function in patients with active psoriatic arthritis (PsA) in relation to healthy controls (HC) and to disease activity. METHOD: The study population comprised 23 patients with active polyarticular PsA and 33 age- and gender-matched HC. Immunoglobulin (Ig) levels, inflammatory laboratory parameters, patient-reported outcomes of joint disease activity, skin scoring (Psoriasis Area and Severity Index, PASI), and joint status were determined in the patients. Monocytes were analysed for the expression of FcRs for IgG (FcγR) class I (CD64), IIa (CD32a), IIb (CD32b), and III (CD16), the FcR for IgA (FcαR) (CD89), and surface-bound IgG. The monocytic FcγR function was assessed by evaluating IgG immune complex (IC) binding and tumour necrosis factor (TNF)-α production following IgG-IC stimulation. The monocytes were further subdivided and analysed according to their CD14 and CD16 expression. RESULTS: The PsA patients presented elevated serum levels of IgG1, 2, and 3 and increased numbers of CD64(+) monocytes. Furthermore, the PsA monocytes exhibited increased cell-bound IgG, and the FcγR function was affected in terms of reduced IgG-IC-mediated TNF-α release. These findings correlated significantly with different markers of joint disease activity. PsA was also accompanied by an increase in the CD16 low-expressing monocyte subset. CONCLUSIONS: An intensified humoral immune response affects monocytes and their FcR status in active polyarticular PsA. The up-regulated CD64(+) monocytes seem to be have an important role in psoriatic joint inflammation. These cells may prove to be a useful target in future PsA therapeutic interventions.

FcgammaRIII-expressing macrophages are essential for development of collagen-induced arthritis.

IgG-binding Fc receptors, and in particular FcgammaRIII, are crucial for induction of collagen-induced arthritis (CIA), as FcgammaRIII-deficient mice are highly protected to arthritis. However, which of the FcgammaRIII-expressing cells that is responsible for induction of arthritis is not known. In this study, we have addressed this question by purifying different FcgammaRIII(+) cell populations, transferred them to FcgammaRIII-deficient mice and studied if the recipient mice can develop arthritis. The cell populations were isolated from spleen, bone marrow and the peritoneal cavity. Our results show that FcgammaRIII(+) CD11b(+) peritoneal macrophages can render FcgammaRIII-deficient mice susceptible to CIA. In contrast, FcgammaRIII(-) peritoneal macrophages or FcgammaRIII(+) spleenocytes, bone marrow cells, mast cells or monocytes could not mediate this effect. To further evaluate the contribution of the FcgammaRIII(+) macrophages in arthritis, we investigated the cytokine profile in these cells during CIA. The arthritic macrophages exhibited significantly higher mRNA levels of TNFalpha and IL-12p35 compared with macrophages from normal mice. We conclude that FcgammaRIII-expressing macrophages, producing pro-inflammatory cytokine and T helper type 1 differentiating factor, are the major effector cells in the induction of CIA.

IgG immune complex-binding in macrophages from arthritis-susceptible and arthritis-resistant mice following collagen type II immunization.

Occupancy of Fc gamma receptors (FcgammaR) by immune complexes (IC) induces secretion of various inflammatory mediators and cytokines. Therefore, knowledge of the FcR function is fundamental for understanding inflammatory processes. Here, we report an alteration in the FcR function in collagen-induced arthritis (CIA). The FcgammaR-binding activity of peritoneal macrophages from arthritis-susceptible DBA/1 mice following collagen type II (CII)/CFA immunization was assessed by Fc rosetting of SRBC opsonized with different IgG subclasses. A progressive reduction of IgG1 IC-binding was observed after immunization, and by the time of arthritis onset, the IgG1 IC-binding was abolished. Binding of IgG2a or IgG2b IC, however, was not affected. The blocked IgG1 IC-binding was reversed by a prior mild acid wash of the CIA macrophages, indicating receptor occupancy as the cause of the blocked binding. The impaired IgG1 IC-binding was associated with arthritis development, as macrophages from CII/CFA-immunized, arthritis-resistant SWR mice or DBA/1 mice, immunized with CFA alone, did not show this effect. Normal DBA/1 macrophages, blocked with a monoclonal antibody to FcgammaRIIB/FcgammaRIII, and macrophages from FcgammaRIII-deficient mice did not bind IgG1 IC, indicating FcgammaRIII as responsible for IgG1 IC-binding. Our data suggest that an increased degree of saturation of FcgammaRIII precedes the development of CIA, which is reflected by a reduced IgG1 IC-binding in macrophages of CII/CFA-immunized DBA/1 mice.

Gene targeting by ribozyme against TNF-alpha mRNA inhibits autoimmune arthritis.

Ribozymes are catalytic RNA that bind and cleave specific regions of target RNA. Therefore, protein synthesis by the target RNA may be specifically inhibited by ribozymes. In this study, we have investigated if ribozymes possess therapeutic activity on inflammatory processes in vivo, as judged from effects on an arthritis model. A hammerhead ribozyme against TNF-alpha was designed and its catalytic activity in vitro was verified. The ribozyme was employed in vivo without any delivery system, as the plasmid-based ribozyme was taken up adequately by various tissues in mice by intravenous injection. The ability of the ribozyme to regulate the development of collagen-induced arthritis (CIA), a model largely dependent on TNF-alpha, was investigated. Systemic administration of the ribozyme to mice immunized with collagen type II in CFA significantly reduced the development of CIA. No effect was observed with a catalytically inactive variant of the ribozyme. Furthermore, the ribozyme efficiently blocked cartilage and bone destruction in the joints and ameliorated established CIA. These data demonstrate for the first time that gene targeting by a ribozyme to inactivate TNF-alpha in vivo is highly efficient in suppressing autoimmune arthritis, thus providing proof of concept that it may be used as therapeutic tool for TNF-alpha-dependent chronic inflammatory disorders.

The impact of Fc receptors on the development of autoimmune diseases.

Immune complexes are major pathogenic factors contributing to tissue injury in a number of autoimmune diseases. Immune complexes consist of autoantibodies to self-antigens, which initiate an inflammatory response through binding to immunoglobulin Fc receptors (FcRs) or via complement components. The significance of FcRs as regulators of inflammatory reactions was defined when mice with genetically engineered deficiencies of FcRs became available, giving new insight into the mechanisms involved in the pathogenesis of immune complex-mediated inflammation and autoimmunity. In this review, some of the most prominent work on inflammation and autoimmunity in animal models, will be presented. This body of knowledge has not only contributed to our understanding of the importance of FcRs in inflammatory autoimmune responses, but more specifically, has paved a new way for designing novel therapeutic compounds in treating autoimmune diseases.

Fc receptors are critical for autoimmune inflammatory damage to the central nervous system in experimental autoimmune encephalomyelitis.

Multiple sclerosis (MS) is simulated by various forms of experimental autoimmune encephalomyelitis, in which T cells, antibodies, cytokines and complementary factors interact with the central nervous system (CNS) myelin proteins and lead to inflammatory damage. We investigated the role of Fc receptors (FcRs), which link the cellular and humoral branches of the immune system, in myelin oligodendrocyte glycoprotein (MOG)-induced experimental autoimmune encephalomyelitis (EAE), using two different FcRgamma knockout DBA/1 mice. The first knockout were the FcRgamma chain-deficient mice, which lack FcgammaRI, FcgammaRIII and Fc(epsilon)RI, while the second knockout mice lack only FcgammaRII. The lack of FcgammaRII enhanced the disease susceptibility with associated increased CNS demyelination. While FcRgamma+/+ DBA/1 mice also developed pronounced CNS infiltration and myelin destruction, FcRgamma-/- littermates were protected despite initial peripheral autoimmune responses to MOG. In vitro analyses revealed equivalent potentials of fluid phase phagocytosis of myelin and MOG in bone-marrow macrophages derived from both FcRgamma+/+ and FcRgamma-/- mice, while MOG-immunoglobulin (Ig)G immune complexes were only internalized by FcRgamma+/+ macrophages. This was associated with cellular activation in FcRgamma+/+ but not FcRgamma-/- macrophages, as assessed by the activation of intracellular mitogen activated protein (MAP)-kinase signalling elements. We propose that protection from EAE in FcRgamma-deficient mice is due to the inefficient antigen processing/presentation of myelin proteins during the induction of secondary immune responses locally in the CNS, which leads to demyelination. This demonstrates the importance of FcR in the promotion of autoimmune inflammation of the CNS and highlights the therapeutic possibility of treatment of MS with FcR-directed modalities.

Immune complex-mediated enhancement of antibody responses without induction of delayed-type hypersensitivity.

Immunoglobulin (Ig)G and IgE antibodies enhance the humoral response in vivo to soluble antigens with which they form complexes. In vitro, antigen is targeted to B cells by IgE antibodies and to macrophages and dendritic cells (DCs) by IgG, thus leading to increased antigen presentation to specific T cells. Possibly these mechanisms are also responsible for antibody-mediated enhancement in vivo. We now address the question of whether IgG- and/or IgE-antigen complexes can prime for delayed-type hypersensitivity (DTH), a reaction known to require primed T helper (Th)1 cells. Mice were immunized with IgG-anti-2,4,6-trinitrophenyl (TNP)/BSA-TNP or IgE-anti-TNP/BSA-TNP. Mice given BSA-TNP alone or BSA-TNP in complete Freund's adjuvans (CFA) were used as controls. DTH and IgG-anti-BSA levels were measured after subsequent challenge with BSA. A potent BSA-specific antibody response was induced by IgE- or IgG-complexed antigen as well as by CFA/antigen but DTH-reactions were only observed in mice immunized with CFA/antigen. Both IgE and IgG enhanced the production of BSA-specific IgG1, IgG2a and IgG2b, although the most pronounced enhancement was seen in the production of IgG1. These findings suggest that Th2 cells rather than Th1 cells are involved in the immune response to IgG- and IgE-immune complexes.

Induction and suppression of collagen-induced arthritis is dependent on distinct fcgamma receptors.

Receptors for immunoglobulin (Ig)G (FcgammaRs) are important for the antibody-mediated effector functions of the immune system. FcgammaRI and FcgammaRIII trigger cell activation through a common gamma chain, whereas FcgammaRII acts as a negative regulator of antibody production and immune complex-triggered activation. Here we describe the in vivo consequences of FcgammaR deficiency in a mouse model of human rheumatoid arthritis. FcRgamma chain-deficient mice on arthritis-susceptible DBA/1 background were immunized with collagen for induction of collagen-induced arthritis. The DBA/1 mice lacking FcRgamma chain were protected from collagen-induced arthritis in contrast to wild-type mice, although both groups produced similar levels of IgG anticollagen antibodies. In comparison, DBA/1 mice lacking FcgammaRII developed an augmented IgG anticollagen response and arthritis. These observations suggest a crucial role of FcgammaRI and FcgammaRIII in triggering autoimmune arthritis.

Importance of CD23 for collagen-induced arthritis: delayed onset and reduced severity in CD23-deficient mice.

Increased expression of the low affinity receptor for IgE, FcepsilonRII/CD23 has been observed in rheumatoid arthritis. In view of this, we have investigated the expression and influence of CD23 in collagen-induced arthritis (CIA), an animal model for rheumatoid arthritis. CD23+ cells were analyzed in lymph nodes of DBA/1 mice immunized with bovine collagen type II (BCII) in CFA or with CFA only. The percentage of CD23+ lymph node cells was increased in both BCII/CFA- and CFA-immunized mice at 1, 3, and 7 wk after immunization compared with unimmunized mice, indicating a role for the adjuvant to trigger general inflammation and CD23 expression. To investigate the functional role of CD23 in CIA, CD23-deficient mice on the DBA/1 genetic background were studied. After immunization with BCII/CFA, these mice developed CIA with delayed onset and reduced severity compared with wild-type mice. These findings suggest that an increased number of CD23+ cells is part of an inflammatory response and that CD23 expression is of pathogenic importance in the arthritic process.

The inflammatory reaction in elective flap surgery.

The inflammatory response in three different flap procedures was investigated by measuring the preoperative and postoperative levels of C-reactive protein, leukocyte count, and body temperature. Patients scheduled for delayed breast reconstruction were operated on with the lateral thoracodorsal flap, the latissimus dorsi flap, or the pedicled TRAM flap. All patients received 2 gm of intravenous cloxacillin for antibiotic prophylaxis and 1 gm of paracetamol four times a day as basic treatment for postoperative pain. Within each treatment group, significant postoperative changes in C-reactive protein levels, leukocyte count, and body temperature were noted when compared with preoperative values. The highest C-reactive protein level (130 mg/ml) was found in the TRAM group on the third postoperative day. The kinetic pattern of C-reactive protein was similar for the latissimus dorsi flap and lateral thoracodorsal flap procedures, but the maximum C-reactive protein levels were significantly lower, 74 and 44 mg/ml respectively. Small (0.5 to 0.9 degrees C) but significant differences in body temperature were also noted on the second and third postoperative day. The TRAM flap group had the highest, the latissimus dorsi flap group intermediate, and the lateral thoracodorsal flap group the lowest value. The postoperative C-reactive protein levels seem to reflect the extent of the surgical trauma.

Common commercial cosmetic products induce arthritis in the DA rat.

Many different agents, including mineral oil and silicone, have the capacity to act as immunological adjuvants, i.e., they can contribute to the activation of the immune system. Some adjuvants, including mineral oil, are known to induce arthritis in certain strains of rats after intradermal injection or percutaneous application. The aim of this study was to determine if common commercial cosmetic products containing mineral oil could induce arthritis in the highly susceptible DA (Dark Agouti) rat. Intradermal injection of five out of eight assayed cosmetic products without further additives resulted in arthritis with synovitis. One of the products induced a very aggressive arthritis, which had declined after 5-9 weeks. When this product was also assayed for arthritogenicity upon percutaneous administration, it induced a mild and transient arthritis in 5 out of 10 DA rats, whereas control animals showed no clinical signs of joint involvement. No arthritic reaction was seen in rats after peroral feeding with the most arthritogenic product or by intravaginal application of Freund's adjuvants. Silicone gel implants in DA rats did not cause arthritis. We conclude that mineral oils included in common commercially available products retain their adjuvant properties and are arthritogenic in the presently investigated arthritis-prone rat strain. There is yet no evidence that mineral oils present in cosmetics may contribute to arthritis in humans, but we suggest that this question should be subject to further investigation.

Characterization of adhesion molecule expression in the pathogenesis of homologous collagen-induced arthritis in the DA rat.

We have investigated the inflammatory cell infiltrates and adhesion molecule expression in the synovial fluid (SF) and synovial membranes (SM) of rats with homologous collagen-induced arthritis. Immunohistochemical staining revealed that the majority of the cells in the SF were granulocytes, expressing CD11b and CD11c. In SM, the majority of the cells were monocytes/macrophages. CD49d and CD49f were expressed mainly in the erosion zone in SM, and ICAM-1 was expressed in the lining layer, in the capillaries, and in the erosion zone. In SF 7% of the cells were ICAM-1 positive. CD2 was more abundant in SM than in SF. These findings might explain the difference in granulocyte counts between SF and SM. CD49d and CD49f expression might retain lymphocytes and monocytes in SM, while granulocytes not expressing CD49d and CD49f are not retained.

Polyclonal Th1 cells transfer oil-induced arthritis.

T-cells play a critical role in oil-induced arthritis (OIA) in DA rats. The present study focuses on the involvement of CD4/CD8 T cells in OIA by using adoptive transfer. Mitogen-activated T cells from DA rats previously injected with incomplete Freund's adjuvant (IFA) were depleted of CD4+ T cells or CD8+ T cells before transfer to irradiated naive receipients. The results indicate that CD4+ T cells are essential for the induction of passively induced OIA. However, in vitro blocking experiments with monoclonal antibodies (mAb) to the CD4 molecule of the T cells before transfer did not affect the passive OIA. Neither was passive OIA inhibited by treating the CD4+ T cells with mAb to intracellular adhesion molecule-1 (ICAM-1) in order to block cell-cell interactions or migration. The arthritogenic CD4+ T cells were sensitive, however, to in vitro treatment with mAb to the interleukin-2 receptor, which inhibited the disease or delayed the onset of passive OIA in recipients. The arthritogenic CD4+ T cells were also analysed for expression of specific T-cell receptor (TCR) variable (V) beta chains, critical for recognition of autoantigen, by utilizing V beta gene-specific polymerase chain reaction (PCR). The results show a heterogeneous expression of V beta segments of the TCR, indicating a polyclonal origin of the pathogenic cells. Moreover, an investigation of the T helper (Th)1/Th2 status of the CD4+ T cells, defined by cytokine expression, was made at the mRNA level by using in situ hybridization. High numbers of interleukin-2 (IL-2) mRNA expressing cells and also interferon-gamma (IFN-gamma) and tumour necrosis factor-alpha (TNF-alpha)-expressing cells could be identified. We conclude from this study that non-immunogenic IFA triggers polyclonal, IL-2-dependent Th1 cells which induce arthritis. The contribution of the CD4 or ICAM-1 molecules for arthritis induction seem to be of minor importance.

Altered Th1/Th2 balance associated with non-major histocompatibility complex genes in collagen-induced arthritis in resistant and non-resistant rat strains.

Collagen-induced arthritis (CIA) is a T cell-dependent disease in which susceptibility is controlled by genes both within and outside the major histocompatibility complex (MHC). In the present study, we compared the humoral responses and kinetics of cytokine secretion patterns in the draining lymph nodes of arthritis-susceptible DA rats and arthritis-resistant F344 and DA MHC congenic PVG.1AV1 rats immunized with rat type II collagen (RCII) in incomplete Freund's adjuvant. The results demonstrate a marked humoral RCII response and a Th1 cytokine profile, with expression of interferon-gamma and interleukin (IL)-2 mRNA in DA rats; a limited humoral RCII response and a Th2 cytokine profile, with expression of IL-4 mRNA in arthritis-resistant F344 rats; and a marked humoral RCII response in arthritis-resistant PVG.1AV1 rats. However, in contrast to DA rats, PVG.1AV1 rats produce IgG1 autoantibodies which, together with strong expression of IL-4 mRNA, indicates the involvement of Th2 subsets. From these data, we conclude that non-MHC gene(s) determines the direction of the anti-RCII response towards a Th1 disease-promoting, or a Th2 disease-limiting response.

Dynamic expression of transforming growth factor-betas (TGF-beta) and their type I and type II receptors in the synovial tissue of arthritic rats.

A well characterized animal model that shares many characteristic features with rheumatoid arthritis (RA) is collagen-induced arthritis (CIA) in DA rats. Recent studies have demonstrated that TGF-beta, a multifunctional cytokine, is an important modulator of the immune response in CIA, and possibly also in RA. In this study we have investigated the expression of the precursor forms of TGF-beta1, TGF-beta2, TGF-beta3, as well as TGF-beta type I receptor (TGF-betaRI) and TGF-beta type II receptor (TGF-betaRII) in the synovial tissue of arthritic rats during the course of the disease. By using immunohistochemical techniques, an abundant expression of all three TGF-beta isoforms and their receptors was observed in the arthritic synovia, an expression that increased with time after onset of disease. Antibodies to TGF-beta1, TGF-beta2, TGF-betaRI and TGF-betaRII stained blood vessels intensively, already at the early onset of inflammation, whereas the synovial lining layer and chondrocytes expressed strong immunoreactivity later on in the inflammatory process. The most intense staining with these antibodies was detected in fibroblasts within fibrotic tissue, in particular at the cartilage pannus junction. Interestingly, TGF-beta3 only stained macrophage-like cells and chondrocytes in the synovia. The data suggest that the abundant expression of TGF-beta1, TGF-beta2, TGF-beta3 as well as TGF-betaRI and TGF-betaRII in the synovia, is of pathogenic importance in the development of CIA, although the question of how the different TGF-beta isoforms may enhance or counteract different arthritogenic events remains open.

Specific and long-lasting protection from collagen-induced arthritis and oil-induced arthritis in DA rats by administration of immunogens.

DA rats develop transient arthritis after subcutaneous immunization with adjuvant-oil, while chronic arthritis and collagen autoreactivity ensues when collagen is added to the oil. We show here that DA rats can be protected from oil-induced arthritis (OIA) and rat collagen-induced arthritis (rCIA) by addition of antigen to these arthritogenic inocula. We have investigated this remarkable phenomenon and demonstrate that both foreign and self antigens can be protective, apparently provided they are immunogenic; hence HSP-65kDa, ovalbumin, rat myelin basic protein, rat IgG and bovine albumin are effective while rat albumin is not. This protection is long-lasting and disease-specific because rats protected from rCIA resist a later attempt to induce arthritis, but not experimental autoimmune encephalomyelitis (EAE). Protection from rCIA depends neither on the blocking of humoral autoreactivity to collagen nor on a change in the isotype profile of anti-collagen antibodies. We demonstrate that immunogens can also be protective when injected intraperitoneally only a few days before onset of arthritis. Our results indicate that protection is mediated through bystander immune reactions towards the co-immunized antigen and that the arthritogenicity of a given provocation, be it adjuvants, microbes or autoantigens, may be a complex net result of arthritogenic and contra-arthritogenic immune reactions.

TNF-alpha dominates cytokine mRNA expression in lymphoid tissues of rats developing collagen- and oil-induced arthritis.

Experimental arthritis can be induced in the DA rat strain with rat type II collagen (RCII) administered in Freund's incomplete adjuvant oil (FIA) or with only FIA. If ovalbumin (Ova), is added to these arthritogens the development of arthritis is blocked. To investigate the mechanisms responsible for induction of arthritis, as well as inhibition of arthritis, a kinetic study of the local cytokine expression in lymph nodes has been performed after immunization with the above mentioned agents. By using in situ hybridization techniques, mRNA expression of TNF-alpha, IL-2, IFN-gamma and IL-4 was determined. The results show a rapid and pronounced accumulation of TNF-alpha mRNA expression, in RCII/FIA and FIA immunized rats. This pronounced expression of TNF-alpha mRNA was not recorded in the Ova/FIA immunized animals, which instead were the only animals in which the IL-4 gene was expressed. The expression of IFN-gamma mRNA was limited in RCII/FIA- and FIA-immunized rats, whereas IL-2 mRNA expression was detected only after RCII/FIA injection. Lymph node cells from RCII-immunized animals generated a high amount of TNF-alpha mRNA after restimulation with RCII, whereas restimulation with the mitogen Con A generated a cytokine mRNA response dominated by IL-2 and IFN-gamma. These and other results indicate that a strong local expression of TNF-alpha, induced by arthritogenic stimuli, may be important for the induction of arthritis. Moreover, the elicitation of an immune reaction against Ova, may inhibit arthritis development by contributing to a shift in the initial arthritogenic cytokine response.

Oil-induced arthritis in DA rats: tissue distribution of arthritogenic 14C-labelled hexadecane.

As part of the study of the pathogenesis of oil-induced arthritis in DA rats, the tissue dissemination of arthritogenic oil labelled with 14C has been determined. Rats received labelled hexadecane in arthritogenic doses by the intradermal route and were killed at 1 and 6 h, 2, 10, 14, 18 and 27 days after injection. Whole-body autoradiography was performed and localization of radioactivity in different organs was investigated. With the exception of the injection site, the lymph nodes showed the highest content of radioactivity throughout the study. Radioactivity could be seen in the popliteal, inguinal and axillary lymph nodes. Detailed examination of lymph nodes revealed at 1 h radioactivity in the subcapsular sinus of the lymph node. By 10 days the activity had spread to the cortex and paracortex, apart from the subcapsular sinus. The activity in other lymphoid organs, such as the bone marrow and spleen, was more transient, with a peak at 2 days. Oil disseminating to joints was not prominent: very little radioactivity was observed in the knee joints of both non-arthritic and arthritic animals. From these data it is suggested that the adjuvant oil exerts its major proarthritogenic activity in the lymph nodes rather than directly in the joints.

Role of adjuvants in turning autoimmunity into autoimmune disease.

Autoreactive B- as well as T-lymphocytes are often integral parts of the normal immune system and are not per se pathogenic. Also activation of these autoreactive lymphocytes may in many cases occur without ensuing pathology. In the current paper is discussed two situations whereby unspecific inflammatory stimuli in the form of immunological adjuvants may either by themselves cause arthritis, or may change a non-pathogenetic to a destructive and chronic inflammatory disease.

Role of the bowel flora for development of immunity to hsp 65 and arthritis in three experimental models.

An infectious aetiology in rheumatoid arthritis (RA) has for long been suggested, although no conclusive evidence for this is at present available. Lately a large interest has been devoted to the involvement of heat shock proteins (hsps) in autoimmune disorders due to their conserved structure and immunogenic properties. Immunity to hsps has been observed both in human autoimmune conditions and in experimental models of autoimmune disease. We have studied the role of the bacterial flora and hsp immunity in the arthritic response in three experimental models of arthritis; type II collagen arthritis (CIA), adjuvant arthritis (AA) and oil induced arthritis (OIA); by using germ free and conventional DA rats. A high incidence of severe arthritis developed in all the models evaluated irrespectively of whether the animals were in the conventional or germ free state. This confirms earlier results which show a minor effect of the bacterial flora in CIA and AA in high responder strains. These results also show that a severe OIA can develop in germ free animals. Despite the severe arthritic response induced, no serum antibody levels to hsp 65 could be detected in the germ free animals, which was in contrast to the conventional animals where a positive anti-hsp 65 serum response was seen in 35-80% of the animals with CIA, AA or OIA. These results show that development of a humoral response to hsp 65 in these models of arthritis is dependent on the presence of a bacterial flora. Further, the lack of humoral immunity in germ free animals despite a severe arthritic response indicates that humoral immunity to hsp 65 is not involved in development of disease in these three models of experimental arthritis.