HUMAN COMPLEMENT ACTIVATION BY SELF-ASSOCIATED IgG RHEUMATOID FACTORS

IgG rheumatoid factors (IgG–RFs) undergo concentration-dependent self-association into dimers and higher polymers, as previously reported. The interactions of purified IgG–RF from the plasma of 3 patients with rheumatoid arthritis, with guinea pig and human complement were studied. Self-associating IgG–RFs were isolated by affinity columns and gel filtration. These preparations contained no detectable IgM and were composed only of IgG subclasses known to fix complement. Complement utilization of IgG–RF was compared with that of monomeric IgG, heat-aggregated IgG, and soluble rabbit IgG immune complexes. Although incubation of IgG–RF or monomeric IgG with 3 units of guinea pig or human complement resulted in decreased hemolysis of sheep erythrocytes sensitized with IgM hemolysin, these substances were less than 100 times as effective as heat-aggregated IgG or soluble immune complexes. The ability of human or guinea pig complement that had been incubated with IgG–RF to restore hemolytic activity to C4-deficient guinea pig serum served to distinguish Clq binding from complement cascade activation. IgG–RFs and monomeric IgG did not activate guinea pig complement cascade in contrast to aggregated IgG. IgG–RFs, however, activated human complement cascade; monomeric IgG only bound human Clq. These results indicate that self-associated IgG–RFs can activate human complement in fluid phase, but less effectively than aggregated IgG or large-latticed immune complexes.     

Rheumatoid factor isotypes and circulating immune complexes in rheumatoid arthritis

Solid phase enzyme immunoassays were here used to quantify rheumatoid factors (RF) of the IgM, IgG and IgA classes and the immune complexes (IK) by their ability to bind to C1q or conglutinin in both the serum and synovial fluid of patients with rheumatoid arthritis (RA). Elevated serum levels of any RF isotype could be found in all patients with seropositive RA (IgM: 63%, IgG: 87%, IgA: 90%). Seronegative patients with RA presented to a significantly lesser extent with elevated levels of all the RF isotypes tested (IgM: 0%, IgG: 40%, IgA: 32%). Synovial fluid RF levels were significantly higher in SPRA patients than in SNRA patients with the exception of IgG-RF. All of the RF classes in both RA groups, however, were elevated when compared to RF in the synovial fluid of patients with osteoarthrosis. Both C1q binding and conglutinin binding immune complexes were significantly higher in the synovial fluid than in the serum of RA patients. The erythrocyte sedimentation rate and plasma iron levels were correlated with the levels of C1q binding immune complexes (IC) in the synovial fluid; total iron binding capacity showed an inverse relationship to synovial fluid IgG-RF levels. A radiographic index was also correlated with IgG-RF levels in the synovial fluid. Extraarticular manifestations were significantly more frequent in patients with elevated serum levels of IgM-RF or conglutinin binding IC. These findings indicate that IgG-RF in the synovial fluid and the formation of IC determined by their ability to bind C1q seem to be closely related to clinical features of local disease.(ABSTRACT TRUNCATED AT 250 WORDS)     

In vitro secretion of human IgM rheumatoid factor. Evidence for distinct rheumatoid factor populations in health and disease

The production of antibodies that react with the Fc fragment of IgG, i.e., rheumatoid factors (RF), is now regarded as a normal host immune response. It is not clear, however, if such putative physiologic RF are different from their counterparts which characterize pathologic states like rheumatoid arthritis (RA). Using Staphylococcus aureus Cowan I as an in vitro stimulant of RF production, we now report that the IgM-RF secreted by blood mononuclear cells obtained from healthy newborn infants and healthy adults can be distinguished not only from classic monoclonal RF and polyclonal RA serum RF, but also from the RF secreted by blood mononuclear cells obtained from RA patients. Whereas the Fc-binding activity of all RF secreted in vitro was easily inhibited by aggregated human IgG, only the RF produced by the normal umbilical cord cells and the normal adult cells were inhibited by monomeric Fc(IgG). The normal RF were also selectively inhibited by monomeric rabbit and guinea pig (Fc(IgG). The RF secreted by umbilical cord blood cells utilized lambda and kappa light chains, with a disproportionate use of lambda light chains relative to the total IgM secreted. Together, these data provide evidence for distinct subsets of RF in health and in disease.     

Activation of the classical pathway of complement by rheumatoid factors

A simple, sensitive solid-phase radioimmunoassay to quantitate the activation of the classical pathway of complement by rheumatoid factor (RF) is described. RF (purified, in serum or synovial fluid) was bound to reduced and alkylated IgG adsorbed to polyvinyl chloride microtiter plates and reacted with diluted normal human serum (complement). The activation and binding of C4 were quantitated with ¹²⁵I-Fab'₂-anti-C4. Purified, polyclonal IgM–RF was 100- to 1,000-fold more effective than purified IgG–RF in activating complement. The amount of complement activation produced by RF in each of the 57 sera and 2 synovial fluid samples correlated directly with the amount of IgM–RF present. The complement activating abilities of polyclonal IgM–RF in the sera of 15 rheumatoid arthritis patients were homogeneous. This novel technique is readily applicable to the investigation of complement activation by RF in disease.     

Release of soluble immune complexes from immune adherence receptors on human erythrocytes is mediated by C3b inactivator independently of β1H and is accompanied by generation of C3c

Antigen.antibody complexes (Ag.Ab) prepared from (125)I-labeled bovine serum albumin and guinea pig anti-albumin were incubated at 37 degrees C for 30 min with normal human serum diluted optimally for binding (1:16) and then with autologous erythrocytes (RBC). After washing, RBC-bearing antigen.antibody.complement complexes (Ag.Ab.C) were resuspended in serum reagents or solutions of purified complement components, and the kinetics of dissociation were analyzed. Ag.Ab.C dissociated in serum heated at 56 degrees C for 30 min (SDelta30) but not in serum heated for 120 min (SDelta120). Dissociation in SDelta30 markedly decreased after adsorption with anti-C3b inactivator but not anti-beta1H or anti-C4 binding protein (C4bp), and dissociation in SDelta120 markedly increased after addition of C3b inactivator. Hemolytic assays revealed that SDelta30 retained inactivator activity whereas SDelta120 lacked significant activity. Ag.Ab.C dissociated in the presence of purified inactivator or C3b but not beta1H or C3. Dissociation was more rapid with inactivator than with C3b and occurred at 0 degrees C as well as at 37 degrees C. Treatment with inactivator inhibitor abolished dissociation in SDelta30; dissociation in inactivator deficient serum was markedly reduced. Addition of beta1H did not enhance inactivator-mediated dissociation at limiting dilutions of inactivator, and adsorption of Ag.Ab.C with anti-beta1H or preparation of Ag.Ab.C with serum adsorbed with anti-beta1H did not diminish dissociation. After dissociation with inactivator, Ag.Ab.C were unchanged in size but were no longer able to bind to fresh RBC and gave enhanced binding to Raji and Daudi lymphoblastoid cells. NaDodSO(4)/polyacrylamide gel electrophoresis of Ag.Ab.C prepared with (125)I-labeled C3 revealed that, after binding to RBC, dissociation with inactivator was accompanied by generation of a C3 fragment the size of C3c. Preincubation of Ag.Ab.C with excess inactivator did not prevent subsequent binding of Ag.Ab.C to RBC but, immediately after binding, Ag.Ab.C dissociated rapidly. These findings indicate that C3b inactivator can release immune complexes from immune adherence receptors on human RBC, that release occurs independently of beta1H, alters cell binding properties of immune complexes, and involves multiple cleavages of the C3b alpha' chain, and that receptors in human RBC membrane are required for this C3b inactivator-mediated breakdown.     

Measurement of complement activation products in patients with chronic rheumatic diseases

Summary Measurement of the complement activation products C1s:C1-inh, C3bP and C5b-9 by ELISA in plasma samples from normals, rheumatoid arthritis (RA) and systemic lupus erythematosis (SLE) patients showed significantly elevated levels in the two patient groups (P<0.0001 for C1s:C1-inh, C3bP and C5b-9) compared to normals. In seropositive RA patients there were significant correlations between the levels of the three complement activation complexes and IgM-RF, IgG-RF and IgA-RF. However, IgM-RF did not interfere with any of the ELISA systems. Mean levels of C1s:C1-inh, C3bP and C5b-9 were the same in paired plasma and synovial fluids; however, C3bP levels in the paired samples did not correlate with one another by rank. Our conclusions are that: (a) elevated plasma levels of these complement activation products are detectable in rheumatic diseases; (b) plasma levels of these complement activation products are related to Rheumatoid factor (RF) levels in seropositive RA patients; and (c) IgM-RF does not influence these solid-phase ELISA procedures.     

Clinical significance of rheumatoid factor--its complement activating property and isotype]

We investigated the relationship between complement-activating properties of rheumatoid factors (RF) and their isotypes. Active and inactive RA patients without extra-articular symptoms (EAS) and those with EAS were studied. Patients with SLE, liver cirrhosis (LC) and normal volunteers were served as controls. Isotype of RF was measured by ELISA and complement activation (CA) of RF was measured by hemolytic assay. The CA values were significantly higher in RA patients with EAS than those in RA patients without EAS and with other diseases. The IgG and IgM-RF values were significantly higher in active RA without EAS than in inactive RA. Positive correlations between IgM-RF values and CA values were observed in RA with or without EAS, SLE and LC. The Sephadex G-200 gel filtration analysis of sera revealed CA in only 19S IgM fraction. Additionally, we purified 19S IgM-RF from sera of RA, SLE and LC patients by affinity chromatography, and found 19S IgM-RF had CA. These data suggest that serum IgM-RF may play a critical role in the pathogenesis of RA, especially in RA with EAS.     

Complement mediated inhibition of immune precipitation in rheumatoid arthritis: studies on interaction of heat aggregated IgG with IgM rheumatoid factor.

Serum samples from patients with seropositive rheumatoid arthritis contain an inhibitor of complement mediated inhibition of immune precipitation (CMIP). This inhibitory effect can be produced by the addition of either purified monoclonal or polyclonal IgM rheumatoid factor (RF) to human serum. The specificity of the rheumatoid factor influences the degree of inhibition, and when precipitation occurs the rheumatoid factor coprecipitates with the antigen-antibody complex. In rheumatoid sera there was a significant positive correlation between IgM RF concentration and inhibitory activity, though the range of inhibitory activity seen for the same concentration of rheumatoid factor was considerable. Small quantities of heat aggregated IgG (HAGG) had a much greater effect on the measurement in an enzyme linked immunosorbent assay (ELISA) of IgM RF than they did on the inhibitory activity of IgM RF in the CMIP assay. Larger quantities of HAGG initiated complement activation and increased the precipitation of immune complexes. IgM RF reduced the complement activating properties of HAGG by reducing the amount of Clq which bound to the aggregate. The mechanisms by which IgM RF overcomes CMIP in rheumatoid sera may involve its inhibitory effects on the binding of Cl to the antigen-antibody complex.     

Serum IgG and IgM rheumatoid factors and complement activation in extraarticular rheumatoid disease

Rheumatoid factors (RF) may participate in both synovial and extraarticular (EA) inflammation in rheumatoid disease (RD). The relative roles of serum IgG RF and IgM RF in extraarticular rheumatoid disease (EARD) are unclear, as is the importance of complement (C) activation by these proteins. To investigate the relation of C activating properties of IgM RF (RF CAP) and total IgG RF and IgM RF to EARD we compared 18 patients with only articular disease to 27 patients with various EA manifestations (nodules, cutaneous vasculitis, neuropathy, Felty's syndrome) using radioimmunoassays for IgG and IgM RF and an established hemolytic assay for C activation by IgM RF. We calculated RF CAP by determining the mean hemolysis of sensitized SRBC/ml of RF serum (MH/ml). Normal volunteers and patients with other inflammatory arthritides served as controls. Controls had negligible amounts of IgG RF, IgM RF, and RF CAP. Mean IgG and IgM RF levels and RF CAP values were significantly higher in patients with EARD than those in the arthritis-only (AO) group. Mean IgM RF concentrations and IgM RF CAP correlated with each other and EARD. IgG RF also correlated with IgM RF and EARD, but did not contribute to RF CAP or EARD when adjusted for IgM RF. Further, some patients had high RF CAP values despite modest IgM RF levels. These data suggest that quantitative differences in IgM RF CAP and total IgM RF may be more important than IgG RF as determinants of EARD.     

Immunologic factors affecting the in-vivo and in-vitro survival of the Legionnaires' disease bacterium.

Immunologic factors affecting viability of the Legionnaires' disease (LD) bacterium were studied in vitro and in vivo in mice and guinea pigs. In bactericidal tests, fresh human serum quickly killed LD cells. Heating fresh serum to 56 degrees C for 30 min destroyed bactericidal activity; absorbing it with bentonite had little effect. Fresh normal human serum was more effective than guinea pig serum. Adding LD cells to fresh normal human serum caused a greater than 50% depletion in functional complement activity, apparently by activating the classic C-142 pathway, because human serum deficient in C4 was not bactericidal. Antibodies to the Knoxville 1 LD strain in guinea pigs showed enhanced complement-mediated bactericidal activity. Without complement, immune guinea pig or human sera prolonged in-vitro LD cell survival. Antibodies to Knoxville 1 in mice depressed in-vitro bactericidal activity of human complement against Knoxville 1. In-vitro bactericidal tests support in-vivo studies in subcutaneous chambers. Complement-deficient mice immunized with Knoxville 1 were (P less than 0.01) less resistant to homologous challenge than nonimmunized mice. Immunized guinea pigs had a greater than 80-fold increase in resistance to subcutaneous-chamber infection.     

Immunoglobulin Binding Properties of the Prosorba Immunadsorption Column in Treatment of Rheumatoid Arthritis

Abstract: Studies of the humoral effects of the Prosorba column were conducted in conjunction with the Phase 3 trial of Prosorba versus sham therapy for rheumatoid arthritis (RA). When perfused with normal human plasma in vitro, Prosorba bound predominantly IgG with a maximal capacity of approximately 462 g of Ig per Prosorba column, equal to about 1.5% of circulating IgG. Prosorba treatment did not alter the concentrations of albumin, IgG, IgM, and IgA in 3 RA patients, except for a small dilutional effect. Kinetic studies demonstrated that Prosorba removed IgG > IgM, IgA, and IgM rheumatoid factor (RF) during the initial moments of apheresis and almost exclusively IgM RF after 15 min. No net protein removal occurred at 60 min. Mean values of circulating immune complexes (CICs) were not significantly decreased by 12 weekly treatments. Complement was activated by the apheresis system upstream of the Prosorba column without changing C3 or C4 levels. We conclude that the Prosorba mechanism of action in RA is not bulk removal of Ig, but might involve modification of the CIC repertoire and could include, but not be limited to, effects related to complement activation.     

IgM rheumatoid factor in Lyme disease: correlation with disease activity, total serum IgM, and IgM antibody to Borrelia burgdorferi

We tested the sera of 50 patients with Lyme disease for IgM-rheumatoid factor (IgM-RF) using a sensitive ELISA. Levels of IgM-RF greater than 3 SD above the mean of normal subjects were found in 2 of 15 patients with erythema chronicum migrans, 7 of 10 with neurologic abnormalities, and 7 of 25 with Lyme arthritis (p = 0.038). Only 2 of these sera were positive by latex agglutination. In contrast, none of the 23 control patients with osteoarthritis, ankylosing spondylitis, or Reiter's syndrome had positive tests. The levels of IgM-RF correlated with disease activity (p = 0.002), total serum IgM levels (p = 0.002), and specific IgM antibody titers to Borrelia burgdorferi (p = 0.006). IgM-RF reactivity was absorbed with heat aggregated IgG (HAGG), but the titer of specific IgM antibody was insignificantly affected by this procedure. Thus, small amounts of RF are produced at certain times in many patients with Lyme disease, and IgM-RF production appears to be linked to the specific IgM response.     

Mononuclear cell complement receptor blockade in primary biliary cirrhosis.

Peripheral blood monocyte and lymphocyte receptors for Fc and C3b fragments were examined in vitro in patients with primary biliary cirrhosis and other chronic liver diseases using sheep red blood cells coated with anti-SRBC IgG1 (to detect Fc receptors) and with anti-SRBC IgM and complement (to detect C3b receptors). The number of C3b receptors detected on 100 monocytes was significantly lower in patients with primary biliary cirrhosis (23.0 +/- 12.0, mean +/- 1 SD) compared with normal controls (57.4 +/- 16.9) and other chronic liver disease (HBsAg negative chronic active hepatitis 62.0 +/- 17.0, alcoholic cirrhosis 50.9 +/- 4.0), while the number of Fc receptors detected on 100 monocytes was not significantly different in all the groups (primary biliary cirrhosis 72.8 +/- 28.6, chronic active hepatitis 74.7 +/- 14.0, alcoholic cirrhosis 58.0 +/- 13.5 and normal controls 69.6 +/- 19.9). When mononuclear cells isolated from normal individuals were pre-incubated with serum from patients with primary biliary cirrhosis before testing their receptor function there was a significant reduction in the number of C3b receptors detected per 100 monocytes (27.6 +/- 10.8) compared with pre-incubation with normal serum (72.0 +/- 18.0). This reduction in C3b-receptor function was again observed when the serum used for pre-incubation was depleted of circulating immune complexes; but when complement was further depleted from these sera, the number of C3b-receptors detected after pre-incubation was similar to normal values (64.0 +/- 11.8). Lymphocyte receptors showed a similar pattern of results. This implies a specific C3b receptor blockade on monocytes and lymphocytes from patients with primary biliary cirrhosis which appears to be because of blocking by serum factor(s) including complement fragments.     

The effects of components of guinea pig serum on lymphosarcoma 6C3HED in C3H mice

The effects of various preparations of guinea pig serum on the Gardnerlymphosarcoma 6C3HED in C3H mice were studied. Complete regression oftumor as well as marked retardation of growth were noted as a result of treatment with the following sera: (1) normal; (2) depleted of complement 1(R1); of complement 3 (R3), and of complement 4 (R4); (3) depleted ofproperdin (RP); (4) heated at 56 C. for 20 minutes and for 60 minutes andat 66 C. for 30 minutes; and (5) supernatant from centrifuged at 35,000 Gat 2 C. for 3 hours. No tumor inhibitory effect was noted as a result of treatment with (1) serum depleted of complement 2 (R2) and (2) pellet (containing lipopolysaccharides) from serum centrifuged at 35,000 G at 2 C. for3 hours. These results indicate that the tumor inhibitory principle (TIP) inguinea pig serum is probably not complement, properdin, or lipopolysaccharide.

Properdin titers of sera from mice treated with each of the preparations listedabove disclosed low properdin levels when the tumors were growing or werelarge, and normal or elevated properdin levels when the tumors were regressing or disappeared completely. In addition, normal nontumor-bearingmice showed as much as a threefold increase in properdin levels when treatedwith normal or heated guinea pig serum.

While it is possible that the tumor inhibitory principle (TIP) in guinea pigserum may exert its effect through the animal’s own properdin system, such arelationship has not as yet been demonstrated.

Submitted on November 14, 1957 Accepted on January 28, 1958     

Circulating immune complexes and rheumatoid arthritis

Circulating immune complexes (CIC), as detected by the Clq binding assay (ClqBA) in sera from patients with rheumatoid arthritis (RA) were not demonstrable on analysis by ultracentrifugation on sucrose gradients. This discrepancy could be explained by the finding that polyethylene glycol 6000(PEG), used in the ClqBA to separate free radiolabelled Clq from complex bound Clq, increased the avidity of rheumatoid factor (RF), resulting in the formation of Clq binding RF IgM IgG complexes. Addition of purified RF IgM to normal human serum generated a positive ClqBA in a dose dependent way. The increased complex formation between RF IgM and IgG by PEG was also demonstrated in an enzyme linked immunoabsorbent assay and with sucrose gradients, where complexes became detectable when PEG was present. On the other hand RF IgM IgG Clq complexes obtained from the ClqBA dissociated upon removal of PEG. We conclude that high amounts of immune complexes, detected in RA sera by the ClqBA, are at least partly the result of in vitro complex formation between RF IgM and IgG. Therefore the results of this assay do not reflect the situation in the circulation in vivo.     

Complement-activating properties of IgM rheumatoid factors reacting with IgG subclasses

Summary To estimate the complement-activating property (CAP) of IgM rheumatoid factor (RF), which was purified from synovial fluids of patients with rheumatoid arthritis, in a reaction with each IgG subclass, the activation and binding of C4 in the classical pathway of complement by IgM RF was measured in an enzyme-linked immunosorbent assay using biotinylated F(ab')₂ antibody to human C4. The CAP of IgM RF reacting with IgG3 was significantly higher than that of IgM RFs bound to the other IgG subclasses (P<0.01). These results suggest that IgM RF reacting with IgG3 in synovial fluid could induce a greater degree of complement-dependent inflammation in RA synovium than IgM RF reacting with other IgG subclasses.     

Cell-free synthesis of the fourth component of guinea pig complement (C4): identification of a precursor of serum C4 (pro-C4).

Polysomes (S-20) from homogenates of guinea pig liver synthesized serum albumin and a precursor of the fourth component of guinea pig complement (C4) in vitro. The C4 precursor (pro-C4) accounted for approximately 0.2% and albumin 4% of the radiolabeled protein precipitable by trichloroacetic acid and not bound to polysomes. Pro-C4 is a single polypeptide chain (molecular weight 200,000) which is then converted to C4, a three-chain (molecular weights 95,000, 78,000 and 31,000) structure linked by interchain disulfide bridges. Pro-C4 was also detected intracellularly in short-term tissue cultures of guinea pig liver. C4 was found in the medium harvested from these cultures.     

IgE deposition in normal skin of patients with rheumatoid arthritis in relation to clinical and laboratory findings.

Biopsy specimens from apparently normal skin of 28 patients with classical or definite rheumatoid arthritis (RA) were examined for the presence of IgE deposition by a direct immunofluorescence technique. IgE deposition was found in 12 patients (43%) and in none of the 10 controls. This deposition was mainly localised on mast cells, and in three patients perivascular IgE staining was also noted. The skin from nine of the 12 patients also showed deposition of IgM and complement C3 or C4 factor, or both. All 12 patients with skin IgE deposition had raised levels of IgM rheumatoid factor (RF) in the serum. Nine of these also had IgE RF. IgE-containing circulating immune complexes (IgE CIC), raised serum IgE levels, and extra-articular (EA) manifestations were present in respectively 10, nine, and eight skin IgE positive patients. It is suggested that IgE and IgE CIC may be involved in the pathogenesis of RA and its EA manifestations.     

Terminal differentiation of spontaneous rheumatoid factor–secreting B cells from rheumatoid arthritis patients depends on endogenous interleukin-1 0

Objective. The presence of serum rheumatoid factor (RF) and spontaneous RF-secreting B cells is a common feature in most patients with rheumatoid arthritis (RA). This study analyzed the cytokine(s) that controls the final maturation of B cells capable of spontaneous IgM-RF secretion in vitro. Methods. RA patients' peripheral blood mononuclear cells (PBMC), as well as adherent and nonadherent cell fractions, were cultured, and spontaneous IgM-RF and interleukin-1 0 (IL-1 0) secretion were determined by enzyme-linked immunosorbent assay. Results. The RF+ RA PBMC, but not PBMC from RF– RA patients or healthy controls, actively produced IgM-RF in a linear manner for 14 days. This activity depended on the presence of fetal calf serum and did not require cellular DNA synthesis. Spontaneous IgM-RF secretion depended on IL-1 0, as deduced from the following findings: 1) IL-1 0, but not a variety of cytokines including IL-6, restored missing IgM-RF secretion by PBMC in serum-free supplemented cultures; 2) the addition of anti–IL-1 0, but not anti–IL-6, blocking antibodies inhibited PBMC IgM-RF secretion, and this effect could be reversed by exogenous IL-1 0; and 3) RA PBMC actively produced IL-1 0 in vitro. The cells responsible for endogenous IL-1 0 production were found in the adherent cell fraction. Finally, IL-1 0 induced IgM-RF, but not total IgM, secretion by RA PBMC. Conclusion. In patients with RA, circulating B cells capable of spontaneous IgM-RF secretion require IL-1 0 production by adherent cells to reach terminal maturation.     

Comparison of serological markers between ACPA<Superscript>+</Superscript> and ACPA<Superscript>−</Superscript> of RA patients

Anti-citrullinated protein/peptide antibodies (ACPA) have recently emerged as sensitive and specific serological markers of rheumatoid arthritis (RA), providing superior alternative of the rheumatoid factor (RF) test in the laboratory diagnostics of RA. We compare the change of serum RF, CRP, IgG, IgM, IgA, total complement, C3 and C4. The sera sample was collected from 123 patients with RA. ACPA were detected with ELISA, and RF, CRP and total complement (Ct), C3 and C4 were examined by automatic biochemical analyzer. Serum RF and total complement concentrations were significantly higher in ACPA+ than in ACPA−, but there were no correlation between ACPA and RF and Ct. Between ACPA+ and ACPA−, there were no significant difference of CRP, IgG, IgM, IgA, total complement, C3 and C4. While there were significant correlation between the concentration of C3 and IgM and ACPA in ACPA+. Conclusion: This is the first study to show that ACPA concentration in ACPA+ patients with RA is positively related to serum IgM and C3 levels.