| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Pathology and Immunology and
Department of Radiology, Washington University School of Medicine, St. Louis, MO 63110
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
are essential for disease development (16, 17), as are mast cells, neutrophils, and macrophages (18, 19, 20). FcRs are important for disease induction, with the low affinity Fc
RIII, present on macrophages, neutrophils, and mast cells, being the critical player (21, 22). The alternative pathway of complement is also required, including the C5a receptor (21, 23). Many of the same innate immune system components have also been shown to be important in the CIA arthritis model (reviewed in Refs. 24, 25, 26). A fascinating aspect of the K/BxN model is the exquisite joint specificity of the disease despite the GPI autoantigen being ubiquitously expressed. We previously described a novel detection system for determining the localization patterns of arthritogenic anti-GPI IgG in the joints of normal healthy mice, using rodent-scale positron emission tomography (microPET). The microPET R4 scanner permits dynamic noninvasive high resolution imaging of radiolabeled GPI-specific IgG in mice at multiple time points. Using this system, we observed that anti-GPI IgG rapidly localized within minutes to distal joints of the front and rear limbs (27) and remained there for at least 24 h. These kinetics were consistent with direct Ab recognition of GPI in the joints, and in support of this model, extracellular GPI was shown to be present on the cartilaginous surfaces of normal joints in naive mice (28).
The present studies were undertaken to delineate the order of involvement of these critical cells, receptors, and molecules in the development of arthritis induced by autoantibodies. Using microPET, we demonstrate that neutrophils, mast cells, and immune complexes (ICs) are essential for getting the Ab into the joints, whereas C5 acts at a later stage. From these results we were able to propose a four-stage model for how autoantibody-induced arthritis occurs.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
KRN TCR Tg mice (C57BL/6 background) were provided by D. Mathis and C. Benoist (Harvard Medical School, Boston, MA). Nonobese diabetic mice were obtained from Taconic Farms (Germantown, NY), and BALB/c mice were obtained from National Cancer Institute (Frederick, MD). K/BxN mice were obtained by breeding homozygous KRN transgenic mice to nonobese diabetic mice in our mouse colony (Washington University, St. Louis, MO). Fc
R1 deficient (FcRKO), B6/129F1, WBB6F1/J-KitW/KitW-v, and WBB6F1/J-Kit+/Kit+ mice were obtained from The Jackson Laboratory (Bar Harbor, ME). All animals were housed in accordance with National Institutes of Health and American Association for Accreditation of Laboratory Animal Care regulations, and animal protocols were reviewed and approved by the Washington University animal studies committee.
Measurement of ankle thickness
Mice were examined for clinical signs of inflammation, and ankle thickness was determined with a pocket thickness gauge (Ralmikes Tool-A-Rama, Middlesex, NJ) by measuring the ankle across the malleoli as previously described (18). Ankle thickness measurements were rounded to the nearest 0.05 mm and reported as the average of a group of identically treated mice ± 1 SD.
Serum and polyclonal Abs
K/BxN serum was harvested and processed as described previously (18). Polyclonal IgG from K/BxN serum was purified by protein G-Sepharose 4 Fast Flow (Amersham Pharmacia Biotech, Piscataway, NJ). Bound IgG was washed extensively with PBS, eluted using 0.1 M glycine HCl, pH 2.8, and immediately neutralized with 1 M Tris, pH 8.0, before dialysis against PBS. GPI-specific polyclonal Abs (anti-GPI IgG) were purified from protein G-purified total K/BxN IgG on affinity columns of recombinant polyhistidine-tagged mouse GPI conjugated to cyanogen bromide-activated Sepharose 4B (Sigma-Aldrich, St. Louis, MO), eluted, and dialyzed as described above. Abs were concentrated with Centricon-10 concentration devICs (Millipore, Bedford, MA), quantitated by optical absorption measured at 280 nm, and sterile-filtered through 0.22-µm pore size filters. The arthrogen/CIA arthritis-inducing mAb mixture was purchased from Chemicon International (Temecula, CA). A mixture of control mAbs composed of the same isotype proportion (3:1 ratio of IgG2a:IgG2b) as the arthrogen/CIA mixture was created by mixing two irrelevant Abs. mAb 4E12 (IgG2a, anti-3.L2 TCR clonotype) was purified from ascites over protein G-Sepharose, and mAb GIR94 (IgG2b) was donated by Drs. K. Sheehan and R. Schreiber (Washington University). For in vivo depletion, the anti-neutrophil rat IgG2b mAb RB6.8C5 was purified using protein G-Sepharose 4 Fast Flow affinity matrix (Amersham Pharmacia Biotech) from ascites produced in SCID mice. The isotype control Ab GK1.5 mAb against mouse CD4 (rat IgG2b) was purified from ascites by saturated ammonium sulfate precipitation (45% final concentration) and dialyzed against PBS, pH 7.4, before being stored at –70°C.
Arthritic serum transfer challenge
Before injection, serum aliquots were thawed, centrifuged, and diluted with PBS as described previously (18). All mice received 250 µl i.p. because it consistently produced disease induction in 100% of C57BL/6 and BALB/c mice.
Ab depletion in vivo
For in vivo neutrophil depletion, 250 µg of RB6.8C5 mAb was diluted in PBS and injected i.p. 48 h before use in microPET and biodistribution experiments. Control GK1.5 mAb (50 µl of ascites) was diluted and injected in the same manner. These doses of Ab were previously shown to completely eliminate neutrophils or CD4+ T cells for 3–5 days (18).
Conjugation of 1,4,7,10-tetraazacyclododecane-N,N',N'',N'''-tetraacetic acid (DOTA) linker to Abs
All Abs were conjugated to DOTA using the modified N-hydroxysuccinimide method (29) as previously described (27). After conjugation, the reaction mixtures were centrifuged repeatedly through a Centricon-30 with 30 mM ammonium citrate buffer, pH 6.5, to eliminate unconjugated small molecules. The purified conjugates were concentrated to 2 mg/ml in PBS buffer and stored at –70°C until further use. The concentrations of Ab conjugates were determined by UV spectrophotometer. Addition of DOTA did not affect the biological function or distribution of the Abs (27).
64Cu preparation and radiolabeling of conjugated Abs
64Cu (64CuCl2 in 0.1 M HCl; radionuclide purity, >99%) was produced with an in-house cyclotron from enriched 64Ni targets by a previously described method (30). Typically, 100 µg of DOTA-conjugated Ab and 1 mCi of 64Cu were incubated in 30 mM ammonium citrate, pH 6.5, at 43°C for 45 min. The reaction was terminated by addition of 5 µl 10 mM diethylenetrinitrilopenta-acetic acid solution. Labeled IgG was separated from unincorporated label by a size-exclusion Bio-Spin6 column (Bio-Rad, Hercules, CA).
MicroPET data collection
The microPET-R4 rodent scanner (Concorde Microsystems, Knoxville, TN) provides a 10 x 8-cm field of view with a reconstructed resolution of 2.25 mm in the central 40 mm of the field of view. Images are reconstructed using Fourier rebinning, followed by two-dimensional, filtered back projection. Mice were anesthetized with 1–2% vaporized isofluorane, and a microcatheter (Harvard Apparatus, Holliston, MA) was inserted into the external jugular vein. Pairs of mice were immobilized in a supine position upon custom-built support beds with attached anesthetic gas nose cones for data collection in the microPET-R4 scanner. 64Cu-labeled Abs in 150 µl of PBS were injected via a jugular vein catheter simultaneously with initiation of data collection, and the lines were flushed with saline to insure complete delivery of Abs. Data were collected continuously for 30–45 min. The delivered activity of each dose was determined by counting each sample syringe before and after injection using a dose calibration instrument (Radioisotope Calibrator CRC-12; Capintec, Ramsey, NJ). For later time points, mice were reanesthetized, immobilized, and scanned as described above. Quantitation of regions of interest was performed by viewing regions of interest over the selected tissues and averaging the activity concentration over the contained voxels. Images were reconstructed with AsiPro (Concorde Microsystems, Knoxville, TN), and region of interest analysis was performed with Analyze AVW 3.0 (Biomedical Imaging Resource, Mayo Foundation, Rochester, MN).
Biodistribution of labeled Abs
The biodistribution of 64Cu-DOTA-labeled anti-GPI IgG was determined in age- and sex-matched groups of BALB/c mice treated with either GK1.5 anti-CD4 mAb or RB6.8C5 anti-GR-1 mAb (250 µg was injected i.p. into each mouse 48 h before experimental use) and untreated FcRKO, B6129PF/1, WBB6F1W/Wv, and WBB6F1+/+ mice. After tail vein injection of the labeled Abs (250 µg for anti-GPI and 2 mg each for anti-collagen and isotype control Abs), the animals were allowed free access to food and water. At the specified time point postadministration, the mice were sacrificed, and the organs and tissues of interest were removed by dissection and weighed. The lower limbs were removed mid-femur and divided into two pieces: the knee sections (from mid-tibia/fibula to mid-femur) and ankle/paw sections (mid-tibia/fibula and entire rear paw) and were counted separately. The upper limbs were counted as a single unit after transecting them mid-humerus. The amount of radioactivity present in each sample was quantified by counting using an automatic well-type counter ( 8000; Beckman Coulter, Fullerton, CA). The percent injected dose per organ was calculated by comparison with a weighed and counted sample of the injectate. SDs are indicated with error bars. Student’s t tests were performed, and values for anti-GPI IgG samples, which significantly differ from normal IgG mouse values, are indicated with an asterisk in the figures (p < 0.01).
Preformed ICs
Soluble murine peroxidase-anti-peroxidase (mPAP) ICs consisting primarily of two HRP molecules bound to three anti-peroxidase IgG (31) were obtained from DAKO (Carpinteria, CA), dialyzed extensively against PBS to remove sodium azide, and sterile-filtered. A dose of 50 µl (250 µg of total protein) was selected for these experiments based upon pilot studies in which the minimum dose necessary to cause detectable changes in vascular permeability was determined using Evans Blue dye injected systemically (data not shown). For microPET experiments, mPAP ICs were injected during dynamic scanning via the external jugular catheter 15 min after labeled Abs were injected, and scanning was continued for an additional 30 min. For biodistribution experiments with mPAP, the ICs were injected via tail vein 15 min after labeled Ab injection i.v.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We wanted to identify the steps involved in the localization of anti-GPI Abs to the affected joints and the subsequent development of arthritis. To this end, we used mice deficient in FcR, C5, mast cells, and neutrophils, which previously had been shown to be resistant to disease induction, and examined using microPET the localization of anti-GPI Abs. Mice with a targeted disruption of the common FcRI-chain (32) lack expression of FcRI and FcRIII (hereafter referred to as FcRKO mice) and are completely resistant to joint inflammation after transfer of anti-GPI serum (21, 22) (Fig. 1A). FcRKO mice exhibited a complete block in localization of anti-GPI IgG to the front and rear limb joints compared with wild-type B6129F1 control mice (Fig. 1, E and I), indicating that FcR(s) are critical for proximal events leading to localization of arthritogenic Abs in the joints. The FcRIII was the FcR that involved in this localization, as shown by complete lack of joint accumulation in FcRIII-deficient mice (data not shown).
|
Differences in Ab joint localization identified by microPET analysis are confirmed by classical biodistribution studies
To confirm that Ab localization and disease resistance were distinct events and to test larger numbers of mice, we analyzed Ab localization by biodistribution studies following a similar protocol as that used in the microPET experiments. Mice were injected i.v. with anti-GPI IgG or control IgG and were killed after 30 min, and the front and rear limbs and knees were collected and measured in a gamma counter. The amount of localized Ab was determined by comparison with known standards of labeled Ab to yield the percentage of input dose per organ. Thirty minutes after injection of the Abs, groups of FcRKO, W/Wv, and RB6-depleted mice all exhibited decreased localization in the entire front and rear limbs relative to control mice (p < 0.01; Fig. 2). The amount of Ab present in the knee joints of all the mice was equivalent. C5-deficient mice were not tested by biodistribution, as they showed no defect in localization by microPET. These studies allowed us to confirm statistically our microPET findings, but lack the latter’s ability to focus precisely on the specific areas that develop disease. This highlights the power of the microPET in measuring more precise areas of Ab localization within the joints. Thus, FcRKO, mast cell-deficient, and neutrophil-deficient mice all show a dramatic reduction in anti-GPI IgG localization by both microPET and biodistribution analysis.
|
Our studies had shown that FcR were critical for the localization of anti-GPI Abs to the joints. GPI is found in three locations in the mouse: in the cytoplasm of all cells, in the circulation, and on the articular cartilage surface. We reasoned that FcR could be involved in the joint localization of the Abs in three different ways: 1) the anti-GPI Abs are binding to FcR-bearing cells in the blood, and these cells then bring the complexes into the joints; 2) the anti-GPI Abs directly bind GPI on the articular cartilage, and FcR cells then bind to these complexes; and 3) the anti-GPI Abs form ICs in the serum, which facilitate entry of the Abs into the joints. Because GPI is present in both serum and joints, we could not distinguish the precise role of FcRs and ICs. Type II collagen, in contrast, is relatively sequestered; it is found only in joint, tracheal and bronchial cartilage and in the vitreous humor of the eye. More importantly, it is not detected in the circulation and therefore is precluded from forming ICs in the serum. Immunization with type II collagen induces arthritis in the well-characterized model of CIA. In addition, a commercial mixture of four mAbs against distinct portions of murine type II collagen (34) (Arthrogen-CIA), can induce joint inflammation very similar to that induced by transfer of GPI-specific Abs (e.g., edema, synovitis, inflammation, cellular infiltrates, and eventual erosion of cartilage and bone) (11, 12, 34).
To determine the trafficking and localization of the anti-collagen Abs, pairs of BALB/c mice were injected with an arthritogenic dose of 2 mg of labeled anti-collagen mAb mixture or an isotype control mixture of irrelevant Abs and dynamically scanned for 1 h on the microPET. Surprisingly, no joint localization was observed in either mouse within the first hour or at 24 h (data not shown). Thus, the presence of circulating joint-specific Abs in the serum was not sufficient to induce joint localization and subsequent disease without some additional trigger.
IC administration facilitates anti-collagen and control Ab localization to the joints
Because anti-GPI IgG is capable of localizing to the joints without additional triggers, the presence of low levels of circulating GPI Ag (28, 35) may be an important factor in the localization of anti-GPI IgG. Injection of anti-GPI IgG could potentially provoke the formation of small numbers of ICs in the circulation, and formation of these ICs could play a significant role in modifying the distribution and accessibility of specific Abs to the joints. To test this hypothesis, the microPET/anti-CII experiments were repeated with the addition of irrelevant ICs after establishing a baseline scan for the labeled anti-collagen and control Abs. Fifteen minutes after beginning the scan and simultaneous injection of anti-collagen and control Abs, soluble preformed mPAP (31) were injected i.v. during continued scanning. Representative results from average ankle ROI analysis of three pairs of mice (Fig. 3) demonstrate that irrelevant mPAP complexes can exert dramatic effects on the local joint concentration of labeled Ab. Approximately 5 min after mPAP injection, there was a rapid increase in the concentration of labeled Ab within the ankle regions for both anti-collagen and control Abs (4- to 5-fold over baseline levels before mPAP injection). Ab accumulation reached plateau levels 10–15 min after mPAP injection (25–30 min into the scans) and remained stable thereafter. The timing and kinetics of Ab accumulation in the ankles in response to systemic mPAP injection are almost identical with those observed with anti-GPI IgG without any additional ICs (27), implying that a similar underlying mechanism is responsible. We next wanted to ascertain whether this joint localization following mPAP administration was specific for the anti-collagen Abs. There was no difference in the kinetics or overall level of Ab localization for specific (anti-collagen) and irrelevant (control) Abs (Fig. 3). These findings are consistent with soluble ICs inducing an overall change in vascular permeability that nonspecifically permits all Abs entry into the joint tissues. This could be occurring through either complement-mediated or FcR-mediated changes in vascular permeability. The maximal density of labeled anti-collagen and control Ab in the ankle region was lower than that for anti-GPI IgG (
0.015 vs 0.03 µCi/µl, on the average), and this may be due either to the difference in the radiospecificity of the two Abs injected (350 mCi/2.0 mg of 64Cu-anti-collagen compared with 350 mCi/250 µg of 64Cu-anti-GPI IgG) or the involvement of other factors that may augment localization of anti-GPI IgG. In our initial microPET studies, we reported that control Abs did not localize to the joints. Based on these findings with ICs, it is now clear that the lack of localization was due to the absence of ICs, not the lack of a target Ag in the joint. Thus, by studying the localization of anti-collagen Abs using microPET, we made the important observation that soluble ICs in the blood are required for arthritogenic Abs to gain access to their target organs.
|
To confirm the distribution and kinetics of anti-collagen localization with larger groups of mice and by a different method, we again conducted classical biodistribution studies with the same 2-mg dose of anti-collagen Abs or an isotype control mixture in BALB/c mice. Separate groups of three animals for each Ab type were treated with either PBS or mPAP at 15 min to parallel the microPET experimental protocol. Mice were injected i.v. with anti-collagen or control IgG, and separate groups of animals for each Ab type were treated with either PBS or mPAP at 15 min to parallel the microPET experiments. Animals were sacrificed after 45 min or 24 h, and a panel of organs and tissues was collected and measured in a gamma counter. As in our previous studies (27), biodistribution results (Fig. 4A) closely correlated with earlier microPET findings (Fig. 3), with greatly enhanced joint localization for both Ab types at 45 min compared with PBS-treated mice that received either anti-collagen or control Abs. Mock-treated mice (PBS at 15 min) lacked specific joint localization regardless of whether they received anti-collagen or control Abs (Fig. 4A) and did not show any difference either in the initial scan (data not shown) or the 24 h postinjection scan (Fig. 4B).
|
Identification of stages subsequent to Ab localization to the joint
The microPET and biodistribution studies clearly demonstrated that simply getting an Ab into the joint via IC-mediated increased permeability was not sufficient to cause disease, as control Abs entered the joints to the same level as the arthritogenic anti-collagen Abs, but did not cause arthritis (data not shown). Thus, a second stage in arthritis development is that once the Abs enter the joints, they must subsequently recognize a target Ag on the joint surface, such as collagen II or bound GPI to the articular cartilage.
A third stage must also exist after the autoantibodies enter the joint and bind to their Ag, which involves activation of the innate immune system and inflammation. We obtained evidence for this stage when we determined that CII Abs along with ICs injected into BALB/c mice did not result in any arthritis symptoms (data not shown) despite efficient localization in the joints (Figs. 3 and 4). Control mice, which received CII Abs plus LPS instead of ICs, did develop arthritis. Thus, there is another series of steps required for arthritis to develop, which LPS is capable of triggering. LPS is a pleiotropic mediator of inflammation and can activate several different molecules/pathways that could be directly involved, including IL-1, TNF, and Toll-like receptor-4. Thus, we have identified a third stage in disease induction, which involves the activation of innate effector mechanisms and inflammation.
Four-stage model of how autoantibodies induce arthritis
From our studies we have been able to experimentally delineate three stages involved in the initiation of arthritis by autoantibodies. We propose a four-stage working model in which we have added one additional stage. Some of the effector molecules and cells in this model are most likely involved in multiple stages. For example, neutrophils can be involved in stages 1 and 3. Importantly, the progression to these various stages is reversible, thus allowing several places where progression to arthritis can be prevented or reversed. The details of the stages are given below and are diagrammed in Fig. 5.
|
Initially, the injected Ab encounters Ag in the circulation, forms small soluble ICs, and engages FcRIII on neutrophils in the bloodstream, thus activating the neutrophils (37). The neutrophils, through the release of cytokines or other soluble mediators, cause a small change in the local vasculature, allowing ICs to access the mast cells found in close proximity to the microvasculature in the synovium. Next, the FcRIII ligation on mast cells causes the degranulation and release of mediators such as histamine and TNF- that are capable of causing rapid and more widespread changes in vascular permeability (38, 39, 40, 41, 42). Increased permeability amplifies the vascular change through increased mast cell triggering, and Ab enters the joints through mass flow of fluid phase proteins. After a short period (i.e., 20 min), vascular integrity is restored through exhaustion of the preformed mast cell contents and/or a refractory phase for the vasculature. Restoration of the vascular integrity would thus explain the observation that the amount of Ab levels off after 20 min (27). In the CII model, LPS administration must result in production and/or release of some vasoactive molecules, leading to the anti-CII Abs gaining access to the joint. There is most likely some low level trafficking of Abs into a normal joint, the level of which normally falls below that needed for any observable pathology. In the case of the administration of the 4-mg dose of anti-CII or in susceptible strains such as DBA/1J mice, there is sufficient trafficking of the Abs into the joints to account for the observed arthritis. This movement most likely occurs at a slower rate than what we observed with the ICs. What the ICs do essentially is greatly facilitate the rapid influx of the Abs into the joints above a threshold level. Thus, in stage 1, ICs cause entry of Abs into the joints through sequential interaction with PMNs and mast cells.
Stage 2: Recognition of target Ag in the joints. The second essential stage involves the binding of the Abs to target autoantigens in the joint (Fig. 5, stage 2). Once inside the joint, the anti-GPI and anti-CII Abs can bind to GPI or CII on the articular cartilage surface, respectively. The nonspecific Abs would not bind to any specific Ags, and would be cleared from the joints. It is possible to have some transient arthritis in the absence of a target Ag in the joint, when a high level of Abs get into the joint, as can be seen with serum sickness. Stage 2 provides an important checkpoint for the control of arthritis and shows that getting autoantibodies into the joints is necessary, but not sufficient, for arthritis to occur.
Stage 3: Activation of innate effector mechanisms and inflammation.
Once the Abs have bound to their respective Ags, they then would serve as substrates for activation of the innate immune system (Fig. 5, stage 3). This would involve both activation of the alternative pathway of complement, and potential activation of PMNs, NK cells, mast cells, and macrophages through the FcRs. The activation of these pathways and cells would result in the production of proinflammatory cytokines (IL-1 and TNF-) and chemokines (e.g., macrophage inflammatory protein-1 and monocyte chemoattractant protein-1). This is the step that LPS most likely acts upon in the CII model, as supported by the observation that either IL-1
or TNF- is sufficient to induce arthritis with anti-CII Abs in the absence of LPS (43). ICs alone in the blood must not activate the innate immune system to produce a threshold level of these cytokines/chemokines. After the production of the proinflammatory cytokines and chemokines, a strong inflammatory response then ensues in the joints. This would occur around 24–48 h, at which time the joint swelling is observed, and neutrophils are detectable histologically (18). The initiation of the inflammatory response would also cause long term changes in vascular permeability and improved access of Abs into the joint, as has been found in infectious foci of inflammation (44, 45, 46, 47), contributing to continued influx of autoantibodies and potential epitope spreading. Arthritis at this stage can also be reversed, as observed in the anti-GPI serum transfer model. The acute arthritis that develops after a single injection of anti-GPI Abs wanes over time, and by 14 days no inflammation is evident. Continual injection of Abs will maintain the arthritis; however, once the injections are stopped, the paws return to normal. Thus, stage 3 involves innate immune system activation, inflammation, and arthritis development.
Stage 4: Chronic disease. Once stages 1–3 have occurred, the chronic phase would occur in stage 4 (Fig. 5). This stage involves macrophages, synoviocytes, and lymphocytes. The pathology in the joints would include synovial hyperplasia, pannus formation and bone/cartilage erosion. Some of the same effector molecules involved in earlier stages could also be operating in this stage, i.e., TNF, whereas, other molecules may be unique to this stage, i.e., MMPs. Once this stage has occurred, the disease is not reversible, but some of the on-going inflammatory processes can be ameliorated with treatments such as TNF inhibitors.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our findings also may provide a link between infection and autoimmune disease. The process of postinfectious arthritis can be explained by our findings. An individual could contain circulating Abs against joint-specific Ags such as collagen, which under normal circumstances do not have access to the joints or cause disease. During infection, non-joint-specific ICs are generated, which could increase the joint vascular permeability, thus permitting the autoantibodies to gain access to the target tissue (stage 1). The autoantibodies would bind their target Ag (stage 2), provoking a transient inflammatory process that under many circumstances resolves spontaneously (stage 3). In some other individuals with higher levels of autoantibodies or other factors, however, the disease could become chronic rheumatoid arthritis (stage 4). In this manner, nonspecific ICs could initiate an autoimmune process. A similar mechanism may be responsible for the accumulation of ICs in the joints of patients suffering from serum sickness. Resolution of the temporary arthritis observed in serum sickness occurs through clearance of the ICs that lack specific targets in the joints.
Our inability to observe localization of monoclonal anti-CII to the joints without the addition of ICs is in contrast to a previous report by Stuart and Dixon (48). In that study they tracked the localization of coinjected 125I-labeled, affinity-purified, polyclonal anti-CII and 131I-labeled normal IgG in DBA/1J mice. They observed that at 30 min there was specific localization of the anti-CII to the hind feet, as determined by the CII/nIgG ratio, which peaked at 1 h. At 30 min, there was, however, a significant accumulation of nIgG in the joints, 50% of the anti-CII. There are several possible explanations for the differences in our two studies. The most likely explanation involves the mouse strains used. DBA/1J mice are the most susceptible strain to collagen-induced arthritis via immunization or transfer of antibodies. In our studies we tested BALB/c mice, which are less susceptible. In the anti-CII transfer model, DBA/1J mice also do not require LPS to induce disease, whereas all other strains do. Thus, one component of this enhanced susceptibility of DBA/1J mice may involve the enhanced ability of Abs to gain access to the joints via an unidentified mechanism, thereby, accounting for their observed localization. Another, but not mutually exclusive, possibility relates to a difference in trafficking to the joints between a pool of monoclonal and polyclonal anti-CII Abs. Future studies on the trafficking differences of Abs in DBA/1J and other strains may provide important insights into basis for the genetic susceptibility of DBA/1J to collagen-induced arthritis.
Why only distal joints are targeted by the formation of ICs in the serum remains to be determined. Several possibilities exist, including the number or location of mast cells surrounding the joint vasculature, the presence of activated neutrophils in the vasculature of the joints, or a fundamental difference in the joint endothelium itself. However, determination of these factors may allow for development of successful interventions for various joint inflammation involving soluble ICs, such as, reactive arthritis following bacterial or viral infection, serum sickness, systemic lupus erythematosus, and RA.
There is a large body of data implicating mast cells in rapid and dramatic alterations of vascular integrity and recruitment of inflammatory cells in response to FcR-mediated signals, including asthma (49), bullous pemphigoid (42), and peritonitis models (50, 51). This occurs through the release of preformed granule contents, including histamine, serotonin, TNF-, and proteases such as tryptase (52, 53, 54, 55). Cross-linking of FcRIII on mast cells has been shown to cause degranulation (56) and changes in vascular permeability (albeit in tissues other than the distal joints). More recently, mast cells were implicated in the K/BxN serum transfer model (20), although their precise role was not described. Neutrophils are not typically implicated in mediating rapid changes in vascular permeability, but the microPET and biodistribution data in neutrophil-depleted mice clearly indicated a proximal role in Ab localization. This could occur through the release of soluble mediators or proteases that act upon nearby endothelium. Thus, multiple cell types play specific and partially overlapping roles in the changing vascular permeability of the joints leading to autoantibody accessing the joints.
Overall, from our studies we were able to provide direct evidence in support of a four-stage model for the development of arthritis. This model positions ICs, neutrophils, and mast cells in an essential initial stage of the joint localization of Abs and provides a conceptual framework for the design and interpretation of therapeutic interventions of Ab-mediated types of RA.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Elan Pharmaceuticals, Inc., 800 Gateway Boulevard, Building 800, Room 154B, South San Francisco, CA 94080.
3 Address correspondence and reprint requests to Dr. Paul M. Allen, Department of Pathology and Immunology, Washington University School of Medicine, Campus Box 8118, 660 South Euclid Avenue, St. Louis, MO 63110. E-mail address: pallen{at}wustl.edu
4 Abbreviations used in this paper: RA, rheumatoid arthritis; CIA, collagen-induced arthritis; CII, type II collagen; DOTA, 1,4,7,10-tetraazacyclododecane-N,N',N'',N'''-tetraacetic acid; FcRKO, FcR1 deficient; GPI, glucose-6-phosphage isomerase; IC, immune complex; mPAP, mouse anti-peroxidase immune complex; PET, positron emission tomography.
Received for publication November 20, 2003. Accepted for publication April 9, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in antibody-induced arthritis. J. Exp. Med. 196:77.R signaling in the K/B x N serum transfer model of arthritis. J. Immunol. 169:6604. chain deletion results in pleiotrophic effector cell defects. Cell 76:519.[Medline]
-adrenoceptor stimulant. Am. J. Physiol. 253:H1586.
in neutrophil recruitment. Inflammation 19:119.[Medline]
and TNF-, and the noninvolvement of IL-6, in the development of monoclonal antibody-induced arthritis. J. Immunol. 169:1459./cachectin. Nature 346:274.[Medline]
RIII (CD16) deficient mice. Immunity 5:181.[Medline]Related articles in The JI:
This article has been cited by other articles:
|
|
 |
|
  T. N. Mayadas, G. C. Tsokos, and N. Tsuboi Mechanisms of Immune Complex-Mediated Neutrophil Recruitment and Tissue Injury Circulation, November 17, 2009; 120(20): 2012 - 2024. [Full Text] [PDF]
|
|
|
|
 |
|
 N. Rajasekaran, S. Solomon, T. Watanabe, H. Ohtsu, M. Gajda, R. Brauer, and H. Illges Histidine decarboxylase but not histamine receptor 1 or 2 deficiency protects from K/BxN serum-induced arthritis Int. Immunol., November 1, 2009; 21(11): 1263 - 1268. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Studelska, L. Mandik-Nayak, X. Zhou, J. Pan, P. Weiser, L. M. McDowell, H. Lu, H. Liapis, P. M. Allen, F. F. Shih, et al. High Affinity Glycosaminoglycan and Autoantigen Interaction Explains Joint Specificity in a Mouse Model of Rheumatoid Arthritis J. Biol. Chem., January 23, 2009; 284(4): 2354 - 2362. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Boross, P. L. van Lent, J. Martin-Ramirez, J. van der Kaa, M. H. C. M. Mulder, J. W. C. Claassens, W. B. van den Berg, V. L. Arandhara, and J. S. Verbeek Destructive Arthritis in the Absence of Both Fc{gamma}RI and Fc{gamma}RIII J. Immunol., April 1, 2008; 180(7): 5083 - 5091. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. D. McCall-Culbreath, Z. Li, and M. M. Zutter Crosstalk between the {alpha}2{beta}1 integrin and c-met/HGF-R regulates innate immunity Blood, April 1, 2008; 111(7): 3562 - 3570. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. K. Banda, K. Takahashi, A. K. Wood, V. M. Holers, and W. P. Arend Pathogenic Complement Activation in Collagen Antibody- Induced Arthritis in Mice Requires Amplification by the Alternative Pathway J. Immunol., September 15, 2007; 179(6): 4101 - 4109. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bergtold, A. Gavhane, V. D'Agati, M. Madaio, and R. Clynes FcR-Bearing Myeloid Cells Are Responsible for Triggering Murine Lupus Nephritis J. Immunol., November 15, 2006; 177(10): 7287 - 7295. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Fusello, L. Mandik-Nayak, F. Shih, R. E. Lewis, P. M. Allen, and A. S. Shaw The MAPK Scaffold Kinase Suppressor of Ras Is Involved in ERK Activation by Stress and Proinflammatory Cytokines and Induction of Arthritis J. Immunol., November 1, 2006; 177(9): 6152 - 6158. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. K. Banda, J. M. Thurman, D. Kraus, A. Wood, M. C. Carroll, W. P. Arend, and V. M. Holers Alternative Complement Pathway Activation Is Essential for Inflammation and Joint Destruction in the Passive Transfer Model of Collagen-Induced Arthritis J. Immunol., August 1, 2006; 177(3): 1904 - 1912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Scatizzi, J. Hutcheson, E. Bickel, J. M. Woods, K. Klosowska, T. L. Moore, G. K. Haines III, and H. Perlman p21Cip1 Is Required for the Development of Monocytes and Their Response to Serum Transfer-induced Arthritis Am. J. Pathol., May 1, 2006; 168(5): 1531 - 1541. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. D. Kim, R. C. Chou, E. Seung, A. M. Tager, and A. D. Luster A unique requirement for the leukotriene B4 receptor BLT1 for neutrophil recruitment in inflammatory arthritis J. Exp. Med., April 17, 2006; 203(4): 829 - 835. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. van Gaalen, A. Ioan-Facsinay, T. W. J. Huizinga, and R. E. M. Toes The Devil in the Details: The Emerging Role of Anticitrulline Autoimmunity in Rheumatoid Arthritis J. Immunol., November 1, 2005; 175(9): 5575 - 5580. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Del Nagro, R. V. Kolla, and R. C. Rickert A Critical Role for Complement C3d and the B Cell Coreceptor (CD19/CD21) Complex in the Initiation of Inflammatory Arthritis J. Immunol., October 15, 2005; 175(8): 5379 - 5389. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and mice deficient in FcR, C5, mast cells, and neutrophils (
) were determined from microPET scans as described in Materials and Methods. Mice were i.v. injected with 250 µg of 64Cu-labeled anti-GPI IgG and dynamically scanned for 30 min. Each data curve represents the combined average concentration of 64Cu activity within the region of interest for the left and right ankles of an individual mouse and is presented as the percentage of maximum 64Cu activity within the region of interest (for each experiment. Each plot is a separate experiment, and each strain of mouse was individually tested three or four times.
C'), producing the anaphylatoxin, C5a; i, C5a then activates multiple cell types (neutrophils, mast cells, macrophages, and endothelial cells) to produce proinflammatory molecules (i.e., TNF-