Splenectomy attenuates streptococcal cell wallinduced arthritis and alters leukocyte activation.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 48, No. 12, December 2003, pp 3557–3567 DOI 10.1002/art.11424 © 2003, American College of Rheumatology Splenectomy Attenuates Streptococcal Cell Wall–Induced Arthritis and Alters Leukocyte Activation Donald Kimpel, Tim Dayton, John Fuseler, Laura Gray, Krishnaswamy Kannan, Robert E. Wolf, and Matthew Grisham Objective. To investigate the role of the spleen in the pathogenesis of streptococcal cell wall (SCW)– induced arthritis and determine the impact of splenectomy on monocytes and T cells involved in the arthritis. Methods. Female Lewis rats were separated into 4 groups: 1) saline-injected, sham-operated; 2) salineinjected, splenectomized; 3) peptidoglycan– polysaccharide (PG-PS)–injected, sham-operated; and 4) PG-PS–injected, splenectomized. After a 10-day recovery period, rats received a single intraperitoneal injection of saline or PG-PS (25 g rhamnose/gm body weight). We evaluated the effect of splenectomy on joint inflammation, histopathology, leukocyte subtypes in blood and lymph nodes, cytokines, and cell surface expression of CD44 and CD45RC in the chronic phase of the disease (day 28). Results. Splenectomy dramatically decreased chronic joint inflammation and histopathologic damage as well as altered cell types in lymph nodes and peripheral blood, as analyzed by flow cytometry. Nitric oxide (NO) production, levels of interleukin-1␤ (IL-1␤), IL-6, tumor necrosis factor ␣, and a biomarker of Th1 cell predominance correlated with the level of joint inflammation. Surprisingly, in splenectomized animals, increased expression of adhesion molecules thought to track T cells to inflamed tissue were observed in lymph nodes. Conclusion. The result of splenectomy was attenuation of SCW-induced arthritis and changes in mediators of inflammation, including T cell subsets, proinflammatory cytokines, and NO production. Splenectomy may remove an important antigen reservoir and alter immune cell activation in the SCW-induced arthritis model. Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by joint inflammation and joint destruction, but encompassing a wide spectrum of features, which may cause joint deformity and extraarticular damage. It is currently held that genetic and environmental factors pattern this inflammatory response (1). Current models for understanding the pathogenesis of RA highlight interactions between T cells and professional antigen-presenting cells in initiating responses to antigens found in synovial tissue (2). Clinical evidence also indicates that the presence of articular cartilage is essential for the perpetuation of arthritis, although specific autoantigens have not yet been identified or characterized (3). Moreover, there is evidence that cartilage degradation is associated with the development of cartilage-responsive T cells (4). Therapies targeted to the immune system are effective in the treatment of RA, both in humans and in animal models (5–7). In the case of streptococcal cell wall (SCW)–induced arthritis, joint inflammation is suppressed by cyclosporin A, FK-506, or depletion of T cells (8–10). Much evidence from animal models of arthritis and human studies suggests that a Th1 mechanism is involved in inflammatory arthritis (11). However, evidence also suggests important roles for other cell types, such as mononuclear phagocytes (12). Recently, we have shown that significant changes occur in both T cells and mononuclear phagocytes during the development of Supported by the Center of Excellence for Arthritis and Rheumatology. Dr. Kimpel’s work was supported by a research grant from the NIH (R01-AR-46976) and by a Louisiana Education Quality Support Fund grant from the State of Louisiana Board of Regents (1988-01-RD-A-11). Donald Kimpel, MD (current address: University of Virginia, Charlottesville), Tim Dayton, MS, John Fuseler, PhD, Laura Gray, BS, Krishnaswamy Kannan, PhD, Robert E. Wolf, MD, PhD, Matthew Grisham, PhD: Louisiana State University Health Sciences Center, Shreveport. Address correspondence and reprint requests to Donald Kimpel, MD, Associate Professor of Medicine, Division of Rheumatology and Immunology, PO Box 800412, Charlottesville, VA 229080412. E-mail: firstname.lastname@example.org. Submitted for publication December 3, 2002; accepted in revised form September 4, 2003. 3557 3558 KIMPEL ET AL SCW-induced chronic inflammatory arthritis (13). These cellular changes were most evident in the spleen but also occurred in lymph nodes and peripheral blood. Together, these findings highlight the fact that an autoimmune disease such as RA is a systemic inflammatory process that uses several lymphoid organs in addition to the inflamed tissue itself. Use of these widely distributed lymphoid organs, such as the spleen and peripheral lymph nodes, requires a highway network (blood and lymphatics) for lymphocytes and other inflammatory cells, as well as appropriate traffic signals for them to reach their target. The adhesion molecules expressed on the inflammatory cell and on the endothelium of vessels provide these signals, which guide the trafficking and development of cells such as lymphocytes as they mature and mediate the inflammation process. The complex sequence of events in inflammatory arthritis as it transitions from an acute to a chronic process is poorly defined at present. However, the spleen as a central organ in the immune system may play a prominent role in the development of this systemic inflammatory process. Splenectomy has been used as an effective therapy in several autoimmune diseases, including idiopathic thrombocytopenic purpura (ITP) and Felty’s syndrome (FS) (14–17). Because the results of our previous studies (13,18) suggested a central role for the spleen, we hypothesized that splenectomy may alter the course of SCW-induced chronic arthritis. We set out to test this hypothesis and at the same time determine the impact of splenectomy on the cell populations we previously showed to be altered by the development of SCWinduced arthritis. In this study, we showed that splenectomy indeed reduces the chronic phase of arthritis in association with changes in cellular and cytokine parameters. This is the first study to show quantifiable immunologic changes and clinical effects of splenectomy on the development of chronic inflammatory arthritis. MATERIALS AND METHODS Reagents. The monoclonal antibodies used in this study were as follows: fluorescein isothiocyanate (FITC)–, phycoerythrin (PE)–, or biotin-conjugated mouse anti-rat antibodies to CD3, CD4, CD8b, CD11b/c, CD44, CD45R, CD45RC, and NKR-P1A (all from PharMingen, Mountain View, CA). Second-step reagent (streptavidin–allophycocyanin [APC]) was obtained from Biomeda (Foster City, CA). Appropriate murine isotype-matched negative controls were used to establish background fluorescence for each fluorochrome. The 10S fraction of peptidoglycan–polysaccharide (PG-PS) was purchased from Lee Laboratories (Grayson, GA). The material was briefly sonicated before use. All other chemicals and reagents used in this study were purchased from Sigma (St. Louis, MO) unless specified otherwise. Splenectomy. Female Lewis rats weighing a mean ⫾ SD of 100 ⫾ 10 gm were purchased from Harlan SpragueDawley (Indianapolis, IN). Rats were anesthetized by inhalation of isoflurane (fraction of inspired oxygen 0.35 1/1 isoflurane [IsoFlo] Abbott Laboratories, North Chicago, IL). Under sterile conditions, a midline laparotomy was performed. For a splenectomy, the spleen was gently mobilized, exteriorized, and the vascular supply was cut off by 2 ligatures (4-0 resorbable suture) placed around the vessels on the upper and lower poles of the spleen. The abdominal wall was closed in 2 layers, each by a running suture (4-0 resorbable suture). In shamoperated animals, the spleen was exteriorized after a midline laparotomy and was gently mobilized before closing the abdominal wall. Rats were clinically monitored during the immediate recovery phase following inhalation of anesthesia. During postoperative monitoring, rats were observed daily. All animal surgical procedures, including administration of anesthesia, laparotomy, and splenectomy, were approved by the Institutional Animal Care and Use Committee (IACUC). Groups of animals used in the study were as follows: 1) saline-injected, sham-operated (C/Sh); 2) saline-injected, splenectomized (C/Spl); 3) PG-PS–injected, sham-operated (PG/ Sh); and 4) PG-PS–injected, splenectomized (PG/Spl). In each experiment, 16 animals were randomly divided into 4 groups composed of 4 animals/group. For histology, joint tissues were processed from 1 or 2 representative animals per group. Splenectomy studies were repeated 4 times with various combinations of antibodies. The arthritis responses and cellular changes between experiments were similar. Unless stated otherwise, the data reported here are from a single experiment using a single lot of PG-PS. During the surgical procedure, 1 rat from the C/Sh group died; thus, the C/Sh group had only 3 rats. Induction of rat arthritis by PG-PS injection. The standard protocol for SCW-induced arthritis was followed as previously described (8,19). Briefly, experimental rats were given a single intraperitoneal (IP) injection of PG-PS (25 g rhamnose/gm body weight). Control animals were injected with an equal volume of saline. Rats were observed daily, and the development of arthritis was assessed by objective and histopathologic criteria. For all splenectomy experiments, surgical procedures involving sham operation and splenectomy were performed using sterile techniques as applicable. Animals were allowed to recover from surgery for 10 days. At that point, controls received saline only, whereas PG-PS groups received a single IP injection of PG-PS. The day of PG-PS injection was day 0, and all rats were killed on day 28. Animals were fasted overnight prior to killing to allow for measurement of endogenous nitrate and nitrite (NOx). All animal protocols described in this study were approved by the IACUC. Evaluation of arthritis. Paw edema was measured by immersion of the shaved paw to a marked line above the ankles in a water plethysmometer, as previously described (8,13). The mean swelling in both hind paws was added together to yield a single data point. The change from the initial paw volume (preinjection or day 0) was calculated. Data are presented as the mean change in paw volume per group. Joint swelling was scored (arthritis index) according to IMPACT OF SPLENECTOMY ON SCW-INDUCED ARTHRITIS a standardized method by an experienced observer (TD). Briefly, a score of 0–4 was assigned as follows: 0 ⫽ no evidence of hyperemia and/or inflammation; 1 ⫽ hyperemia with little or no paw swelling, 2 ⫽ swelling confined predominantly to the ankle region, with modest hyperemia, 3 ⫽ increased paw swelling and hyperemia of the ankle and metatarsal regions, and 4 ⫽ maximal paw swelling and hyperemia involving the ankle, metatarsal, and tarsal regions. The scores for each paw were summed, for a maximum possible score of 16. For histologic evaluation, ankle (hind paw) joints were fixed at 4°C in Zamboni’s fixative and decalcified in an extraction buffer, as previously described (8). The joints were split, placed in cryomolds containing tissue-freezing medium (Triangle Biomedical Sciences, Durham, NC), and slow-frozen in liquid nitrogen. Serial sections (10–12 m thick) were cut parallel to the long axis of the joint. The sections were stained with Masson’s trichrome stain (Sigma). Images were captured using a Nikon E600 microscope equipped with a SynSys digital camera (Photometrics, Tucson, AZ). Exposure times were automatically determined using the Trichrome Image dialog of MetaMorph software (Universal Imaging, Downingtown, PA). Flow cytometry analysis. Peripheral blood and lymph node cells were isolated as previously described (13). For immunofluorescence staining, cells were washed twice in phosphate buffered saline supplemented with 1% bovine serum albumin (fluorescence-activated cell sorter [FACS] buffer), incubated on ice for 30 minutes with saturating concentrations of appropriate monoclonal antibodies or isotype controls, washed 3 times in FACS buffer, fixed in 2% paraformaldehyde, and resuspended in FACS buffer. When appropriate, cells were incubated with biotin-conjugated monoclonal antibodies, washed 3 times, incubated for 30 minutes with the relevant streptavidin conjugate, washed, and then fixed in paraformaldehyde. Cell fractions were gated on viable cells, and samples were analyzed using a FACSVantage flow cytometer (Becton Dickinson, Franklin Lakes, NJ). Fluorescence was detected at 525 nm (FITC), 590 nm (PE), and 660 nm (APC); data were analyzed using CellQuest software (Becton Dickinson). Typically, 10,000 cells were analyzed based on forward versus side scatter gating. Phenotype analysis. Lymph node and blood cells were analyzed in the lymphocyte gate using forward and side scatter patterns. After gating, fluorescence was analyzed for single-, dual-, or 3-color analysis using a FACSCalibur flow cytometer (Becton Dickinson) to detect the cell surface expression of various markers. Immunostaining results were expressed in most cases as the percentage of positive cells in blood and lymph nodes. Cell types were defined in lymph nodes and peripheral blood by flow cytometry using well-established lineage-specific markers. All T cells were analyzed in the lymphocyte gate for the presence of CD3⫹, CD4⫹, and CD8⫹ cells, and T cell subsets were evaluated for CD44 and CD45RC expression. B cells were identified by the expression of pan–B cell marker CD45R (B220). Natural killer (NK) cells were identified by the expression of NKR-P1A. Monocytes in the lymphocyte gate were identified by CD11b/c positivity and by CD3 negativity; the identity of the monocytes in the lymphocyte gate was confirmed by backgating, by comparison with the CD3⫺ population, and by analysis of cells in the monocyte gate. Leukocyte counts and differential counts were confirmed morphologically by automated cell counting and by light microscopy. 3559 Figure 1. Effect of splenectomy on hind paw volume in streptococcal cell wall (SCW)–induced arthritis. Experimental arthritis was induced by a single intraperitoneal injection of peptidoglycan-polysaccharide (PG-PS) in Lewis rats, as outlined in Materials and Methods. The increase in hind paw volume (joint swelling) is characteristic of SCW-induced experimental arthritis in Lewis rats during a 28-day period. Rats in the PG-PS ⫹ splenectomy group had significantly reduced chronic arthritis compared with sham-operated, PG-PS– injected rats. Values are the mean ⫾ SD (n ⫽ 15). Cytokine measurement. Concentrations of tumor necrosis factor ␣ (TNF␣) and interleukin-1␤ (IL-1␤) in plasma samples from 2 experiments were determined using rat enzyme-linked immunosorbent assay kits (R&D Systems, Minneapolis, MN) according to the manufacturer’s recommendations. Plasma IL-6 levels were measured by 7TD1 bioassay, as previously described (20,21). Both negative and positive controls were included in each run. TNF␣ and IL-1␤ were expressed in pg/ml per million cells, whereas IL-6 levels were expressed in units/ml. Measurement of nitrate and nitrite. Rats were fasted overnight before collecting plasma samples from individual rats. All samples were stored at ⫺70°C until analyzed. Levels of NOx were measured by Griess reaction, in which plasma nitrate was converted to nitrite by nitrate reductase (22,23). Statistical analysis. Using GraphPad Prism software (GraphPad Software, San Diego, CA), each data set was subjected to analysis of variance followed by post hoc analysis with Bonferroni adjustment for multiple comparisons. Individual results are expressed as the mean ⫾ SD. Differences between groups were considered statistically significant when the P value was less than 0.05. RESULTS Effect of splenectomy on the development of SCW-induced arthritis. SCW-induced arthritis was produced in female Lewis rats by a single IP injection of PG-PS, as described previously (8). All rats in the PG/Sh group developed acute and chronic joint inflammation. Over a period of 28 days, the joint inflammation fol- 3560 KIMPEL ET AL Figure 2. Reduction of the arthritis index in peptidoglycan–polysaccharide (PG-PS)–injected, splenectomized (PG/Splenect.) rats. Splenectomy significantly reduced the arthritis index in rats in the PG/Splenect. group compared with that in the PG-PS–injected, sham-operated (PG/Sh) group. Initially, there was acute inflammation both in PG/Sh and in PG/Splenect. rats. However, from day 13 onward, the arthritis index and paw edema were stable in the PG/Splenect. rats. The arthritis index was 0 in all saline-treated controls. Values are the mean ⫾ SD (n ⫽ 15). lowed a typical pattern, progressing from the initial acute phase to the remission phase followed by spontaneous reactivation of persistent chronic arthritis (8,13). PG/Sh rats were compared with PG/Spl rats. Splenectomy had a striking effect on the development of arthritis. The data in Figure 1 clearly show that after day 11 post–PG-PS injection, paw joint volume was significantly reduced in PG/Spl rats. Similar changes were observed in the arthritis index, as shown in Figure 2. Effect of splenectomy on joint histopathology. Figure 3 shows the effect of splenectomy on PG-PS– induced joint histopathology. In the normal joint (Figure 3A), the articular cartilage was characterized by an intact smooth surface and exhibited metachromatic (a mixture of red and green) staining with Masson’s trichrome stain. Minimal bone remodeling was present at the articular cartilage–synovium interface, and the synovial fluid was clear and acellular. The synovial membrane was typically composed of ⬃2 cell layers (synoviocytes), which stained intensely red. The synovial membrane was supported by the connective tissue of the capsule, which stained green. Synovitis, extensive pannus formation, and erosion of the articular cartilage and subchondral bone characterized the histopathology of chronic joint inflammation in this model of polyarthritis (Figure 3B). Numerous small, red-staining inflammatory cells were densely localized in the synovium, the synovial space, Figure 3. Joint histology evaluated on day 28 in sham-operated and splenectomized rats injected with saline or PG-PS. A, Sham-operated, saline-treated animal with normal joint morphology. B, PG/Sh rat showing typical PG-PS–induced damage to the joint, characterized by synovitis, pannus development, and erosion of the articular cartilage and subchondral bone. C, Saline-injected, splenectomized control rat. The morphology of the joint in this group has the same appearance as that of the normal control animals. D, PG/Splenectomy animal. Joint histopathology is predominantly characterized by synovitis accompanied by some erosion of the marginal bone and articular cartilage. Development of pannus is absent, and there is minimal erosion of the subchondral bone. a ⫽ articular cartilage; sb ⫽ subchondral bone; p ⫽ pannus; sv ⫽ synovial villus. See Figure 2 for other definitions. (Original magnification ⫻ 200.) IMPACT OF SPLENECTOMY ON SCW-INDUCED ARTHRITIS 3561 and in and along the trailing edge of pannus. Additionally, there were large multinucleated red-staining cells (presumed to be osteoclasts), which appeared to be closely associated with erosion of marginal bone and cartilage at the synovium–bone interface and especially with erosion of the subchondral bone (Figure 3B). The pannus also contained green-staining loose connective tissue fibers. The joint morphology of the C/Spl animals (Figure 3C) showed no difference from that in the C/Sh animals (Figure 3A). In the PG/Spl animals, joint inflammation appeared to be confined to the synovium– bone interface at the joint margins (Figure 3D). The inflammation seen was predominantly synovitis. There was some marginal bone erosion and loss of articular cartilage, but these did not appear to be extensive. Interestingly, there was suppression of the development of pannus, and subchondral bone erosion was also conspicuously absent. Effect of splenectomy on lineage markers of leukocytes. Peripheral blood. In PG-PS–treated rats, a decrease in the percentage of CD3⫹ cells was previously Figure 5. Effect of splenectomy on the CD11⫹ cell population in peripheral blood (top) and lymph nodes (LN) (bottom). On day 28 after peptidoglycan–polysaccharide (PG-PS) injection, the monocyte population was analyzed using CD11b/c monoclonal antibodies (see Materials and Methods). Values are the mean and SD (n ⫽ 15). C/Sh ⫽ saline control, sham-operated; C/Spl ⫽ saline control, splenectomized; PG/Sh ⫽ PG-PS–injected, sham-operated; PG/Spl ⫽ PG-PS–injected, splenectomized. Figure 4. Effect of splenectomy on T cell subsets in peripheral blood (top) and lymph nodes (LN) (bottom). Lewis rats were sham operated or splenectomized 10 days prior to injection of saline or peptidoglycan–polysaccharide (PG-PS). On day 28, rats were killed and cells were analyzed using lineage-specific markers (see Materials and Methods). T cells were identified as CD3⫹, and the T cell subset identified as CD3⫹,CD4⫹ was analyzed as a percentage of the CD3⫹ population. Values are the mean ⫾ SD (n ⫽ 15). observed due to the large increase in CD11⫹ cells, some of which were captured in the lymphocyte gate. Splenectomy also produced a nonsignificant decrease in CD3⫹ cells; however, subsequent subset analysis did not show any differences in the percentage of CD3⫹,CD4⫹ (Figure 4) or CD3⫹,CD8⫹ (data not shown) cells. There was no difference in the percentage of B cells or NK cells (data not shown). One of the striking changes in peripheral blood was the increase in CD11⫹ cells in both PG/Spl and PG/Sh rats (Figure 5). Between the PG-PS groups, PG/Spl rats had a lower percentage of CD11⫹ cells than did PG/Sh rats (P ⬍ 0.05). Lymph nodes. In lymph nodes, the percentage of total CD3⫹ cells was lower in both of the splenectomized groups (C/Spl and PG/Spl) compared with the sham-operated comparator groups. When the percentage of CD3⫹,CD4⫹ cells was analyzed, there were no significant differences between groups (Figure 4). Further subset analysis did not reveal changes in the percentages of CD3⫹,CD8⫹ or NK cells (data not shown). 3562 KIMPEL ET AL Figure 6. Expression of CD45RC in peripheral blood and lymph nodes. Expression of CD45RC was analyzed using Ox-22 monoclonal antibodies, and mean channel fluorescence was determined. CD45RChigh-expressing cells are associated with a Th1 phenotype (see Results). Values are the mean ⫾ SD (n ⫽ 15). PG-PS ⫽ peptidoglycan–polysaccharide; Splenect. ⫽ splenectomized. However, B cell levels were increased in C/Spl and PG/Spl rats compared with their respective controls (P ⬍ 0.01) (data not shown). A significant decrease was also found in the CD11⫹ monocyte population in PG/ Spl rats compared with PG/Sh rats (P ⬍ 0.05), although the total percentage of CD11⫹ cells was very low, as expected (Figure 5). Effect of splenectomy on T lymphocyte subsets. The role of Th1 lymphocytes in RA pathogenesis is one of the most intensely researched areas (24); however, the characterization of the particular subset of T cells that drives the inflammatory process is a subject of controversy. In this regard, characterization of CD45RC expression by CD4⫹ T cells deserves mention because there is a direct correlation between these cells and the pathogenic process in several autoimmune models (25). To better understand the role of these cells in SCWinduced arthritis, CD3⫹,CD4⫹ T cells were analyzed for the expression of CD45RC. PG/Sh animals demonstrated increased CD45RC in association with SCWinduced arthritis in both blood and lymph nodes (Figure 6). Splenectomy resulted in decreased CD45RC expression in T cells isolated from peripheral blood and lymph nodes in both saline- and PG-PS–treated rats (Figure 6). However, in PG-PS–treated rats, the difference between sham-operated and splenectomized groups was very pronounced (P ⬍ 0.01). These results are consistent with Figure 7. Effect of splenectomy on CD44 expression in T cells in peripheral blood (top) and lymph nodes (LN) (bottom). Intensity of CD44 antigen expression was analyzed as the mean channel fluorescence in T cells (CD3⫹), CD4⫹ T cells (CD3⫹,CD4⫹), and CD8⫹ T cells (CD3⫹,CD4⫺). CD44 is an adhesion and homing molecule that is expressed at high levels on lymphocytes in joint fluid (see Results). Values are the mean ⫾ SD (n ⫽ 15). PG-PS ⫽ peptidoglycan–polysaccharide. IMPACT OF SPLENECTOMY ON SCW-INDUCED ARTHRITIS our understanding that CD45RChigh expression is associated with proinflammatory Th1 potential and demonstrate a correlation between the degree of arthritis and CD45RC expression (26,27). Effect of splenectomy on CD44 expression. Several studies have suggested that CD44high phenotype is a marker of activation, is associated with aggressive inflammatory phenomena in various autoimmune models, and has a propensity to traffic to joints (28–30). We had previously seen that CD4⫹ T cells isolated from the joints of PG-PS–treated animals expressed high levels of CD44 compared with those from the lymph nodes (13). Here we evaluated the effect of an arthritis-suppressing therapy (splenectomy) on the cell surface expression of CD44 by various T cell subsets in lymph nodes and peripheral blood. In the blood, there was no difference in CD44 mean channel fluorescence in any T cell subsets (Figure 7). However, in the lymph nodes, the expression of CD44 on T cell subsets (CD3⫹, CD3⫹,CD4⫹, and CD3⫹,CD4⫺) increased significantly in PG/Spl rats compared with PG/Sh rats. In saline-treated rats, the same trend was present but reached significance only in the CD3⫹,CD4⫺ population. In both cases (saline and PG-PS treatment), splenectomy appeared to result in accumulation of T cells expressing high levels of CD44 in the lymph nodes, although these rats had no arthritis (saline) or had attenuated arthritis (PG-PS). Splenectomy suppression of nitrate and nitrite production in SCW-induced arthritis. NOx in plasma and serum samples served as a measure of nitric oxide (NO) production in vivo (21–23). As previously observed, increased NO production was demonstrated in PG/Sh rats, the most profoundly arthritic group (P ⬍ 0.01) (Figure 8). Compared with the C/Sh rats, C/Spl rats showed a small elevation of NOx levels, which was not statistically significant. However, splenectomy resulted Figure 8. Alteration of nitrate and nitrite (NOx) levels in plasma by PG-PS and by splenectomy (Splenect.). Plasma was obtained from all fasted animals on day 28, and NOx was measured (see Materials and Methods). NOx levels correlated with the severity of arthritis. Values are the mean and SD (n ⫽ 15). See Figure 5 for other definitions. 3563 Figure 9. Changes in plasma cytokine levels in sham-operated and splenectomized rats after PG-PS treatment. Plasma concentrations of interleukin-1␤ (IL-1␤) and tumor necrosis factor ␣ (TNF␣) were measured by enzyme-linked immunosorbent assay (n ⫽ 32). The IL-6 level was determined by 7TDI bioassay (n ⫽ 15). Cytokine levels correlated with the arthritis severity. Values are the mean and SD. See Figure 5 for other definitions. in significant reductions in NOx levels in PG/Spl rats, down to the level in the saline controls (P ⬍ 0.01). Thus, NO production paralleled the severity of the arthritis. Effect of splenectomy on cytokines. Several studies have shown that proinflammatory cytokines play a pivotal role in driving the inflammatory activity in arthritis (31,32). To elucidate the underlying mechanisms involved in the attenuation of chronic inflammation in PG/Spl rats, plasma levels of TNF␣, IL-1␤, and IL-6 were measured. As shown in Figure 9, the PG/Sh group had significantly elevated plasma levels of TNF␣, IL-1␤, and IL-6 compared with those in the C/Sh group. In PG/Spl animals, TNF␣, IL-1␤, and IL-6 concentra- 3564 KIMPEL ET AL tions dropped to the basal levels similar to those in the C/Sh group. DISCUSSION This is the first study to show a suppression of chronic joint inflammation as measured by changes in the arthritis index and paw volume in Lewis rats when splenectomy was performed prior to the induction of experimental arthritis. Joint histopathology showed significantly reduced pannus and conspicuously absent subchondral bone erosion. Our previous studies of cellular changes in arthritis showed that PG-PS induced an increase in CD11⫹ monocytes with a concomitant decrease in T cells, while activation markers increased in both cell types, most obviously in the spleen (13). Activated neutrophils, monocytes, and lymphocytes are present in inflamed joint tissues, but their journey may be initiated by events at distant sites, such as the lymph nodes and spleen. Splenectomy, however, induced cellular changes in PG-PS–treated animals, including a reduction in CD11⫹ cells, a reduction in T cells, and an increase in B cells. Our previous time course studies showed that both monocytes and T cells in the spleen are activated by day 5 after injection of PG-PS (13). It is therefore difficult to determine if one cell type leads to activation of the other or if there is synergistic coactivation of T cells and CD11⫹ cells. In any case, splenectomy decreased the numbers and the activation of both T cells and CD11⫹ cells. Splenectomy also led to a reduction in plasma levels of NOx and proinflammatory cytokines compared with those in the PG/Sh animals. The spleen, a secondary lymphoid organ, is a major site of immune surveillance, antigen recognition, activation, and clonal proliferation. We hypothesized that splenectomy may alter the dynamics of the cellular and humoral elements that play a vital role in joint inflammation (14,15). Splenectomy as a therapeutic intervention has been used in autoimmune diseases such as chronic ITP and FS (14–17). Unlike these primary antibody–mediated disorders, antibodies in SCWinduced arthritis are generally not considered the primary driving force (13). Anecdotally, patients with sickle cell disease have a low incidence of autoimmune diseases, and RA coexistent with sickle cell disease has rarely been reported (33). Although it is difficult to separate the role of genetic differences, this observation indirectly suggests that the autosplenectomy that occurs in patients with sickle cell disease may protect against the development of autoimmune disorders. Splenectomy could be beneficial to arthritis patients, based on our data and the beneficial effects specific to FS, but clinical evidence is contradictory. A recent literature search found one case of long-term improvement in arthritis following splenectomy in a patient with FS (34). However, in another case, a patient who had features of FS but without arthritis who was treated with splenectomy later developed inflammatory arthritis (35). It is possible that selected patients with RA could benefit from splenectomy. Schwab and coworkers investigated the role of the spleen as a site of antigen deposition in SCWinduced arthritis by demonstrating the presence of polysaccharide complex within splenic macrophages, synovial macrophages, and neutrophils during various stages of inflammation in SCW-induced arthritis (36–39). These studies suggest that splenic macrophages and Kupffer cells of the liver can act as antigen reservoirs for PG-PS. It has been shown that degradation of PG-PS by mutanolysin reduced the amount of complexes in the spleen and the liver, and that the smaller-sized complexes did not induce chronic arthritis (38). More recently, Schrijver and other investigators (11,40) have proposed that PG, a major component of the cell wall of gram-positive bacteria, plays a role in the pathogenesis of RA. In the studies described here, it is likely that splenectomy resulted in the loss of antigen reservoirs and a loss of cell populations that are very important for joint inflammation, thus resulting in altered immune kinetics and attenuation of joint damage in the PG/Spl group. To evaluate for gross hematologic alterations, we performed a total white blood cell (WBC) count, total lymphocyte count, and hematocrit in all 4 groups of rats. Splenectomy in Lewis rats did not cause any major changes in the total WBC count, total lymphocyte count, and hematocrit in saline- and PG-PS–injected rats. In RA, neutrophils and mononuclear phagocytes play a prominent role in joint destruction (5,12), but the predominance of T cells in synovial tissue with an activated/memory phenotype has led to the conviction by many that T cells play a prominent role in the disease process. In fact, T cells, neutrophils, and monocytes all appear to play a role (5,41). This view is supported in the SCW-induced arthritis model in which the evidence for a T cell role is strong (18,42), while neutrophils and monocytes are also associated with disease activity (2,36,43,44). Consistent with this view, we have quantitated changes in both T lymphocytes and monocytes during the course of development of SCW-induced arthritis (13). IMPACT OF SPLENECTOMY ON SCW-INDUCED ARTHRITIS The percentage of CD3⫹ T cells in the lymph nodes of both saline and PG-PS groups decreased after splenectomy. One possible explanation for this is altered homeostasis in peripheral compartments, leading to accumulation in other secondary lymphoid tissues after such as Peyer’s patches. A second possibility is that altered regulation leads to an overall reduction in the T cell repertoire. In contrast, the B cell population was found to be increased in splenectomized rats. Analysis of the monocyte population produced 2 observations. As compared with the C/Sh rats, the PG/Sh rats showed a significant increase in the percentage of monocytes in both blood and lymph nodes, a finding consistent with our previous observations. However, upon splenectomy, there was a significant drop in the monocyte population; this may be influenced by the changes in T cells or may be due to other factors, such as the loss of the splenic PG-PS reservoir. Synovial T cells from RA patients have been described as CD44bright, and CD44 is thought to be important in trafficking to inflamed joints (29). In the SCW-induced arthritis model, we have also seen that CD4⫹ T cells in inflamed joints express high levels of CD44, and expression in blood and the spleen is increased by PG-PS treatment (13). Surprisingly, expression of CD44 was increased even more dramatically in lymph nodes after splenectomy, although the arthritis was attenuated. There are several possible explanations for this. CD44 may be one of several adhesion molecules necessary for entry into joints, or it may be an inexact marker of adhesion capability. In any case, the role of CD44 in the trafficking of cells to sites of inflammation may be more complex than previously appreciated. To further characterize the T cells involved in this arthritis model, we used an antibody that recognizes one epitope of CD45, CD45RC, the expression of which defines two subpopulations of functionally distinct, mature CD4⫹ T cells. The CD45RChigh cells produce a predominance of interferon-␥ and IL-2, provide help to B cells during primary immune responses, are active in graft-versus-host responses, can cause autoimmunity, and are the precursors of the CD45RClow population (25,45). In contrast, the CD45RClow population contains the majority of T helper cells for secondary humoral responses, produces more IL-4 than the CD45RChigh subpopulation, proliferates to recall antigens, and contains cells that suppress some autoimmune manifestations (27,46,47). Similar cell populations have been described in a mouse model of inflammatory bowel disease (48–50). Consistent with this pattern, PG/Sh rats exhibited 3565 high expression of CD45RC antigen in concert with active inflammation. In comparison, we observed a reduction in CD45RC mean fluorescence in PG/Spl rats to the levels in saline-treated rats. Taken together, our data also support the current belief that CD45RC is an important marker of inflammation in the rat model and an indicator of a Th1-type inflammatory response. To examine noncellular indicators of inflammation, we measured levels of NO, IL-1␤, IL-6, and TNF␣. Although characterization of NO as pro- or antiinflammatory is often difficult, the involvement of NO in immune activation and inflammatory processes is now well established (for review, see ref. 51). TNF␣ and IL-1 are well-established proinflammatory cytokines in RA and are important targets of current treatment strategies, while IL-6 is a pluripotent inflammatory cytokine produced by numerous cell types during inflammation (31,52,53). Furthermore, proinflammatory cytokines such as TNF␣ and/or IL-1 can potentiate the production of NO in cultured synoviocytes, synovial fibroblasts, and articular chondrocytes (8,54). Our results demonstrated that levels of NOx, IL-1␤, IL-6, and TNF␣ correlate with the degree of arthritis and are decreased in parallel with the arthritis in PG/Spl animals. We have also observed that there is a clear difference between splenocytes and lymph node cells isolated from naive rats when cultured in the presence of 10 g/ml PG-PS. Lymph node cells responded minimally, whereas the splenocytes produced ⬃50-fold more TNF␣ (Kimpel D, et al: unpublished observations). Joint histopathology of the PG/Sh rats demonstrated extensively developed pannus and the associated erosion of articular cartilage and subchondral bone, as seen previously in SCW-induced arthritis (8). In PG/Spl rats, joint histopathology revealed that there was a suppression of pannus formation and minimal erosion of the articular cartilage and subchondral bone. The results of the histopathologic evaluation strongly correlated with the clinical, serologic, and cellular inflammatory changes that occurred in response to splenectomy. In summary, splenectomy prior to induction of arthritis has a pronounced effect on the development of chronic inflammation, and several humoral (NOx, IL1␤, IL-6, and TNF␣) and cellular (CD45RC, CD11⫹) biomarkers correlate with the degree of inflammation. The impact of splenectomy is likely mediated by removal of a critical antigen reservoir of PG-PS, with diminished immune cell activation and attenuated joint inflammation. Finally, an important role of secondary lymphoid organs in the systemic process of inflammation in RA is 3566 KIMPEL ET AL emphasized, with the spleen playing an important role in T cell–dependent arthritis induced by PG-PS. ACKNOWLEDGMENTS We thank Dr. Robert Chervenak and Ms Deborah Chervenak of the Core Facility for Flow Cytometry for technical assistance and thoughtful discussion. 16. 17. 18. 19. REFERENCES 1. Goronzy JJ, Weyand CM. Rheumatoid arthritis: A. Epidemiology, pathology, and pathogenesis. In: Klippel JH, Weyand CM, Wortmann RL, editors. Primer on the rheumatic diseases. Atlanta: Arthritis Foundation; 1997. p. 155–61. 2. Janossy G, Panayi G, Duke O, Bofill M, Poulter LW, Goldstein G. Rheumatoid arthritis: a disease of T-lymphocyte/macrophage immunoregulation. Lancet 1981;2:839–42. 3. Laskin RS. Total condylar knee replacement in patients who have rheumatoid arthritis: a ten-year follow-up study. J Bone Joint Surg Am 1990;72:529–35. 4. Alsalameh S, Mollenhauer J, Hain N, Stock KP, Kalden JR, Burmester GR. Cellular immune response toward human articular chondrocytes: T cell reactivities against chondrocyte and fibroblast membranes in destructive joint diseases. Arthritis Rheum 1990;33: 1477–86. 5. Allen JB, Wong HL, Costa GL, Bienkowski MJ, Wahl SM. Suppression of monocyte function and differential regulation of IL-1 and IL-1ra by IL-4 contribute to resolution of experimental arthritis. J Immunol 1993;151:4344–51. 6. Butler DM, Malfait AM, Maini RN, Brennan FM, Feldmann M. Anti-IL-12 and anti-TNF antibodies synergistically suppress the progression of murine collagen-induced arthritis. Eur J Immunol 1999;29:2205–12. 7. Moreland LW. Potential biologic agents for treating rheumatoid arthritis. Rheum Dis Clin North Am 2001;27:445–91. 8. Fuseler JW, Hearth-Holmes M, Grisham MB, Kang D, Laroux FS, Wolf RE. FK506 attenuates developing and established joint inflammation and suppresses interleukin 6 and nitric oxide expression in bacterial cell wall induced polyarthritis. J Rheumatol 2000;27:190–9. 9. Van den Broek MF, van Bruggen MCJ, Stimpson SA, Severijnen AJ, van de Putte LBA, van den Berg WB. Flare-up reaction of streptococcal cell wall induced arthritis in Lewis and F344 rats: the role of T lymphocytes. Clin Exp Immunol 1990;79:297–306. 10. Lens JW, van den Berg WB, van de Putte LB, Berden JH, Lems SP. Flare-up of antigen-induced arthritis in mice after challenge with intravenous antigen: effects of pre-treatment with cobra venom factor and anti-lymphocyte serum. Clin Exp Immunol 1984;57:520–8. 11. Gaston JS. The involvement of the gut in the pathogenesis of inflammatory synovitis [editorial]. Br J Rheumatol 1995;34:801–2. 12. Burmester GR, Stuhlmuller B, Keyszer G, Kinne RW. Mononuclear phagocytes and rheumatoid synovitis: mastermind or workhorse in arthritis? Arthritis Rheum 1997;40:5–18. 13. Kimpel D, Dayton T, Kannan K, Wolf RE. Streptococcal cell wall arthritis: kinetics of immune cell activation in inflammatory arthritis. Clin Immunol 2002;105:351–62. 14. Kuwana M, Okazaki Y, Kaburaki J, Kawakami Y, Ikeda Y. Spleen is a primary site for activation of platelet-reactive T and B cells in patients with immune thrombocytopenic purpura. J Immunol 2002;168:3675–82. 15. Louwes H, Vellenga E, Houwerzijl EJ, de Wolf JT. Effects of prednisone and splenectomy in patients with idiopathic thrombo- 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. cytopenic purpura: only splenectomy induces a complete remission. Ann Hematol 2001;80:728–32. Rashba EJ, Rowe JM, Packman CH. Treatment of the neutropenia of Felty syndrome. Blood Rev 1996;10:177–84. Sandler SG. The spleen and splenectomy in chronic immune (idiopathic) thrombocytopenic purpura. Semin Hematol 2000;37 Suppl 1:10–2. Kimpel D, Dayton T, Kannan K, Wolf RE. Streptococcal cell wall induced arthritis: leukocyte activation in extra-articular lymphoid tissue. Inflammation 2003;27:59–70. Schwab JH. Bacterial cell-wall induced arthritis: models of chronic recurrent polyarthritis and reactivation of monoarticular arthritis. In: Henderson B, Edwards JCW, Pettipher ER, editors. Mechanisms and models in rheumatoid arthritis. San Diego: Academic Press; 1995. p. 431–46. Frei K, Leist TP, Meager A, Gallo P, Leppert D, Zinkernagel RM, et al. Production of B cell stimulatory factor-2 and interferon ␥ in the central nervous system during viral meningitis and encephalitis: evaluation in a murine model infection and in patients. J Exp Med 1988;168:449–53. Fuseler JW, Conner EM, Davis JM, Wolf RE, Grisham MB. Cytokine and nitric oxide production in the acute phase of bacterial cell wall-induced arthritis. Inflammation 1997;21:113–31. Grisham MB, Johnson GG, Lancaster JR Jr. Quantitation of nitrate and nitrite in extracellular fluids. Methods Enzymol 1996; 268:237–46. Nims RW, Cook JC, Krishna MC, Christodoulou D, Poore CM, Miles AM, et al. Colorimetric assays for nitric oxide and nitrogen oxide species formed from nitric oxide stock solutions and donor compounds. Methods Enzymol 1996;268:93–105. Panayi GS, Corrigall VM, Pitzalis C. Pathogenesis of rheumatoid arthritis: the role of T cells and other beasts. Rheum Dis Clin North Am 2001;27:317–34. Ramirez F, Mason D. Recirculatory and sessile CD4⫹ T lymphocytes differ on CD45RC expression. J Immunol 2000;165:1816–23. Hoffmann JC, Herklotz C, Zeidler H, Bayer B, Westermann J. Anti-CD2 (OX34) MoAb treatment of adjuvant arthritic rats: attenuation of established arthritis, selective depletion of CD4⫹ T cells, and CD2 down-modulation. Clin Exp Immunol 1997;110: 63–71. Powrie F, Mason D. OX-22high CD4⫹ T cells induce wasting disease with multiple organ pathology: prevention by the OX-22low subset. J Exp Med 1990;172:1701–8. Brennan FR, Mikecz K, Glant TT, Jobanputra P, Pinder S, Bavington C, et al. CD44 expression by leucocytes in rheumatoid arthritis and modulation by specific antibody: implications for lymphocyte adhesion to endothelial cells and synoviocytes in vitro. Scand J Immunol 1997;45:213–20. Haynes BF, Hale LP, Patton KL, Martin ME, McCallum RM. Measurement of an adhesion molecule as an indicator of inflammatory disease activity: up-regulation of the receptor for hyaluronate (CD44) in rheumatoid arthritis. Arthritis Rheum 1991;34: 1434–43. Kevil CG, Bullard DC. Cell adhesion molecules in the rheumatic diseases. In: Koopman WJ, editor. Arthritis and allied conditions: a textbook of rheumatology, 14th ed. Philadelphia: Lippincott Williams & Wilkins; 2001. p. 478–89. Brennan FM, Maini RN, Feldmann M. Role of pro-inflammatory cytokines in rheumatoid arthritis. Springer Semin Immunopathol 1998;20:133–47. Feldmann M, Brennan FM, Maini RN. Role of cytokines in rheumatoid arthritis. Annu Rev Immunol 1996;14:397–440. Marino C, McDonald E. Rheumatoid arthritis in a patient with sickle cell disease. J Rheumatol 1990;17:970–2. Uluhan A, Sager D, Jasin HE. Juvenile rheumatoid arthritis and common variable hypogammaglobulinemia. J Rheumatol 1998;25: 1205–10. IMPACT OF SPLENECTOMY ON SCW-INDUCED ARTHRITIS 35. Bradley JD, Pinals RS. Felty’s syndrome presenting without arthritis. Clin Exp Rheumatol 1983;1:257–9. 36. Dalldorf FG, Cromartie WJ, Anderle SK, Clark RL, Schwab JH. The relation of experimental arthritis to the distribution of streptococcal cell wall fragments. Am J Pathol 1980;100:383–402. 37. Eisenberg R, Fox A, Greenblatt JJ, Anderle SK, Cromartie WJ, Schwab JH. Measurement of bacterial cell wall in tissues by solid-phase radioimmunoassay: correlation of distribution and persistence with experimental arthritis in rats. Infect Immun 1982;38:127–35. 38. Janusz MJ, Esser RE, Schwab JH. In vivo degradation of bacterial cell wall by the muralytic enzyme mutanolysin. Infect Immun 1986;52:459–67. 39. Schwab JH. Phlogistic properties of peptidoglycan-polysaccharide polymers from cell walls of pathogenic and normal-flora bacteria which colonize humans. Infect Immun 1993;61:4535–9. 40. Schrijver IA, Melief MJ, Tak PP, Hazenberg MP, Laman JD. Antigen-presenting cells containing bacterial peptidoglycan in synovial tissues of rheumatoid arthritis patients coexpress costimulatory molecules and cytokines. Arthritis Rheum 2000;43:2160–8. 41. Thomas R. Antigen-presenting cells in rheumatoid arthritis. Springer Semin Immunopathol 1998;20:53–72. 42. DeJoy SQ, Ferguson KM, Sapp TM, Zabriskie JB, Oronsky AL, Kerwar SS. Streptococcal cell wall arthritis: passive transfer of disease with a T cell line and crossreactivity of streptococcal cell wall antigens with Mycobacterium tuberculosis. J Exp Med 1989; 170:369–82. 43. Gaston JS. Rheumatoid arthritis: cellular immunity in RA. In: Klippel JH, Dieppe PA, editors. Rheumatology. 2nd ed. London: Mosby; 1998. p. 10.1–6. 44. McInnes IB. Rheumatoid arthritis: from bench to bedside. Rheum Dis Clin North Am 2001;27:373–87. 45. Mason D, Powrie F. Memory CD4⫹ T cells in man form two 3567 46. 47. 48. 49. 50. 51. 52. 53. 54. distinct subpopulations, defined by their expression of isoforms of the leukocyte common antigen, CD45. Immunology 1990;70: 427–33. Ericsson PO, Linden O, Dohlsten M, Sjogren HO, Hedlund G. Functions of rat CD4⫹ T cell subsets defined by CD45RB. Cell Immunol 1991;132:391–9. Fowell D, McKnight AJ, Powrie F, Dyke R, Mason D. Subsets of CD4⫹ T cells and their roles in the induction and prevention of autoimmunity. Immunol Rev 1991;123:37–64. Laroux FS, Grisham MB. Immunological basis of inflammatory bowel disease: role of the microcirculation. Microcirculation 2001; 8:283–301. Powrie F, Correa-Oliveira R, Mauze S, Coffman RL. Regulatory interactions between CD45RBhigh and CD45RBlow CD4⫹ T cells are important for the balance between protective and pathogenic cell-mediated immunity. J Exp Med 1994;179:589–600. Read S, Mauze S, Asseman C, Bean A, Coffman R, Powrie F. CD38⫹ CD45RBlow CD4⫹ T cells: a population of T cells with immune regulatory activities in vitro. Eur J Immunol 1998;28: 3435–47. Grisham MB, Kawach S, Laroux FS, Gray L, Hoffman J, van der Heyde H. Role of nitric oxide in chronic gut inflammation. In: Salvemini D, Billiar TR, Vodovotz Y, editors. Nitric oxide and inflammation. Basel: Birkhauser Verlag; 2001. p. 161–71. Feldmann M, Maini RN. The role of cytokines in the pathogenesis of rheumatoid arthritis. Rheumatology (Oxford) 1999;38 Suppl 2:3–7. Fox DA. Cytokine blockade as a new strategy to treat rheumatoid arthritis: inhibition of tumor necrosis factor. Arch Intern Med 2000;160:437–44. Stichtenoth DO, Frolich JC. Nitric oxide and inflammatory joint diseases. Br J Rheumatol 1998;37:246–57.