To characterize the expression of CCL19 and CCL21 in rheumatoid arthritis (RA) synovial tissue (ST) and to examine their regulation and pathogenetic role in macrophages and RA ST fibroblasts.
To characterize the expression of CCL19 and CCL21 in rheumatoid arthritis (RA) synovial tissue (ST) and to examine their regulation and pathogenetic role in macrophages and RA ST fibroblasts.
Expression of CCL19 and CCL21 in RA and normal ST was demonstrated by immunohistochemistry analysis. CCL19 and CCL21 levels in synovial fluid (SF) from patients with osteoarthritis (OA), juvenile idiopathic arthritis, psoriatic arthritis (PsA), and RA were quantified by enzyme-linked immunosorbent assay (ELISA). Regulation of CCL19 and CCL21 expression in in vitro–differentiated RA peripheral blood macrophages as well as RA ST fibroblasts was determined by real-time reverse transcription–polymerase chain reaction. Proangiogenic factor production in CCL19- and CCL21-activated in vitro–differentiated peripheral blood macrophages and RA ST fibroblasts was examined by ELISA.
CCL19 and CCL21 were elevated in RA ST compared to tissue from normal controls. Levels of CCL19 and CCL21 were greatly increased in RA and PsA SF versus OA SF. In RA macrophages and fibroblasts, expression of CCL19 was increased by stimulation with lipopolysaccharide, tumor necrosis factor α (TNFα), and interleukin-1β (IL-1β). However, CCL21 expression was modulated only by IL-1β in RA fibroblasts, and by TNFα and RA SF in RA macrophages. CCL19 and CCL21 activation induced vascular endothelial growth factor and angiotensin I (Ang I) production in RA ST fibroblasts and secretion of IL-8 and Ang I from macrophages.
The findings of the present study identify, for the first time, regulators of CCL19 and CCL21 in RA fibroblasts and in vitro–differentiated RA peripheral blood macrophages and demonstrate a novel role of CCL19/CCL21 in angiogenesis in RA.
Rheumatoid arthritis (RA) is a chronic systemic disorder characterized by the development of new capillaries that are involved in the infiltration of inflammatory cells which results in synovial hyperplasia and progressive destruction of cartilage and bone. Synovial tissue (ST) lining consists of macrophages and fibroblasts that have profound effects in the destructive process in RA, via production of proinflammatory cytokines, chemokines, and proangiogenic factors (1, 2).
CCL19 and CCL21 and their corresponding receptor CCR7 are involved in organizing the thymic architecture and homing of various subpopulations of T cells and antigen-presenting dendritic cells to lymph nodes (3). CCL19 is expressed in lymph nodes and fibroblastic reticular cells in the T cell–rich area, whereas CCL21 is secreted from fibroblastic reticular cells and high endothelial venules (4, 5). Although CCL19 and CCL21 have similar affinity for CCR7, ligation of these chemokines mediates different signaling effects. Previous studies have shown that while CCL19 was chemotactic for RA ST fibroblasts, CCL21 was unable to attract these cells (6). Consistently, CCL19-activated RA ST fibroblasts have been shown to produce vascular endothelial growth factor (VEGF), while this effect was not noted with CCL21 stimulation (6).
Expression of CCL19 on lymphatic endothelium induces CCR7+ monocyte/macrophage migration (7), and CCL19 can also synergize with CCL2-mediated monocyte migration (8). CCL19 and CCL21 are also involved in the migration of dendritic cells to lymph nodes and in the ability of dendritic cells to prime T cells (9).
Findings of previous in vivo studies suggest that CCL19 and CCL21 play an important role in a number of autoimmune diseases, such as multiple sclerosis, atherosclerosis, and RA. Mice lacking CCL19/CCL21 or their receptor CCR7 were resistant to experimental autoimmune encephalomyelitis due to reduced numbers of Th17 and Th1 cells (10). Other studies have shown that antibodies to CCL19 and CCL21 preserve lesion size and foam cell content in apolipoprotein E–deficient mice, suggesting that these chemokines play a crucial role in cell trafficking to the site of inflammation (11, 12). Recent studies using CCR7-deficient mice with antigen-induced arthritis demonstrated lessened arthritis due to impaired development and organization of tertiary lymphoid tissue, indicating the importance of CCR7 and its corresponding ligands in lymphoid neogenesis (13). However, the role of CCL19 and CCL21 in RA angiogenesis is undefined.
This study was performed to characterize the expression pattern of CCL19 and CCL21 in RA ST and synovial fluid (SF), to determine the factors that modulate the expression of these chemokines in RA macrophages and fibroblasts, and to characterize their pathogenetic role in RA. Our results demonstrate that CCL19 and CCL21 are elevated in RA ST lining and endothelial cells. These chemokines are also significantly elevated in RA SF compared to osteoarthritis (OA) SF. Macrophages from RA SF exhibit higher levels of CCL19 and CCL21 compared to RA and normal peripheral blood cells. In in vitro–differentiated RA peripheral blood macrophages and RA fibroblasts, CCL19 and CCL21 are differentially modulated by proinflammatory factors and/or RA SF. These chemokines are capable of inducing proangiogenic factors in cells in the RA joint. Thus, this study documents a novel role of CCL19 and CCL21 in RA pathogenesis, suggesting that therapy directed against CCR7 ligation may reduce leukocyte migration in RA by inhibiting angiogenesis.
The studies were approved by the Northwestern University Institutional Review Board, and all donors provided written informed consent. Formalin-fixed paraffin-embedded RA and normal ST specimens were sectioned at the Pathology Core Facility of Northwestern University. ST samples were immunoperoxidase stained using Vector Elite ABC Kits, with diaminobenzidine (Vector) as a chromogen. Slides were deparaffinized in xylene for 20 minutes at room temperature, followed by rehydration by transfer through graded alcohol. Antigens were unmasked by first incubating slides in boiling citrate buffer for 15 minutes, followed by type II trypsin digestion for 30 minutes at 37°C. Endogenous peroxidase activity was blocked by incubation with 3% H2O2 for 5 minutes. Nonspecific binding of avidin and biotin was blocked using an avidin–biotin blocking kit (Vector). Tissue samples were incubated with antibodies to mouse anti-human CCL19 or goat anti-human CCL21 (1:100 and 1:67, respectively; R&D Systems) or an IgG control antibody (Beckman Coulter). Slides were counterstained with Harris' hematoxylin and treated with lithium carbonate for bluing. Each slide was evaluated by one of the authors (AAM), under blinded conditions. Tissue sections were scored for staining of the lining and endothelium (14–17). Cell staining was scored on a 0–5 scale in which 0 = no staining, 1 = few of the cells positively stained, 2 = some (fewer than half) of the cells stained, 3 = approximately half of the cells stained, 4 = more than half of the cells stained, and 5 = all cells stained. Scored data were pooled, and the mean ± SEM was calculated in each data group.
RA, OA, and normal ST was homogenized as described previously (18, 19) in 1 ml of complete Mini protease inhibitor cocktail homogenization buffer (Roche) on ice, followed by sonication for 30 seconds. Homogenates were centrifuged and filtered through a 0.45-μm pore size filter before quantification of CCL19 and CCL21 levels by enzyme-linked immunosorbent assay (ELISA). The final concentration of these chemokines in ST was normalized to the protein concentration in each tissue sample.
Normal and RA peripheral blood and RA SF mononuclear cells were isolated by Histopaque gradient centrifugation (Sigma) as previously described (20, 21). Monocyte/macrophages were isolated from normal and RA peripheral blood or RA SF, using a negative selection kit according to the instructions of the manufacturer (StemCell Technologies). Monocytes were subsequently differentiated to macrophages in 20% fetal bovine serum (FBS)/RPMI for 7 days. Heparinized SF was centrifuged at 800g for 10 minutes at room temperature to obtain cell-free SF. CCL19 and CCL21 levels in SF from patients with RA, OA, juvenile idiopathic arthritis (JIA), and psoriatic arthritis (PsA) were analyzed.
Human CCL19, CCL21, VEGF, interleukin-8 (IL-8), and angiotensin I (Ang I) were measured by ELISA, according to the instructions of the manufacturer (R&D Systems).
Fibroblasts from fresh RA ST were isolated by mincing and digestion in a solution of Dispase, collagenase, and DNase (21). Cells were used between passages 3 and 9 and cultured in Dulbecco's modified Eagle's medium supplemented with 10% FBS.
In vitro–differentiated RA peripheral blood macrophages and RA ST fibroblasts were treated with lipopolysaccharide (LPS) (10 ng/ml; Sigma), tumor necrosis factor α (TNFα) (10 ng/ml; R&D Systems), IL-1β (10 ng/ml; R&D Systems), IL-17 (50 ng/ml; R&D Systems), IL-6 (10 ng/ml; R&D Systems), IL-8 (10 ng/ml; R&D Systems), or RA SF (1:4 dilution). Cells were harvested after 6 hours, and levels of CCL19 and CCL21 messenger RNA were quantified by real-time reverse transcription–polymerase chain reaction (RT-PCR). RA ST fibroblasts, in vitro–differentiated normal peripheral blood macrophages, or human microvascular endothelial cells (HMVECs) were treated with 10 ng/ml CCL19 or CCL21 (both from R&D Systems), and cell-conditioned medium was harvested following treatment for 24 or 48 hours.
Total cellular RNA from the different cell types was extracted using TRIzol (Invitrogen). Subsequently, RT and real-time RT-PCR were performed to determine CCL19 and CCL21 expression levels as described previously (20–22). Relative gene expression was determined by the ΔΔCt method, and results were expressed as fold increases.
Data were analyzed using Student's 2-tailed t-tests for paired and unpaired samples. P values less than 0.05 were considered significant.
RA and normal ST were stained with antibodies against CCL19 and CCL21 in order to characterize their expression pattern in RA patients compared to healthy individuals. We found that expression of CCL19 (Figures 1A–C) and CCL21 (Figures 1D–F) was significantly increased in lining and endothelial cells in RA ST compared to normal ST. Consistently, ST lining and endothelial cells were positively stained for CCR7, and the immunostaining was much higher in RA ST compared to normal ST (P < 0.05) (data not shown). Since CCR7 and its ligands are coexpressed on the same cell types, this suggests that cells producing these chemokines may be responsive to their activation. Further, up-regulation of CCL19 and CCL21 on endothelial cells may play an important role in mediating transendothelial migration.
Next, CCL19 and CCL21 levels were quantified by ELISA and compared among RA, OA, and normal ST. Our results demonstrate that CCL19 expression was 3-fold higher in RA ST compared to OA and normal ST and was similar in OA and normal ST (Figure 2A). Although the highest concentration of CCL21 was detected in RA ST, CCL21 levels were 2-fold higher in OA ST compared to normal ST, in contrast to the findings for CCL19 (Figure 2B). Levels of CCL21 were generally higher than levels of CCL19 in RA and OA ST. However, normal ST exhibited similar levels of CCL19 and CCL21.
Levels of CCL19 and CCL21 in RA SF were compared with levels in SF obtained from patients with other rheumatic diseases. The data demonstrated that both CCL19 (Figure 2C) and CCL21 (Figure 2D) were up-regulated in RA SF (2.5-fold and 1.5-fold, respectively) and PsA SF (2-fold and 1.6-fold, respectively) compared to OA SF. Levels of CCL19 and CCL21 were comparable in OA and JIA SF. While levels of CCL21 were higher than levels of CCL19 in RA ST, CCL19 levels were significantly greater than CCL21 levels in RA SF (P < 0.0005). These findings suggest that expression of CCR7 ligands is greatly increased in RA ST and SF. Hence, these chemokines may play an important role in the pathogenesis of RA.
Based on the histologic findings, we investigated whether expression of CCL19 and CCL21 was elevated in RA ST fibroblasts compared to normal ST fibroblasts. Real-time RT-PCR demonstrated that both CCL19 expression (Figure 3A) and CCL21 expression (Figure 3B) were ∼4-fold greater in RA fibroblasts than in normal fibroblasts. We further showed that, while LPS, TNFα, and IL-1β up-regulated the expression of CCL19 in RA fibroblasts, stimulation with IL-17, IL-6, IL-8, and RA SF had no effect on this process (Figure 3C). In RA fibroblasts, expression of CCL21 was increased only with IL-1β activation, and other proinflammatory factors were ineffective (Figure 3D). These results suggest that, although both CCL19 and CCL21 are elevated in RA fibroblasts, their expression is differentially regulated.
Since expression levels of CCL19 and CCL21 were elevated in RA ST lining, which consists of ST fibroblasts and macrophages, we investigated whether expression of these chemokines was increased in RA peripheral blood monocytes and SF macrophages compared to normal peripheral blood monocytes and macrophages. We found that concentrations of CCL19 were significantly higher in RA SF macrophages compared to in vitro–differentiated RA peripheral blood macrophages (146-fold increase) and in vitro–differentiated normal peripheral blood macrophages (73-fold increase), as demonstrated by real-time RT-PCR (Figure 4A). RA peripheral blood monocytes and macrophages exhibited elevated levels of CCL21 compared to normal peripheral blood monocytes and macrophages (increases of 6-fold and 2-fold, respectively) (Figure 4B). Further, RA SF macrophages expressed higher levels of CCL21 compared to RA and normal peripheral blood macrophages (2.5-fold and 6-fold, respectively). These results indicate that RA SF macrophages may be an important source of CCL19 and CCL21 production.
To determine which factors modulate expression of CCL19 and CCL21 in in vitro–differentiated RA peripheral blood macrophages, cells were either left untreated or treated with LPS, TNFα, IL-1β, IL-17, IL-6, IL-8, or RA SF. The results demonstrated that CCL19 expression was highly up-regulated by activation of RA macrophages with LPS (145-fold), TNFα (43-fold), IL-1β (82-fold), and IL-8 (12-fold), compared to untreated cells (Figure 4C). However, only TNFα and RA SF stimulated up-regulation of CCL21 in macrophages (6.5-fold and 6-fold, respectively) (Figure 4D). These findings suggest that expression levels of CCL19 and CCL21 are differentially modulated in RA macrophages and that CCL19 is more responsive to stimuli.
To examine the mechanism by which CCL19 and CCL21 mediate RA pathogenesis, CCL19- and CCL21-activated macrophages, RA ST fibroblasts, and HMVECs were screened for proinflammatory factors. CCL19 and CCL21 induced production of VEGF by RA ST fibroblasts (Figure 5A) and production of IL-8 by macrophages (Figure 5B), as well as production of Ang I by RA ST fibroblasts (Figure 5C) and macrophages (Figure 5D). However, CCL19 and CCL21 were unable to induce production of IL-6, CCL2/monocyte chemotactic protein 1 (MCP-1), CCL3/macrophage inflammatory protein 1α, or Ang II by RA ST fibroblasts or macrophages (data not shown). Furthermore, production of TNFα and IL-1β by macrophages was not mediated by CCL19 and CCL21 stimulation. CCL19 and CCL21 stimulation in HMVECs did not result in production of any proangiogenic factor. These results indicate that the pathogenetic role of CCL19 and CCL21 in RA is mediated through production of proangiogenic factors.
In this study, we have shown that the ST lining and endothelium in RA patients express higher levels of CCL19 and CCL21 compared to tissue from normal controls. Consistently, levels of CCL19 and CCL21 in RA ST and SF are elevated compared to those in OA ST and SF. Our data demonstrate that macrophages are an important source of CCL19 and CCL21 production in RA SF, and levels of these chemokines are greatly up-regulated in RA SF compared to RA and normal peripheral blood macrophages. CCL19 and CCL21 expression levels were also elevated on RA fibroblasts. We further showed that, while CCL19 expression was modulated similarly in RA macrophages and fibroblasts, expression of CCL21 was differentially regulated in these cell types. Finally, we demonstrated that CCL19 and CCL21 can activate key proangiogenic factors in macrophages and RA ST fibroblasts. Taken together, our results suggest that ligation of CCR7 by CCL19 and CCL21 can activate angiogenesis in RA.
Previous studies have demonstrated that CCR7 and CCL21 are localized in lymphocytic infiltrates and dendritic cells in RA ST and play an essential role in maturation and homing of dendritic cells to lymphocytic aggregates (23). Others have shown that in mice, CCL19 is produced by fibroblastic cells in lymph nodes, whereas CCL21 is mainly secreted from high endothelial venules (4, 5). In this report, we present the novel finding that CCL19 and CCL21 are coexpressed on the endothelium and RA ST lining, where macrophages and RA fibroblasts are in close proximity and interact with one another. Since CCL19 and CCL21 and their receptor, CCR7 (data not shown), are present on all 3 cell types, this may suggest that these cells can be directly activated through ligation. Activation of macrophages and RA fibroblasts by CCL19 and CCL21 may be important in production of proinflammatory factors in the synovium. Further, expression of these chemokines on the vessel walls may play a role in facilitating transendothelial migration of CCR7+ cells.
Angiogenesis is an early and critical event in the pathogenesis of RA (2) and PsA (24), which promotes ingress of leukocytes and pannus formation. Angiogenesis does not play a vital role in the pathogenesis of OA, and CCL19 and CCL21 levels are significantly lower in OA SF compared to SF from RA and PsA patients, suggesting that these factors may contribute to RA and PsA neovascularization.
Expression of CCL19 and CCL21 in RA SF macrophages was substantially higher than that in peripheral blood macrophages from either normal subjects or RA patients. While CCL19 expression levels were similar in RA and normal peripheral blood monocytes, CCL21 concentrations were significantly higher in RA peripheral blood monocytes compared to normal cells. Interestingly, the expression and function of CCL19 and CCL21 in RA macrophages has not been described previously. Earlier studies have shown that monocytes activated with prostaglandin E2 express higher levels of CCR7 compared to unstimulated cells, which allows the cells to migrate toward CCL19 and CCL21 (25). A recent study also demonstrated that CCL19 is a chemoattractant for both dendritic cells and monocytes (26). Conversely, others have shown that in the presence of CCL19 and CCL21, monocytes can migrate in response to lower concentrations of CCL2/MCP-1; however, these chemokines were unable to induce monocyte migration alone (8).
Since fibroblasts and macrophages in RA ST express CCL19 and CCL21, we performed experiments to determine the factors that modulate CCL19 and CCL21 expression in these cells. We found that in RA ST fibroblasts, CCL19 expression is induced by LPS, TNFα, and IL-1β stimulation, whereas CCL21 expression is modulated by IL-1β activation. Interestingly, in vitro–differentiated RA peripheral blood macrophages responded similarly to RA fibroblasts in that LPS, TNFα, and IL-1β were able to regulate the expression of CCL19, but RA SF was not. In contrast, CCL21 expression levels were differentially regulated in RA macrophages and RA fibroblasts, with macrophages responding only to TNFα and RA SF stimulation and fibroblasts responding only to IL-1β. Since TNFα- and RA SF–activated RA macrophages express comparable levels of CCL21, TNFα may be one of the major factors present in RA SF that mediates this effect. These results suggest that the expression of the CCR7 ligands is differentially modulated in the different cell types present in the RA ST lining layer.
Others have shown that lymphotoxins α and β induce production of CCL19 and CCL21 in stromal cells within the T cell zones, through activation of the NF-κB pathway (27). Furthermore, expression of CCL19 and CCL21 in mouse iris was induced by LPS (28). Results from previous studies suggest that activation of Toll-like receptors through NF-κB can directly or indirectly modulate the expression and function of CCL19 and CCL21 at sites of inflammation (29).
We investigated the pathogenetic role of CCL19 and CCL21 in RA fibroblasts, macrophages, and endothelial cells, and screened activated cells for a variety of inflammatory factors. The results demonstrate that CCL19 and CCL21 were able to induce production of VEGF and Ang I from RA ST fibroblasts and secretion of IL-8 and Ang I from macrophages. Production of potent proangiogenic factors from lining cells activated by CCL19 and CCL21 is consistent with the role of CCR7 in tumor cell invasion and metastasis. Cyclooxygenase 2 and hypoxia-inducible factor 1 have been shown to induce CCR7 expression in breast (30, 31) and lung (32) tumor cells, which facilitates their metastasis to lymph nodes, where CCL19 and CCL21 are abundantly expressed. The data presented herein highlight the importance of CCR7 and its ligands in the pathogenesis of RA.
In conclusion, CCL19 and CCL21 are elevated on RA ST lining and endothelium. Fibroblasts from RA ST and macrophages from RA SF and ST express higher levels of CCL19 and CCL21 compared to control cells. Nevertheless, the expression of CCL19 and CCL21 is differentially regulated in RA fibroblasts and macrophages. Potent proangiogenic factors are secreted from CCL19- and CCL21-activated RA ST fibroblasts and macrophages, suggesting that these chemokines play an important role in angiogenesis in RA.
All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Shahrara had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study conception and design. Pickens, Shahrara.
Acquisition of data. Pickens, Chamberlain, Volin, Mandelin.
Analysis and interpretation of data. Pickens, Chamberlain, Volin, Pope, Shahrara.
We are grateful to Sharon Tymkiw and Elizabeth Randall for providing phlebotomy services as well as to Brian Zanotti and Nikolay Zlatarov for their technical expertise.