Deletion of calponin 2 in macrophages attenuates the severity of inflammatory arthritis in mice

Abstract

Calponin is an actin cytoskeleton-associated protein that regulates motility-based cellular functions. Three isoforms of calponin are present in vertebrates, among which calponin 2 encoded by the Cnn2 gene is expressed in multiple types of cells, including blood cells from the myeloid lineage. Our previous studies demonstrated that macrophages from Cnn2 knockout (KO) mice exhibit increased migration and phagocytosis. Intrigued by an observation that monocytes and macrophages from patients with rheumatoid arthritis had increased calponin 2, we investigated anti-glucose-6-phosphate isomerase serum-induced arthritis in Cnn2-KO mice for the effect of calponin 2 deletion on the pathogenesis and pathology of inflammatory arthritis. The results showed that the development of arthritis was attenuated in systemic Cnn2-KO mice with significantly reduced inflammation and bone erosion than that in age- and stain background-matched C57BL/6 wild-type mice. In vitro differentiation of calponin 2-null mouse bone marrow cells produced fewer osteoclasts with decreased bone resorption. The attenuation of inflammatory arthritis was confirmed in conditional myeloid cell-specific Cnn2-KO mice. The increased phagocytotic activity of calponin 2-null macrophages may facilitate the clearance of autoimmune complexes and the resolution of inflammation, whereas the decreased substrate adhesion may reduce osteoclastogenesis and bone resorption. The data suggest that calponin 2 regulation of cytoskeleton function plays a novel role in the pathogenesis of inflammatory arthritis, implicating a potentially therapeutic target.

macrophage-mediated inflammation is a critical part of the pathogenesis of inflammatory diseases, such as systemic lupus erythematosus and rheumatoid arthritis (28, 29). Rheumatoid arthritis is a multifactorial autoimmune disease characterized by chronic joint inflammation, which leads to progressive joint destruction and ultimately to significant disability and severely reduced quality of life. Macrophages are mobile cells, and their actin cytoskeleton plays a vital role in motility-based functions, such as migration, adhesion, and phagocytosis. The functions of macrophage cytoskeleton are, therefore, essential for the activities of macrophages in health and in the development and resolution of inflammation (29).

Calponin is an actin filament-associated protein of 34–37 kDa (292–330 amino acids) in size. It was originally found in smooth muscle as a regulatory protein potentially analogous to troponin in the striated muscle (59). Subsequent studies found that calponin is also present in abundance in many types of nonmuscle cells (27, 70). Biochemical studies demonstrated that calponin binds multiple cytoskeleton-related proteins, including actin (59) [also cross-linking actin filaments (30)], tropomyosin (6), myosin (34), the regulatory light chain of myosin (58), Ca²⁺-calmodulin (59), caldesmon (11), desmin (38), tubulin (9), and gelsolin (8). Together with a high-affinity binding to F-actin, calponin inhibits the actin-activated myosin MgATPase (1, 16, 68, 69), myosin motor activity (12, 53), and the force development and shortening velocity of smooth muscle cells (65) to play a role in the regulation of smooth muscle contractility (2) and nonmuscle cell motility (70).

Encoded by three homologous genes (CNN1, CNN2, and CNN3), three isoforms of calponin are present in vertebrates [previously named h1, h2 (45, 56, 60), and h3 or acidic (3, 64) calponins]. Amino acid sequences of the three calponin isoforms are largely conserved (27, 70) except for the COOH-terminal variable region (27, 70) that produces differentiated features, such as their isoelectric points (9.4 for calponin 1, 7.5 for calponin 2, and 5.2 for calponin 3). Calponin 1 is expressed specifically in differentiated smooth muscle cells and functions in modifying smooth muscle contractility (18, 27, 44, 65, 70). Calponin 3 was first reported in smooth muscle (3) and brain (64). Later studies found calponin 3 in B lymphocytes (33), trophoblasts (52), and myoblasts (51) with a function in cell fusion. Calponin 2 is expressed in a broad range of cell types, including smooth muscle cells (18), endothelial cells (62), epithelial cells, fibroblasts (17, 19), B lymphocytes (33), and myeloid leukocytes, i.e., granulocytes, monocytes, and macrophages (21) with functions in regulating cell proliferation, migration, and phagocytosis.

Consistent with earlier findings in epidermal keratinocytes (17), lung alveolar cells (19), and fibroblasts, we have demonstrated that calponin 2 is an inhibitory regulator of macrophage migration and phagocytosis (21). The phagocytotic function of macrophages is essential for the clearance of dead and apoptotic cells and other inflammatory materials, presenting a therapeutic mechanism for promoting the resolution of inflammation (13, 29, 35, 54). Defective macrophage phagocytosis has been related to the pathogenesis of systemic lupus erythematosus (32) and rheumatoid arthritis (46, 63).

In the present study, we investigated the effect of calponin 2 gene (Cnn2) knockout (KO) on the pathogenesis and pathology of inflammatory arthritis in the anti-glucose-6-phosphate isomerase (GPI) serum-transfer mouse model (41). The results showed that the development of arthritis was attenuated in Cnn2-KO mice with significantly reduced inflammation and bone erosion compared with that in wild-type (WT) mice. In vitro differentiation of calponin 2-null bone marrow cells produced fewer osteoclasts with decreased bone resorption. The data suggest that calponin 2 regulation of cytoskeleton function plays a novel role in the development of inflammatory arthritis.

MATERIALS AND METHODS

Human specimens.

Patients with rheumatoid arthritis were recruited at the Northwestern Medical Faculty Foundation, the Rehabilitation Institute of Chicago, and Northwestern Memorial Hospital. Rheumatoid arthritis was diagnosed according to the 2010 ACR-EULAR classification criteria for rheumatoid arthritis. Peripheral blood and the inflamed joint fluid were obtained, and monocytes and macrophages were obtained using a CD14+ isolation kit (negative selection without CD16 deletion; StemCell Technologies, Vancouver, British Columbia, Canada) as previously described (22, 25). Healthy control blood was obtained from Buffy coat purchased from The American Red Cross (Minneapolis, MN) or healthy donors. Control human macrophages were obtained from normal peripheral blood monocytes, isolated by elutriation, followed by in vitro differentiation for 7 days, as previously described (21, 22, 25).

Synovial tissues were obtained from patients with rheumatoid arthritis at the time of arthroplasty. Synovial tissues from arthritis-free controls were obtained at the time of autopsy or surgery for injury from the National Disease Research Interchange. All human specimens were obtained with written, informed consent and studied under protocols approved by the Institutional Review Board of Northwestern University Feinberg School of Medicine.

Immunohistochemistry.

The expression of calponin 2 in patient synovial tissues was examined using immunohistochemistry. Four-μm-thin paraffin sections were stained with an anti-calponin 2 monoclonal antibody (mAb) 1D2 using a standard immunohistochemical method with horseradish peroxidase-labeled anti-mouse IgG second antibody (Sigma, St. Louis, MO) and H₂O₂-diaminobenzidine substrate reaction (17). 1D2 hybridoma cultural supernatant was used to avoid background from anti-cytokeratin autoantibodies in animal body fluid. SP2/0 myeloma culture supernatant was used as control. The slides were read in a blinded manner as previously described (25).

SDS-PAGE and Western blotting.

As described previously (17), tissue and cell extracts or control protein samples were homogenized in SDS gel sample buffer and analyzed with PAGE using 12% Laemmli gels. The gels were fixed and stained with Coomassie blue R250 to confirm sample integrity and protein inputs. Protein bands in unfixed duplicate gels were transferred to nitrocellulose membrane for Western blotting using anti-calponin 2 polyclonal anti-calponin 2 antibody RAH2, followed by alkaline phosphatase-labeled anti-rabbit IgG second antibody (Sigma) and 5-bromo-4-chloro-3-indolyl phosphate and nitroblue tetrazolium substrate reactions.

Cnn2-KO mouse lines.

Mice were bred and maintained in Wayne State University animal care facility. All experimental procedures were approved by the Institutional Animal Care and Use Committees of Wayne State University and Northwestern University.

The generation and initial characterization of Cnn2-floxed (Cnn2^f/f) and systemic Cnn2-KO mice have been described previously (21). The Cnn2-targeted alleles have been transferred to C57BL/6 background by nine or more generations of backcross breeding. Age-matched WT littermates were used as control for the systemic Cnn2-KO mice in phenotype studies.

Cnn2^f/+ mice were crossed with lysM^cre mice, a transgenic mouse line bearing a Cre recombinase gene driven by the lysM promoter (purchased from Jackson Laboratories, Bar Harbor, ME) (24) to generate myeloid cell-specific Cnn2-deletion mice (Cnn2^f/f,lysM^cre). Genotypes of the offspring were identified on tail biopsies using PCR as described previously (21). The conditional Cnn2 deletion was confirmed in elicited peritoneal macrophages using Western blotting as described below. Age matched Cnn2^+/+,lysM^cre, or Cnn2^f/+,lysM^cre littermates were used as control in phenotype studies.

K/BxN anti-GPI serum transfer mouse arthritis model.

Anti-GPI serum transfer induces immune complex-based arthritis in mice, a disease model in which macrophages play important roles in the pathogenesis and resolution of inflammation (20, 23). K/BxN mice were generated as described, and anti-GPI sera were collected at 8–9 wk of age (41). The K/BxN anti-GPI sera were injected intraperitoneally in adult mice of both sexes at 100 μl/20 gm body wt to induce arthritis (20, 22). The development of inflammatory arthritis was evaluated after induction by measuring the swelling (width) of two hind ankles using a digital caliper and grading the clinical index of all four paws on a scale of 0–3/each paw (maximum score = 12), according to a previously described method (22, 41, 42).

Histopathological studies were performed for ankle tissues harvested on a predetermined postinduction day during the course of arthritis. Hematoxylin and eosin (H and E)-stained decalcified paraffin sections were examined under a light microscope and scored by blinded observers for inflammation (0–5), bone erosion (0–5), pannus formation (0–5), and median synovial lining thickness as previously described (48–50).

The level of calponin 2 in ankle homogenates was determined by using Western blotting. The intensity of calponin 2 band was quantified by densitometry analysis.

Immunophenotyping and peripheral blood profile.

Spleens were harvested from Cnn2 KO and WT mice to isolate cells. Myeloid cell types were determined by immunophenotyping, employing multicolor fluorochrome-conjugated antibodies against surface markers CD64, CD11b, F4/80, Ly6G, and Ly6C and markers for dead and alive cells. Mouse peripheral blood leukocytes were analyzed with flow cytometry using multicolor fluorochrome-conjugated antibodies to CD45, CD11b, CD115, Ly6G, Ly6C, and CD62L. The osteoclast precursors (OCPs) in bone marrow cells were analyzed by flow cytometry with anti-CD45, anti-c-Kit, anti-CD11b, and anti-c-Fms antibodies. Flow immunophenotyping data were acquired employing a BD LSR II flow cytometer (BD FACSDIVA software; BD Biosciences, San Jose, CA) and analyzed using Flowjo software (TreeStar, Ashland, OR). Complete blood counts and differentials were performed employing a Hemavet 950 analyzer (Drew Scientific, Waterbury, CT) (24).

Osteoclastogenesis and bone resorption assays.

Bone marrow cells were flushed out from the femur and tibia of Cnn2-KO mice and WT littermates and seeded in 96-well plates at 5 × 10⁴ cells/well in α-MEM containing 10% FBS, 2% macrophage colony-stimulating factor (M-CSF)-producing cell line-conditioned medium, and 10 ng/ml RANKL (R&D Systems, Minneapolis, MN). The cells were cultured to generate osteoclasts as described previously (72).

To examine osteoclast formation, the cells were cultured for 5–7 days and stained for tartrate-resistant acid phosphatase (TRAP) activity to identify osteoclasts. The number and spreading area of TRAP+ osteoclasts were measured. For osteoclast function, the cells were seeded on bovine cortical bone slices and cultured for 10–12 days. The bone slices were then stained with 1% toluidine blue after removal of osteoclasts. The number of resorption pits per bone slice was assessed as described previously (72, 73).

Histology and histomorphometric analysis of bone cross sections were carried out on postmortem mouse knees fixed in buffered 3.7% formalin, decalcified in 10% EDTA, and embedded in paraffin. Four-μm sections were stained with H and E and for TRAP activity. Analyses of the number of TRAP+ osteoclasts per millimeter of bone surface and bone volume/tissue volume were carried out using Osteometrics (Atlanta, GA) image analysis software.

Cell adhesion assay.

Thioglycollate-elicited peritoneal macrophages from Cnn2-KO mice and WT littermates were examined for adhesion to cultural dishes. The adherent cells were fixed with glutaradhyde 10, 15, 20, 30, and 40 min after seeding on plastic dishes and gentle washing to remove nonadherent cells. The adhered cells stained with Crystal Violet to measure the efficiency of substrate adhesion.

Cell migration assay.

To investigate the role of calponin 2 in regulating macrophage migration, peritoneal macrophages were elicited from calponin 2 KO and WT littermate mice by injection of 2 ml of 3% thioglycollate broth 72 h before lavage. The elicited peritoneal cells were collected in RPMI 1640 medium containing 10% FBS, 2 mM l-glutamine, 100 IU/ml penicillin, and 50 IU/ml streptomycin. Cells (1 × 10⁶) in 200 μl of media were seeded on the 14-mm-diameter glass of a 35-mm glass-bottom culture dish (Fisher Scientific, Rockford, IL). One hour after the seeding, 2 ml of culture media was added to the dish. The cells were then incubated in 5% CO₂ at 37°C for 24 h to form a confluent monolayer. The macrophage monolayer was wounded by being scratched with a thin pipette tip (21). Maintained in an environmental stage incubator, phase-contrast microscopic images were taken every 10 min for 90 min. Migration distance of single cells was measured from the photographs.

Data analysis.

Densitometry quantification of SDS gels and Western blots was carried out using ImageJ software (NIH, Bethesda, MD). Statistical significance of quantitative data was examined using Student's t-test. Correlations were determined by Pearson's linear-regression test. Significance levels were set at P < 0.05.

RESULTS

Increased level of calponin 2 in macrophages of patients with rheumatoid arthritis.

mAb 1D2 immunohistochemistry staining of paraffin sections of joint tissues from patients with rheumatoid arthritis detected significantly higher levels of calponin 2 than that in samples from arthritis-free control subjects (Fig. 1A). The calponin 2 stain was mainly in infiltrated inflammatory cells in the lining and sublining of arthritis joints, which are morphologically macrophage-like.

Fig. 1.

Increased calponin 2 in monocytes and macrophages of patients with rheumatoid arthritis (RA). A: immunohistochemical staining of paraffin sections using monoclonal antibody (mAb) 1D2 showed a significantly increased number of calponin 2-positive cells in the joint lining and sublining regions of synovial tissue from a representative patient with RA than that in a control joint. Nonspecific mouse IgG was used in the negative controls. The majority of cells expressing high levels of calponin 2 had the morphological appearance of macrophages. B and C: the expression of calponin 2 in monocytes and macrophages was determined by Western blotting and densitometry quantification normalized to the level of actin. Higher levels of calponin 2 were detected in peripheral blood monocytes from RA than that from control subjects (B) and in macrophages from the synovial fluid of patients with RA than that in macrophages in vitro differentiated from monocytes of control subject (C). Quantification of the Western blots by normalization against total protein produced similar results (data not shown). *P < 0.05 in Student's t-test. Data are representative of 3 samples in each group.

To assess the nature of the cells with increased calponin 2, we examined purified peripheral blood CD14⁺ monocytes and synovial fluid macrophages from patients with rheumatoid arthritis compared with that in normal human peripheral blood monocytes and in vitro differentiated macrophages. The Western blot data in Fig. 1B demonstrated higher calponin 2 in peripheral blood monocytes from patients with rheumatoid arthritis compared with that in monocytes from normal subjects. The data in Fig. 1C further demonstrate that synovial fluid macrophages of patients with rheumatoid arthritis had significantly higher calponin 2 than that in macrophages in vitro differentiated from peripheral blood monocytes of normal subjects. These findings implicate that increased levels of calponin 2 in macrophages may contribute to the pathogenesis of inflammatory arthritis.

Increased levels of calponin 2 in mouse joint tissue during the development of anti-GPI serum-induced arthritis.

We examined the level of calponin 2 in tissue homogenates of ankle joints of WT mice during the course of anti-GPI serum transfer-induced arthritis (22). Western blots and densitometry quantification showed that calponin 2 increased significantly during the development of arthritis 5–14 days after induction (Fig. 2, A and B). The level of calponin 2 in the mouse ankle homogenates was significantly correlated with the degree of joint inflammation (Fig. 2C). This positive correlation supports the observation that the high levels of calponin 2 in synovial macrophages of patients with rheumatoid arthritis may contribute to the development of pathological inflammation (Fig. 1).

Fig. 2.

Increase of calponin 2 in arthritic mouse ankles. A and B: Western blot using anti-calponin 2 antibody RAH2 (A) and densitometry quantification (B) demonstrated a rise of calponin 2 in total protein extracts from the ankle tissue of wild-type (WT) mice 3–4 mo of age during the development of anti-glucose-6-phosphate isomerase (GPI) serum transfer-induced arthritis. The relative level of calponin 2 exhibited a significant peak from days 5 to 14 after the induction of arthritis (**P < 0.005 compared with day 0). C: levels of calponin 2 during the course of anti-GPI serum transfer-induced arthritis showed a significantly positive correlation with the clinical signs (represented by the degree of ankle swelling).

Significantly attenuated severity of anti-GPI serum-induced arthritis in systemic Cnn2-KO mice.

After induction by intraperitoneal injection of anti-GPI serum (41, 43), the systemic Cnn2-KO mice developed significantly less severe arthritis compared with that in WT littermates. Compared for the difference in ankle swelling (presented as the delta-ankle circumference) and clinical arthritis scores determined at days 0, 2, 4, 7, and 9 after the injection of anti-GPI sera, the data clearly showed that Cnn2-KO attenuated the severity of inflammatory arthritis (Fig. 3A).

Fig. 3.

Systemic Cnn2-knockout (KO) mice exhibited significantly attenuated development of arthritis after anti-GPI serum transfer. A: development of clinical arthritis in Cnn2-KO (Cnn2^−/−) and WT mice after anti-GPI serum injection was determined by the swelling of both hind ankles (Δ-ankle thickness) and inflammation score (maximum 12). The results demonstrated significantly attenuated development and earlier remission of arthritis in the Cnn2^−/− group compared with that of WT mice. B and C: histopathology studies (B) and quantitative analysis (C) showed that the Cnn2^−/− group had attenuated inflammation and bone erosion (indicated by the arrows) than that in WT group in ankles collected at day 9 after anti-GPI serum injection; n = 9 mice in Cnn2^−/− and n = 8 mice in WT groups. B, bone; C, cartilage. All mice used were 3–4 mo of age. Data are means ± SE. *P < 0.05 and **P < 0.01 between the two groups in Student's t-test.

Mouse ankles collected on day 9 after the induction of arthritis were examined for histopathology. H and E-stained joint sections (Fig. 3B) and quantitative analysis demonstrated significantly less inflammation and bone erosion in Cnn2-KO mice than that in WT mice (Fig. 3C). The results indicate that calponin 2 plays a role in the pathogenesis of inflammatory arthritis, and, therefore, Cnn2-KO attenuates the severity of disease.

Myeloid cell-specific deletion of calponin 2 effectively attenuated the severity of anti-GPI serum-induced arthritis.

We further performed experiments using mice with a conditional calponin 2 depletion specifically in myeloid cells to determine whether calponin 2-dependent functions of macrophages contribute to the pathogenesis of inflammatory arthritis. Cnn2^f/f,lysM^cre mice were generated by crossing Cnn2-floxed mice (21) with lysM^cre mice. lysM^cre mice carry a transgene encoding Cre recombinase under the myeloid cell-specific lysM promoter that is expressed in granulocytes, monocytes, macrophages, and osteoclasts but not T and B lymphocytes, dendritic cells (7, 24), or fibroblasts. In addition to PCR genotyping (Fig. 4A), the deletion of calponin 2 in macrophages was verified using Western blots of protein extracts from elicited peritoneal macrophages. The result in Fig. 4B shows that calponin 2 was undetectable in Cnn2^f/f,lysM^cre mouse macrophages in contrast to the significant expression in macrophages elicited from Cnn2^f/f mice and unaffected in the skin (a representative control tissue) of Cnn2^f/f,lysM^cre mice.

Fig. 4.

Attenuated inflammatory arthritis in myeloid cell-specific Cnn2-KO mice. A and B: conditional deletion of Cnn2 in myeloid cells of Cnn2^f/f,lysM^cre mice was demonstrated by PCR genotyping (A) and Western blot analysis of calponin 2 expression in elicited peritoneal macrophages (B). In contrast to the significant levels of calponin 2 in macrophages elicited from Cnn2^f/f mice and in the skin of Cnn2^f/f,lysM^cre mice, macrophages of Cnn2^f/f,lysM^cre mice showed no detectable calponin 2. C: the development of clinical arthritis in Cnn2^f/f,lysM^cre and control mice after anti-GPI serum injection was determined by the swelling of both hind ankles (Δ-ankle thickness) and the clinical inflammation score (maximum 12). The results demonstrate a significantly attenuated development of inflammatory arthritis in Cnn2^f/f,lysM^cre mice in comparison with that in control mice. n = 7 mice at 3–4 mo of ages in each group. *P < 0.05; **P < 0.005 vs. control in Student's t-test.

Confirming the phenotype of systemic Cnn2-KO mice, myeloid cell-specific Cnn2-KO very effectively attenuated the severity of inflammatory arthritis, as shown by the significantly lower clinical arthritis scores and less ankle swelling (presented as the delta-ankle circumference, Fig. 4C) than that of control mice. Little calponin 2 is normally expressed in mature granulocytes (data not shown). Therefore, the attenuated development of arthritis in myeloid cell-specific Cnn2-KO mice strongly supports that calponin 2 contributes to the pathogenesis of inflammatory arthritis via the function of macrophages.

Cnn2-KO did not alter the baseline immunophenotype of mouse spleen and peripheral blood myeloid cells.

Flow cytometry studies profiled splenocytes of Cnn2-KO and WT mouse, such as living granulocytes (CD64⁻ Ly6G⁺CD11b⁺), tissue residential macrophages (CD64⁺F4/80^hiCD11b^loLy6C⁻), and monocyte-derived macrophages (CD64⁺F4/80^loCD11b^hiLy6C⁺). The peripheral neutrophils, lymphocytes, and monocytes were analyzed by blood count, and CD11b⁺CD115⁺ monocytes were further analyzed by flow cytometry for the subsets of classic monocytes (Ly6C⁺CD62L⁺) and nonclassic monocytes (Ly6C⁻CD62L⁻). The results in Fig. 5 show that the deletion of calponin 2 did not cause detectable change in the baseline myeloid cells, except for slightly increased F4/80^loCD11b⁺Ly6C⁺ macrophages (Fig. 5C). Although no change was seen in the populations of monocytes (Fig. 5, D and E), this phenotype may be an implication for a role of calponin in monocyte-macrophage differentiation, and further investigation is required.

Fig. 5.

No significant baseline phenotypic alterations of spleen and blood monocytes, macrophages, and granulocytes in Cnn2-KO mice. A: representative flow cytometry analysis of mouse spleen myeloid cells and the gating strategy. After exclusion of doublets and debris and gating on living cells, granulocytes were defined as CD64⁻ Ly6G⁺CD11b⁺. Tissue residential macrophages were identified as CD64⁺F4/80^hiCD11b^loLy6C⁻, and the monocyte-derived macrophages are identified as CD64⁺F4/80^loCD11b^hiLy6C⁺. SSC-A, side scatter area; FSC-H, forward scatter height. B and C: granulocytes (B) and macrophages (C) were analyzed in WT and Cnn2-KO mice (n = 4). D: profiles of neutrophils, lymphocytes, and monocytes in peripheral blood were analyzed by blood count (n = 9/group). E: flow cytometry analysis of peripheral blood monocytes, which were defined as CD11b⁺CD115⁺ly6G⁻ (n = 4). The classic monocytes were identified as Ly6C⁺CD62L⁺ and nonclassic monocytes as Ly6C⁻CD62L⁻. All mice used were 3–4 mo of age. The data are presented as means ± SE. *P < 0.05 by unpaired Student's t-test.

Calponin 2 deletion decreased the formation and activity of osteoclasts.

Osteoclasts are derived from myeloid cell lineage and are effector cells for bone erosion in arthritis joints (4). To examine whether calponin 2 plays a role in osteoclast formation, we cultured mouse bone marrow cells that contain osteoclast precursors (OCPs) in the presence of M-CSF and RANKL, cytokines essential for osteoclast formation in an osteoclast formation assay. The results show that Cnn2-KO mouse bone marrow cells formed significantly fewer osteoclasts than that from cells of WT littermates (Fig. 6A). The results in Fig. 6B further show that, on bone slices, osteoclasts derived from Cnn2-KO mouse bone marrow cells produced fewer resorption pits than that produced by WT osteoclasts.

Fig. 6.

Cnn2-KO reduced osteoclastogenesis and bone resorption. A: representative microscopic images of tartrate-resistant acid phosphatase (TRAP)-stained osteoclasts (OCs) in plastic dish employing bone marrow cells from WT and Cnn2-KO mice and quantitative analysis demonstrated significantly reduced OC formation from calponin 2-null precursor cells compared with controls (n = 3 mice in each group). B: microscopic images and quantification analysis of the generation of resorption pits (indicated by arrows) on bone slices demonstrated significant less OC formation and fewer and smaller resorption pits on bone slices with Cnn2-KO than WT cells (n = 3 mice in each group). C: bone marrow cells were subjected to flow cytometric analysis after they were stained with anti-CD45, anti-c-Kit, anti-CD11b, and anti-c-Fms antibodies. The percentage of osteoclast precursors (OCPs, CD45⁺/c-Kit⁺/c-Fms⁺/CD11b^+/−) was determined. Values are means ± SD of 3 pairs of mice at 3–4 mo of age. D: bones from 3-mo-old Cnn2-KO mice and WT littermates were analyzed. Photos show hematoxylin and eosin (H&E)- and TRAP-stained tibial sections (right) and histomorphometric analyses (left) of bone volume/tissue volume (BV/TV) (%) and number of TRAP⁺ osteoclasts. *P < 0.05 vs. WT mice in Student's t-test.

We previously reported that both CD45⁺c-Kit⁺c-Fms⁺CD11b⁺ and CD45⁺c-Kit⁺c-Fms⁺CD11b⁻ cells give rise to osteoclasts and represent bone marrow OCPs (31). To determine whether the decreased osteoclast formation in Cnn2-KO mice is associated with reduced OCPs, we examined the frequency of OCPs in bone marrow cells by flow cytometry and found a slightly decreased OCP frequency in the Cnn2-KO group (Fig. 6C). We performed bone morphometric analysis on hind limb sections and demonstrated that bone volume and osteoclast numbers were similar in 3-mo-old Cnn2-KO mice (Fig. 6D), indicating that deletion of calponin 2 did not significantly affect basal osteoclast formation and function. However, calponin 2 may be required for stimulated osteoclast generation and activation, both of which are critical factors in the joint destructions in rheumatoid arthritis.

Deletion of calponin 2 alters macrophage adhesion.

Cell adhesion is a critical factor in macrophage differentiation and activation (57). The results of our macrophage adhesion studies in Fig. 7 show that the deletion of calponin 2 caused slower adhesion of macrophages to the cultural dish. The maximum number of adherent cells at 40 min was, however, not affected.

Fig. 7.

Deletion of calponin 2 decreased the velocity of macrophage adhesion. Calponin 2-null mouse peritoneal macrophages showed slower adhesion to plastic cultural dish compared with that of WT cells, whereas the total numbers of adherent cells after 40 min were not different between the two groups. ***P < 0.001 vs. WT cells; n = 3 mice at 3–4 mo of age in each group.

The effect of calponin 2 deletion on decreasing the adhesion velocity of macrophages is consistent with its similar effects in previous studies on the adhesion of prostate cancer cells (40) and platelets (15). In addition to the previously demonstrated functions of calponin 2 in regulating macrophage migration and phagocytosis (57), the new results suggest that calponin 2-dependent cell adhesion may be critical to the differentiation and activation of macrophages, as well as the bone resorption activity of osteoclasts (61). Therefore, the function of calponin 2 in regulating mechanointeractions between macrophages and the tissue environment may play a novel role in the pathogenesis of inflammatory arthritis.

Deletion of calponin 2 increases migration velocity of macrophages.

We examined the migration velocity of calponin 2-null and WT mouse macrophages in the absence of chemotactic stimulation using in vitro wound-healing assay. Time lapse microscopy demonstrated that deletion of calponin 2 increased migration velocity of macrophages (Fig. 8), reflecting an intrinsic effect on the dynamics of cytoskeleton activity. This observation is consistent with the effect of calponin 2 deletion in other cell types (40).

Fig. 8.

Deletion of calponin 2 increased the velocity of macrophage migration. Confluent monolayer cultures of elicited mouse peritoneal macrophages were manually scratch wounded using a thin pipette tip, and the healing was monitored using time-lapse phase-contrast microscopy. A: phase-contrast pictures taken every 10 min after the wound. B: distance of single cell (pointed by the arrowheads) movement measured from the time-lapse micrographs showed that calponin 2-null mouse macrophages migrated significantly faster than that of WT cells. **P < 0.001; n = 3 mice at 3–4 mo of ages in each group.

DISCUSSION

Extended from our previous finding that calponin 2 regulates macrophage migration and phagocytosis (21), the present study demonstrated a novel role of macrophage calponin 2 in the pathogenesis of inflammatory arthritis. The new data suggested the following observations.

Upregulation of calponin 2 in macrophages in the joint of inflammatory arthritis.

A finding of our study is the increases in the level of calponin 2 in macrophage-like cells and the number of calponin 2-positive cells in joint tissues of patients with rheumatoid arthritis (Fig. 1A). The increases in calponin 2 in inflammatory arthritis were further demonstrated in macrophages purified from synovial fluid of patients with rheumatoid arthritis compared with that in macrophages differentiated in vitro from normal peripheral blood monocytes (Fig. 1C).

Interestingly, peripheral blood monocytes from patients with rheumatoid arthritis also expressed higher levels of calponin 2 than that in peripheral blood monocytes from normal subjects (Fig. 1B), suggesting a possibly systemic increase of calponin 2 expression in myeloid cells in inflammatory disease. It would be worth further investigating whether the higher level of calponin 2 in myeloid cells is a precondition for increased susceptibility to the development of inflammatory diseases or an initial change in the pathogenesis of rheumatoid arthritis.

Anti-GPI antibodies are found in many patients with rheumatoid arthritis in association with more severe forms of disease (39). Macrophages play a critical role in the pathogenesis of the mouse model of anti-GPI serum transfer-induced arthritis. Our data showed a peak expression of calponin 2 in the mouse arthritis joints 5–14 days after GPI serum transfer, concurrent with the development of arthritic inflammation (Fig. 2). This finding further supports the close correlation between increased calponin 2 and inflammatory arthritis.

Fibroblasts also play a role in the development of inflammatory arthritis (71). We observed increased calponin 2-positive cells located at the synovial lining in patients with rheumatoid arthritis (Fig. 1), where both macrophages and synovial fibroblasts are usually located. Therefore, synovial fibroblasts may also have increased calponin 2 in rheumatoid arthritis. Although our present study focuses on macrophages, the regulation and function of calponin 2 in fibroblasts during the pathogenesis of inflammatory arthritis are worth further investigation.

Deletion of calponin 2 in myeloid cells effectively attenuated the development of inflammatory arthritis.

We performed multiple experiments to reproducibly demonstrate that anti-GPI serum transfer-induced arthritis was significantly attenuated in systemic Cnn2-KO mice (Fig. 3). The selective deletion of calponin 2 in myeloid cells in Cnn2^f/f,lysM^cre mice also highly effectively attenuated the development of inflammatory arthritis (Fig. 4). Together with the finding that macrophages in the joints of patients with rheumatoid arthritis had elevated levels of calponin 2 (Fig. 1C), the data indicate that calponin 2 in macrophages may promote rheumatoid arthritis and that its reduction could be a novel therapeutic approach to facilitating the resolution of arthritic inflammation.

Actin cytoskeleton-based cell motility plays an essential role in the functions of macrophages (5). On the basis of its inhibitory regulation of the actin-activated myosin motor function (37), deletion of calponin 2 increases macrophage migration and phagocytosis (21). Demonstrating the correlation between the increase of calponin 2 in macrophages and the development of inflammatory arthritis and the dramatic effect of calponin 2 deletion on attenuating the severity of disease, our study suggests a hypothesis that the increased level of calponin 2 in rheumatoid arthritic macrophages may reduce their motility and phagocytotic activities, both of which are essential for trafficking out of the joint and the clearance of cellular debris, apoptotic cells, and inflammatory products. A high level of calponin 2 would result in reduced scavenging of inflammatory stimuli and hindering the resolution of inflammation, whereas the deletion of calponin 2 in macrophages would enhance these functions to facilitate the resolution of arthritic inflammation.

Increased phagocytosis has been implicated for a role in promoting macrophage differentiation toward an anti-inflammatory phenotype (10), suggesting another mechanism for the deletion of calponin 2 to attenuate arthritic inflammation.

Deletion of calponin 2 inhibits osteoclastogenesis and bone resorption activity of osteoclasts.

The finding that, in the anti-GPI serum transfer model, calponin 2-null mice developed milder joint inflammation with significantly less bone erosion compared with that in WT mice (Fig. 3) is supported by the fact that deletion of calponin 2 decreased osteoclastogenesis and bone resorption activity of osteoclasts (Fig. 6). The expression of calponin 2 is dependent on cell adhesion (17, 26, 38), and calponin 2 enhances cell adhesion to the extracellular substrate (Fig. 7). The high level of calponin 2 in inflammatory macrophages may, therefore, increase the accumulation of macrophages in the joint (Fig. 1A) and promote macrophage activation and osteoclast formation.

An established function of calponin 2 is the stabilization of actin cytoskeleton (33), which may be a mechanism for its inhibitory effects on cell proliferation, migration, and other motility-related activities. Osteoclast-mediated bone erosion in arthritic joints is based on adhesion and the formation of an actin ring (14). Therefore, the deletion of calponin 2 may have a direct effect on reducing the ability of osteoclasts to form the actin ring and resorption pits (Fig. 6), thus attenuating bone erosion, as seen in anti-GPI serum-induced mouse arthritis (Fig. 3).

Potential role of calponin 2-regulated cell adhesion in macrophage activation and differentiation.

It is widely observed that cell adhesion plays a critical role in macrophage differentiation (36). Interaction of monocytes with extracellular matrix affects their differentiation into residential macrophages (67). Monocyte-macrophage differentiation is accompanied by integrin/CD11b expression and cell adhesion. Treatment with CD11b antisense RNA reduced adhesion and attenuated differentiation (47). The effect of calponin 2 deletion on decreasing the velocity of macrophage adhesion (Fig. 7) suggests a hypothesis that the increased calponin 2 in rheumatoid arthritic macrophages (Figs. 1 and 2) would facilitate cell adhesion and may promote pathological activation and differentiation into proinflammatory phenotypes. Supporting this hypothesis, deletion of calponin 2 in macrophages resulted in effective attenuation of arthritis (Figs. 3 and 4), as well as increased phagocytosis and the rate of proliferation (55, 66), both of which are anti-inflammatory phenotypes.

Calponin 2 as a novel cytoskeleton regulator of macrophage function in inflammatory arthritis.

Calponin 2 regulation of the cytoskeleton function of macrophages is proposed to underlie the effect of Cnn2-KO on attenuating arthritic inflammation. The macrophage is a key player in immune responses and is regulated tightly under the homeostasis of a network of signaling pathways. Many of the present therapies for rheumatoid arthritis target upstream receptors and pathways in the signaling hierarchy of inflammation response, which have broad effects on multiple cell types and may cause multiple functional changes in macrophages. In contrast, the targeted regulation of calponin 2 on cytoskeleton and cell motility-based functions of macrophages demonstrates a downstream effector mechanism with relatively restrictive consequences.

Supporting the hypothesis that deletion of calponin 2 in macrophages is a downstream cellular mechanism to attenuate inflammatory arthritis, the levels of representative pro- and anti-inflammatory cytokines IL-1β, TNF-α, and IL-10 in the tissue extracts of the inflammatory joints of Cnn2-KO mice were similar to that in WT controls (data not shown). Therefore, targeting the cytoskeleton-based function of macrophages through reducing the expression or facilitating the degradation of calponin 2 may steer macrophage function toward the resolution of inflammation for the treatment of inflammatory arthritis without altering the systemic proinflammatory environment, a potentially novel mechanism that merits further investigation.

GRANTS

This study was supported in part by grants from the National Institutes of Health (HL086720 and AR048816 to J.-P. Jin, AR055240 to R. Pope, and AR048697 to L. Xing).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Q.-Q.H., R.M.P., and J.-P.J. conception and design of research; Q.-Q.H., M.M.H., W.S., and L.X. performed experiments; Q.-Q.H., M.M.H., W.S., L.X., R.M.P., and J.-P.J. analyzed data; Q.-Q.H., L.X., R.M.P., and J.-P.J. interpreted results of experiments; Q.-Q.H., M.M.H., W.S., L.X., and J.-P.J. prepared figures; Q.-Q.H., M.M.H., L.X., and J.-P.J. drafted manuscript; Q.-Q.H., M.M.H., R.M.P., and J.-P.J. edited and revised manuscript; Q.-Q.H., M.M.H., W.S., L.X., R.M.P., and J.-P.J. approved final version of manuscript.