Osteopontin deficiency protects joints against destruction in anti-type II collagen antibody-induced arthritis in mice

Abstract

Rheumatoid arthritis is one of the most critical diseases that impair the quality of life of patients, but its pathogenesis has not yet been fully understood. Osteopontin (OPN) is an extracellular matrix protein containing Arg-Gly-Asp (RGD) sequence, which interacts with alpha(v)beta3 integrins, promotes cell attachment, and cell migration and is expressed in both synovial cells and chondrocytes in rheumatoid arthritis; however, its functional relationship to arthritis has not been known. Therefore, we investigated the roles of OPN in the pathogenesis of inflammatory process in a rheumatoid arthritis model induced by a mixture of anti-type II collagen mAbs and lipopolysaccharide (mAbs/LPS). mAbs/LPS injection induced OPN expression in synovia as well as cartilage, and this expression was associated with joint swelling, destruction of the surface structures of the joint based on scanning electron microscopy, and loss of toluidine blue-positive proteoglycan content in the articular cartilage in wild-type mice. In contrast, OPN deficiency prevented the mice from such surface destruction, loss of proteoglycan in the articular joint cartilage, and swelling of the joints even when the mice were subjected to mAbs/LPS injection. Furthermore, mAbs/LPS injection in wild-type mice enhanced the levels of CD31-positive vessels in synovia and terminal deoxynucleotidyltransferase-mediated UTP end labeling-positive chondrocytes in the articular cartilage, whereas such angiogenesis as well as chondrocyte apoptosis was suppressed significantly in OPN-deficient mice. These results indicated that OPN plays a critical role in the destruction of joint cartilage in the rheumatoid arthritis model in mice via promotion of angiogenesis and induction of chondrocyte apoptosis.

Figure 1

OPN protein expression in the knee joints of wild-type mice with arthritis. OPN protein expression in the knee joints was detected by OPN immunostaining in saline (a) and mAb/LPS-injected, wild-type mice (b). OPN expression was enhanced in the superficial layer (arrow) and deep layer (*) of cartilage by injection with mAb/LPS (a vs. b).

Figure 2

Surface morphology of articular cartilage examined by scanning electron microscopy. Surface of articular cartilage of finger joints of wild-type (a, b, e, and f) or OPN-deficient (c, d, g, and h) mice was examined by scanning electron microscopy on day 18 after saline (a, e, c, and g) or mAb (b, f, d, and h) injection. (i) Quantification of the destructed area in articular cartilage surface. The rough areas of cartilage in saline-injected, wild-type (n = 5); mAb/LPS-injected, wild-type (n = 5); saline-injected, OPN-deficient (n = 5); and mAb/LPS-injected, OPN-deficient (n = 6) mice were measured with an image analyzer equipped with the Luzex system (Nireco, Tokyo). Data are expressed as means ± SEM. *, P < 0.05; asterisk indicates that the difference between saline vs. mAb/LPS-injected, wild-type mice was statistically significant.

Figure 3

Proteoglycan-staining levels in articular cartilage. The sections of articular cartilage of finger joints in hind paws of wild-type (a and b) or OPN-deficient (c and d) mice were stained with toluidine blue to visualize proteoglycan levels after injection with saline (a and c) or mAbs/LPS (b and d). Significant loss of proteoglycan staining (arrows) was observed in mAb/LPS-injected, wild-type mice (b), but such loss was not observed in mAb/LPS-injected, OPN-deficient mice (d).

Figure 4

TUNEL signals in cartilage. The levels of TUNEL-positive cells were low in the knee joint cartilage in saline-injected mice (a), but they were enhanced significantly by the injection of mAbs/LPS (b arrows) in wild-type mice. In contrast, such up-regulation was not observed in OPN-deficient mice even after mAb/LPS injection (c and d). (e) Quantification of TUNEL-positive cells (n = 4). Data are expressed as means ± SEM. *, P < 0.05, saline vs. mAb/LPS-injected, wild-type mice.

Figure 5

The swelling of joints in wild-type and OPN-deficient mice after injection with mAbs/LPS. Wild-type (n = 5) or OPN-deficient (n = 6) mice ages 6 to 7 weeks were injected i.p. with mAbs on day 0, followed by i.p. injection of 50 μg of LPS on day 3. (a) Arthritis score was graded on a 1–4 scale as described in Materials and Methods. Data are means ± SEM. *, P < 0.05, saline vs. mAb/LPS-injected, wild-type mice. (b) Photographs of swelling in fore and hind paws in wild-type and OPN-deficient mice on day 9. (c) Soft x-ray photographs of fore paws on day 14. (d) Thickness of soft tissues at finger joints (n = 3); quantification of the thickness of the soft tissues at finger joints was conducted by using soft x-ray pictures. The levels of the thickness of the soft tissues at the phalangeal joints were expressed as averaged soft tissue width of middle three fingers (2nd, 3rd, and 4th) in fore paw of three mice. Data are expressed as means ± SEM. *, P < 0.05; asterisk indicates that the difference between saline vs. mAb/LPS-injected, wild-type mice was statistically significant.

Figure 6

Inflammatory cell infiltration in the knee joint. The sections of knee joints were stained with hematoxylin/eosin. Sections were prepared from wild-type (a and b) or OPN-deficient (c and d) mice injected with saline (a and c) or mAbs/LPS (b and d). Inflammatory cells accumulated in synovia were abundant (arrows) in mAb/LPS-injected, wild-type mice (b). In contrast, they were less severe in OPN-deficient mice even after injection with mAbs/LPS (c and d).

Figure 7

Angiogenesis in knee joints. The levels of PECAM-1 (CD31)-positive cells were assessed by staining, using MEC 13.3 rat monoclonal anti-mouse PECAM-1 (CD31) antibody in saline-injected (a and c) or mAb/LPS-injected (b and d), wild-type (a and b), or OPN-deficient (c and d) mice. (e) PECAM-1 (CD31)-positive cells (n = 3) in the area (0.14 × 0.16 mm²) of synovium were counted. Data are expressed as means ± SEM. *, P < 0.05; asterisk indicates that the difference between saline- and mAb/LPS-injected, wild-type mice was statistically significant.

Figure 8

Serum TNF-α levels in mice injected with LPS and urinary D-Pyr levels in mAb/LPS-injected mice. (a) Serum TNF-α levels 1 h after i.p. LPS (0111:B4; 50 μg per body) injection (n = 5) were examined. Serum TNF-α levels induced by LPS injection were similar in wild-type and OPN-deficient mice. (b) Urinary D-Pyr levels (nM/mM creatinine) were measured on day 14 after the mAb injection. Data are expressed as means ± SEM. *, P < 0.05; asterisk indicates that the difference between saline- and mAb/LPS-injected mice was statistically significant.