Attenuation of the progression of articular cartilage degeneration by inhibition of TGF-β1 signaling in a mouse model of osteoarthritis

Abstract

Transforming growth factor beta 1 (TGF-β1) is implicated in osteoarthritis. We therefore studied the role of TGF-β1 signaling in the development of osteoarthritis in a developmental stage-dependent manner. Three different mouse models were investigated. First, the Tgf-β receptor II (Tgfbr2) was specifically removed from the mature cartilage of joints. Tgfbr2-deficient mice were grown to 12 months of age and were then euthanized for collection of knee and temporomandibular joints. Second, Tgfbr2-deficient mice were subjected to destabilization of the medial meniscus (DMM) surgery. Knee joints were then collected from the mice at 8 and 16 weeks after the surgery. Third, wild-type mice were subjected to DMM at the age of 8 weeks. Immediately after the surgery, these mice were treated with the Tgfbr2 inhibitor losartan for 8 weeks and then euthanized for collection of knee joints. All joints were characterized for evidences of articular cartilage degeneration. Initiation or acceleration of articular cartilage degeneration was not observed by the genetic inactivation of Tgfbr2 in the joints at the age of 12 months. In fact, the removal of Tgfbr2 and treatment with losartan both delayed the progression of articular cartilage degeneration induced by DMM compared with control littermates. Therefore, we conclude that inhibition of Tgf-β1 signaling protects adult knee joints in mice against the development of osteoarthritis.

Copyright © 2015 American Society for Investigative Pathology. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1

Genotyping strategy for the removal of the exon 2 of Tgfbr2.

Figure 2

The removal of the exon 2 of Tgfbr2 in articular chondrocytes of mouse knee joints. A: Before the Cre-recombination, a size of 1-kb PCR product is generated by the primers Tgfbr2-F4 and Tgfbr2-R and after the Cre-recombination, a size of 608-bp PCR product is generated by the primers Tgfbr2-F and Tgfbr2-R. B: Cells with intense X-gal staining (blue color) appear above the tidemark (dashed line) of the articular cartilage of knee joints in AgcCreERT2+/−,Rosa26+/floxlacZ mice at 2 months of age. A counterstain was performed with Safranin O staining of proteoglycan. C: Every articular chondrocyte in AgcCreERT2+/−;Tgfbr2+/+ littermates contains the exon 2 of Tgfbr2. However, <14% of articular chondrocytes with the exon 2 of Tgfbr2 are observed in the conditionally Tgfbr2-knockout mice. Significant difference was found in the number of the exon 2 of the Tgfbr2 cells between the two groups, ∗∗P < 0.01 (t-test).

Figure 3

Structural characteristics of articular cartilage in mouse knee joints. No overt observable structural characteristic changes were found in the articular cartilage of knee joints in the Tgfbr2-deficient mice at the ages of 6 and 12 months (Safranin O/Fast Green). Scale bar = 100 μm.

Figure 4

Immunohistostaining for collagen type X. Collagen type X is present in the extracellular matrix of the HZ of growth plates in mice at 1 month of age (arrowheads). The collagen type X is absent in the PZ of the growth plate. No positive staining of the collagen type X was found in the AC of both Tgfbr2-deficient mice and their AgcCreERT2+/−;Tgfbr2+/+ littermates at 12 months of age. The dashed line in the growth plate indicates the boundary between PZ and HZ. The dashed lines in ACs indicate the boundary between the articular cartilage and subchondral bone. AC, articular cartilage; HZ, hypertrophic zone; PZ, proliferative zone; SB, subchondral bone.

Figure 5

X-Gal staining in the condylar cartilage of TM joints in mice. Intensive X-gal staining cells (blue color) are observed in the bottom layer of condylar cartilage in TM joints of AgcCreERT2+/−,Rosa26+/floxlacZ mice at the ages of 1 (B and F) and 6 (D and H) months. The black dashed lines in F and H indicates the border between the condylar cartilage and subchondral bone. No X-gal staining cells are in the top layer of the condylar cartilage in TM joints, which is consistent with the structural characteristics of the condylar cartilage of TM joints by Safranin O/Fast Green staining (I and J). X-Gal staining cells are absent in the cartilage of TM joints in the control mice (A, C, E, and G). TM, temporomandibular.

Figure 6

Structural characteristics of condylar cartilage in mouse TM joints. No overt structural characteristic changes are seen in the condylar cartilage of TM joints in the Tgfbr2-deficient mice at the ages of 6 and 12 months. Safranin O/Fast Green. Scale bar = 100 μm. TM, temporomandibular.

Figure 7

Structural characteristic evaluation of the articular cartilage of knee joints from the Tgfbr2-deficent mice after DMM surgery. A: Fibrillation (arrow) in the articular cartilage of AgcCreERT2+/−;Tgfbr2+/+ littermates at 8 weeks after the surgery. However, only localized absence of the proteoglycans is observed in AgcCreERT2+/−;Tgfbr2−/− mice at 8 weeks after the surgery. At 16 weeks after the surgery, a complete loss of the articular cartilage (lesion reaching the tidemark and beyond) is evident in both femoral and tibia condyles of the AgcCreERT2+/−;Tgfbr2+/+ littermates. The missing cartilage is extended >75% of the surface area. In the AgcCreERT2+/−;Tgfbr2−/− mice, the fibrillation reaching to the tidemark is evident, and the missing cartilage is only extended <25% of the surface area. A delay in the degenerative process is seen in Tgfbr2-deficent mice after surgery. B: The mean scores were 2.19 ± 0.79 in AgcCreERT2+/−;Tgfbr2+/+ mice and 0.88 ± 0.22 in AgcCreERT2+/−;Tgfbr2−/− mice at 8 weeks after DMM surgery. C: The mean scores were 4.17 ± 0.70 in AgcCreERT2+/−;Tgfbr2+/+ mice and 2.25 ± 0.66 in AgcCreERT2+/−;Tgfbr2−/− mice at 16 weeks after DMM surgery. Data are expressed as means ± SD. n = 8 mice. ∗P < 0.05 (t-test). Scale bar = 100 μm. DMM, destabilization of the medial meniscus.

Figure 8

Immunohistostaining of p-Smad2/3 in the articular cartilage of the Tgfbr2-deficent mice. p-Smad2/3 protein-positive staining cells (17%) are present in the articular cartilage of AgcCreERT2+/−;Tgfbr2+/+ mice after DMM surgery. However, the p-Smad2/3+ staining cells are only detected at approximately 2% in the Tgfbr2-defecient mice. The white dashed line separates the articular cartilage from the subchondral bone. The number of the positive staining cells is significantly different between the two groups (P < 0.05). Scale bar = 50 μm. DMM, destabilization of the medial meniscus.

Figure 9

Structural characteristic evaluation of the articular cartilage of knee joints from losartan-treated mice at 8 weeks after DMM surgery. A: The structural characteristic conditions of articular cartilages in mice are similar to Tgfbr2-deficient mice. Fibrillation (arrow) in the articular cartilage of mice without treatment of losartan; no fibrillation is observed in mice with treatment of losartan (Safranin O/Fast Green). B: The mean scores were 2.14 ± 0.58 in mice without the treatment of losartan and 0.64 ± 0.23 in mice with the treatment of losartan. C: Immunostaining for p-Smad2/3 protein is similar to Tgfbr2-deficient mice. The p-Smad2/3 positive-staining cells (16%) are seen in the articular cartilage of mice without treatment of losartan after DMM, whereas the p-Smad2/3 positive-staining cells are detected at approximately 1% in the losartan-treated mice after DMM surgery. The white dashed line separates the articular cartilage from the subchondral bone. Data are expressed as means ± SD. n = 8 mice. n = 7 mice. ∗P < 0.05 (t-test). Scale bars: 100 μm (A); 50 μm (B). DMM, destabilization of the medial meniscus.

Figure 10

A schematic illustration of the molecular pathways underlying articular cartilage degeneration. DDR2, discoidin domain receptor 2; MMP, matrix metalloprotease; TGF, transforming growth factor.