Application of in vivo micro-computed tomography in the temporal characterisation of subchondral bone architecture in a rat model of low-dose monosodium iodoacetate-induced osteoarthritis

Abstract

Introduction: Osteoarthritis (OA) is a complex, multifactorial joint disease affecting both the cartilage and the subchondral bone. Animal models of OA aid in the understanding of the pathogenesis of OA and testing suitable drugs for OA treatment. In this study we characterized the temporal changes in the tibial subchondral bone architecture in a rat model of low-dose monosodium iodoacetate (MIA)-induced OA using in vivo micro-computed tomography (CT).

Methods: Male Wistar rats received a single intra-articular injection of low-dose MIA (0.2 mg) in the right knee joint and sterile saline in the left knee joint. The animals were scanned in vivo by micro-CT at two, six, and ten weeks post-injection, analogous to early, intermediate, and advanced stages of OA, to assess architectural changes in the tibial subchondral bone. The articular cartilage changes in the tibiae were assessed macroscopically and histologically at ten weeks post-injection.

Results: Interestingly, tibiae of the MIA-injected knees showed significant bone loss at two weeks, followed by increased trabecular thickness and separation at six and ten weeks. The trabecular number was decreased at all time points compared to control tibiae. The tibial subchondral plate thickness of the MIA-injected knee was increased at two and six weeks and the plate porosity was increased at all time points compared to control. At ten weeks, histology revealed loss of proteoglycans, chondrocyte necrosis, chondrocyte clusters, cartilage fibrillation, and delamination in the MIA-injected tibiae, whereas the control tibiae showed no changes. Micro-CT images and histology showed the presence of subchondral bone sclerosis, cysts, and osteophytes.

Conclusions: These findings demonstrate that the low-dose MIA rat model closely mimics the pathological features of progressive human OA. The low-dose MIA rat model is therefore suitable to study the effect of therapeutic drugs on cartilage and bone in a non-trauma model of OA. In vivo micro-CT is a non-destructive imaging technique that can track structural changes in the tibial subchondral bone in this animal model, and could also be used to track changes in bone in preclinical drug intervention studies for OA treatments.

Figure 1

In vivo micro-CT imaging of a rat hind limb; anaesthetized rat on the scanner bed in supine position, with the hind limb secured in a customised leg fixative device (inset). Micro-CT, micro-computed tomography.

Figure 2

Axial micro-CT image of a rat tibia, with the region of interest (solid red line) in the medial (M) and lateral (L) subchondral bone compartment. Micro-CT, micro-computed tomography.

Figure 3

Plots of morphometric parameters of subchondral trabecular bone determined by micro-CT in the control tibiae and in the MIA-injected tibiae, at two, six, and ten weeks post-MIA injection. There was a statistically significant increase over time in BV, BV/TV, Tb.Th, and Tb.N, and a decrease in Tb.Sp, in both the control and MIA-injected tibiae. Error bars = SD. BV, bone volume; BV/TV, bone volume fraction; CT, computed tomography; MIA, monosodium iodoacetate; TB.N, trabecular number; Tb.Sp, trabecular separation; Tb.Th, trabecular thickness. * P < 0.05, ** P < 0.01, *** P < 0.001, between control tibiae and MIA-injected tibiae.

Figure 4

Coronal and axial micro-CT images of an MIA-injected tibia at two weeks (a), six weeks (b) and ten weeks (c) post injection, together with a control tibia at ten weeks (d). The MIA-injected tibia showed altered subchondral bone architecture, with sclerosis on the medial tibial compartment (M) (indicated by arrow) at six and ten weeks after injection, whereas the control tibia showed no sclerosis at ten weeks. Micro-CT, micro-computed tomography; MIA, monosodium iodoacetate.

Figure 5

Coronal micro-CT image (A) and histology section (B) of tibia from an MIA-injected knee, showing subchondral plate breach by the activated chondrocytes as a result of abnormal repair, as indicated by arrows (A, B). Micro-CT image revealed the presence of focal subchondral bone lesions in the medial tibial plateau (C) at ten weeks post injection (rectangle). The safranin O and fast green stained section of the MIA-injected knee at ten weeks post injection (D) confirmed the presence of subchondral bone cyst in the areas of bone lesions as predicted by micro-CT image. The cysts were found to be surrounded by sclerotic bone with fibrous tissue containing debris of necrotic bone. Note the presence of large active osteoblasts (arrow) lining the trabeculae adjacent to the sites of bone resorption. Original magnification × 100. Micro-CT, micro-computed tomography; MIA, monosodium iodoacetate.

Figure 6

Micro-CT coronal images of tibia from a control knee (A) and a MIA-injected knee (C) at ten weeks post injection. Micro-CT images revealed the presence of marginal tibial osteophytes in all the rats in the MIA-injected knee (C, arrow), whereas the control knee injected with saline showed no osteophyte-like structure formation (A). The normal control tibia (B) and the marginal tibial osteophytes in the MIA-injected knee (D, arrow) were confirmed by histology at ten weeks post injection. The osteophytes contained marrow spaces filled with fibrous bone marrow cells. Original magnification × 40. Micro-CT, micro-computed tomography; MIA, monosodium iodoacetate.

Figure 7

Three-dimensional surface rendering obtained from micro-CT images of a control knee (A) and of a MIA-injected knee (B) at ten weeks after injection. The control knee maintained the subchondral plate integrity with a smooth contour (A). The MIA-injected knee showed erosion and pitting of the tibial subchondral plate, which was more severe in the medial tibial plateau, as indicated by arrow (B). Micro-CT, micro-computed tomography; MIA, monosodium iodoacetate.

Figure 8

Macroscopic images of a control tibia (A-C) and of a MIA-injected tibia (D-F), at ten weeks post injection. The control tibia (A) had no cartilage lesions on the medial compartment (M) and lateral compartment of the tibial plateau, whereas the MIA-injected tibia (D) had severe cartilage lesions on the medial tibial plateau (M). Fluorochrome images of the MIA-injected tibia showed accumulation of calcein and xylenol orange along the margins of the tibial plateau (E, F), suggesting osteophyte formation whereas there was no accumulation of fluorochrome labels along the margins of control tibia (B, C). MIA, monosodium iodoacetate.

Figure 9

Coronal sections stained with Safranin O and fast green of a control tibia (A) and of a MIA-injected tibia (B-D) showing the cartilage on the medial tibial plateau, at ten weeks post injection. The control tibia (A) showed normal healthy cartilage with normally distributed chondrocytes. The MIA-injected tibia (B) showed loss of proteoglycans, loss of viable chondrocytes, chondrocyte proliferation (arrow), and chondrocyte cluster formation of variable sizes (arrowhead). The cartilage showed fibrillation, vertical fissures (C, arrow) and delamination (D). The subchondral trabecular bone architecture was altered with sclerosed bone (C) and the cellular bone marrow was replaced by loosely arranged spindle cells in a fine fibrous stroma (D, arrow). Original magnification ×100. MIA, monosodium iodoacetate.