Quantitative assessment of murine articular cartilage and bone using X-ray phase-contrast imaging

Abstract

Murine models for rheumatoid arthritis (RA) research can provide important insights for understanding RA pathogenesis and evaluating the efficacy of novel treatments. However, simultaneously imaging both murine articular cartilage and subchondral bone using conventional techniques is challenging because of low spatial resolution and poor soft tissue contrast. X-ray phase-contrast imaging (XPCI) is a new technique that offers high spatial resolution for the visualisation of cartilage and skeletal tissues. The purpose of this study was to utilise XPCI to observe articular cartilage and subchondral bone in a collagen-induced arthritis (CIA) murine model and quantitatively assess changes in the joint microstructure. XPCI was performed on the two treatment groups (the control group and CIA group, n = 9 per group) to monitor the progression of damage to the femur from the knee joint in a longitudinal study (at 0, 4 and 8 weeks after primary injection). For quantitative assessment, morphologic parameters were measured in three-dimensional (3D) images using appropriate image analysis software. Our results showed that the average femoral cartilage volume, surface area and thickness were significantly decreased (P<0.05) in the CIA group compared to the control group. Meanwhile, these decreases were accompanied by obvious destruction of the surface of subchondral bone and a loss of trabecular bone in the CIA group. This study confirms that XPCI technology has the ability to qualitatively and quantitatively evaluate microstructural changes in mouse joints. This technique has the potential to become a routine analysis method for accurately monitoring joint damage and comprehensively assessing treatment efficacy.

Figure 1. Schematic of the experimental setup of the XPCI system.

Femur samples were positioned vertically on the sample stage. During the CT imaging, the sample stage rotated 180° automatically. A CCD camera was used to record the imaging results. In our experiment, the distance between the sample stage and CCD detector was set to 15 cm.

Figure 2. Before and after CT image via image preprocessing.

(A) 2D CT image without image preprocessing. (B) 2D CT image after image preprocessing.

Figure 3. Segmentation methods for quantifying cartilage and bone morphology in the mouse femur.

2D XPCI image of a femoral axial section (A). The method of region growing and manual drawing was used to semi-automatically segment the cartilage, which is indicated in green (B). According to the grayscale histogram (C), an appropriate threshold, indicated by the arrowhead, was selected to separate the bone from the femur. Through the use of the surface rendering method, the 3D morphologies of the femoral cartilage (D) and bone (E) were visualised to calculate the 3D morphologic parameters.

Figure 4. Histological images of femur sections stained with Safranin-O.

(A) Femur sections from control group at 0, 4 and 8 weeks after the primary injection. (B) Femur sections from CIA group at 0, 4 and 8 weeks after the primary injection. (C) Average thickness of femoral cartilage assessed by histology. Data are presented as the the mean±SD. 2D CT image before image preprocessing. 2D CT image before image preprocessing. #: P<0.05 for differences between the control group and the CIA group. Scale bar: 20 µm.

Figure 5. 2D XPCI sagittal images of femora.

(A) Femur from a normal mouse at 0, 4 and 8 weeks after the primary injection. (B) Femur from a CIA mouse at 0, 4 and 8 weeks after the primary injection. The red arrowheads indicate the damaged regions in the femora. Scale bars: 100 µm.

Figure 6. The 3D morphologic images of femoral cartilage.

(A) Femoral cartilage from a mouse in control group at 0, 4 and 8 weeks after the primary injection. (B) Femoral cartilage from a mouse in CIA group at 0, 4 and 8 weeks after the primary injection.

Figure 7. The 3D morphologic images of subchondral bone sufaces.

(A) subchondral bone from a mouse in control group at 0, 4 and 8 weeks after the primary injection. (B) subchondral bone from a mouse in CIA group at 0, 4 and 8 weeks after the primary injection. The black arrowheads indicate the damaged regions of the subchondral bone surface.

Figure 8. The 3D morphologic images of Trabecular bone.

(A) Trabecular bone from a mouse in control group at 0, 4 and 8 weeks after the primary injection. (B) Trabecular bone from a mouse in CIA group at 0, 4 and 8 weeks after the primary injection. The red arrowheads indicate fragments of trabecular bone.

Figure 9. Evaluation of femoral morphology changes between control and CIA groups.

(A) Cartilage volume. (B) Cartilage surface area. (C) Average cartilage thickness. (D) Bone volume to tissue volume ratio (BV/TV). (E) Bone surface to bone volume ratio (BS/BV). (F) Trabecular bone thickness (Tb.Th). (G) Trabecular bone number (Tb.N). (H) Trabecular bone space (Tb.Sp). Data are presented as the mean±SD. *: P<0.05 for differences between the control group and the CIA group; #: P<0.05 for differences within the CIA group.