A model of chronic enthesitis and new bone formation characterized by multimodal imaging

Abstract

Enthesitis is a key feature of several different rheumatic diseases. Its pathophysiology is only partially known due to the lack of access to human tissue and the shortage of reliable animal models for enthesitis. Here, we aimed to develop a model that mimics the effector phase of enthesitis and reliably leads to inflammation and new bone formation. Enthesitis was induced by local injection of monosodium urate (MSU) crystals into the metatarsal entheses of wild-type (WT) or oxidative-burst-deficient (Ncf1**) mice. Quantitative variables of inflammation (edema, swelling) and vascularization (tissue perfusion) were assessed by magnetic resonance imaging (MRI), bone-forming activity by [¹⁸F]-fluoride positron emission tomography (PET), and destruction of cortical bone and new bone formation by computed tomography (CT). Non-invasive imaging was validated by histochemical and histomorphometric analysis. While injection of MSU crystals into WT mice triggered transient mild enthesitis with no new bone formation, Ncf1** mice developed chronic enthesitis accompanied by massive enthesiophytes. In MRI, inflammation and blood flow in the entheses were chronically increased, while PET/CT showed osteoproliferation with enthesiophyte formation. Histochemical analyses showed chronic inflammation, increased vascularization, osteoclast differentiation and bone deposition in the affected entheseal sites. Herein we describe a fast and reliable effector model of chronic enthesitis, which is characterized by a combination of inflammation, vascularization and new bone formation. This model will help to disentangle the molecular pathways involved in the effector phase of enthesitis.

Keywords: Enthesitis; Gout; Mouse model; New bone formation; Spondyloarthropathy.

Fig. 1.

Chronic enthesitis with massive new bone formation after injection of monosodium urate (MSU) crystals in Ncf1** mice. (A,B) Paw swelling elicited by enthesitis after subcutaneous injection of MSU crystals onto the metatarsal ligament insertion sites of BALB/c.Ncf1** and wild-type (WT) BALB/c mice. Plot shows the relative thickness of the MSU-crystal-injected foot normalized to the contralateral PBS-injected foot, as determined by measurement with an electronic caliper (A) or magnetic resonance imaging (MRI) (B). P-values in A refer to comparisons at day 1 or day 21, respectively. P-values in A and B were calculated by unpaired two-tailed Student's t-test. n=8-10. H&E stainings (C), immunohistochemical stainings for Ly6G-positive neutrophils (D) and tartrate-resistant acid phosphatase (TRAP) stainings (E) of paws from BALB/c and BALB/c.Ncf1** mice 21 days after injection of MSU crystals or PBS (contralateral control). Figures show representative images chosen from at least five paws. Black arrows in E mark TRAP-positive osteoclasts, red arrowheads indicate the original surface of cortical bone (before new bone formation occurred). Scale bars: 100 µm or 500 µm.

Fig. 2.

New bone formation in enthesitis of BALB/c and BALB/c.Ncf1** mice as evaluated by PET/CT and histomorphometry. MSU crystals were injected into the metatarsal ligament insertion sites of WT BALB/c and BALB/c.Ncf1** mice, and new bone formation was evaluated 2 and 22 days after by PET/CT using ¹⁸F or by histomorphometry on TRAP-stained sections. (A) Representative PET images showing the uptake of ¹⁸F in the metatarsal space (not quantitative) and (B) quantification as maximum standard uptake values (SUV max). (C) Representative images of paws from BALB/c and BALB/c.Ncf1** mice injected weekly with calcein until the end of the experiment 21 days after injection of MSU crystals or PBS (contralateral control). The green calcein signal indicates newly formed and mineralized bone. (D) Safranin-O–Fast-Green stainings of paws from BALB/c.Ncf1** mice 21 days after injection of MSU crystals or PBS (contralateral control). Pictures show representative images from four paws. (E) Representative 3D-volume rendered CT images of MSU-crystal-injected paws and (F) CT quantification of volumes of newly formed bone along the metatarsals. The arrow in E indicates a massive enthesiophyte. (G) Histomorphometrical analysis of newly formed bone in MSU-crystal-induced enthesitis. Boxplots in B, F and G are visualized as follows: horizontal lines show medians, boxes represent interquartile ranges, whiskers display extreme values. n=6-11. Dashed lines in B indicate the range of values in non-arthritic (PBS-injected contralateral) paws. (H) Correlation between SUV max at day 2 and 22 days after MSU-crystal injection, and volume of newly formed bone. Scale bars: 100 μm or 500 μm.

Fig. 3.

Bone destruction during enthesitis in BALB/c and BALB/c.Ncf1** mice as assessed by CT and histomorphometry. (A) Representative CT images showing the cortical bone and enthesophytes during MSU-crystal-induced enthesitis. Arrows point at lesions where bone density is severely decreased, arrowheads to newly formed bone. (B) Volumes of areas of lowered bone density in the cortical bone as evaluated by CT. (C) Histomorphological analysis of eroded bone and (D) number of TRAP⁺ osteoclasts (OCs) in MSU-crystal-injected paws from BALB/c WT and BALB/c.Ncf1** mice. Horizontal lines show medians, boxes represent interquartile ranges and whiskers display extreme values. n=10 (B) or 4-7 (C,D), respectively. (E) Correlation analysis between erosions of the cortical bone measured by CT and histomorphometry.

Fig. 4.

Inflammatory changes during enthesitis in BALB/c and BALB/c.Ncf1** mice as assessed by MRI. MSU crystals were injected onto the metatarsal enthesial insertion sites of BALB/c WT and BALB/c.Ncf1** mice. Paws were evaluated 1 or 21 days after by MRI. (A) Free water content visualized as hyperintense areas (arrow) on representative short tau inversion recovery (STIR) MRI images. (B) Volume of hyperintense areas (STIR volume) and (C) soft-tissue volume, as calculated from MRI. Mice were then sacrificed and areas of inflammatory infiltrates were histomorphometrically analyzed on H&E-stained paw sections (D). Horizontal lines show medians, boxes represent interquartile ranges and whiskers display extreme values. n=9-10. Dashed lines indicate the range of values in non-arthritic (PBS-injected contralateral) paws. P-values were calculated using unpaired two-tailed Student's t-test. (E,F) Correlation of MRI-assessed features of inflammation (STIR volume and soft-tissue volume) with each other (E) and with histomorphometric assessment of inflammation (F) in BALB/c.Ncf1** mice.

Fig. 5.

Increased blood flow and enhanced vascularization in enthesitis. MSU crystals were injected into the enthesial insertion sites of BALB/c WT and BALB/c.Ncf1** mice. Paws were evaluated by MRI after 1 or 21 days. (A) Colour-coded maps of DCE-MRI depict the distribution of the intravenously injected MR contrast agent (not quantitative). Areas with high blood volume are displayed in red (arrow). From DCE-MRI data, peak enhancement (B), area under the curve (AUC, C), time to peak (TTP, D) and wash out (E) were calculated. Horizontal lines show medians, boxes represent interquartile ranges and whiskers display extreme values. n=8-9. Dashed lines indicate the range of values in non-arthritic (PBS-injected contralateral) paws. P-values were calculated using unpaired two-tailed Student's t-test.