Prometheus, Division of Skeletal Tissue Engineering Leuven, KU Leuven, O&N I Herestraat 49 PB813, B-3000 Leuven, Belgium 3

Transcription

1 Contrast-enhanced nanofocus computed tomography: a powerful technique enabling the combined visualization of the tissue architecture of cartilage and bone G. Kerckhofs 1,2, J. Sainz 3, M. Wevers 1, T. Van de Putte 3, J. Schrooten 1,2 1 Department of Metallurgy and Materials Engineering, KU Leuven, Kasteelpark Arenberg 44 PB2450, B-3001 Leuven, Belgium, greet.kerckhofs@mtm.kuleuven.be 2 Prometheus, Division of Skeletal Tissue Engineering Leuven, KU Leuven, O&N I Herestraat 49 PB813, B-3000 Leuven, Belgium 3 TiGenix NV, Haasrode Researchpark 1724, Romeinse straat 12 PB2, B-3001 Leuven, Belgium Introduction and aim To date, magnetic resonance imaging is, due to its high sensitivity for discriminating soft tissues [1,2], the standard technique for the visualization and quantification of cartilage damage. For bone and mineralized tissue, X-ray computed tomography (CT) is the best imaging technique [3-5]. However, for pre-clinical research, where small animal models are typically used to evaluate the efficacy of different therapies for bone and cartilage repair, both techniques suffer from shortcomings in the offered spatial image and/or contrast resolution. In addition, the use of two different techniques does not allow a detailed analysis of the structure, function and pathophysiology of the chondro-osseous junction, which is key for the understanding of the pathogenesis of joint diseases like osteoarthritis (OA). As a solution, the combination of Hexabrix, a radio-opaque negatively charged ionic iodated dimer, with microfocus CT (i.e. equilibrium partitioning of an ionic contrast agent via micro-ct or EPIC-CT) has been employed for simultaneously imaging bone and cartilage in 3D [6-8]. For example, Piscaer et al. [6] applied Hexabrix to in vivo visualize and quantify cartilage degeneration in rats. Palmer et al. [7] performed similar experiments by means of ex vivo micro-ct. However, EPIC-CTs spatial image resolution is limited, and therefore cannot be used for assessment of the mouse model, where the thickness of the hyaline cartilage is only some tens of microns. This would exclude the analysis of the architecture of calcified and non-calcified cartilage and the underlying bone, while the quantification of the thickness and quality of these different layers may be of importance to understand the early effects of OA [9]. In this study, a new methodology, named contrast-enhanced nanofocus X-ray computed tomography (CE-nano-CT), is introduced, aiming at simultaneously imaging the (sub-)tissue architecture of cartilage and bone in mouse models at high contrast and spatial image resolution. High contrast differentiation between bone and cartilage as well as other soft tissues is obtained by the use of an anionic contrast agent (i.e. Hexabrix). When combined with the resolution of nano-ct, this provides the necessary spatial image resolution to reveal differences in structure and quality between the calcified and non-calcified cartilage and the underlying bone in small rodents. To demonstrate the validity and potential of this new methodology, CE-nano-CT was used to visualize the articular cartilage of the knee joint of mice before and after ex vivo treatment with papain. This is known to remove the noncalcified cartilage layer, but keeping the calcified cartilage intact [10]. Validation by comparison with histological stainings was performed. Additionally, the use of this technique was evaluated in a mouse model with meniscus destabilization, were OA manifests itself by various degrees of cartilage lesions [11]. The thickness of both the non-calcified and the calcified cartilage layer in the knee joint of mice that underwent joint destabilization was visualized and quantified in 3D and compared to unaffected mice.

2 Materials and methods - Mouse model In this study, knee joints from adult nude mice (Nude outbred, Rj:NMRI-Foxn1 nu /Foxn1 nu ) were harvested, the joint capsule and synovial membrane were dissected and removed to expose the cartilage layers of the tibial plateau and the femoral condyles. The knee joints were divided in three groups. The first group (WT) consisted of knee joints that were left untreated. The second group (G1) was composed of knee joints that were treated ex vivo with papain from papaya latex, [Sigma Aldrich] after exposure of the cartilage, for enzymatic digestion of the non-calcified cartilage layer (treatment: papain 1% - Cysteine HCl for 3 days at 60 C) according to the protocol described in Hughes et al. [10]. The third group (G2) contained knee joints that were harvested from mice that underwent a medial meniscus destabilization surgical procedure, which resulted in mechanically-induced cartilage lesions, and were sacrificed 8 weeks later. The tibias and/or femurs were then processed for nano- CT imaging and histological analyses. - Contrast-enhanced nanofocus computed tomography (CE-nano-CT) With X-ray nano-ct a complete 3D set of images is acquired to visualize the internal architecture of a sample at the submicron level in a non-destructive way. Nano-CT imaging requires the use of a nanofocal spot X-ray tube, combined with high geometric magnification, and hence differs from conventional medical CT scanning and micro-ct in its ability to resolve details as small as a few hundreds of nanometers in size. The system applied in this study was a Phoenix NanoTom S [GE Measurement and Control Solutions, Germany] with a 180 kv / 15 W high-performance nanofocus X-ray tube. It was equipped with a tungsten target and was operated at a voltage of 60 kv and a current of 140 µa. A 1 mm filter of aluminum was used. The exposure time was 500 ms and a frame averaging of 1 and image skip of 0 was applied, resulting in a scanning time of 20 minutes. The reconstructed images had an isotropic voxel size of (2 µm)³. The contrast agent used in this study was an injectable solution of ioxaglate meglumine 39.3% and ioxaglate sodium 19.6% [Hexabrix 320, Guerbet Nederland B.V., Gorinchem, The Netherlands]. This is a radio-opaque dye which contains ioxaglate, a radio-opaque negatively charged ionic iodated dimer. The negatively charged ioxaglate will be locally repulsed by the negative fixed charge density of the cartilage, which results from the glycosaminoglycans (sgag) in the cartilage. As a consequence, ioxaglate accumulation in the cartilage is inversely related to the sgag content. To obtain this, the joints of the three different groups were, before imaging, immersed for 30 minutes in a solution of Hexabrix 320 diluted 4/5 in PBS, were then wrapped in parafilm and were stably positioned in the nano-ct system. - 3D image analysis For 3D visualization of the bone and cartilage of the femurs and/or tibias, Mimics [Materialise, Haasrode, Belgium] was applied. To quantify the thickness of the non-calcified and calcified cartilage layers, the cartilage region was manually selected every 20 images using CTAn [SkyScan NV, Kontich, Belgium]. By selecting a manually chosen global threshold, the non-calcified and calcified cartilage were segmented and analyzed in 3D by calculating the volume, average thickness and thickness distribution.

3 Figure 1: (A, D, G, J) A representative coronal histological section (Safranin-O/Fast green staining) through the femoral condyles, (B, E, H, K) the corresponding CE-nanoCT section and (C, F, I, L) the same CE-nanoCT section with the traced regions of interest (i.e. total (non-calcified and calcified) hyaline cartilage) of (A-F) a wild type mouse condyle and (G-L) a mouse condyle treated with papain to remove the non-calcified cartilage layer, but preserving the calcified cartilage layer. (D-F) white rectangle indicated in (J-L). Scale bars = 200 µm. Magnification of the white rectangle indicated in (A-C), and (G-I) a blow-up b of the

4 - Histology After nano-ct imaging, the femurs and/or tibias were processed for histology. They were fixed in 10% formalin, decalcified in 4% formic acid to enable tissue sectioning, and embedded in paraffin. Sections of 5 µm were made throughout the joint and stained every 60 µm with Safranin-O and Fast Green, resulting in images where bone stains light blue, calcified cartilage stains light red and non-calcified cartilage stains dark red, as can be seen in figure 1. Results - Comparison to histology Figures 1A and 1D show a typical coronal section upon Safranin-O/Fast green staining of a wild type femoral condyle, where the bone is colored light blue, the calcified cartilage light red and the non-calcified cartilage dark red. Figures 1B and 1E present the corresponding CE-nano-CT images. The non-calcified cartilage (dark grey), and the calcified cartilage (light grey porous layer) could be distinguished from each other, as clearly, if not more easily, on the CE-nano-CT images than on the histological sections. The porous structure of the calcified cartilage allowed a clear visualization of transition between the subchondral bone and the calcified cartilage, the chondro-osseous junction. In figures 1C and 1F, the cartilage region-of-interest (ROI) is highlighted. This ROI was used for the quantification of the volume and thickness of the non-calcified and calcified cartilage layers. Figures 1G to 1L are representative histological sections (fig. 1G and 1J) and their corresponding CE-nano-CT images (fig. 1H, 1I, 1K and 1L) for the papain-treated group of femoral condyles (G1). From this, it is apparent that the papain treatment effectively removed the non-calcified cartilage layer. The calcified cartilage layer was still clearly visible on both the histological and CE-nano-CT images (highlighted in fig. 1I). Comparison between fig 1F and 1L confirmed the identification of the different cartilage layers visualized by CE-nano-CT. - 3D visualization and quantification In figures 2A, 2B and 2C, typical cross-sectional (respectively coronal, axial and sagital) CEnano-CT images of a tibia of mice that had undergone an in vivo joint destabilization are shown. Figure 2D represents a 3D model of the tibia. The non-calcified cartilage layer on the lateral side of the tibia is colored in green, the non-calcified cartilage layer on the medial side in red. Both in the 2D cross-sectional images and the 3D image, a severe lesion within the cartilage is apparent on the medial side. This location is in line with the literature on the pathology as it is caused by meniscus destabilization [11]. 3D quantification of the average non-calcified and calcified cartilage thickness (fig. 3A and B respectively) showed significant differences between the wild type mice (n = 4) and the mice with a destabilized meniscus (n = 4) for the non-calcified cartilage, but not for the calcified cartilage. The destabilization of the meniscus for 8 weeks in vivo thus resulted in a significant decrease in average non-calcified cartilage thickness, leaving the underlying calcified cartilage more unaffected.

5 Figure 2: A representative (A) coronal, (B) axial and (C) sagital CE-nano-CT imagee of a tibia off group 2 (mice with a destabilized meniscus). (D) A representative 3D imagee of the same tibia, where the bone and calcified cartilage is colored in yellow, the non-calcified hyaline cartilage layer on the lateral side of the tibia in green and the non-calcified hyaline cartilage on the medial m side in red. Scale bars = 1 mm. Figure 3: (A) The average thickness of non-calcified cartilage on the medial m tibial plateau of wild type mice (WT; n = 4) and mice with a destabilized meniscus (G2; n = 4), and (B) the average thickness of calcified cartilage on the medial tibial plateau of the same WT mice andd operated mice from G2. [* p < 0.05 = significantly different]. Conclusion CE-nano-Cimage resolution, the tissue architecture of bone and hyaline cartilage in knee joints of small is a powerful tool to simultaneously visualize, with high contrast and spatial rodents. With this technique, non-calcified d and calcified cartilage can be distinguished from each other and both can be clearly discriminated from the subchondral bone on the base of one single scan. Based on such high-contrast and high-resolution images, a 3D visualization of the joint surface and quantitative measurementss of the volume, thickness and quality (presencee of sgags) of the different tissue types can be performed. This allows to investigate in mouse models the consequences of OA on the joint surface architecture and to assess the efficacy of different treatmentss for cartilage and bonee repair faster and in a more quantitative manner than it is now possible with the current state-of-the-s -art histochemical techniques.

6 Acknowledgements This study was financed by the Agency for Innovation by Science and Technology in Flanders (IWT/OZM/090655). We are very grateful to Gert Vanderlinden, Joyce Op De Beeck, Johan Neys and Muriel Hemelaers for the excellent technical assistance in the histological sectioning and staining, and in the drawing of the ROIs in the CE-nano-CT images. References: 1. Xu H.H., Othman S.F., Monitoring Tissue Engineering Using Magnetic Resonance Imaging. J. Biosci. Bioeng. 106 (6): p , Kuperman V., Magnetic resonance imaging: physical principles and applications, Academic Press, San Diego, Cartmell S., Huynh K., Lin A., Nagaraja S. and Guldberg R. Quantitative microcomputed tomography analysis of mineralization within three-dimensional scaffolds in vitro. J Biomed Mater Res A 69A (1): p , Hagenmueller H., Hofmann S., Kohler T., Merkle H.P., Kaplan D.L., Vunjak- Novakovic G., Mueller R. and Meinel L. Non-invasive time-lapsed monitoring and quantification of engineered bone-like tissue. Ann. Biomed. Eng. 35 (10): p , Porter B.D., Lin A.S.P., Peister A., Hutmacher D. and Guldberg R.E. Noninvasive image analysis of 3D construct mineralization in a perfusion bioreactor. Biomaterials 28 (15): p , Piscaer T.M., Waarsing J.H., Kops N., Pavljasevic P., Verhaar J.A.N., van Osch G.J.V.M. and Weinans H. In vivo imaging of cartilage degeneration using [mu]ctarthrography. Osteoarthritis and Cartilage 16 (9): p , Palmer A.W., Guldberg R.E. and Levenston M.E. Analysis of cartilage matrix fixed charge density and three-dimensional morphology via contrast-enhanced microcomputed tomography. Proceedings of the National Academy of Sciences of the United States of America 103 (51): p , Xie L., Lin A.S.P., Levenston M.E. and Guldberg R.E. Quantitative assessment of articular cartilage morphology via EPIC-mu CT. Osteoarthritis and Cartilage 17 (3): p , Botter S.M., van Osch G., Waarsing J.H., van der Linden J.C., Verhaar J.A.N., Pols H.A.P., van Leeuwen J., Weinans H., Cartilage damage pattern in relation to subchondral plate thickness in a collagenase-induced model of osteoarthritis, Osteoarthritis and Cartilage 16: p , Hughes L.C., Archer C.W., Gwynn I. ap, The ultrastructure of mouse articular cartilage: collagen orientation and implications for tissue functionality. A polarised light and scanning electron microscope study and review, Eur Cell Mater 9: p.68-84, Glasson S.S., Blanchet T.J., Morris E.A., The surgical destabilization of the medial meniscus (DMM) model of osteoarthritis in the 129/SvEv mouse. Osteoarthritis Cartilage 15(9): p , 2007.