If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Wallace H. Coulter Department of Biomedical Engineering, Parker H. Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, USA
Wallace H. Coulter Department of Biomedical Engineering, Parker H. Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, USA
Wallace H. Coulter Department of Biomedical Engineering, Parker H. Petit Institute for Bioengineering and Bioscience, Georgia Institute of Technology, USA
Address correspondence and reprint requests to: R.E. Guldberg, Parker H. Petit Institute for Bioengineering and Bioscience, 315, Ferst Drive, Georgia Institute of Technology, Atlanta, GA 30332-0405, USA. Tel: 1-404-894-6589; Fax: 1-404-385-1397.
Current histological scoring methods to evaluate efficacy of potential therapeutics for slowing or preventing joint degeneration are time-consuming and semi-quantitative in nature. Hence, there is a need to develop and standardize quantitative outcome measures to define sensitive metrics for studying potential therapeutics. The objectives of this study were to use equilibrium partitioning of an ionic contrast agent via Equilibrium Partitioning of an Ionic Contrast-Microcomputed tomography (EPIC-μCT) to quantitatively characterize morphological and compositional changes in the tibial articular cartilage in two distinct models of joint degeneration and define localized regions of interest to detect degenerative cartilage changes.
Materials and Methods
The monosodium iodoacetate (MIA) and medial meniscal transection (MMT) rat models were used in this study. Three weeks post-surgery, tibiae were analyzed using EPIC-μCT and histology. EPIC-μCT allowed measurement of 3D morphological changes in cartilage thickness, volume and composition.
Results
Extensive cartilage degeneration was observed throughout the joint in the MIA model after 3 weeks. In contrast, the MMT model showed more localized degeneration with regional thickening of the medial tibial plateau and a decrease in attenuation consistent with proteoglycan (PG) depletion. Focal lesions were also observed and 3D volume calculated as an additional outcome metric.
Conclusions
EPIC-μCT was used to quantitatively assess joint degeneration in two distinct preclinical models. The MMT model showed similar features to human Osteoarthritis (OA), including localized lesion formation and PG loss, while the MIA model displayed extensive cartilage degeneration throughout the joint. EPIC-μCT imaging provides a rapid and quantitative screening tool for preclinical evaluation of OA therapeutics.
. However, there are currently no clinically approved therapeutics that can halt the progression of OA by controlling cartilage erosion and chondrocyte loss
Developing disease modifying osteoarthritis drugs (DMOADs) remains an unmet need for OA patients. Animal models used for assessing therapeutic effects of these drugs, such as matrix metalloproteinase inhibitors (MMPi), have traditionally used histology as an endpoint readout
A new class of potent matrix metalloproteinase 13 inhibitors for potential treatment of osteoarthritis: evidence of histologic and clinical efficacy without musculoskeletal toxicity in rat models.
Discovery and characterization of a novel inhibitor of matrix metalloprotease-13 that reduces cartilage damage in vivo without joint fibroplasia side effects.
Biochemical and metabolic abnormalities in articular cartilage from osteo-arthritic human hips. II. Correlation of morphology with biochemical and metabolic data.
. The screening rate of drugs is currently limited by the sensitivity of the scoring method (semi-quantitative pathology scoring), requiring a large number of samples and extensive time commitment. As additional disadvantages, histology is destructive, time consuming and typically represents only a 2D analysis. Therefore, there is a need to develop quantitative outcome measures in animal models of joint degeneration to provide platforms to better understand joint diseases and to evaluate the efficacy of potential therapeutics.
The need to develop new imaging technologies is not limited to just studying potential therapeutics. Anatomical and chemical composition of structures can be characterized using these new imaging modalities allowing for long-term in-vivo monitoring of development and progression of joint diseases. These techniques have been critical to progress in fields such as cancer research and cardiology, but have thus far been limited in orthopedics mainly due to presence of multiple different tissue structures, resolution and acquisition time of images
Microcomputed tomography (μCT) has traditionally provided 3D, quantitative analysis of hard tissues. This imaging methodology has recently been modified to facilitate application to soft tissues, such as cartilage, using contrast agents. Equilibrium Partitioning of an Ionic Contrast agent via μCT (EPIC-μCT), allows for non-destructive evaluation of cartilage morphology and composition
. Cartilage degeneration results in a depletion of negatively charged proteoglycans (PGs). This depletion allows for more of the negatively charged contrast agent to diffuse into the cartilage matrix compared to healthy tissue. The X-ray attenuation of equilibrated tissue has previously been shown to have an inverse relationship with sulfated glycosaminoglycans (sGAG) content (validated by Safranin-O optical density and the 1,9-dimethylmethylene blue colorimetric assay) resulting in higher attenuation values for degenerated tissue
. Morphological measurements with EPIC-μCT have been validated compared to measurements of cartilage thickness (histomorphometry and needle probe testing) and demonstrated a strong linear relationship
. This presents a quantitative metric for analyzing cartilage composition and detecting changes in cartilage quality in addition to 3D morphology. Previous studies have used different variations of μCT, such as EPIC-μCT and in vivo μCT arthrography to image cartilage in multiple joint degeneration models
Several surgically and chemically induced models of OA have been developed in different species as platforms to evaluate disease pathology as well as to screen therapeutics
. In this study, two joint degeneration rat models commonly used to screen OA therapies were characterized—chemical induction of joint degeneration using monosodium iodoacetate (MIA) and surgical induction of joint degeneration via medial meniscal transection (MMT)
. Each method presents advantages for use in small animal models of joint degeneration. The MIA model is quick, easy, reproducible, and animals exhibit signs of OA related pain
. The MMT model, is characterized by progressive cartilage degeneration involving chondrocyte and PG loss, osteophyte and lesion formation, collagen degradation and cartilage fibrillation
. A major difference in the two models is that the MIA model leads to global degenerative changes whereas the MMT model leads to more localized damage. The MIA model is also typically more aggressive and associated with increased joint pain. While a high dose MIA model—where PGs are nearly completely depleted—has previously been characterized using EPIC-μCT
, EPIC-μCT has not been used to characterize a less aggressive low dose MIA model or the MMT model.
EPIC-μCT allows for high-resolution 3D quantification of cartilage morphology and composition. It is a relatively high throughput technique, provides 3D volumetric data, and enables increased sensitivity and 3D visualization. The objectives of this study were to use EPIC-μCT to quantitatively characterize morphological and compositional changes in the tibial articular cartilage in two distinct models of joint degeneration and define localized regions of interest to detect degenerative cartilage changes.
Methods
Induction of joint degeneration
Georgia Institute of Technology Institutional Animal Care and Use Committee approved all experimental animal procedures (IACUC protocol #A09007). Two groups of rats were used in this study – 8-week old Wistar rats injected with MIA and weight matched Lewis rats (300–325 g) that underwent MMT surgery
Eight 8-week old male Wistar rats (Charles River, Wilmington, MA) were acclimated for 1 week. After anesthetizing with isoflurane, 0.3 mg MIA (Sigma–Aldrich, St. Louis, MO) in 50 μl saline was injected into the joint space through the infrapatellar ligament of the left knee. Right knees were injected with 50 μl saline as a contralateral control.
Seven weight matched adult male Lewis rats (Charles River, Wilmington, MA; 300–325 g) were acclimated for 1 week. The animals were anesthetized with isoflurane, and a small incision through the skin was made on the medial aspect of the left femoro-tibial joint. The medial collateral ligament (MCL) was exposed by blunt dissection and transected to visualize the joint space and medial meniscus. The meniscus was transected completely at its narrowest point. The skin was closed with 4.0 silk sutures and then stapled using wound clips. All surgical sites were observed daily for signs of abnormal healing.
Assessment of cartilage
Rats were euthanized via CO2 inhalation at 3 weeks after MIA injection or MMT surgery. Tibiae were harvested, dissected free of surrounding tissues, and fixed in 10% neutral buffered formalin (NBF, EMD Chemicals, Gibbstown, NJ) for 3–4 days and then transferred to 70% ethanol
For EPIC-μCT, the proximal end of each tibia was immersed in 2 ml of 30% Hexabrix 320 contrast agent (Covidien, Hazelwood, MO) and 70% ion-free phosphate buffer saline (PBS) at 37 °C for 30 min, an incubation period known to result in equilibration of the agent
. There was no difference in the average X-ray attenuation levels, thickness, volume or surface area determined for fresh and formalin-fixed cartilage at that time point (Fig. S1). Proximal tibiae were scanned using a μCT 40 (Scanco Medical, Brüttisellen, Switzerland) at 45 kVp, 177 μA, 200 ms integration time, and a voxel size of 16 μm
Following scanning, tibiae were decalcified in Cal-Ex II (Fisher Scientific, Waltham, MA) for 14 days. Samples were routinely paraffin embedded. For comparison with EPIC-μCT images, sagittal sections were cut at 5 μm thickness. Sections were stained for sGAGs with 0.5% Safranin-O and 0.2% Fast Green counterstain or hematoxylin & eosin. For each sample, three sagittal sections were used for thickness histomorphometry analysis and digital images of each section were captured. The cartilage thickness of each section was defined as the average value of five manual thickness measurements at regular intervals perpendicular to the cartilage surface and the values for the three sections were then averaged.
μCT regional analysis
Scanco evaluation software was used to assess 3D morphology and composition. Raw scan data were automatically reconstructed to 2D grayscale tomograms. These were rotated to sagittal sections, and the cartilage was contoured to separate it from trabecular bone and surrounding air
. The cartilage was segmented using a bandpass filter with a minimum threshold value of 175 to eliminate air and 225 to eliminate bone from the raw image following which 3D images were generated. The threshold values were globally applied for both the left and right tibiae of all animals.
The MMT surgery produces localized cartilage degeneration on the medial plateau of the tibia. To observe localized effects of joint degeneration, we evaluated the medial tibial plateau in the MMT group. The regional analyses were replicated for the MIA tibiae for comparison. Four regional analyses were performed using different volumes of interest (VOI). For the MMT model, images were evaluated for: (1) Full articular cartilage in the proximal tibia including medial and lateral aspects (2) Medial plateau only (3) Medial 1/3 of the medial tibial plateau (4) Focal lesions on the medial plateau only (Fig. 1). For the MIA model, only analyses (1) and (2) were performed for overall tibial articular cartilage and tibial medial articular cartilage respectively. Two outcome measures were defined for evaluation of focal defects: erosion (defect extended to less than 50% of cartilage thickness) and lesion (defect extended to more than 50% of cartilage thickness)
. To define VOIs of focal lesions, manual contouring was performed in the isolated lesion area (typically ∼8 slices). Within this VOI, segmented cartilage volume was subtracted from the total volume evaluated to compute lesion volume.
Fig. 1Representative image illustrating EPIC-μCT image analysis (tibial cartilage overlaid on bone) and volume of interest (VOI) designations. Pseudocolor coding (indicated at bottom) displays larger thicknesses in red and smaller thicknesses in blue/green. B: Representative grayscale EPIC-μCT images of cartilage showing erosions (B) and lesion (C) (Inset shows zoomed in view). Lesion locations were identified by manually examining sagittal grayscale images slice by slice.
All data were expressed as mean ± standard deviation of the mean. The means of cartilage morphology variables—cartilage thickness, volume, and attenuation—fit a Gaussian distribution (Kolmogorov–Smirnov test) and were evaluated using paired t-tests to compare experimental and control tibiae for MMT and MIA groups separately. Each individual region (overall cartilage, medial tibial plateau and medial 1/3 of medial tibial plateau) was evaluated independently with separate paired t-tests for each of the cartilage morphology variables listed above. All analysis was done using GraphPad Prism (GraphPad Software Inc. 5.01, La Jolla, CA). Statistical significance was set at a 95% confidence interval (P < 0.05). A P-value correction factor was not used in this analysis as direct comparisons between the three regions were not performed. For reference, the results with the incorporation of a P-value correction factor are presented in the supplementary material (Figs. S2 and S3).
Results
MIA articular cartilage analysis
Representative images of tibial articular cartilage sections stained with Safranin-O in comparison with EPIC-μCT images are shown [Fig. 2(A,C,D and F)]. At 3 weeks post injection, the MIA cartilage displayed extensive PG loss (2D), which was not observed in the saline injected contralateral control (2A). Images of sagittal sections from the medial tibial plateau of the tibia generated by EPIC-μCT for MIA samples are also shown [Fig. 2(B and E)].Using the EPIC-μCT technique, the overall tibial cartilage was quantitatively evaluated for thickness, volume, and attenuation. At 3 weeks post injection, cartilage thickness in MIA tibiae decreased by 30% (P < 0.0001) compared to saline contralateral controls [Fig. 3(A)]. Cartilage attenuation in MIA tibiae increased by 18% (P = 0.043) compared to saline contralateral controls [Fig. 3(C)].
Fig. 2Representative images of tibial articular cartilage from saline contralateral control and MIA injected knees at 3 weeks post injection. A–C: Representative saline contralateral control section, Safranin-O stained, 4× (A), corresponding EPIC-μCT thickness map from the same sample (B) and 10× Safranin-O section (C). D–F: Representative MIA injected sections, Safranin-O stained, 4× (D), corresponding EPIC-μCT thickness map from the same sample (E) and 10× Safranin-O section (F).
Following overall cartilage analysis, the medial tibial articular cartilage was evaluated separately. Average cartilage thickness and volume were 31% (P < 0.0001) and 18% (P = 0.0033) lower, respectively, compared to saline contralateral controls [Fig. 3(D and E)]. Average attenuation was 22% (P = 0.0005) higher than saline contralateral controls [Fig. 3(F)]. The decrease in cartilage volume was not observed when assessing the overall tibial plateau.
MMT tibial articular cartilage analysis
Representative images of tibial articular cartilage sections stained with Safranin-O in comparison with EPIC-μCT images are shown [Fig. 4(A,C,D and F)]. The MMT cartilage displayed erosion on the histology image [Fig. 4(D and F)], which was not observed on the contralateral control. The EPIC-μCT image of the MMT tibia also displayed both an erosion and a lesion on the sagittal section of the medial tibial plateau [Fig. 4(E)]. Cartilage histomorphometry measurements similarly showed a significant increase in the thickness of the medial tibial plateau in the MMT tibiae compared to the contralateral controls (Fig. S4).
Fig. 4Representative images of tibial articular cartilage in control and MMT knees at 3 weeks post surgery. A–C: Representative Safranin-O stained control section, 4× (A), corresponding EPIC-μCT thickness map from the same sample (B) and 10× Safranin-O section (C). D–F: Representative Safranin-O stained MMT section, 4× (D), corresponding EPIC-μCT thickness map from the same sample (E) and 10× Safranin-O section (F). Green arrow in D & E indicates fissure site and red arrow in D & E indicates lesion site.
Using the EPIC-μCT technique, the overall tibial cartilage was quantitatively evaluated for thickness, volume, and attenuation for the medial and lateral plateaus combined. The MMT joints did not show any difference in thickness [Fig. 5(A)]. Average volume was significantly higher in the MMT tibiae (26.6%, P < 0.0001) compared to the contralateral controls [Fig. 5(B)]. No significant difference in cartilage attenuation was observed when analyzing the overall tibial cartilage [Fig. 5(C)].
Fig. 5Tibial articular cartilage properties quantified for MMT and contralateral control knees (paired comparisons). A–C: Overall tibial articular cartilage parameters - Cartilage thickness (A); cartilage volume (B); cartilage attenuation (C). D–F: Medial tibial articular cartilage parameters – Cartilage attenuation (D); cartilage thickness (E); cartilage volume (F). At 3 weeks post surgery, overall and medial cartilage volume from MMT joints was significantly higher than contralateral control cartilage. Medial tibial cartilage thickness and attenuation from MMT joints were significantly higher than contralateral controls. * = P < .05, n = 7.
The MMT medial tibial cartilage had significantly higher thickness and volume by 31% (P = 0.0028) and 70% (P = 0.0259) compared to contralateral controls [Fig. 5(D and E)]. In the analyses of cartilage in the full tibial plateau, cartilage attenuation was 25% (P = 0.0003) higher in the MMT medial tibiae than contralateral controls [Fig. 5(E)]. Though no change in average attenuation was observed for cartilage in the full tibial plateau, isolating the medial plateau resulted in detection of a significant difference in articular cartilage attenuation for MMT tibiae compared to controls.
MMT medial 1/3 of medial articular cartilage analysis
Following analysis of the medial tibial plateau, further localized analysis was performed for the medial 1/3 of the medial tibial plateau. Evaluating this region, similar results were observed as with the whole medial plateau. Average cartilage thickness and volume in the MMT tibiae were 50% (P < 0.0001) and 90% (P < 0.0001) higher, respectively, compared to contralateral controls; average cartilage attenuation was 35% (P < 0.0001) higher compared to controls [Fig. 6(A,C)].
Fig. 6Medial 1/3 medial tibial articular cartilage properties quantified for MMT and contralateral control knees (paired comparisons). A: Cartilage thickness; B: cartilage volume; C: cartilage attenuation. Cartilage thickness and volume from MMT surgery tibiae were significantly higher than contralateral control cartilage. Average cartilage attenuation in MMT surgery tibiae was significantly higher than the respective contralateral control tibiae. * = P < .05, n = 7.
The grayscale 2D images from the medial plateau of the MMT animals were viewed slice by slice to locate, visualize, and quantify focal defects. Pseudo-color attenuation maps were generated based on these images. Sagittal sections from the attenuation maps showed higher attenuation on the MMT tibiae near the lesion sites (Fig. 7). In addition to differences in attenuation, surface defects on the cartilage were also observed. Two MMT samples had three lesions and six MMT samples had one lesion each on the medial tibial articular cartilage. Six MMT samples had one erosion and one MMT sample had two erosions on the medial tibial articular cartilage. The average number of lesions was 1.57 ± 0.37 and average number of erosions was 1.14 ± 0.14 for the MMT tibiae, and no defects were observed on the controls. The average lesion volume was 0.0393 ± 0.003 mm3 for MMT tibiae, and no lesions were found on any of the contralateral controls (Fig. 7). The average focal lesion volume was only 7.2% of the average medial 1/3 medial cartilage volume and 2.1% of the average medial plateau cartilage volume. The average focal lesion volume accounted for only 0.13% of the full cartilage volume when medial and lateral plateaus were combined. Because lesion volumes represented such a small region in proportion to the full cartilage volume, localized assessment clearly enhanced detection of morphometric differences between MMT and controls.
Fig. 7Representative EPIC-μCT images of tibial articular cartilage in control and MMT knees at 3 weeks post surgery. A–B: Representative sagittal attenuation maps of medial tibial plateau for control (A) and MMT (B) (Inset shows zoomed in view). Red displays higher attenuation levels and green indicates lower attenuation. Red arrow indicates lesion site in MMT. Higher levels of attenuation were seen surrounding the lesion site for MMT tibiae. C: Focal lesion volume quantification for MMT tibiae.
The study of animal models of joint degeneration is essential for evaluating potential therapeutics and understanding the pathogenesis of cartilage loss. Studying the progression of OA in humans is challenging because of poor evaluative tools and limited access to diseased tissue during early development and through disease progression. Suitable animal models for research are consistent, reproducible, have an appropriate time frame of progression for high-throughput screening and recapitulate human pathology
. Histopathology was the primary analytical method for these previous studies. EPIC-μCT allows for efficient, non-destructive 3D imaging and localized analysis, providing quantitative methods to assess cartilage thickness, attenuation and volume among other variables. For this study, the MIA and MMT models were quantitatively characterized via EPIC-μCT to measure localized variables of attenuation, thickness and volume.
The results from the current low dose MIA study indicated an overall thinning of the cartilage layer. These are consistent with a previous dose response study for MIA, which showed that 0.3 mg was the lowest dose that caused articular cartilage damage covering the entire articular surface
. In another study, Safranin-O staining indicated hypertrophied and disorganized chondrocytes as well as degeneration of the cartilage surface at 3 weeks at the 0.3 mg dose and more extensive damage at a 3 mg dose
. It is to be noted that these previous studies did not assess global and regional degeneration of the cartilage nor did they use quantitative techniques. In the current study, attenuation was significantly increased in the MIA joints, which was indicative of decreased PG content. Similarly, the Safranin-O staining showed severely depleted sGAGs. In regional analyses and the overall tibial plateau, the low-dose MIA model showed reduced cartilage thickness, volume as well as increased attenuation.
The MIA model remains the current standard to assess pain therapeutics, is easy to use and is reproducible. Hence, some studies have tested DMOADs and NSAIDs in this model due to its widespread prevalence
Weight bearing as a measure of disease progression and efficacy of anti-inflammatory compounds in a model of monosodium iodoacetate-induced osteoarthritis.
. Our results showed uniform thinning and PG loss even in this low dose MIA model. This is not a typical presentation of human OA and does not correspond with many of the OARSI recommended OA histopathology grading features. The MIA model represents biochemically induced OA and disease progression is expedited in the low dose model. This can be an issue regarding the suitability of this model to test disease modifying therapeutics for knee OA. Additionally, transcriptional profiling of gene expression has shown little overlap with this model and human OA and the disease process induced by MIA causes rapid cartilage and subchondral bone damage (at <3 weeks). This extensive damage suggests that the time-frame of disease progression is much faster in the MIA model than OA progression in humans
. Thus, the MIA model shows different genetic, compositional and morphological changes than normally observed in human OA. Despite this, the MIA model does have an advantage in detecting joint pain-related behaviors due to the rapid damage caused to the articular cartilage
. It remains the more popular model for testing therapeutics targeted towards OA symptoms but has limited utility for testing DMOADs and as an entry point to the chronic OA cycle.
In the MMT model, overall and regional changes in cartilage parameters were detected. While changes were detectable at the overall level, most were in the medial 1/3 region. This observation that cartilage damage was most extensive in the outer 1/3 was similar to previous reports in the rat MMT model and attributed to the fact that this is the load bearing region that is destabilized
. Though the medial 1/3 of the medial tibial plateau displayed cartilage thickening in the MMT model, focal lesions were also observed. A previous study supported the identification of localized lesions in the MMT model through immunostaining for type II collagen neoepitope antigen (TIINE) antibody
. These lesions had not been previously quantified. Results from the present study showed that the lesion volume accounted for a small percentage (2.1%) of the medial plateau cartilage volume. The lesions compared to the full tibial articular cartilage layer occupied an even smaller fraction of the total volume, making cartilage changes challenging to detect when performing an overall cartilage analysis.
The increase in thickness for the medial plateau contradicts studies that have shown cartilage thickness to be reduced in MMT rats at 3 weeks
. To confirm this thickness increase that was measured via EPIC-μCT, histomorphometric measurements were made and similarly showed a significant increase in cartilage thickness in the medial tibial plateau of the MMT animals compared to contralateral controls (Fig. S4). Other studies have also reported thickening of OA cartilage in humans, beagles and rabbits
. Another study in dogs, reported a transient episode of chondrocyte proliferation and increased matrix production within the lesion site and superficial zone of the adjacent articular cartilage in response to destabilization of joints
In addition to thickening of cartilage, another factor contributing to the increase in thickness/volume could be cartilage-like structures called chondrophytes. Thickening of marginal zone cartilage and generation of chondrophytes and osteophytes has also been reported as a result of destabilization of the joint or injury in many animal models including the MMT model at 3 weeks
Analysis of early changes in the articular cartilage transcriptisome in the rat meniscal tear model of osteoarthritis: pathway comparisons with the rat anterior cruciate transection model and with human osteoarthritic cartilage.
Resemblance of osteophytes in experimental osteoarthritis to transforming growth factor beta-induced osteophytes: limited role of bone morphogenetic protein in early osteoarthritic osteophyte formation.
. A previous study hypothesized that the observed peripheral thickening could be an adaptive or compensatory mechanism to increase articular cartilage surface though no data was presented on PG content
. The cartilage volume from the MMT tibiae in the medial 1/3 region increased by 90% compared to the control tibiae. Formation of these chondrophyte/osteophyte structures could augment measured cartilage volume and thickness because diffusion of contrast agent into the chondrophyte would be expected to be similar to that of articular cartilage. Even though we observed higher volume and thickness, the average attenuation remained higher for the MMT tibiae indicating lower cartilage quality and depleted PG content.
OA is a multifactorial disease and there are several contributing factors to disease progression. Focal lesions on the articular cartilage have been observed in humans with early stage OA, and are considered to be among factors involved in OA progression
. The MMT model displayed focal lesions when regional analysis was performed using EPIC-μCT. While histology can identify lesions, the 2D nature of the technique means that lesion incidence and overall volumes cannot be quantified. EPIC-μCT allowed for both slice-by-slice visualization and volumetric quantification of these lesions. Our ability to further discern the sensitivity of the EPIC-μCT technique to measure more subtle PG changes was limited by the fact that sham surgery was not performed. Contralateral control limbs were used as negative controls as sham surgeries do not develop lesions
, and this additionally allowed us to minimize animal numbers. The MMT model displayed several characteristics similar to OA characteristics in humans such as peripheral thickening, loss of PGs, erosion of cartilage and localized lesions.
Though the objectives of this study were not to directly compare MIA and MMT models, some interesting overall observations exist. The data show widespread cartilage degeneration in the low dose MIA model and localized degradation in the MMT model. The MIA model appears to cause greater damage in full tibial plateau cartilage compared to controls; whereas, this difference at the full tibial plateau level was not detected in the MMT model at 3 weeks. It should be noted that two different strains of rats were used in this study. Based on previous studies, the standard rat strain for the MIA model is Wistar and for the MMT model is Lewis
. Direct comparison of absolute quantitative values are not appropriate and thus, not performed. Although there may be variations in the response of different strains to the two methods used to induce joint degeneration, this limitation does not alter the conclusions of the study. Another limitation of our analysis was that all measurements were performed in the sagittal plane as established previously in the EPIC-μCT protocol, and this limited our ability to perform comparisons to the OARSI histomorphometric standards
. Future work may involve developing a new protocol to analyze coronal sections to compare results to current histological standards.
In summary, this study has presented a powerful technique for quantitative assessment of localized changes associated with OA in two important preclinical models of the disease. The data show that the MMT model caused less extensive global damage than the MIA model at 3 weeks. The resemblance of physical characteristics produced in the rat MMT model to that of human OA progression indicates that the MMT model could be a better predictive tool for examining OA progression. The EPIC-μCT technique facilitated rapid, quantitative assessments of cartilage morphology, cartilage attenuation, lesions and erosions over full rat tibial articular cartilage as well as in delineated regions. It is a powerful technique with the potential to significantly improve preclinical throughput for screening therapeutics and to provide substantially more information than traditional histopathology reports.
Author contributions
TT, NJW, AL, HS, RK, MH and REG contributed to conception and design of the study. Article draft was written by TT and critically revised by all authors. YR performed histology experiments and SM assisted with μCT experiments. NJW, REG, and AL assisted with manuscript editing and preparation. All authors read and approved the final manuscript.
Conflict of Interest
The authors have no conflicts of interest to disclose.
Role of funding source
This research was funded by the National Institute of Health (NIH). The study sponsors were not involved in the study design, data collection or analysis, or in the writing of the manuscript.
Acknowledgements
This research was supported by NIH grant R21AR053716. The authors would like to thank Dr Brani Vidakovic for his input on statistical methods.
Appendix A. Supplementary data
The following are the supplementary data related to this article:
Fig. S1No effect due to fixation with 10% NBF on average volume (P = 0.42), thickness (P = 0.40), attenuation (P = 0.65) or surface area (P = 0.68), compared to fresh samples in 10-week-old male rats. (Data from Wang, Guldberg et al., ORS poster “Effects of Tissue Preservation Methods on EPIC-μCT Measurement of Articular Cartilage Morphology and Composition”, Feb 2009).
Fig. S2Tibial articular cartilage properties quantified for MIA and saline contralateral control knees after incorporation of P-value correction factor. A–C: Overall tibial articular cartilage parameters – Cartilage thickness (A); cartilage volume (B); cartilage attenuation (C). D–F: Medial tibial articular cartilage parameters – Cartilage attenuation (D); cartilage thickness (E); cartilage volume (F). At 3 weeks post injection, overall and medial tibial cartilage thickness from MIA-injected joints was significantly lower than saline contralateral control cartilage. Medial tibial cartilage attenuation from MIA-injected joints was significantly higher than saline contralateral controls. Medial tibial cartilage volume from MIA-injected joints was significantly lower than saline contralateral controls. Red box indicates data that was previously significant but with incorporation of P-value correction, is not significantly different. * = P < 0.0167, n = 8.
Fig. S3Tibial articular cartilage properties quantified for MMT and contralateral control knees after incorporation of P-value correction factor. A–C: Overall tibial articular cartilage parameters – Cartilage thickness (A); cartilage volume (B); cartilage attenuation (C). D–F: Medial tibial articular cartilage parameters - Cartilage attenuation (D); cartilage thickness (E); cartilage volume (F). At 3 weeks post surgery, overall cartilage volume from MMT joints was significantly higher than contralateral control cartilage. Medial tibial cartilage thickness and attenuation from MMT joints were significantly higher than contralateral controls. Red box indicates data that was previously significant but with incorporation of P-value correction, is not significantly different. * = P < 0.0167, n = 7.
Fig. S4Histology thickness measurements for medial tibial plateau. A–B: H & E stained sagittal sections of medial tibial plateau from contralateral control knee (A) and MMT knee (B). Insets show zoomed in region at 4× indicated by black box C: Histomorphometry measurements showing a significant increase in medial tibial cartilage thickness in MMT knees compared to contralateral control. Red lines on A & B indicate thickness measurements. 5 measurements were taken per slice and each sample had 3 sagittal slices. Individual measurements from each slice were averaged and then measurements for all 3 slices were averaged together to calculate medial tibial cartilage thickness (n = 4, *** = P < .001).
A new class of potent matrix metalloproteinase 13 inhibitors for potential treatment of osteoarthritis: evidence of histologic and clinical efficacy without musculoskeletal toxicity in rat models.
Discovery and characterization of a novel inhibitor of matrix metalloprotease-13 that reduces cartilage damage in vivo without joint fibroplasia side effects.
Biochemical and metabolic abnormalities in articular cartilage from osteo-arthritic human hips. II. Correlation of morphology with biochemical and metabolic data.
Weight bearing as a measure of disease progression and efficacy of anti-inflammatory compounds in a model of monosodium iodoacetate-induced osteoarthritis.
Analysis of early changes in the articular cartilage transcriptisome in the rat meniscal tear model of osteoarthritis: pathway comparisons with the rat anterior cruciate transection model and with human osteoarthritic cartilage.
Resemblance of osteophytes in experimental osteoarthritis to transforming growth factor beta-induced osteophytes: limited role of bone morphogenetic protein in early osteoarthritic osteophyte formation.
00833-9&id=gr1.jpg){kind=link}
00833-9&id=gr2.jpg){kind=link}
00833-9&id=gr3.jpg){kind=link}
00833-9&id=gr4.jpg){kind=link}
00833-9&id=gr5.jpg){kind=link}
00833-9&id=gr6.jpg){kind=link}
00833-9&id=gr7.jpg){kind=link}
00833-9&id=figs1.jpg){kind=link}
00833-9&id=figs2.jpg){kind=link}
00833-9&id=figs3.jpg){kind=link}
00833-9&id=figs4.jpg){kind=link}