Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance

Abstract

Objectives

Osteoarthritis is a painful joint disease responsible for walking impairment. Its quantitative assessment by gait analysis in mice may be a relevant and noninvasive strategy to assess the disease severity. In this study, we aimed to determine the severity of osteoarthritis at the tissular and gait levels in unilateral and bilateral posttraumatic murine osteoarthritis.

Methods

Twenty-four C57BL/6 male mice were randomly assigned to 3 groups (n = 8/group): controls, unilateral surgery, and bilateral surgery. Posttraumatic osteoarthritis was induced unilaterally or bilaterally by destabilization of the medial meniscus. Gait analysis was performed weekly with the CatWalk^TM XT system until the 16th week after surgery. After animal sacrifices, histological and micro–computed tomographic assessment was performed.

Results

Operated knees showed a significant increase in the histological score compared with controls (P < 0.001). Calcified anterior medial meniscal bone volume was higher on the ipsilateral side after unilateral destabilization of the medial meniscus (P < 0.001) and on both sides after bilateral intervention (P < 0.01). One week after surgery, the mice mean speed decreased significantly in both operated groups (P < 0.001 and P < 0.05). In the unilateral group, a significant increase in the contralateral hind print area appeared from week 4 to week 16.

Conclusions

While bilateral destabilization of the medial meniscus induced no detectable gait modification except 1 week after surgery, unilateral model was responsible for a gait disturbance on the contralateral side. Further studies are needed to better define the place of the CatWalk^TM in the evaluation of mouse models of osteoarthritis.

Keywords

posttraumatic osteoarthritis murine model gait analysis destabilization of the medial meniscus

Introduction

Osteoarthritis (OA), the most prevalent joint disease, is a multifaceted inflammatory and degenerative disease that affects 10% of men and 18% of women aged above 60 years¹ and constitutes a worldwide leading cause of pain and disability.² OA is characterized by cartilage degradation, subchondral bone remodeling, osteophyte development, and synovial inflammation.³ The knee is one of the most frequently affected joints, making knee OA responsible for more than 80% of the disease’s total burden.^3,4 Knee OA implies walking disability due to pain and joint dysfunction (e.g., loss of range of motion) during loading and joint motion.^4,5 Human OA development is usually slow over many years and variable from one subject to another, leading to the difficulty of longitudinal studies. To overcome this time-related issue, animal models that generate OA in a few weeks are useful.⁶ While numerous mouse models are used to assess the functional consequences of OA,^7-9 very few faithfully reproduce the human OA pathogenic mechanisms that dampen their pathomimicry level and limit the relevance of their human translatability. Indeed, chemical models with intra-articular agent injection, such as collagenase or sodium mono-iodoacetate, provide painful and fast end-stage OA. They are particularly used to study anti-inflammatory properties and efficiency of analgesic drugs.^10,11 Some specific mouse strains are also used to investigate OA development,⁶ for example, STR/ort mice develop OA spontaneously early in life and C57BL/6 are predisposed to age-related OA.^12,13 Surgically induced models, such as the destabilization of the medial meniscus (DMM) or anterior cruciate ligament transection (ACLT), generate knee joint instability responsible for progressive articular degeneration mimicking human posttraumatic OA.¹⁴ Although in human knee OA both knees are involved in more than 75% of the cases,¹⁵ bilateral mouse models of OA seem to have been rarely used.¹⁶

Walking disturbances are one of the main consequences of pain associated with OA. These walking disorders are responsible for patient’s decreased mobility and physical activities, leading to restriction of participation.¹⁷ As one of the main symptoms of knee OA is walking disturbance, this parameter should be taken into account in the evaluation of OA in animal models or preclinical tests of disease-modifying OA drugs (DMOADs). Indeed, walking assessment can be an accurate predictive tool to evaluate the potential effectiveness of a therapy in humans, using common clinical criteria rather than radio-histological criteria, because symptoms do not often correlate to the degree of the disease progression on imaging.¹⁸ To assess rodent walking behavior, several gait analysis systems have been developed using either freely walking methods or treadmill systems. Treadmills exhibit the advantage of controlling velocity to reduce speed variability between mice, but they prevent natural gait and the ability for the animal to adapt walking velocity.¹⁹ On the contrary, freely walking systems allow assessment of natural gait, including natural speed velocity which may be modified with pain. Freely walking systems appear relevant to assess OA consequences on natural gait behavior.¹⁹ Gait analysis systems may provide a noninvasive method of joint function assessment in murine posttraumatic OA, which could help scientists evaluate in vivo knee function and preclinical efficiency of DMOAD, with criteria closer to those used in human clinical practice.

In this study, we aimed to measure, in parallel analyses, gait disturbances, tissular changes, and micro–computed tomodensitometric alterations of knee joints in murine posttraumatic OA models, with the ultimate goal of determining whether gait disturbance could be a noninvasive and longitudinal method to assess posttraumatic OA. To answer this question, we have evaluated gait disturbance in mice after unilateral and bilateral DMM as a model of posttraumatic OA, using the CatWalk^TM XT system. This system allows the evaluation of free-moving and weightbearing walking. To confirm the OA onset in unilateral and bilateral DMM in mice, we have also evaluated histological and micro–computed tomography (µCT) joint changes in both models.

Methods

Animals

All the animal experiments were performed according to the European directive 2010/63/EU governing ethic and regulations in animal experiments. These experiments have been approved under the project N°2015033111162660. Mice were housed in the animal facility of Nantes Faculty of Medicine (Agreement No. C44015). Our present work adheres to the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines 2.0²⁰ (Table S1).

Surgical Induction of Posttraumatic OA

Twenty-four 10-week-old male C57BL/6 mice, weighing 25.3 ± 1.3 g, were obtained from Janvier Labs (Saint-Berthevin, France). All animals had reached full skeletal maturity at the time of the study. After 1 week of acclimatization, mice were randomly assigned to 3 groups (n = 8 per group): a control group with nonoperated animals, a group with a unilateral DMM (Uni DMM), and a group with a bilateral DMM (Bi DMM). DMM surgery was performed on anesthetized mice according to previously published description.^21,22 Briefly, a 3-mm longitudinal incision from the patella to the tibial plateau was performed at the right knee (Uni DMM group) or both knees (Bi DMM group). Then, the joint capsule was incised medially near the patella. The patella was dislocated laterally and the medial menisco-tibial ligament was sectioned with #11 blade. After surgery, mice were monitored daily for the first week and weekly at the time of gait analysis for signs of physical distress.

Gait Analysis

Gait analysis was performed using the CatWalk^TM XT gait analysis system (Noldus Information Technology, The Netherlands).²³ Before surgery and gait recording, mice were pretrained to cross the 42-cm walkway for 2 weeks.⁷ Animals were individually allowed to walk freely on an 8-cm-wide corridor walkway with a glass floor in a dark compartment. Gait was recorded according to a video-based system. Upon contact with the glass plate, the paw reflected the green light to a video camera below (100 Hz), which recorded the entire run. Then, gait was recorded for each group before surgery (baseline) and every week for 16 weeks after OA induction. Gait analysis experiments were conducted blindly by the same investigator. The scheme of the study is illustrated in Figure 1 . The hind and the front paws were assessed for data analysis. Gait parameters were acquired with the CatWalk^TM XT according to the method previously described by our lab.²⁴ Briefly, the criteria for compliant runs were a run duration between 1 and 5 seconds and a maximum speed variation of 80%. Runs with backing, sniffing, cleaning, or long stops were also discarded during the acquisition phase by the investigator. Five compliant runs were necessary for statistical analysis. In most cases, the first 5 or 6 runs produced the 5 compliant runs. We studied the following parameters based on their frequent use in rodent gait analysis recommendations^19,25: velocity, paw print area, paw print intensity, stance phase, swing phase, swing speed, and duty cycle (the stance phase as a percentage of the step cycle: stance / [stance + swing] × 100).

Figure 1.

Experimental scheme of gait analysis and articular assessment. After 1 week of acclimatization, mice were randomly assigned to 3 groups: controls, unilateral DMM, and bilateral DMM. Mice were pretrained to cross the walkway for 2 weeks. Then, DMM surgery was performed. After 1 week of recovery, gait was recorded weekly for 16 weeks. At the end of the protocol, mice were euthanized and both knee joints were analyzed with micro–computed tomography (µCT) followed by histological scoring. DMM = destabilization of the medial meniscus.

µCT Assessment

At the end of the gait analysis protocol, mice were euthanized and both knee joints were analyzed with µCT according to Bouxsein et al.’s²⁶ guidelines. Joints were fixed during 48 hours in a 4% solution of paraformaldehyde and then placed in 70° ethanol. Joints were scanned with a Micro Computer Tomograph Skyscan 1272^® (Bruker, Germany). The µCT acquisitions have been performed at a resolution of 5 µm without filters. For the acquisitions, we used a voltage of 50 kV and an intensity of 200 µA, with a rotation angle of 0.45°, an exposure time of 620 ms and a frame averaging 2, and a mean scan duration of 17 minutes 50 seconds. The images obtained during the acquisition were processed with NRecon® software (Bruker) to obtain a 3-dimensional reconstruction of the different samples. The reconstructions of each sample were then spatially oriented identically, using Dataviewer^® software (Bruker), so that they could be compared to each other. In addition, the Dataviewer^® software makes it possible to obtain images of the samples in the sagittal, coronal, and transverse planes, which constitute the starting point for the selection of the area of interest. Images allowing an overview of each sample were produced with the CTVox^® software (Bruker). Then, the final selection of the area of interest was carried out using the CTAnalyser® software (Bruker) (Fig. S1). According to Shin et al.,²⁷ we have focused on the subchondral bone of the tibial epiphysis and the anterior medial meniscus area to assess bone volume (mm³), subchondral bone thickness (mm), and trabecular bone thickness (mm). The anterior medial meniscus has been isolated after 3 different successive stages (Fig. S2) on a coronal plane by 2 different operators (M.M. and B.M.): (1) exclusion of the femur, the tibia, and the patella; (2) selection of the remaining medial structures; and (3) isolation of the anterior part of the medial meniscus.

Histology

After µCT assessment, knee joints were dissected, skin and muscle were removed, and knee joints were placed in 0.5 M (pH 7.4) EDTA solution at 4 °C for 2 weeks for decalcification. Then, samples were dehydrated in a graded series from 70° to 100° of ethanol solution before paraffin embedding. Paraffin sections of 5 µm were stained with safranin O-fast green with an automaton, Varistain Gemini S (Thermo scientific, USA). The severity of OA was evaluated using a scoring system adapted from the Osteoarthritis Research Society International (OARSI) as previously described ( Table 1 ).²² The 4 quadrants of the joints, the medial femoral condyle, the lateral femoral condyle, the medial tibial plateau, and the lateral tibial plateau, were analyzed, and scores were assessed by calculating the total sum of the quadrants. Scoring was performed blindly and independently by 2 different trained readers (A.F.C. and B.M.) on at least 2 or 3 levels, from the anterior area to the mid-weightbearing region of the joint for each sample to obtain enough interpretable quadrants.

Table 1.

Adapted OARSI Grading System Used to Evaluate OA Severity.

Chondrocyte death	None	0
	≤10%	1
	≤25%	2
	≤50%	3
	≤75%	4
	≤100%	5
Hypertrophy	None	0 1
Hypertrophy	Presence	0 1
Clusters	None	0
	Superficial zone	1
	Middle zone	2
	Complex structures	3
	Mild layer cyst formation	4
Loss of safranin-O staining	None	0
	Superficial zone of noncalcified cartilage	1
	Middle zone of noncalcified cartilage	2
	Noncalcified cartilage and intense staining around clusters	3
	Absent with staining around clusters	4
	Absent	5
Surface	Intact	0
	Superficial fibrillation or abrasion	1
	Deep fibrillation	2
	Vertical fissures	3
	Delamination / excavation	4
Bone	None	0
	Denudation	4
	Microfracture	5
	Remodeling	6
Total		25

Adapted from Mével et al.²²

OARSI = Osteoarthritis Research Society International; OA = Osteoarthritis.

Statistical Analysis

Statistical analyses were performed with the SPSS 23.0^® software (Chicago, IL). Results are expressed as mean (SEM). We used repeated-measures analysis of variance (ANOVA) followed by Bonferroni corrections to compare longitudinal changes in each gait parameter within each group. We also calculated the area under the curve for the parameter “paw print area” in each group. To compare histological and µCT scores among groups, Kruskal-Wallis tests followed by Dunn tests were used. The significant difference was determined at P < 0.05. Graphics and figures were performed with GraphPad Prism 7.0 ( GraphPad Software, San Diego, CA).

Results

Mice recovered quickly after surgery and healed in less than a week, which enabled us to start the gait analysis protocol at 1 week after surgery for each group. All groups were comparable concerning body weight (controls = 24.88 ± 1.55 g; unilateral DMM = 25.38 ± 1.19 g; bilateral DMM = 25.75 ± 1.28 g; P = 0.31).

Histology

To assess OA following unilateral and bilateral DMM, we performed histological analyses of knee joints 16 weeks after surgery. Our data ( Fig. 2A ) revealed articular cartilage degenerative lesions at the right knee of the unilateral DMM group and both knees of the bilateral DMM group ( Fig. 2A ). Moreover, meniscal morphological modifications were also observed in both DMM groups, with thickening of the meniscus compared with the control group which exhibited a thin and homogeneous meniscus. Histological scoring showed that the right knee of the mice with unilateral DMM and both knees of the mice with bilateral DMM exhibited a significant increase in the OARSI score compared with the control group (P < 0.001 and P < 0.05, respectively) ( Fig. 2B ). No significant difference was found in the OARSI score between the right knee in the unilateral group and both knees of the bilateral DMM group (P = 0.09). Similar degenerative changes were found in the case of DMM either in bilateral or in unilateral DMM. No difference was found between the control group and the left side of the unilateral DMM group.

Figure 2.

Histological analyses of knee joints after surgical OA induction: (A) Histological sections in the right and left medial compartments of control, unilateral DMM (UNI DMM), and bilateral DMM (BI DMM) operated knees at 16 weeks. Safranin-O and Fast-Green staining. Arrows point to cartilage degenerative lesions observed on operated sides in the unilateral and bilateral DMM groups. Stars (*) show thickening of the meniscus in both knees in bilateral DMM and in the right knee in unilateral DMM. Circles (°) exhibit a thin and homogeneous meniscus in the control group and in the left knee of the DMM group. DMM = destabilization of the medial meniscus; MM = medial meniscus; MFC = medial femoral condyle; MTP = medial tibial plateau. (B) Bilateral adapted OARSI scores according to experimental groups. Data are shown as mean (SEM). Comparison between controls and unilateral DMM: ***P < 0.001 (operated side in unilateral DMM = right side), comparison between controls and bilateral DMM: °P < 0.05. R = right knee; L = left knee; OARSI = Osteoarthritis Research Society International.

µCT Assessment

To further document the joint tissue damage induced by the unilateral and bilateral DMM, we performed µCT analyses on subchondral bone. Three-dimensional images of coronal sections of the joints showed morphological modification of right meniscus in the unilateral DMM ( Fig. 3A ) and both menisci in bilateral DMM ( Fig. 3A ). Calcified medial meniscal volume was higher in DMM, on the right side for unilateral DMM (P < 0.001) and on both sides for bilateral DMM (P < 0.01) ( Fig. 3A and 3B ). Subchondral bone thickness was significantly higher on the operated side of mice with unilateral DMM (P < 0.05). However, no difference was found in the subchondral bone thickness and trabecular bone thickness ( Fig. 3A and 3B ) in mice with bilateral DMM.

Figure 3.

Micro–computed tomography (µCT) analyses of knee joints after surgical OA induction: (A) Representative 3-dimensional µCT images of coronal sections of the right and left knee joints of controls, unilateral DMM (UNI DMM), and bilateral DMM (BI DMM) at 16 weeks. Arrows (→) show meniscal morphological modifications of right meniscus in the unilateral DMM and right and left menisci in the bilateral DMM. OA = osteoarthritis; DMM = destabilization of the medial meniscus. (B) µCT assessment of the calcified medial meniscal volume, subchondral bone thickness, and trabecular bone thickness according to experimental groups. Data are shown as mean (SEM). Comparison between controls and unilateral DMM: *P < 0.05, ***P < 0.001(operated side in unilateral DMM = right side), comparison between controls and bilateral DMM: °°P < 0.01. R = right knee; L = left knee.

Gait Analysis

After having confirmed that unilateral and bilateral DMM-induced knee joint damage attributed to the development of OA, we sought to compare the walking disturbance of controls and OA-affected mice. The mean speeds (velocity) are significantly inferior to the control group in both unilateral and bilateral DMM groups in the first week after surgery (P < 0.001 and P < 0.05, respectively) and recovered similar values to the control group as early as the second week until the 16th week ( Fig. 4A ). No statistical difference in the stance phase duration, swing phase duration, and swing speed was found between controls and unilateral or bilateral DMM, on both the right and the left side and on both the front and hind paws (Fig. S3). The first week after surgery, the duty cycle was significantly increased on the right side of the hind paws only for the bilateral DMM group (P < 0.001) and on the left side of the hind paws for both DMM groups (P < 0.001) ( Fig. 4B and 4C ). On the front paws, the duty cycle was only increased on the right side of unilateral DMM (P < 0.001) ( Fig. 4D and 4E ). The right paw print area was not significantly different between the control group and the 2 DMM groups, on both the front and hind paws ( Fig. 4F and 4H ). In the unilateral DMM group, a significant increase in the left hind paw print area appeared 1 week after surgery compared with controls (P < 0.01); this significant increase was lost during weeks 2 and 3, came back at week 4 after DMM, and was maintained until week 16 ( Fig. 4G ). This significant increase in the left hind paw print area of the unilateral DMM group was also present from the first week after surgery to the end of the study, compared with the right side (Fig. S4). Concerning this parameter, no difference was found between the right and the left side, both in the control and in the bilateral DMM groups. Moreover, the other parameters were also symmetrical between the right and left side in controls and bilateral DMM (data not shown). The area under the curves of the unilateral DMM group concerning the left paw print area was significantly increased compared with controls and unilateral DMM (P < 0.001 and P < 0.001, respectively) (Table 2). Pressure intensity was inconstantly variable between the hind paws of the operated side of the unilateral DMM and controls (Fig. S3).

Figure 4.

Gait parameter changes in mice following surgical OA induction. Velocity (A), bilateral duty cycle (B, C, D, E), and bilateral print area (F, G, H, I) on the front and hind paws were recorded for each group before surgery (baseline: 0) and weekly for 16 weeks after DMM. DMM = destabilization of the medial meniscus. Comparison between bilateral DMM and controls: °P < 0.05, comparison between unilateral DMM and controls: *P < 0.05, **P < 0.01, ***P < 0.001.

Table 2.

Comparison of the Areas Under the Curves Between Groups for the Hind Paw Print Areas (One-Way ANOVA Followed by a Bonferroni Post Hoc Test).

	Area Under the Curve (cm²/week)	95% CI	P Value
Right hind paws
Controls	6.14	6.10–6.20	0.40
Unilateral DMM	6.23	6.20–6.25
Bilateral DMM	6.28	6.20–6.26
Left hind paws
Controls	6.30^{a ***}	6.27–6.33	< 0.0001
Unilateral DMM	7.48^a*,b*	7.44–7.52
Bilateral DMM	6.48^b***	6.45–6.51

ANOVA = analysis of variance; CI = confidence interval; DMM = destabilization of the medial meniscus.

No statistical difference was found between controls and bilateral DMM.

Significant difference between areas under the curve of the left hind paws of controls and the left hind paws of unilateral DMM (nonoperated side).

Significant difference between areas under the curve of the left hind paws of bilateral DMM and the left hind paws of unilateral DMM (nonoperated side).Bonferroni post hoc test: ***P < 0.001.

Discussion

OA is the most prevalent degenerative joint disease in which cardinal symptoms are joint pain and functional disability.^1,2 Therefore, OA strongly impacts patients’ mobility in daily living. While animal models of OA have been widely used to study the pathophysiology and progression of joint damage, only very few studies have focused their attention on the OA-associated gait disturbance. However, in preclinical studies, it appears to be of major interest to assess functional parameters that are common criteria to those used to judge the effectiveness of a treatment in humans. This study aimed to evaluate in a parallel manner the consequences of a unilateral and a bilateral medial meniscus destabilization on gait parameters and tissular changes in mice. Indeed, while in human pathology both sides are often affected by OA in about 80% of cases,^16,28 few studies have investigated the bilateral model of OA in mice.^16,25 Therefore, to reproduce more accurately posttraumatic human OA pathogenesis, we have chosen to use a surgical OA induction model, such as DMM, ACLT, or medial meniscus transection (MMT), which is known to induce a slower progressive OA than chemical models.^21,29,30

Histological analyses showed a significant increase in the OARSI score in the right knee of the unilateral DMM group and the right and left knees of the bilateral DMM group. These results are consistent with previous studies showing a significant increase in the OARSI score in DMM.^21,29,31 Interestingly, the OARSI score appears similar in both groups, and bilateral DMM did not increase the OARSI score, indicating that cartilage degradation and OA are not aggravated when both joints are affected. As expected from literature findings,²⁷ unilateral and bilateral DMM exhibited higher calcified meniscal volume in the operated knees compared with controls or nonoperated sides. Surprisingly, we found an increase in the subchondral bone thickness only in unilateral DMM, whereas no difference was noted in bilateral DMM. Moreover, in both models, no significant difference was measured concerning trabecular bone thickness. This is particularly astonishing because DMM mice could more easily favor the nonoperated knee when walking, and that in human knee OA no obvious radiological difference between the bilateral and unilateral forms seems to exist.²⁸

Using gait analysis, we wanted to assess gait disturbances that could be induced by unilateral or bilateral DMM. Gait analysis allows the assessment of walking impairment, which is one of the main symptoms in human OA and a major therapeutic issue. This technique also allows to perform a longitudinal follow-up with repeated measurements over time and without sacrificing animals at each time point. However, results should be taken with caution because investigated gait disturbance parameters could vary from one study to another. For example, velocity is not always assessed in rodent models because some systems such as DigiGait^TM (Mouse Specifics, Cambridge, MA) or Treadscan^TM (CleverSys, Reston, VA) use treadmills with constant speed to get rid of speed variation.^19,25 These systems may be of interest to eliminate aberrant results and variability due to mouse behavior issues. However, setting apart speed variation may also constitute a limit because natural velocity is not taken into account and may induce gait variations due to inappropriate speed. Therefore, walkways such as CatWalk^TM XT or GaitScan^TM allowing free walking appear as interesting tools to perform gait analysis. Moreover, a comparison between CatWalk^TM XT and DigiGait^TM seemed to be in favor of the use of CatWalk^TM XT, due to its technical advantages in data analysis with high-quality images.³² In our study, speed variations appeared only during the first week after surgery, and that could be explained by postoperative phenomena such as knee pain or knee swelling despite an apparent skin healing. This speed parameter is not systematically taken into account, notably because of the disparate data from one study to another. On the one hand, some studies on ACLT rat models found no speed variation in a daily analysis during a short period after surgery or a monthly analysis performed up to 6 months.^30,33 On the other hand, in other studies using DMM models and CatWalk^TM XT gait analysis, the velocity was not provided.^8,29,31,34 In a systemic mouse model of OA with a collagen IX deficiency, using a custom gait arena, Allen et al.³⁵ observed a decrease in the velocity compared with controls. At only 1 week after surgery, we have noted an increase in the hind paw duty cycle on the DMM-operated side and on the ipsilateral front paw in the unilateral group and on both sides at the hind paws in the bilateral group, which implies a relative prolonged stance phase and/or a reduced swing phase for each walking cycle, even if these 2 parameters are not significantly modified taken independently. Thus, it may be coherently associated with the reduction in speed during this same period. Yet, results regarding the duty cycle seemed variable among the previous studies. Indeed, using CatWalk^TM XT, Ferland et al.³⁰ found no duty cycle modification in their ACLT group during the 26 days of analysis, whereas Muramatsu et al.²⁹ and Zhu et al.³⁶ observed a decrease in the duty cycle ratio at different time points after DMM induction in mice. Moreover, Kloefkorn et al.,³⁷ with a custom gait arena, found no duty cycle modification in the first 6 weeks after OA induction by MMT in rats. In the present work, we noted no difference regarding the swing and stance phases between the 3 groups. Once again, both parameters are variably modified according to the studies. On the one hand, Ferland et al.³⁰ exhibited no modification of the swing phase without mentioning the stance phase, and on the other hand, Muramatsu et al.²⁹ reported shortening of the stance phase and elongation of the swing phase. Allen et al. found a reduction of the ipsilateral stance phase, 23 days after surgery in MMT mice, but did not mention results concerning the swing phase.³⁷ In our unilateral DMM group, we showed an increase in the footprint area of the nonsurgical side, whereas bilateral DMM mice presented similar footprint areas to control mice from the beginning to the end of the study. Poulet et al.³⁸ reported similar results in a mechanical loading model of OA, where they found only contralateral gait modifications, especially an increase in the footprint area. This is of particular interest because they obtained similar results with a different OA model and the use of a treadmill system of gait analysis, suggesting that contralateral footprint areas may be a gait parameter of interest to study in unilateral OA. However, these findings may be astonishing especially because no other modification was found on the contralateral side and the front paws. This may result from a lack of power in the study due to the low number of mice, which do not allow to show a modification of the operated side compared with controls. However, the intragroup analysis (Fig. S4) showed a difference between the 2 legs in the unilateral DMM group, from the first week to the end of the study. Moreover, on an observational point of view (Fig. S4), delta between left and right paw print area seemed to globally increase until week 7 and then seemed stabilized. We can assume that this absence of obvious changes from the seventh week (week 14 excepted) may result from adaptation phenomena despite an increase in OA severity. Unfortunately, this contralateral parameter is not always mentioned in OA surgical models using CatWalk^TM XT for gait analysis.^25,30,33 In this study, the bilateral DMM group exhibited fewer gait modifications than the unilateral DMM group. This closer gait behavior observed between the control and bilateral group could be due to an inability to compensate with a healthy side, leading to a more symmetrical walking in the bilateral DMM group. This observation may be correlated to the increase in subchondral bone thickness, which was observed only in unilateral DMM, whereas no difference was noted in bilateral DMM. In human gait analysis, Creaby et al.³⁹ showed that in patients with knee OA, those with unilateral knee pain and either unilateral or bilateral structural medial knee OA demonstrate disturbances in knee biomechanics while walking. Interestingly, patients with bilateral pain and structural disease exhibited symmetrical walking, close to healthy controls. The conclusion was that gait disturbance was associated with the presence of unilateral pain and not with structural changes. This strongly suggests that pain in OA mice likely deserves to be assessed in parallel to gait modification to determine whether both parameters may be correlated.

According to our results and those of the literature, gait analysis systems allow highlighting some gait disturbances in surgically induced OA animal models. However, as already mentioned, high variability and heterogeneity between studies are observed.^{8,25,29,34,37} Hypotheses may be drawn to explain such variability. First, results may depend on the animal model chosen; mice and rats may have a different behavior after a similar intervention. Second, while DMM, ACLT, or MMT seems to provide comparable histological and radiological OA lesions, we cannot rule out the possibility that rodents present different gait disturbance. Third, the different types of gait analysis systems used appear as a potential issue to compare results from different studies. Indeed, we have already noticed that a free walking system would not provide the same results as a treadmill system which imposed the speed.³² Besides, the reproducibility between 2 different systems using both treadmills appears poor in murine model.⁴⁰ Finally, variability in results could also be the consequence of the gait analysis protocol. Although some studies use similar murine models and identical gait analysis systems, the delay and the frequency of analyses may vary from a daily analysis for a short period to a weekly or monthly analysis of up to 6 months after surgery. These variabilities in the timeline of analysis may also have an impact on animal behavior due to differences in gait training. Moreover, the progression of the OA is compared at a different time of development.^29-31,33 From our point of view, due to the problems of comparability, studies should provide control groups with standardized gait analysis protocol to limit misinterpretations.

Our study has several limitations. First, we did not use the sham surgery group, which might have been interesting to explain the gait modifications observed in the first week after surgery. Second, CatWalk can provide a lot of parameters, more than 30 per paw, and we have chosen to restrain our analyses to 7 of them, so theoretically, we may have missed significant changes in mice. Yet, we have chosen the most commonly used and recommended CatWalk parameters²⁵ to allow us comparisons with previous studies. Third, it has previously been shown that some of the gait modifications observed in mice could be related to pain. This parameter, not assessed in this study, deserves to be further investigated in the pain-related model of murine OA. Fourth, our results showed a highly significant increase in the speed in the control group, and then in all groups over the 16 weeks of the protocol. Although we had performed a 2-week training period for the 3 groups before the beginning of the walking assessment, we cannot exclude a learning effect in our mice. We could therefore discuss the interest of a longer training period to better prevent this effect. Finally, our findings may also be limited due to the relatively low number of mice, which may have been responsible for a lack of power. However, considering the logistical complexity of CatWalk analysis, an increase in the number of animals would have made the study complicated to perform in practice.

This study shows that both the unilateral and the bilateral DMM mouse models provided histological and radiological OA lesions. Bilateral DMM exhibited no gait modification excepted at an early time point after surgery; it could be due either to a better gait adaptation in case of symmetrical OA or to an impossibility to privilege an unpainful side. Unilateral DMM was responsible for a gait disturbance, mainly on the contralateral side, and restricted to few parameters. These changes were relatively stable according to time, which prevents the longitudinal assessment of unilateral OA with gait analysis. Further studies are needed to better define the place of the CatWalk in the evaluation of mouse models of OA.

Supplemental Material

sj-pdf-5-car-10.1177_19476035211053821 – Supplemental material for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance

Supplemental material, sj-pdf-5-car-10.1177_19476035211053821 for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance by Alban Fouasson-Chailloux, Marc Dauty, Benoit Bodic, Martial Masson, Yves Maugars, Benoit Metayer, Joëlle Veziers, Julie Lesoeur, François Rannou, Jérôme Guicheux and Claire Vinatier in CARTILAGE

Supplemental Material

sj-pptx-1-car-10.1177_19476035211053821 – Supplemental material for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance

Supplemental material, sj-pptx-1-car-10.1177_19476035211053821 for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance by Alban Fouasson-Chailloux, Marc Dauty, Benoit Bodic, Martial Masson, Yves Maugars, Benoit Metayer, Joëlle Veziers, Julie Lesoeur, François Rannou, Jérôme Guicheux and Claire Vinatier in CARTILAGE

Supplemental Material

sj-pptx-2-car-10.1177_19476035211053821 – Supplemental material for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance

Supplemental material, sj-pptx-2-car-10.1177_19476035211053821 for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance by Alban Fouasson-Chailloux, Marc Dauty, Benoit Bodic, Martial Masson, Yves Maugars, Benoit Metayer, Joëlle Veziers, Julie Lesoeur, François Rannou, Jérôme Guicheux and Claire Vinatier in CARTILAGE

Supplemental Material

sj-pptx-3-car-10.1177_19476035211053821 – Supplemental material for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance

Supplemental material, sj-pptx-3-car-10.1177_19476035211053821 for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance by Alban Fouasson-Chailloux, Marc Dauty, Benoit Bodic, Martial Masson, Yves Maugars, Benoit Metayer, Joëlle Veziers, Julie Lesoeur, François Rannou, Jérôme Guicheux and Claire Vinatier in CARTILAGE

Supplemental Material

sj-pptx-4-car-10.1177_19476035211053821 – Supplemental material for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance

Supplemental material, sj-pptx-4-car-10.1177_19476035211053821 for Posttraumatic Osteoarthritis Damage in Mice: From Histological and Micro–Computed Tomodensitometric Changes to Gait Disturbance by Alban Fouasson-Chailloux, Marc Dauty, Benoit Bodic, Martial Masson, Yves Maugars, Benoit Metayer, Joëlle Veziers, Julie Lesoeur, François Rannou, Jérôme Guicheux and Claire Vinatier in CARTILAGE

Footnotes

Supplemental Material

Supplementary material for this article is available on the Cartilage website at .

Acknowledgments and Funding

The authors thank the SFR Bonamy animal facility “UTE Medecine” Nantes University. The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the French network “Research on OsteoArthritis Disease” (ROAD network), the ANR projects KLOTHOA and PPAROA, and “Agence de la Biomedecine” project 18GREFFE012.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval

All the animal experiments were performed according to the European directive 2010/63/EU governing ethic and regulations in animal experiments. These experiments have been approved under the project N°2015033111162660. This work adheres to the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines 2.0.

ORCID iD

Alban Fouasson-Chailloux

References

Glyn-Jones

Palmer

AJR

Agricola

Price

Vincent

Weinans

, et al. Osteoarthritis. Lancet. 2015 Jul 25;386(9991):376-87.

Wallace

Worthington

Felson

Jurmain

Wren

Maijanen

, et al. Knee osteoarthritis has doubled in prevalence since the mid-20th century. Proc Natl Acad Sci USA. 2017 Aug 29;114(35):9332-6.

Hunter

Bierma-Zeinstra

. Osteoarthritis. Lancet. 2019 Apr 27;393(10182):1745-59.

Lawrence

Felson

Helmick

Arnold

Choi

Deyo

, et al. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part II. Arthritis Rheum. 2008 Jan;58(1):26-35.

Creamer

Lethbridge-Cejku

Hochberg

. Where does it hurt? Pain localization in osteoarthritis of the knee. Osteoarthr Cartil. 1998 Sep;6(5):318-23.

McCoy

. Animal models of osteoarthritis: comparisons and key considerations. Vet Pathol. 2015 Sep 1;52(5):803-18.

Masocha

Parvathy

Pavarthy

. Assessment of weight bearing changes and pharmacological antinociception in mice with LPS-induced monoarthritis using the Catwalk gait analysis system. Life Sci. 2009 Sep 9;85(11-12):462-9.

Kuang

Chen

Yang

Zeng

W-N

, et al. miR-100-5p-abundant exosomes derived from infrapatellar fat pad MSCs protect articular cartilage and ameliorate gait abnormalities via inhibition of mTOR in osteoarthritis. Biomaterials. 2019 Mar 20;206:87-100.

Allen

Adams

Setton

. Evaluating intra-articular drug delivery for the treatment of osteoarthritis in a rat model. Tissue Eng Part B Rev. 2010 Feb;16(1):81-92.

10.

Parvathy

Masocha

. Gait analysis of C57BL/6 mice with complete Freund’s adjuvant-induced arthritis using the CatWalk system. BMC Musculoskelet Disord. 2013 Jan 8;14:14.

11.

Ferreira-Gomes

Adães

Mendonça

Castro-Lopes

. Analgesic effects of lidocaine, morphine and diclofenac on movement-induced nociception, as assessed by the Knee-Bend and CatWalk tests in a rat model of osteoarthritis. Pharmacol Biochem Behav. 2012 Jun;101(4):617-24.

12.

Mason

Chambers

Flannelly

Gaffen

Dudhia

Bayliss

. The STR/ort mouse and its use as a model of osteoarthritis. Osteoarthr Cartil. 2001 Feb;9(2_suppl):85-91.

13.

Akoum

Tahiri

Corvol

M-T

Borderie

Étienne

Rannou

, et al. Aging Cartilage in wild-type mice: an observational study. Cartilage. 2020 Jun;1:1-5.

14.

Setton

Elliott

Mow

. Altered mechanics of cartilage with osteoarthritis: human osteoarthritis and an experimental model of joint degeneration. Osteoarthr Cartil. 1999 Jan;7(1):2-14.

15.

Castaño Carou

Pita Fernández

Pértega Díaz

de Toro Santos

; Grupo de estudio EVALÚA. Clinical profile, level of affection and therapeutic management of patients with osteoarthritis in primary care: the Spanish multicenter study EVALÚA. Reumatol Clin. 2015 Dec;11(6):353-60.

16.

Ruan

MZC

Patel

Dawson

Jiang

M-M

Lee

BHL

. Pain, motor and gait assessment of murine osteoarthritis in a cruciate ligament transection model. Osteoarthr Cartil. 2013 Sep;21(9):1355-64.

17.

Nüesch

Dieppe

Reichenbach

Williams

Iff

Jüni

. All cause and disease specific mortality in patients with knee or hip osteoarthritis: population based cohort study. BMJ. 2011 Mar 8;342:d1165.

18.

Dieppe

Lohmander

. Pathogenesis and management of pain in osteoarthritis. Lancet. 2005 Mar 12;365(9463):965-73.

19.

Jacobs

Kloefkorn

Allen

. Gait analysis methods for rodent models of osteoarthritis. Curr Pain Headache Rep. 2014 Oct;18(10):456.

20.

Percie

Sert

Ahluwalia

Alam

Avey

Baker

Browne

, et al. Reporting animal research: explanation and elaboration for the ARRIVE guidelines 2.0. PLoS Biol. 2020 Jul;18(7):e3000411.

21.

Glasson

Blanchet

Morris

. The surgical destabilization of the medial meniscus (DMM) model of osteoarthritis in the 129/SvEv mouse. Osteoarthr Cartil. 2007 Sep;15(9):1061-9.

22.

Mével

Merceron

Vinatier

Krisa

Richard

Masson

, et al. Olive and grape seed extract prevents post-traumatic osteoarthritis damages and exhibits in vitro anti IL-1β activities before and after oral consumption. Sci Rep. 2016 19;6:33527.

23.

Hamers

Lankhorst

van Laar

Veldhuis

Gispen

. Automated quantitative gait analysis during overground locomotion in the rat: its application to spinal cord contusion and transection injuries. J Neurotrauma. 2001 Feb;18(2_suppl):187-201.

24.

Dupleichs

Masson

Gauthier

Dutilleul

Bouler

J-M

Verron

, et al. Pain management after bone reconstruction surgery using an analgesic bone cement: a functional noninvasive in vivo study using gait analysis. J Pain. 2018;19(10):1169-80.

25.

Lakes

Allen

. Gait analysis methods for rodent models of arthritic disorders: reviews and recommendations. Osteoarthr Cartil. 2016 Nov;24(11):1837-49.

26.

Bouxsein

Boyd

Christiansen

Guldberg

Jepsen

Müller

. Guidelines for assessment of bone microstructure in rodents using micro-computed tomography. J Bone Miner Res. 2010 Jul;25(7):1468-86.

27.

Shin

Pozzi

Boyd

Clark

. Integrin α1β1 protects against signs of post-traumatic osteoarthritis in the female murine knee partially via regulation of epidermal growth factor receptor signalling. Osteoarthr Cartil. 2016;24(10):1795-806.

28.

Ledingham

Regan

Jones

Doherty

. Radiographic patterns and associations of osteoarthritis of the knee in patients referred to hospital. Ann Rheum Dis. 1993 Jul;52(7):520-6.

29.

Muramatsu

Sasho

Saito

Yamaguchi

Akagi

Mukoyama

, et al. Preventive effects of hyaluronan from deterioration of gait parameters in surgically induced mice osteoarthritic knee model. Osteoarthr Cartil. 2014 Jun;22(6):831-5.

30.

Ferland

Laverty

Beaudry

Vachon

. Gait analysis and pain response of two rodent models of osteoarthritis. Pharmacology Biochemistry and Behavior. 2011 Jan 1;97(3):603-10.

31.

Malfait

Ritchie

Gil

Austin

J-S

Hartke

Qin

, et al. ADAMTS-5 deficient mice do not develop mechanical allodynia associated with osteoarthritis following medial meniscal destabilization. Osteoarthr Cartil. 2010 Apr;18(4):572-80.

32.

Tian

N-X

Bai

Q-Y

Chen

Sun

X-H

Wang

. Gait assessment of pain and analgesics: comparison of the DigiGait^TM and CatWalk^TM gait imaging systems. Neurosci Bull. 2019 Jan 18;35(3):401-18.

33.

Cheuk

Hung

Chan

. Limb Idleness Index (LII): a novel measurement of pain in a rat model of osteoarthritis. Osteoarthr Cartil. 2012 Nov;20(11):1409-16.

34.

Fang

Huang

Welch

Norley

Holdsworth

Beier

, et al. Early Changes of Articular Cartilage and Subchondral Bone in The DMM Mouse Model of Osteoarthritis. Sci Rep. 2018 Feb 12;8(1):2855.

35.

Allen

Griffin

Rodriguiz

Wetsel

Kraus

Huebner

, et al. Decreased physical function and increased pain sensitivity in mice deficient for type IX collagen. Arthritis Rheum. 2009 Sep;60(9):2684-93.

36.

Zhu

Zhen

, et al. Subchondral bone osteoclasts induce sensory innervation and osteoarthritis pain. J Clin Invest. 2019 Mar 1;129(3):1076-93.

37.

Kloefkorn

Jacobs

Loye

Allen

. Spatiotemporal gait compensations following medial collateral ligament and medial meniscus injury in the rat: correlating gait patterns to joint damage. Arthritis Res Ther. 2015 Oct 14;17:287.

38.

Poulet

de Souza

Kent

Saxon

Barker

Wilson

, et al. Intermittent applied mechanical loading induces subchondral bone thickening that may be intensified locally by contiguous articular cartilage lesions. Osteoarthr Cartil. 2015 Jun;23(6):940-8.

39.

Creaby

Bennell

Hunt

. Gait differs between unilateral and bilateral knee osteoarthritis. Arch Phys Med Rehabil. 2012 May;93(5):822-7.

40.

Dorman

Krug

Frizelle

Funkenbusch

Mahowald

. A comparison of DigiGait^TM and TreadScan^TM imaging systems: assessment of pain using gait analysis in murine monoarthritis. J Pain Res. 2013 Dec 24;7:25-35.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.08 MB

0.66 MB

0.07 MB

0.19 MB

0.07 MB