A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis

Osteoarthritis and Cartilage xxx (2016) 1e10

A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis A.J. Ramme y, M. Lendhey y, J.G. Raya z, T. Kirsch y, O.D. Kennedy y x * y Department of Orthopaedic Surgery, New York University School of Medicine, New York, NY, USA z Center for Biomedical Imaging, Department of Radiology, New York University School of Medicine, New York, NY, USA x Department of Mechanical and Aerospace Engineering, New York University Tandon School of Engineering, New York, NY, USA

a r t i c l e i n f o

s u m m a r y

Article history: Received 24 November 2015 Accepted 18 May 2016

Objective: Subchondral microdamage may play an important role in post-traumatic osteoarthritis (PTOA) development following anterior cruciate ligament (ACL) rupture. It remains unknown whether this injury mechanism causes subchondral microdamage, or whether its repair occurs by targeted osteoclastmediated remodeling. If so these events may represent a mechanism by which subchondral bone is involved in PTOA. Our objective was to test the hypothesis that subchondral microdamage occurs, and is co-localized with remodeling, in a novel rat model of ACL rupture. Design: We developed a novel non-invasive rat animal model for ACL rupture and subchondral microdamage generation. By inducing ACL rupture noninvasively rather than surgically, this more closely mimics the clinical injury. MicroCT, MRI and histological methods were used to measure microstructural changes, ligament damage, and cellular/matrix degeneration, respectively. Results: We reproducibly generated ACL rupture without damage to other soft joint tissues. Immediately after injury, increased microdamage was found in the postero-medial aspect of the tibia. Microstructural parameters showed increased resorption at 2 weeks, which returned to baseline. Dynamic histomorphometry showed increased calcein label uptake in the same region at 4 and 8 weeks. Chondrocyte death and protease activity in cartilage was also noted, however whether this was directly linked to subchondral changes is not yet known. Similarly, cartilage scoring showed degradation at 4 and 8 weeks post-injury. Conclusions: This study shows that our novel model can be used to study subchondral microdamage after ACL-rupture, and its association with localized remodeling. Cartilage degeneration, on a similar timescale to other models, is also a feature of this system. © 2016 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Keywords: Non-invasive model Acute injury Subchondral Bone Microdamage Remodeling

Introduction Post-traumatic osteoarthritis (PTOA) makes up 12% of the overall osteoarthritis disease burden, and is defined by its association with a specific acute joint injury. Its associated costs are up to $3 billion annually, and it mostly affects a relatively young demographic1,2. PTOA is the long-term consequence of damage to multiple joint tissues, including subchondral bone, by acute injuries

* Address correspondence and reprint requests to: O.D. Kennedy, Department of Orthopaedic Surgery, New York University School of Medicine, 301 East 17th Street, Suite 1500, New York, NY 10003, USA. Tel: 1-212-598-6563; Fax: 1-212-598-6096. E-mail addresses: [email protected] (A.J. Ramme), Matin.Lendhey@ nyumc.org (M. Lendhey), [email protected] (J.G. Raya), Thorsten.Kirsch@ nyumc.org (T. Kirsch), [email protected] (O.D. Kennedy).

such as anterior cruciate ligament (ACL) rupture. The local tissue specific injury response likely contributes to overall ‘joint failure’3. Subchondral bone microdamage may be an important mechanism in this system, and may contribute to PTOA development. Little is known about microdamage in subchondral tissues. However, it has been linked with bone marrow lesions (BMLs), which are seen in 80% of ACL rupture cases by MRI, and these lesions have been implicated in PTOA progression4,5. Many current animal models of PTOA do not generate subchondral microdamage, and thus are not appropriate for assessing its potential contribution to disease development. We present a novel animal model system that addresses this need, and provides a platform for further studies on the specific role of subchondral bone in PTOA development. Microdamage has been shown to affect mechanical and biological properties of bone tissue. Mechanically, its presence reduces

http://dx.doi.org/10.1016/j.joca.2016.05.017 1063-4584/© 2016 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017

2

A.J. Ramme et al. / Osteoarthritis and Cartilage xxx (2016) 1e10

bone tissue modulus. Biologically, microdamage initiates osteoclast-mediated remodeling which is mediated by osteocyte apoptosis involving cytokine signaling, angiogenesis and matrixderived growth factor release6. Most of our current knowledge on microdamage pathways was established using experimental systems focused on diaphyseal cortical or trabecular bone7e10. However, subchondral tissues likely illicit similar responses, which may have particular consequences unique to that region. Specifically, subchondral microdamage may initiate targeting osteoclastmediated remodeling which may contribute to overall joint failure. Traditionally, surgical models have been used to study cartilage disease in small animals11. However, since there is no pathological load involved in a surgical model, these do not generate subchondral microdamage. Christiansen et al.12 carried out a comprehensive review of various non-invasive mouse models, which have recently been developed13e16. The authors contrast these methods and note that each provides an opportunity to study a particular aspect of PTOA12. In this case, our novel model provides an opportunity to study subchondral microdamage, which has been implicated in BML generation5 and which may in turn may be linked to PTOA initiation and progression. We propose that this microdamage, and its subsequent repair by remodeling are linked. This system could be a candidate mechanism for subchondral bone involvement in overall joint failure. The current study was designed to test the hypothesis that subchondral microdamage occurs after ACL rupture, in a novel non-invasive animal model, and is colocalized with increased subchondral bone remodeling. Materials and methods Animals In total, 64 adult female Sprague Dawley rats (age 14e16 weeks) were purchased from Charles River (Wilmington, MA, USA) for these studies. Two groups were sacrificed on the day of injury (‘Day-0’ groups) and were used for joint injury assessment (n ¼ 8) and microdamage analyses (n ¼ 8). Three survival groups were sacrificed at 2, 4, and 8 weeks post-injury (n ¼ 16/time-point). Time-points were chosen based on reports in the literature showing these represent good approximations of mild-moderate OA development after injury17e19. At each survival time-point, half of the animals (n ¼ 8) were assigned for undecalcified embedding in poly methylmethacrylate (PMMA), and the other half (n ¼ 8) for standard decalcified paraffin embedding. Group sizes were determined by sample size calculations with effect sizes estimated from our previous studies20,21 (desired power ¼ 0.8, a ¼ 0.05). Animals were randomly assigned to groups and were kept in a standard lightedark cycle, with ad libitum access to food/ water. In one Day-0 group, one animal displayed evidence of hemarthrosis prior to ACL rupture and was excluded. During injury protocol, operators were unaware of treatment allocation for any animal. For analyses, an independent person assigned codes to samples, which were not broken until analyses were completed. All procedures were conducted based on ‘animal research: reporting of in vivo experiments’ (ARRIVE) guidelines, and with Institutional Animal Care and Use Committee (IACUC) approval. Non-invasive ACL rupture model Rat knees were injured using a modification of the established tibial-loading model22. The left limb was secured in an electromagnetic loading frame (Electroforce 3200, Bose Corp., MN, USA) using custom-made fixtures. The lower fixture was designed using solid modeling computer-aided design (Solidworks, Dassault mes, MA) and was fabricated by 3D printing (Mojo 3D printer, Syste

Stratasys, MN) using ABSplus, a production grade thermoplastic material. This fixture was supported the foot in 20 dorsiflexion. The standing surface, which accommodates the plantar surface of the foot in the lower fixture, was also angled at 20 to the horizontal to induce valgus moment across the joint. Finally the fixture was rotated in the transverse plane, about its long axis, through 40 from a plane defined by the long axis of the femur to induce external tibial rotation [Fig. 1(A)]. The upper fixture was machined from stainless steel and a ball-end mill was used to create a concave surface to seat the distal femur. Animals were anesthetized and maintained using isoflurane, a pre-load of 0.5 N was used to stabilize the knee, and then a single dynamic axial compressive load at a 1 mm/s was delivered until ACL rupture occurred. Load, displacement and time data were sampled at 200 Hz and values for ultimate failure load, loading curve slope, and area under the curve were calculated along with the coefficient of variation (CV). All survival animals received the fluorochrome bone labeling agent calcein (Sigma, intraperitoneal; 10 mg/kg) on the day of injury and 3 days before sacrifice. Day-0 injury groups Knee injury was assessed by gross dissection and also by confirmatory MRI imaging, to rule out iatrogenic damage during dissection. One group of animals (n ¼ 8) was used for this study. Half of this group (n ¼ 4) was randomly assigned for dissection and other (n ¼ 4) for MRI. Dissection involved a medial parapatellar incision and capsulectomy and the ACL visualized [Fig. 1(B), (D)]. For MRI, knees were stored in perfluoropolyether vacuum oil (Fomblin®, Sigma, USA). MRI was performed on a 7 T Bruker Biospec magnet (Bruker BioSpin GmbH, Germany) with a 20-cm bore and Litz coil (Doty Scientific, Columbia, SC). The protocol included turbo spin echo (TSE) sequences and T2-weighting (TE/TR ¼ 45/ 5000 ms, echo train length ¼ 2, repetitions ¼ 6, acquisition time ¼ 1:23 h), and PD-weighting (TE/TR ¼ 10.7/3583 ms, echo train length ¼ 3, repetitions ¼ 6, acquisition time ¼ 0:39 h). Sagittal images were acquired at a 60  60 mm2 resolution [Fig. 1(C), (E)]. Microdamage analyses Bilateral knees were fixed in NBF, bulk stained in basic fuchsin and then embedded in PMMA. Two sagittal sections of 200 mm thickness per compartment were made using a diamond cutting saw (Model 650, South Bay Technology, CA), and then polished to a thickness of 100 mm. Sections were taken in this plane so posterior and anterior ROIs could be analyzed, to assess spatial distribution of microdamage in this model. Microdamage was measured in two 1  1 mm2 tibial regions of interest (ROI) positioned to include the uppermost subchondral tissues, while excluding the distal compartment boundary (growth plate) [Fig. 2(C)]. ROIs were set 100 mm either side of the joint midline. Dimensional measurements were made of each subchondral region separately (zone of calcified cartilage [ZCC], cortical plate [CP] and subchondral trabecular bone [STB])23. Microcracks were identified by stereotypical shape, depth of field, and stain permeation24,25. Primary measurements included average crack density (Cr.Dn ¼ Crack Number (Cr.N)/Bone Border (B.Bd) [1/mm]), and crack surface density (Cr.S.Dn ¼ Crack Length (Cr.Le)/Bone Border (B.Bd) [mm/mm]). Survival groups The left knees of all survival animals (n ¼ 48, total) were injured as described above followed by normal cage activity. At the time of sacrifice, animals were euthanized by CO2 asphyxiation and both knees were harvested.

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017

A.J. Ramme et al. / Osteoarthritis and Cartilage xxx (2016) 1e10

3

Fig. 1. Illustrative summary of test set-up and injury protocol for noninvasive ACL rupture in rats. (A) Schematic illustration of rat left hind-limb positioning, distal femur is secured in a cup attached to the system actuator, which delivers the load at a rate of 1 mm/s. The hip joint is slightly abducted and the tiba rotated externally through 40 from the plane defined by the long axis of the femur. The ankle cup (circular inset, lower right) was created such that its surface was at an angle to the base surface, thus inducing a valgus moment across the knee joint. A photographic representation of this set-up is shown (photograph inset, upper left). Images from gross dissection of rat knees showing ACL (black arrows) from (B) control and (D) injured knees. Photomicrographs from MRI imaging in the sagittal plane showing (C) intact and (E) ruptured ACL (white arrows).

Fig. 2. Representative microCT and undecalcified histology images. (A) 2D coronal greyscale image showing the ROI boundaries which were used to evaluate the trabecular compartment and the medial and lateral subchondral plate compartments, respectively. (B) Representative 3D reconstruction of segmented trabecular region and the medial and lateral subchondral plate regions. (C) Composite image of undecalcified basic fuchsin stained, plastic embedded histological section through proximal tibia of the rat knee joint showing the approximated midline of the joint and the measurement locations for subchondral microdamage analyses.

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017

4

A.J. Ramme et al. / Osteoarthritis and Cartilage xxx (2016) 1e10

MicroCT analysis Knees of all animals in survival groups were scanned by microCT (SkyScan 1172; Bruker microCT, Belgium) where projections (4000  4000 pixels) were acquired at an isotropic resolution of 9 mm and 10 W power energy setting (100 kV and 100 mA). Scanning parameters were chosen in accordance with the ASBMR guidelines for mCT analysis of rodent bone structure26,27. Images were re-oriented into the coronal plane and the following 3D parameters were calculated: bone volume fraction (BV/TV), structure model index (SMI), connectivity density (CD), trabecular thickness, (Tb.Th.), trabecular spacing (Tb.Sp.) and cortical thickness (Ct.Th.)28 [Fig. 2(A), (B)]. Dynamic histomorphometry At each time-point knees from one group of animals (n ¼ 8) were processed in PMMA for dynamic histomorphometry using Osteomeasure software (Osteometrics, Atlanta, GA) on sagittal sections. Measurements were taken using the same ROIs as described above. The primary measurements and derived indices were based on recommendations of the American Society for Bone and Mineral Research nomenclature committee29. Labels within the cortex or the tidemark were not treated separately, thus labeled surfaces represent total label uptake. Histological tissue analyses Bilateral knees were harvested cleaned of soft tissues, fixed in Neutral Buffered Formalin (NBF), decalcified in 10% formic acid, embedded in paraffin, and 4 mm-thick serial sections were taken in the frontal plane. Chondrocyte death In order to address whether our injury protocol had any direct effect on cartilage, chondrocyte death was assessed in four randomly selected animals at each time-point. Necrosis was assessed using H&E stained sections in the medial tibial plateau since this region displayed the most robust response in other studies. The number of chondrocytes with pyknotic nuclei, along with the number of empty lacunae, was assessed as a percentage of total cells as described previously10,30. Mean values were pooled from each section as an indirect measure of total necrosis. Chondrocyte death by apoptosis was assessed by immunohistochemistry (IHC) staining for activated caspase-3. Sections were deparaffinized, re-hydrated and blocked for endogenous peroxidases. Sections were treated with hyaluronidase for antigen retrieval, blocked and incubated with antibodies against cleaved caspase-3 (#9661, Cell Signaling Technologies, Danvers, MA). Detection was performed using a goat anti-rabbit secondary antibody with a streptavidinebiotin conjugated system, and developed with a DAB substrate chromogen system (Dako, Carpentaria CA). For measurement of both necrosis and apoptosis, two sections per sample, separated by 50 mm, were evaluated and the mean values calculated. Metalloproteinase activity MMP-13 activity in articular chondrocytes was assessed using sections from the same four randomly selected animals used for chondrocyte death studies. Sections were stained using a monoclonal antibody against MMP-13 (ab3208, Abcam, Cambridge, MA). Briefly, using methods similar to those described above for caspase3 staining, sections were prepared, incubated with antibodies against MMP-13 and detection was performed using a secondary

antibody. For this assay, two sections per sample, separated by 50 mm, were evaluated and the mean values calculated. Synovitis assessment Safranin-O stained sections from four randomly selected animals at each survival time-point were used for synovial pathology evaluation. The synovial insertion of the medial tibia was evaluated using a modified form of an established synovitis score for synovial lining thickness31,32. This region was selected as a qualitative pilot study showed the most robust injury responses occurred at this location, also based on other reports in the literature31. Briefly, the enlargement of the synovial lining cell layer was scored as follows: 1e2 cell layers thick ¼ 0 points; 2e4 cell layers thick ¼ 1 point; 4e9 cell layers thick ¼ 2 points, >10 cell layers thick ¼ 3 points. As before, two sections per sample, separated by 50 mm, were evaluated and the mean values calculated. Cartilage degradation scoring Cartilage degradation was assessed using safranin-O and scored using the Osteoarthritis Research Society International (OARSI) recommendations for histologic assessment of osteoarthritis in the rat17. Accordingly, degeneration was scored using the arbitrary scale: 0 e no degeneration, 1 e minimal degeneration; 5e10% of total projected cartilage affected by matrix/chondrocyte loss. 2 e minimal degeneration; 11e25% affected. 3 e moderate degeneration; 26e50% affected. 4 e marked degeneration; 51e75%, 5 e severe degeneration;  75% affected. This score is an evaluation of overall degeneration where sections were divided into three zones (outer, central and inner). A three-zone sum was calculated. Two independent readers carried out all histological analyses. Parameters from all slides, from each location of each sample were averaged, and scores from each reader were averaged for each location. Statistical analyses Microcrack length and density data were assessed for normality using a KolmogoroveSmirnov test and did not meet criteria so were treated with a non-parametric Wilcoxon rank-sum test. MicroCT and histological results knees were compared at each time-point using a two-tailed paired t test. Differences between time-points were determined using a one-way analysis of variance (ANOVA), and post-hoc analysis carried out using a Fisher's Protected Least Significant Difference (PLSD) test, with P < 0.05. Data are presented as mean values ± standard deviation (SD). Results Animals responded well to anesthesia and a veterinary assessment detected no obvious signs of limping, or limb withdrawal after 12 h post-injury. Mechanical parameters were calculated as follows: Failure load (69.8 ± 8.5 N), loading curve slope (24.3 ± 7.3 N/mm), and area under the curve (209.9 ± 31.4 Nmm). The coefficients of variation (CV) for these parameters were 12.21%, 15.33% and 14.97%, respectively. Gross dissection showed all injured knees to have ACL rupture, which was either mid-substance or towards the femoral insertion site, and no avulsion fractures were observed. MRI imaging also showed evidence of ligament rupture in all cases, and no damage to other ligamentous or meniscal structures was noted. Other features of injury included minor swelling and occasional hemarthrosis. ACL rupture caused subchondral micro-architectural changes at all time-points. Trabecular BV/TV was lower in all injured knees at all time-points, although levels were close to baseline by 8 weeks (Fig. 3). At 2 weeks BV/TV was reduced by 12.03%, mainly under the influence of tissue volume (TV). SMI, which is a shape measure of

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017

A.J. Ramme et al. / Osteoarthritis and Cartilage xxx (2016) 1e10

5

Fig. 3. Analysis of bone micro-architectural parameters by microCT in trabecular and CP compartments. (A) Trabecular bone volume fraction (BV/TV) in control and injured knee joints at 2, 4 and 8 week time-points individual data points are shown and a line connects injured knee to contra-lateral control to indicate the direction of change. (B) Trabecular micro-architectural parameters were calculated and expressed as percent changes in injured knee compared to contra-lateral control, BV ¼ Bone Volume, TV ¼ Tissue Volume, SMI ¼ Structural Model Index, CD, Tb.Th., Tb.Sp ¼ Trabecular Spacing. (C) CP thickness of medial and lateral compartments in control and injured knees at 2, 4 and 8 week timepoints. Values are expressed as mean ± SD, *P < 0.05.

trabeculae, remained unchanged and suggests that most trabeculae remained rod-like. While Tb.Th. was markedly decreased at 2 weeks values returned to baseline at 4 and 8 weeks. The number of connections (CD) was lower in injured knees at each time-point. The subchondral plate had reduced thickness at 2 and 4 weeks, which but also returned to baseline by 8 weeks. Subchondral microdamage was region-specific, and was primarily observed in the postero-medial tibia. In the same region microcrack density and length were increased in both ZCC and CP. In addition, microcrack length was greater in CP than ZCC (Table I). Dynamic histomorphometry showed that label uptake was

increased in the posterior regions, the same region where increased microdamage was found (Fig. 4). This parameter was significantly lower in the trabecular compartment at 2 weeks but then returned to baseline. Interestingly in CP, the label uptake in posterior regions was significantly increased at all time-points. This pattern was also evident in ZCC, but only at 4 and 8 weeks. There were no differences in mineral apposition rate at any time-points. Bone formation rate was increased in the injured group at 8 weeks but not at earlier time-points. Evaluation of chondrocyte death by necrosis was assessed using morphological measurements, while apoptosis was assessed using

Table I Regional distribution and lengths of subchondral microdamage in calcified cartilage and CP in the anterior/posterior medial and lateral compartments of the proximal tibia Region Injured (left knee) Calcified cartilage CP Control (right knee) Calcified cartilage CP

Parameter

Antero-medial

Postero-medial

Antero-lateral

Postero-lateral

Cr.Dn, 1/mm Cr.S.Dn, mm/mm Cr.Dn, 1/mm Cr.S.Dn, mm/mm

0.088 9.36 0.025 2.64

± ± ± ±

0.03a 5.84a 0.054a 4.32a

0.619 27.06 0.52 38.63

± ± ± ±

0.215b 2.50b 0.23b 3.98b

0.026 3.15 0.021 13.61

± ± ± ±

0.055a 2.76c 0.208a 3.68c

0.299 25.06 0.318 33.96

± ± ± ±

0.21a 2.03b 0.208b 8.81b

Cr.Dn, 1/mm Cr.S.Dn, mm/mm Cr.Dn, 1/mm Cr.S.Dn, mm/mm

0.044 8.48 0.025 2.64

± ± ± ±

0.071a 5.24a 0.054a 0.05a

0.015 7.01 0.026 11.21

± ± ± ±

0.033b 6.53a 0.026a 6.06a

0.012 3.15 0.066 7.9

± ± ± ±

0.043b 5.04a 0.069a 3.68a

0.025 8.79 0.026 9.21

± ± ± ±

0.054b 2.46a 0.054a 4.18a

Mean ± SD. Within rows parameters with different superscript letters are significantly different (P < 0.005). N.Cr/B.Bd e microcrack boundary density; Cr.S.Dn e crack surface density.

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017

6

A.J. Ramme et al. / Osteoarthritis and Cartilage xxx (2016) 1e10

Fig. 4. Analysis of dynamic histomorphometry parameters in each tissue compartment at each time-point. (A) Mean values for MS/BS, MAR and BFR in the trabecular compartment for each group at each time-point. (B) Mean values for the same parameters in the CP compartment for each group at each time-point. (C) Mean values for the same parameters in calcified cartilage for each group at each time-point. Values are expressed as mean ± SD, *P < 0.05. (DeG) Photomicrographs of osteochondral junction from undecalcified basic fuchsin stained, plastic embedded histological section where brightfield and fluorescence microscopy images of the same site are concatenated to demonstrate the spatial relationship of localized calcein labels to local morphology from control, 2, 4 and 8 week groups, respectively.

IHC techniques. Significantly increased levels of both were found at 2 weeks, and both subsequently returned to baseline. For assessment of synovial pathology, the same trend was at the medio-tibial synovial insertion, with injury resulting in increased scores at 2 weeks. Synovitis scores remained elevated in injury groups at subsequent time-points but differences were not significant. Staining for MMP-13 showed no differences at 2 weeks but subsequently increased at 4 and 8 weeks. Similarly, cartilage scoring of safranin-O sections showed no significant difference between groups at 2 weeks, but was increased at 4 and 8 weeks (Fig. 5). These observations suggest that initial injury ultimately results in cartilage deterioration. However, although these processes occur along a similar timescale to our observed changes in subchondral bone, it is not yet possible to establish a direct link between them. Discussion PTOA pathogenesis is linked to a specific traumatic joint injury, which catalyzes its initiation. Here we show that, subchondral microdamage is co-localized with remodeling in a rat ACL injury model which mimics the clinical injury. The relevance of subchondral involvement in PTOA is supported by MRI data showing BMLs in approximately 80% of ACL injuries4. In many of these cases, arthroscopic surgery soon after injury shows no overt damage to articular cartilage. Furthermore a recent study by Roemer et al. demonstrated that subchondral BMLs predicted OA, with highest odds compared to multiple other radiographic features, 1 year prior to case-defining visit33. The precise make-up of BMLs has not been determined but many reports suggest they are associated with subchondral microdamage5,34,35. In experimental studies, microdamage has been causally linked with osteoclast-mediated bone remodeling. This response serves as a targeted repair mechanism,

and is mediated by osteocyte apoptosis36,37. Based on these observations, the present study was designed to examine the effects of microdamage on subchondral bone remodeling in a novel rat model of ACL rupture. Many OA animal models use surgical approaches to initiate cartilage degeneration38. These methods do not recapitulate the spectrum of damage to other tissues caused in real joint injury. Specifically, to our knowledge, microdamage has not been demonstrated in any other acute knee injury model. It is also important to note that microdamage is not present in surgical models in the acute phase after injury. This highlights the usefulness and uniqueness of our model for studying acute phase responses in subchondral bone in PTOA. Our novel non-invasive ACL injury model achieves this by incorporating important mechanical components of clinical injury. These include torsional and valgus moments about the joint, along with compression through the long axis of the joint39e41. Other non-invasive rodent injury models have been reported in recent years13,16,42e45. A thorough review of this topic by Christiansen et al.12 points out that each method is designed to study a specific aspect of injury or subsequent disease development. Thus our model complements the existing literature by adding a method to study subchondral bone microdamage as a mechanism of disease. Others have reported noninvasive approaches using larger models such as rabbits46e48, which are suitable for studying surgical procedures such as ACL reconstruction and cartilage repair. Microcracks in subchondral compartment were primarily observed in the postero-medial tibia in both ZCC and CP. The spatial distribution of microdamage is important since ultimate cartilage degeneration occurs in a localized manner. Thus, if the two were related their spatial arrangement is an important factor. Also, the medial tibia is often the most affected by cartilage degradation, and

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017

A.J. Ramme et al. / Osteoarthritis and Cartilage xxx (2016) 1e10

7

Fig. 5. Histological and immunohistochemical assessments of joint degradation. (Row A) Necrotic cell death at each time-point as assessed by pyknotic cell morphology white arrows) and empty lacunae (black arrows) in sections stained with H&E. (Row B) Immunostaining for activated Caspase-3 at each time-point to assess caspase-3 positive cells (black arrows) in each group. (Row C) Immunostaining for MMP-13 at each time-point to assess levels protease activity in each group. (Row D) Synovitis scoring to assess levels synovitis at the medio-tibial synovial insertion in each group at each time-point. (D) Safranin-O sections with modified OARSI score to assess cartilage degradation following ACL rupture. Values are expressed as mean ± SD, *P < 0.05.

this was also evident in our model. Thus the presence of microdamage at this location lends some support to our hypothesis that these phenomena may be related. In a study of thoroughbred stifle joints in racehorses Muir et al.49 found high Cr.Dn in the medial condylar groove with values of 1.05 [1/mm]. Those studies did not consider the ZCC and CP separately, thus their reported values are aggregated. Our Cr.Dn values of 0.61 and 0.52 [1/mm], when combined, compare very well with those data. In addition microcracks were longer in CP compared to ZCC. This is likely due to the self-limiting nature of the upper and lower boundaries of that region. Clinically, subchondral bone in OA is often associated with latestage sclerosis. Contrastingly, we report increased resorption at early time-points. Others have also reported this phenomenon and our data supports those findings16,28. Whether our model would eventually develop a sclerotic subchondral phenotype cannot be determined by these data, but will be investigated in future. Others have observed late-stage sclerosis using different animal models. Using a canine ACLT model Dedrick et al. showed increased Ct.Th. (100%) at 54 months post-surgery50. These data support the concept that subchondral changes are involved in PTOA development51e54. We found increased subchondral remodeling after injury, but activity varied between time-points. STB showed differences in label uptake after 2 weeks, but not at later time-points. This may be due to label-escape phenomenon, which occurs if double label intervals are long compared to baseline remodeling rate. This is

particularly true for trabecular bone where the metabolic rate is relatively high. While we pooled labeled surfaces in our measurement, the majority of label at 4 and 8 weeks were discrete ‘tunneling’ events, which were also qualitatively associated with invading blood vessels [Fig. 4(F), (G); white arrows]. In ZCC, instances of discrete ‘tunneling’ remodeling sites were also noted; suggesting that tidemark mineralization is a feature of our model. The mineral apposition rate was not different in any compartment at any time-point, which is likely due to label-escape effects. The bone formation rate was increased at 8 weeks in both CP and ZCC. This derived measure is a product of MS/BS and MAR, it is likely to be influenced by the increase in the former parameter. In order to assess whether our injury protocol had any direct effect on cartilage, we assessed chondrocyte death by necrosis and apoptosis using morphological and IHC methods, respectively. A potential limitation of using morphological methods for necrosis analyses is overestimation due to differential shrinkage during tissue processing. However, this seems unlikely here, since levels were relatively low compared to other studies31,55. Caspase-3 staining showed apoptosis was highest at 2 weeks. These data are in broad agreement with in vitro explant injury studies in the literature, although here our earliest time-point was 2 weeks, thus we may have missed earlier responses. Our assessment of synovitis pathology showed a similar trend, but again earlier time-points would be useful. MMP-13 activity was increased at 4 and 8 weeks, which is in line with expectations given the cartilage deterioration that we observed. These data suggest that

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017

8

A.J. Ramme et al. / Osteoarthritis and Cartilage xxx (2016) 1e10

pathological processes are initiated in cartilage tissue alongside the subchondral bone event that we have observed. However a direct link between the processes in these two tissues has not yet been established. Cartilage deterioration was observed at later timepoints (4 and 8 weeks), and this is comparable to other models, in which focal lesions develop between 6 and 12 weeks post-insult. This study has a number of limitations. Although we demonstrate co-localization of subchondral microdamage and remodeling, it is important to note that we cannot casually link those events to the observed cartilage degeneration directly. We carried out microdamage studies using 100 mm undecalcified sagittal sections, and cartilage studies using decalcified 5 mm sections in the frontal plane. Thus direct comparison between these groups was not possible. It is also important to note we cannot rule out cartilage degeneration results directly from injury to other joint tissues such as synovium or cartilage itself. Previous studies show that above certain stress thresholds, chondrocyte apoptosis occurs and catabolic gene expression is increased56,57. We did not carry out a full stress analysis of the joint under our loading condition; this will be an important development of our model and will be the subject of future investigations. Finally we did not investigate the potential role of meniscal damage in our model, while no damage was apparent radiologically, this does not rule out its presence and this will also be the subject of future work. Taken together, our data demonstrates that subchondral microdamage occurs during ACL injury using a novel animal model for PTOA. Furthermore, subchondral remodeling is co-localized with microdamage. Cartilage deterioration was also observed in temporal association with these responses. These data support the concept that subchondral microdamage and remodeling may contribute to development of PTOA after acute joint injury.

2.

3.

4.

5.

6.

7.

8.

9.

10.

Author contributions 11. Concept (ODK, AJR, ML, TK), Experimental design (ODK, AJR, ML, JGR, TK), Experiments/Imaging (ODK, AJR, ML, JGR), Data analyses and interpretation (ODK, AJR, ML, JGR, TK), Manuscript preparation and editing (ODK, AJR, ML, JGR, TK). Conflict of interest statement All authors report no support or benefits were received from commercial entities and no other interests that cause conflict of interest with this work. Role of funding source The OREF/MTF provided funding for AJR and ODK with a clinicianscientist training grant for this part of this work. The New York University School of Medicine Department of Orthopaedic Surgery also funded part of this work.

12.

13.

14.

15. Acknowledgments Aided by a grant from the Orthopaedic Research and Education Foundation with funding provided by the Musculoskeletal Transplant Foundation. The authors would like to thank the Department of Orthopaedic Surgery, New York University School of Medicine for support and Mr Kevin Voss, Mr Jurinus Lesporis and Ms Rebecca Amorse for help with undecalcified sample preparation, Dr Cheongeun Oh for advice on statistical analyses and Dr You Jin Lee for histological expertise.

16.

17.

References 18. 1. Lawrence RC, Felson DT, Helmick CG, Arnold LM, Choi H, Deyo RA, et al. Estimates of the prevalence of arthritis and

other rheumatic conditions in the United States. Part II. Arthritis Rheum 2008;58(1):26e35. Brown TD, Johnston RC, Saltzman CL, Marsh JL, Buckwalter JA. Posttraumatic osteoarthritis: a first estimate of incidence, prevalence, and burden of disease. J Orthop Trauma 2006;20(10):739e44. Anderson DD, Chubinskaya S, Guilak F, Martin JA, Oegema TR, Olson SA, et al. Post-traumatic osteoarthritis: improved understanding and opportunities for early intervention. J Orthop Res 2011;29(6):802e9. Felson DT, Chaisson CE, Hill CL, Totterman SM, Gale ME, Skinner KM, et al. The association of bone marrow lesions with pain in knee osteoarthritis. Ann Intern Med 2001;134(7): 541e9. Rangger C, Kathrein A, Freund MC, Klestil T, Kreczy A. Bone bruise of the knee: histology and cryosections in 5 cases. Acta Orthop Scand 1998;69(3):291e4. Schaffler MB, Cheung WY, Majeska R, Kennedy O. Osteocytes: master orchestrators of bone. Calcif Tissue Int 2014;94(1): 5e24. Seref-Ferlengez Z, Kennedy OD, Schaffler MB. Bone microdamage, remodeling and bone fragility: how much damage is too much damage? Bonekey Rep 2015;4:644. Torres AM, Matheny JB, Keaveny TM, Taylor D, Rimnac CM, Hernandez CJ. Material heterogeneity in cancellous bone promotes deformation recovery after mechanical failure. Proc Natl Acad Sci USA 2016;113(11):2892e7. Poundarik AA, Diab T, Sroga GE, Ural A, Boskey AL, Gundberg CM, et al. Dilatational band formation in bone. Proc Natl Acad Sci USA 2012;109(47):19178e83. Bentolila V, Boyce TM, Fyhrie DP, Drumb R, Skerry TM, Schaffler MB. Intracortical remodeling in adult rat long bones after fatigue loading. Bone 1998;23(3):275e81. Little CB, Zaki S. What constitutes an “animal model of osteoarthritis”dthe need for consensus? Osteoarthritis Cartilage 2012;20(4):261e7. Christiansen BA, Guilak F, Lockwood KA, Olson SA, Pitsillides AA, Sandell LJ, et al. Non-invasive mouse models of post-traumatic osteoarthritis. Osteoarthritis Cartilage 2015;23(10):1627e38. Furman BD, Strand J, Hembree WC, Ward BD, Guilak F, Olson SA. Joint degeneration following closed intraarticular fracture in the mouse knee: a model of posttraumatic arthritis. J Orthop Res 2007;25(5):578e92. Poulet B, de Souza R, Kent AV, Saxon L, Barker O, Wilson A, et al. Intermittent applied mechanical loading induces subchondral bone thickening that may be intensified locally by contiguous articular cartilage lesions. Osteoarthritis Cartilage 2015;23(6):940e8. Ko FC, Dragomir C, Plumb DA, Goldring SR, Wright TM, Goldring MB, et al. In vivo cyclic compression causes cartilage degeneration and subchondral bone changes in mouse tibiae. Arthritis Rheum 2013;65(6):1569e78. Christiansen BA, Anderson MJ, Lee CA, Williams JC, Yik JH, Haudenschild DR. Musculoskeletal changes following noninvasive knee injury using a novel mouse model of posttraumatic osteoarthritis. Osteoarthritis Cartilage 2012;20(7): 773e82. Gerwin N, Bendele AM, Glasson S, Carlson CS. The OARSI histopathology initiative e recommendations for histological assessments of osteoarthritis in the rat. Osteoarthritis Cartilage 2010;18(Suppl 3):S24e34. Hayami T, Pickarski M, Zhuo Y, Wesolowski GA, Rodan GA, Duong Le T. Characterization of articular cartilage and subchondral bone changes in the rat anterior cruciate ligament

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017

A.J. Ramme et al. / Osteoarthritis and Cartilage xxx (2016) 1e10

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

transection and meniscectomized models of osteoarthritis. Bone 2006;38(2):234e43. Appleton CT, McErlain DD, Pitelka V, Schwartz N, Bernier SM, Henry JL, et al. Forced mobilization accelerates pathogenesis: characterization of a preclinical surgical model of osteoarthritis. Arthritis Res Ther 2007;9(1):R13. Verborgt O, Tatton NA, Majeska RJ, Schaffler MB. Spatial distribution of Bax and Bcl-2 in osteocytes after bone fatigue: complementary roles in bone remodeling regulation? J Bone Miner Res 2002;17(5):907e14. Cardoso L, Herman BC, Verborgt O, Laudier D, Majeska RJ, Schaffler MB. Osteocyte apoptosis controls activation of intracortical resorption in response to bone fatigue. J Bone Miner Res 2009;24(4):597e605. Fritton JC, Myers ER, Wright TM, van der Meulen MC. Loading induces site-specific increases in mineral content assessed by microcomputed tomography of the mouse tibia. Bone 2005;36(6):1030e8. Ham KD, Carlson CS. Effects of estrogen replacement therapy on bone turnover in subchondral bone and epiphyseal metaphyseal cancellous bone of ovariectomized cynomolgus monkeys. J Bone Miner Res 2004;19(5):823e9. Burr DB, Hooser M. Alterations to the en bloc basic fuchsin staining protocol for the demonstration of microdamage produced in vivo. Bone 1995;17(4):431e3. Li J, Sato M, Jerome C, Turner CH, Fan Z, Burr DB. Microdamage accumulation in the monkey vertebra does not occur when bone turnover is suppressed by 50% or less with estrogen or raloxifene. J Bone Miner Metab 2005;23(Suppl):48e54. Bouxsein ML, Boyd SK, Christiansen BA, Guldberg RE, Jepsen KJ, Muller R. Guidelines for assessment of bone microstructure in rodents using micro-computed tomography. J Bone Miner Res 2010;25(7):1468e86. Palacio-Mancheno PE, Larriera AI, Doty SB, Cardoso L, Fritton SP. 3D assessment of cortical bone porosity and tissue mineral density using high-resolution microCT: effects of resolution and threshold method. J Bone Miner Res 2014;29(1):142e50. Botter SM, van Osch GJ, Waarsing JH, Day JS, Verhaar JA, Pols HA, et al. Quantification of subchondral bone changes in a murine osteoarthritis model using micro-CT. Biorheology 2006;43(3e4):379e88. Parfitt AM, Drezner MK, Glorieux FH, Kanis JA, Malluche H, Meunier PJ, et al. Bone histomorphometry: standardization of nomenclature, symbols, and units. Report of the ASBMR Histomorphometry Nomenclature Committee. J Bone Mine Res 1987;2(6):595e610. Dunstan CR, Somers NM, Evans RA. Osteocyte death and hip fracture. Calcif Tissue Int 1993;53(Suppl 1):S113e6. discussion S6e7. Lewis JS, Hembree WC, Furman BD, Tippets L, Cattel D, Huebner JL, et al. Acute joint pathology and synovial inflammation is associated with increased intra-articular fracture severity in the mouse knee. Osteoarthritis Cartilage 2011;19(7):864e73. Krenn V, Morawietz L, Burmester GR, Kinne RW, MuellerLadner U, Muller B, et al. Synovitis score: discrimination between chronic low-grade and high-grade synovitis. Histopathology 2006;49(4):358e64. Roemer FW, Kwoh CK, Hannon MJ, Hunter DJ, Eckstein F, Fujii T, et al. What comes first? Multitissue involvement leading to radiographic osteoarthritis: magnetic resonance imaging-based trajectory analysis over four years in the

34. 35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

9

osteoarthritis initiative. Arthritis Rheumatol 2015;67(8): 2085e96. Deangelis JP, Spindler KP. Traumatic bone bruises in the athlete's knee. Sports Health 2010;2(5):398e402. Mink JH, Deutsch AL. Occult cartilage and bone injuries of the knee: detection, classification, and assessment with MR imaging. Radiology 1989;170(3 Pt 1):823e9. Kennedy OD, Herman BC, Laudier DM, Majeska RJ, Sun HB, Schaffler MB. Activation of resorption in fatigue-loaded bone involves both apoptosis and active pro-osteoclastogenic signaling by distinct osteocyte populations. Bone 2012;50(5): 1115e22. Kennedy OD, Laudier DM, Majeska RJ, Sun HB, Schaffler MB. Osteocyte apoptosis is required for production of osteoclastogenic signals following bone fatigue in vivo. Bone 2014;64: 132e7. Glasson SS, Chambers MG, Van Den Berg WB, Little CB. The OARSI histopathology initiative e recommendations for histological assessments of osteoarthritis in the mouse. Osteoarthritis Cartilage 2010;18(Suppl 3):S17e23. Koga H, Nakamae A, Shima Y, Iwasa J, Myklebust G, Engebretsen L, et al. Mechanisms for noncontact anterior cruciate ligament injuries: knee joint kinematics in 10 injury situations from female team handball and basketball. Am J Sports Med 2010;38(11):2218e25. Arnold JA, Coker TP, Heaton LM, Park JP, Harris WD. Natural history of anterior cruciate tears. Am J Sports Med 1979;7(6): 305e13. Arendt E, Dick R. Knee injury patterns among men and women in collegiate basketball and soccer. NCAA data and review of literature. Am J Sports Med 1995;23(6):694e701. Lockwood KA, Chu BT, Anderson MJ, Haudenschild DR, Christiansen BA. Comparison of loading rate-dependent injury modes in a murine model of post-traumatic osteoarthritis. J Orthop Res 2014;32(1):79e88. Khorasani MS, Diko S, Hsia AW, Anderson MJ, Genetos DC, Haudenschild DR, et al. Effect of alendronate on post-traumatic osteoarthritis induced by anterior cruciate ligament rupture in mice. Arthritis Res Ther 2015;17:30. Maerz T, Kurdziel MD, Davidson AA, Baker KC, Anderson K, Matthew HW. Biomechanical characterization of a model of noninvasive, traumatic anterior cruciate ligament injury in the rat. Ann Biomed Eng 2015;43(10):2467e76. Maerz T, Kurdziel M, Newton MD, Altman P, Anderson K, Matthew HW, et al. Subchondral and epiphyseal bone remodeling following surgical transection and noninvasive rupture of the anterior cruciate ligament as models of posttraumatic osteoarthritis. Osteoarthritis Cartilage 2016;24(4): 698e708. Radin EL, Martin RB, Burr DB, Caterson B, Boyd RD, Goodwin C. Effects of mechanical loading on the tissues of the rabbit knee. J Orthop Res 1984;2(3):221e34. Pauly HM, Larson BE, Coatney GA, Button KD, DeCamp CE, Fajardo RS, et al. Assessment of cortical and trabecular bone changes in two models of post-traumatic osteoarthritis. J Orthop Res 2015;33(12):1835e45. Isaac DI, Meyer EG, Haut RC. Development of a traumatic anterior cruciate ligament and meniscal rupture model with a pilot in vivo study. J Biomech Eng 2010;132(6):064501. Muir P, McCarthy J, Radtke CL, Markel MD, Santschi EM, Scollay MC, et al. Role of endochondral ossification of articular cartilage and functional adaptation of the subchondral plate in the development of fatigue microcracking of joints. Bone 2006;38(3):342e9.

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017

10

A.J. Ramme et al. / Osteoarthritis and Cartilage xxx (2016) 1e10

50. Dedrick DK, Goldstein SA, Brandt KD, O'Connor BL, Goulet RW, Albrecht M. A longitudinal study of subchondral plate and trabecular bone in cruciate-deficient dogs with osteoarthritis followed up for 54 months. Arthritis Rheum 1993;36(10): 1460e7. 51. Zhen G, Wen C, Jia X, Li Y, Crane JL, Mears SC, et al. Inhibition of TGF-beta signaling in mesenchymal stem cells of subchondral bone attenuates osteoarthritis. Nat Med 2013;19(6):704e12. 52. Burr DB. The importance of subchondral bone in osteoarthrosis. Curr Opin Rheumatol 1998;10(3):256e62. 53. Burr DB, Gallant MA. Bone remodelling in osteoarthritis. Nat Rev Rheumatol 2012;8(11):665e73.

54. Goldring MB, Goldring SR. Articular cartilage and subchondral bone in the pathogenesis of osteoarthritis. Ann N Y Acad Sci 2010;1192:230e7. 55. D'Lima DD, Hashimoto S, Chen PC, Colwell Jr CW, Lotz MK. Human chondrocyte apoptosis in response to mechanical injury. Osteoarthritis Cartilage 2001;9(8):712e9. 56. Kuhn K, D'Lima DD, Hashimoto S, Lotz M. Cell death in cartilage. Osteoarthritis Cartilage 2004;12(1):1e16. 57. Hashimoto S, Ochs RL, Rosen F, Quach J, McCabe G, Solan J, et al. Chondrocyte-derived apoptotic bodies and calcification of articular cartilage. Proc Natl Acad Sci USA 1998;95(6): 3094e9.

Please cite this article in press as: Ramme AJ, et al., A novel rat model for subchondral microdamage in acute knee injury: a potential mechanism in post-traumatic osteoarthritis, Osteoarthritis and Cartilage (2016), http://dx.doi.org/10.1016/j.joca.2016.05.017