Concurrent Joint Contact in Anterior Cruciate Ligament Injury induces cartilage micro-injury and subchondral bone sclerosis, resulting in knee osteoarthritis

32 Objective. Anterior Cruciate Ligament (ACL) injury develops the Osteoarthritis (OA) via 33 two distinct processes: initial direct micro-injury of the cartilage surface by compressive force 34 during ACL injury, and secondary joint instability due to the deficiency of the ACL. Using 35 the conventional Compression-induced ACL-R and novel Non-Compression ACL-R models, 36 we aimed to reveal the individual effects on OA progression after ACL injury. 37 Methods. Twelve-week-old C57BL/6 male were randomly divided to three experimental 38 groups: Compression ACL-R, Non-Compression ACL-R, and Intact. We performed the joint 39 laxity test and microscope analysis at 0 days, in vivo imaging with matrix-metalloproteinases 40 (MMPs) at 3 and 7 days, histological and micro-CT analysis at 0, 7, 14, and 28 days. 41 Results. Although no differences in the joint laxity were observed between both ACL-R 42 groups, the Compression ACL-R group exhibited a significant increase of cartilage roughness 43 immediately after injury compared with the Non-Compression group. At 7 days, Compression 44 group increased MMPs-induced fluorescence intensity slightly and MMP-13 positive cell 45 ratio of chondrocytes significantly than that in the Non-Compression group. Moreover, 46 histological cartilage degeneration initiated in the whole joint level of the Compression group 47 at the same time point. Micro-CT analysis revealed that sclerosis of tibial Subchondral bone 48 in the Compression group developed significantly more than in the Non-Compression group 49 at 28 days, especially in the medial tibial compartment. 50 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 4 Conclusions. Concurrent joint contact during ACL rupture caused initial micro-damage 51 on the cartilage surface and early cartilage degeneration with MMP-13 production, leading to 52 later bone formation in the subchondral bone. 53 54 Key Words: 55 Osteoarthritis, Animal model, ACL injury, Mechanical stress, Subchondral bone 56 57 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 5

58 Osteoarthritis (OA) is a major chronic musculoskeletal disease that progresses 59 irreversibly and negatively influences the quality of life of OA patients. The number of OA 60 patients is rapidly increasing, with a 30.8% increase in OA prevalence between 2007 and 61 2017 [1]. OA can be divided into two types depending on the mechanisms of initiation: 62 primary OA, which is idiopathic and generally occurs during aging, and secondary post-63 traumatic OA (PTOA), which is initiated by a traumatic joint injury such as a meniscus tear or 64 ligament injury. PTOA accounts for approximately 12% of symptomatic OA patients [2] and 65 has an increasing incidence rate due to the increasing popularity of high-impact sports [3]. 66 Therefore, elucidating the mechanisms of PTOA initiation and progression is necessary to 67 find therapeutic strategies to prevent or slow the progression of OA. 68 Anterior Cruciate Ligament (ACL) injury is a known initiator of PTOA 69 development. The annual incidence of ACL injury in the general population is 68.6 per 70 100,000 people [4], and the incidence of PTOA following ACL injury is as high as 87% [5]. 71 In the etiology of OA following ACL injury, concomitant injuries of other joint structures 72 along with ACL rupture and increased mechanical stress caused by joint instability are likely 73 contributing factors. Superphysiological compressive force to the joint surfaces are common 74 during ACL injury. Almost half of ACL injury patients suffer from initial damage of articular 75 cartilage [6], and concomitant cartilage injury may increase OA risk up to 2.4 times at 19 76 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 6 years [7]. In addition, 80-90% of patients with an acute ACL injury also show signs of 77 subchondral bone lesions measured using magnetic resonance imaging (MRI). [5] 78 Subchondral bone changes can affect cartilage degeneration via crosstalk in molecular and 79 biomechanical interactions [8], suggesting that acute subchondral bone injuries may also 80 initiate OA progression. 81 Because the ACL restricts the anterior-posterior (AP) translation of the tibia relative 82 to the femur [9], joint instability after ACL injury can change the articulation of the knee joint 83 and induce abnormal mechanical stress, which may stimulate knee joint tissues to produce 84 catabolic enzymes. Since knee PTOA develops several decades after ACL injury, it was 85 assumed that secondary joint instability is the most important factor, and surgical 86 reconstruction of the ACL has been widely used to restore joint stability and reduce the risk 87 of PTOA development [10]. However, a recent systematic review reported that reconstruction 88 surgery may not affect PTOA progression [11], which suggests that the initial response to 89 compressive forces during injury may also plays a significant role in the pathogenesis of 90 PTOA. Since cartilage is an avascular tissue with poor self-healing capability, acute damage 91 to the articular chondral surface and subchondral bone may be involved with the initiation of 92 PTOA. 93 ACL transection (ACL-T) is a common model for inducing knee OA in small 94 animals, but this method also induces unnecessary acute inflammation in articular tissues due 95 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 7 to the invasive surgical procedures [12-15]. In contrast, the Non-Invasive Compression ACL 96 Rupture (ACL-R) model was developed and used to explore mechanisms of PTOA initiation 97 and progression following ACL injury without the intra-articular inflammation associated 98 with surgical procedures [16-20]. ACL-R is induced by applying compressive force to the 99 joint, which causes anterior tibial dislocation and ACL injury. This method largely imitates 100 acute ACL injury and knee PTOA progression after injury. However, it is difficult to separate 101 the effects of initial overload of the articular cartilage and the secondary effects of joint 102 instability. We recently established a novel Non-Compression ACL-R model, which is made 103 without compression force on the cartilage surface and therefore induces no articular injuries 104 other than ACL rupture [21]. Based on the findings of this model, we hypothesized that 105 comparing the Compression ACL-R model and the novel Non-Compression ACL-R model 106 will allow us to investigate the individual effects of initial tissue damage due to compression 107 vs. secondary instability on PTOA progression. 108 109

110 Animals and Experimental Design 111 Twelve-week-old C57BL/6J male mice were randomly divided into three 112 experimental groups: Compression ACL-R (n = 49), Non-Compression ACL-R (n = 48), and 113 Intact (n=21). Mice were euthanized at 0, 3, 7, 14, and 28 days after injury, and knee joints 114 were dissected for analysis (Fig. 1A). In the joint laxity test, microscopic, morphological, and 115 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 8 histological analysis at day 0 after injury, contralateral knee joints from the Non-Compression 116 group were used as the Intact group. Mice were allowed to acclimate for 1-2 weeks in the 117 vivarium before the start of the experiment, and were cared for in accordance with the 118 guidelines set by the National Institutes of Health (NIH) on the care and use of laboratory 119 animals. All procedures were conducted under approved Institutional Animal Care and Use 120 Committee protocols at Saitama Prefectural University and the University of California 121 Davis. 122 123 Creating Compression ACL-R and Non-Compression ACL-R Models 124 All procedures were performed on the right knee joint of each anesthetized mouse 125 under 1-4% inhaled isoflurane. The Compression ACL-R model was created via tibial 126 compression overload as previously described [18, 22]. Right knee joints were positioned in 127 an electromagnetic materials machine (ElectroForce 3200, TA Instruments, New Castle, DE), 128 and a single tibial compressive overload was applied at 1 mm/s to induce ACL rupture. The 129 Non-Compression ACL-R model was created based on our previous study [23]. The knee 130 joint was fixed at 90 degrees using surgical tape on a stand, and a force was slowly applied by 131 the thumb tip of the operator along the long axis of the femur. The applied force was stopped 132 quickly after hearing a distinct popping sound, which indicated ACL rupture. Anterior tibial 133 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 9 translation was manually confirmed to verify ACL injury. For both models, buprenorphine 134 analgesia (0.1 mg/kg) was injected immediately following injury for pain relief. 135 136 Anterior-Posterior Joint Laxity Testing 137 Knee joints were collected immediately post-injury, and femurs and tibias were 138 embedded in brass tubes using polymethylmethacrylate as previously described [22, 24]. 139 Tibias were fixed to the load cell; then femurs were secured so that the angle of the knee joint 140 was either 60 /i4 or 90 /i4 . Tibias were able to freely translate and rotate about the superior-141 inferior axis. Five anterior-posterior (AP) loading cycles were applied perpendicular to the 142 longitudinal axis of the tibia to a target force of ±1.5 N at a loading rate of 0.5 mm/s. The 143 degree of AP joint laxity was quantified based on the difference between displacement at +0.8 144 N and -0.8 N. 145 146 Microscopic Analysis of Cartilage Surface Roughness 147 For samples collected 0 days following injury, soft tissue was removed carefully to 148 expose the tibial plateau, then 3D optical profilograph images of the whole tibial plateau were 149 taken (VR-6000; Keyence, JPN). The region of interest (ROI) was set as an ellipse with 0.6-150 1.2 mm diameters in the center area of the medial and lateral tibial plateau (Figure 2A). The 151 arithmetical mean height and maximum height of the cartilage surface were measured by 152 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 10 normalizing to the contralateral intact knee of each mouse. Methods are further described in 153 the Supplementary Materials. 154 155 Fluorescent Reflectance Imaging (FRI) 156 At 3 and 7 days post-injury, mice were imaged in vivo using an optical imaging 157 system (IVIS Spectrum, PerkinElmer, Waltham, MA), after IV administration of a near 158 infrared probe that is activated by matrix-metalloproteinases (MMPSense 680, PerkinElmer, 159 Waltham, MA). The ROI was set as a circle of 0.7 mm2 that surrounded the tibial tuberosity 160 to the superior border of the patella, and image processing and quantification were performed 161 via IVIS Living Image software as previously described [25]. To unify the mouse-to-mouse 162 variation in the delivery of the fluorescent probe, the radiant efficiency of the ACL-R knee 163 was normalized to the contralateral intact knee of each mouse. Methods are further described 164 in the Supplementary Materials. 165 166 Micro-Computed Tomography Analysis of Osteophyte Formation and Epiphyseal Bone 167 Microstructure 168 Bilateral knees were scanned using micro-computed tomography (µCT 35, 169 SCANCO, Brüttisellen, Switzerland) with the following parameters: nominal voxel size = 10 170 µm, energy = 55 kVp, intensity = 72µA, integration time = 800 ms. For analysis of 171 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 11 osteophyte formation at 7, 14, and 28 days after injury, VOI contouring included all 172 heterotopic mineralized tissue around the joint, as well as the patella, fabellae, and menisci. 173 The average bone volume of the patella, fabellae, and menisci in the Intact group was used as 174 the “baseline” volume to calculate osteophyte formation in the Compression and Non-175 Compression groups. The difference in total bone volume between the injured mouse knees 176 and the baseline volume was calculated to determine the total osteophyte volume for each 177 injured joint. 178 We also assessed the morphological changes in epiphyseal trabecular bone at 0, 7, 179 14, and 28 days after injury. Morphological analysis of trabecular bone in the tibia and femur 180 was performed by manually drawing contours on 2D transverse slices; the ROI was designed 181 as the trabecular bone enclosed by the growth plate and subchondral cortical bone plate 182 (Figure 3B). Additionally, whole subchondral bone including the cortical bone plate and 183 trabecular bone in the medial and lateral tibial compartments were also measured (Figure 4A). 184 Using the manufacturer's analysis software, we quantified apparent bone mineral density 185 (BMD, g/cm3), bone volume per total volume (BV/TV, %), trabecular number (Tb.N, 1/mm), 186 trabecular thickness (Tb.Th, mm), and trabecular separation (Tb.Sp, mm). 187 188 Histological Analysis for Cartilage Degeneration 189 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 12 After performing µCT analysis, knee joints were decalcified in 10% 190 ethylenediaminetetraacetic acid (EDTA) for 2 weeks, dehydrated in 70% and 100% ethanol 191 and xylene, and embedded in paraffin blocks. The samples were cut in the sagittal plane (7 192 µm thickness) using a microtome (ROM-360; Yamato Kohki Industrial Co., Ltd., Saitama, 193 JPN). Then, safranin-O/fast green staining was performed to evaluate articular cartilage 194 degeneration in the medial and lateral tibial compartments using the Osteoarthritis Research 195 Society International (OARSI) histopathological grading system [26]. These assessments 196 were conducted by two independent observers blinded to all other sample information. We 197 initially assessed the whole joint pathology in the knee joint, then assessed the medial and 198 lateral tibial plateau separately. The mean of the observer's scores was used as a 199 representative value. 200 201 Immunohistochemical Analysis 202 Immunohistochemical (IHC) staining was performed using anti-MMP-13 (1:200, bs-203 0575R, Bioss). Detailed protocols are described in the Supplementary Materials. We 204 calculated the ratio between the number of MMP-13 positive cells and the number of 205 chondrocytes in the anterior and posterior area of the articular cartilage with regions of 206 interest of 40,000 µm2 (200 mm × 200 mm). As with the OARSI score, we initially evaluated 207 the whole joint and then subsequently assessed the medial and lateral tibial plateau separately. 208 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 13 209 Statistical Analysis 210 Statistical analysis was performed using RStudio. We conducted the Shapiro-Wilk test 211 to verify the normality of all data. Student’s t-test was used for the osteophyte formation and 212 Wilcoxon rank sum was used for the microscope and FRI data. One-way analysis of variance 213 was performed for the joint laxity test and MMP-13 IHC analysis initially, and then the 214 Tukey-Kramer test was used for post-hoc analysis. The Kruskal-Wallis test was used to 215 compare subchondral bone µCT data and OARSI scores, and the Steel-Dwass method was 216 used for the subsequent multiple comparisons. Parametric data are expressed as the mean ± 217 95% confidence intervals (95% CI); non-parametric data are expressed as the median ± 218 interquartile ranges. Statistical significance was set at p < 0.05. 219 220

221 AP joint Laxity Test of the Knee 222 Compared with the Intact group, both the Compression and Non-Compression ACL-223 R groups had significantly increased AP joint laxity at both 60 /i4 and 90 /i4 (Figure 1B). 224 However, no significant differences were observed between the Compression and Non-225 Compression groups at either angle. This suggests that any differences in OA progression 226 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint between the ACL-R groups can be attributed to the initial effect of compressive force at the 227 time of ACL injury. 228 229 [Insert Figure 1 here] 230 Figure 1. (A) Experimental design. We made the Compression ACL-R, Non-231 Compression ACL-R, and Intact groups. Mice were euthanized at 0, 3, 7, 14, and 28 days 232 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 15 after injury and following analyses were performed; joint laxity test, microscopic, microCT, 233 and histological analysis. At day 0 after injury, contralateral knee joints from the Non-234 Compression group were used as the Intact group. (B) A result of joint laxity test. The 235 Compression and Non-Compression ACL-R groups significantly increased Anterior-Posterior 236 joint laxity at both 60 /i4 and 90 /i4 . However, no significant differences were observed 237 between both injury groups. Data are presented as the mean ± 95% CI. *P< 0.05; ***P< 238 0.001. 239 240 Microscope Analysis of Cartilage Roughness in the Tibial Plateau 241 Surface roughness in the region not covered by menisci was increased in the 242 Compression ACL-R group, especially on the medial tibial plateau. The normalized 243 arithmetical mean height of the medial compartment in the Compression group was 244 significantly higher than that in the Non-Compression group, but no difference was observed 245 in the lateral compartment (Figure 2A). In addition, there were no significant differences in 246 the maximum height of the medial and lateral compartments between groups. These results 247 indicate that compressive force during compression ACL injury caused initial micro-damage 248 in the surface layer of articular cartilage. 249 250 Quantification of MMP Fluorescence I ntensity 251 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 16 At 3 days following injury, MMP-induced fluorescence intensity around ACL-R 252 knees increased slightly in the Compression (Normalized data: 0.31 ± 0.659) and Non-253 Compression groups (0.512 ± 0.472) compared with each contralateral intact knee, and no 254 significant difference was observed between groups (Figure 2B). Whereas, at 7 days, 255 fluorescence intensity increased moderately in the Compression (0.684 ± 0.597) and slightly 256 in the Non-Compression (0.17 ± 0.368) groups. However, there were no significant 257 differences between these groups. 258 259 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 260 Figure 2. (A) Microscope Analysis of Cartilage Roughness in the Tibial Plateau at 261 day 0 after injury. The ROI was set as an ellipse with 0.6-1.2 mm diameters in the center area 262 of the medial and lateral tibial plateau. Compression group significantly increased the 263 normalized arithmetical mean height in the medial compartment compared with the Non-264 Compression group. (B) Quantification of MMP Fluorescence Intensity. Normalized MMP-265 induced fluorescence intensity around ACL-R knees increased in the Compression, especially 266 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 18 at 7 days following injury. However, there were no significant differences between these 267 groups. Data are presented as the median ± interquartile range. *P< 0.05. 268 269 MicroCT Measurement of Osteophyte Formation 270 Mineralized osteophyte volume was quantified at 14 and 28 days because no 271 osteophytes were observed at 7 days following injury. Similarly, only a small amount of 272 mineralized osteophyte formation was observed in both ACL-R groups at 14 days (Figure 273 3A). At 28 days following injury, a greater volume of ectopic mineralized tissue was 274 observed in the medial femoral condyle and tibial plateau of injured joints. However, there 275 were no significant differences between ACL-R groups. 276 277 MicroCT Analysis of Epiphysis Trabecular Bone of the Distal Femur and Proximal 278 Tibia 279 At 0 days post-injury, no micro injuries or fractures were observed in the femoral and 280 tibial epiphysis trabecular bone in either ACL-R group, and there were no significant 281 differences in the microstructure between any experimental groups (Supplementary Figure 282 2A-B). This indicates that compressive force while inducing ACL rupture didn’t cause 283 concomitant structural failures in the subchondral trabecular bone. Femoral epiphysis 284 trabecular bone showed a significant decrease of BMD, BV/TV, and Tb.Th in the 285 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 19 Compression and Non-Compression groups compared with the Intact group at 7 days, which 286 suggests acute bone absorption at an early time point in both ACL-R groups (Figure 3C). 287 These parameters were still significantly lower in the Compression and Non-Compression 288 groups at 14 days. However, the Compression group recovered bone volume moderately at 28 289 days, and no significant difference in the BMD was observed between the Intact group and 290 the Compression group at this time point, although the Non-Compression group was still 291 significantly lower than the Intact group. Tb.N and Tb.Sp results are described in the 292 Supplementary Material (Supplementary Figure 3A). 293 Similarly, tibial epiphysis trabecular bone also exhibited a significant decrease in 294 BMD and BV/TV in the Compression and Non-Compression groups at 7 days following 295 injury (Figure 3D). At 14 days, the significantly decreased BMD and BV/TV was still present 296 in the Non-Compression group, whereas the Compression group slightly increased bone 297 volume and there was no significant difference in BV/TV between the Intact and the 298 Compression groups. Furthermore, in the Compression group at 28 days, the bone volume 299 increased to the same level as in the Intact group. The Non-Compression group still showed 300 deficits in bone microstructure at this time point, therefore the BMD and BV/TV in the Intact 301 and Compression group were significantly greater than in the Non-Compression group. Tb.N 302 and Tb.Sp results are described in the Supplementary Material (Supplementary Figure 3B). 303 304 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 20 305 Figure 3. (A) Reconstructed images obtained by microCT and measurement of 306 osteophyte formation. Only a small amount of mineralized osteophyte at 14 days and a greater 307 volume of ectopic mineralized tissue in the medial femoral condyle and tibial plateau of 308 injured joints at 28 days were observed in the Compression and Non-Compression groups. 309 However, there were no significant differences. Black arrows show the osteophyte. (B) The 310 ROI to measure the tibial and femoral epiphysis trabecular bone and (C) the result of 311 14-days 28-days l ar Bone CompressionNon-Compression Intact 7 days 14 days 28 days /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 BM D (g/cm 3 ) BV / TV (%) Tb.Th (m m ) N o n - C o m p r e s s i o n C o m p r e s s i o n N o n - C o m p r e s s i o n C o m p r e s s i o n e cular Bone CompressionNon-Compression Intact 7 days 14 days 28 days /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 /i1 BMD (g/cm 3 ) BV / TV (%) Tb.Th(mm ) Femoral Epiphysis Trabecular Bone Tibial Epiphysis Trabecular Bone A B D Lateral Medial ROI: Epiphysis Trabecular Bone Femur Tibia C 14-days 28-days Intact Non-Compression Compression 1.0mm 1.0mm 1.0mm 1.0mm 1.0mm 1.0mm .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 21 microCT Analysis. Femoral epiphysis trabecular bone significantly decreased BMD, BV/TV, 312 and Tb.Th in the Compression and Non-Compression groups compared with the Intact group 313 at 7 days. At 28 days, no significant difference in the BMD was observed between the Intact 314 and the Compression groups. However, the Non-Compression group was still significantly 315 lower than the Intact group at the same timepoint. Tibial epiphysis trabecular bone also 316 showed a significant decrease in BMD and BV/TV in both injury groups at 7 days. However, 317 at 28 days, the BMD and BV/TV in the Intact and Compression groups were significantly 318 greater than in the Non-Compression group. Data are presented as the median ± interquartile 319 range. *P< 0.05. **P< 0.01. 320 321 MicroCT Analysis of Subchondral Bone in the Medial and Lateral Tibial Compartment 322 Analysis of subchondral bone microstructure in the medial and lateral tibial 323 compartments at 7, 14, and 28 days (Figure 4 A) showed that in the medial compartment, the 324 Compression and the Non-Compression groups caused bone loss with a significant decrease 325 in BV/TV and Tb.Th compared to the Intact group at 7 days (Figure 4B). Although BV/TV 326 and Tb.Th in the Non-Compression group were still significantly lower than in the Intact 327 group at 14 days, the Compression group increased BV/TV and Tb.Th and no differences 328 were observed compared to other groups. Additionally, the Compression group exhibited 329 significantly increased BV/TV compared to the Non-Compression group at 28 days. 330 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 22 In the lateral compartment, both ACL-R groups exhibited significant bone loss 331 compared with the Intact group at 7 days similar to what was observed in the medial 332 compartment (Figure 4C). At 14 days, the Compression and Non-Compression groups 333 increased BV/TV and Tb.Th, and no differences were observed between any experimental 334 groups. Unlike the medial compartment, the Compression group exhibited a significant 335 increase of BV/TV and Tb.Th compared to the Intact group at 28 days. Tb.N and Tb.Sp 336

for both compartments are described in the Supplementary Materials (Supplementary 337 Figure 4A-B). 338 339 340 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 341 Figure 4. (A) The ROI of medial and lateral tibial trabecular bone and microCT 342 images. (B) Analysis of subchondral bone microstructure in the medial and lateral tibial 343 compartments. In the medial compartment, the Compression and the Non-Compression 344 groups caused bone loss with a significant decrease in BV/TV and Tb.Th compared to the 345 Intact group at 7 days. At 28 days, the Compression group showed significantly increased 346 BV/TV compared to the Non-Compression group. In the lateral compartment, both ACL-R 347 groups exhibited significant bone loss compared with the Intact group at 7 days. Unlike the 348 medial compartment, the Compression group exhibited a significant increase of BV/TV and 349 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 24 Tb.Th compared to the Intact group at 28 days. Data are presented as the median ± 350 interquartile range. *P< 0.05. **P< 0.01. 351 352 Histological Analysis of Cartilage Degeneration in the Medial and Lateral Tibial 353 Compartment 354 Histological analysis did not detect any acute injuries of articular tissues other than 355 the ACL (cartilage, menisci, subchondral bone) in the knee joint at 0 days following injury 356 (Supplementary Figure 5). In the medial tibial plateau at 7 days, over half of the mice in the 357 Compression group and some mice in the Non-Compression group developed a small loss of 358 cartilage around the posterior edge, and cell proliferation and staining of Safranin-O were 359 observed in the posterior region (Figure 5A). At 14 days, fibrillation of the cartilage surfaces, 360 mild to moderate cartilage erosion, and osteophytes on the posterior tibia were observed in 361 both ACL-R groups, especially in the posterior region. At 28 days, erosion extending to the 362 center and posterior region of the subchondral bone and growth plate were observed in both 363 ACL-R groups. These histological changes were observed in the lateral tibial plateau as well; 364 however, the degree of OA progression was milder compared to the medial tibial plateau at all 365 time points (Figure 5B). 366 The OARSI score for the whole joint pathology in the Compression group increased 367 significantly compared with the Intact group at 7 days (Figure 5C). Moreover, OARSI score 368 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 25 in both ACL-R groups increased significantly compared with the Intact group at 14 days and 369 28 days. For the medial tibial plateau, OARSI score in both ACL-R groups were significantly 370 higher than that in the Intact group at all time points (Figure 5D). For the lateral tibial plateau, 371 OARSI score in both ACL-R groups were significantly higher than that in the Intact group at 372 14 days and 28 days (Figure 5E). 373 374 [Insert Figure 5 here] scale bar 500µm 375 Figure 5. (A) Histological images of cartilage degeneration in the medial and lateral 376 tibial compartment. (B) The OARSI score for the whole joint pathology in the Compression 377 group was significantly higher than that in the Intact group at 7 days. At 14 and 28 days, both 378 Intact Non-Compression Compression OARSI Score 7-days 14-days 28-days /i1/i1 /i1 /i1/i1 /i1 Lateral Plateau Intact Non-Compression Compression 7-days 14-days 28-days /i1/i1 /i1/i1 /i1/i1 /i1/i1 /i1/i1 /i1 Medial Plateau Intact Non-Compression Compression OARSI Score OARSI Score 7-days 14-days 28-days /i1/i1 /i1/i1 /i1/i1 /i1/i1 /i1/i1 Whole Joint Intact Non-Compression Compression Intact Non-Compression Compression 7-days 14-days 28-days Medial Compartment Lateral CompartmentA B CD .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 26 ACL-R groups significantly increased the OARSI score compared with the Intact group. (C) 379 In the medial compartments, OARSI score in both ACL-R groups were significantly higher 380 than that in the Intact group at all time points. (D) In the lateral compartment, both ACL-R 381 groups caused a significant increase of OARSI score compared with the Intact group at 14 382 and 28 days. Data are presented as the median ± interquartile range. *P< 0.05. **P< 0.01. 383 Black scale bar, 500 µm. 384 385 Immunohistochemical Analysis of the Articular Cartilage 386 IHC analysis of MMP-13 was performed at 7 and 14 days following injury because 387 cartilage was highly degenerated by 28 days (Figure 6A-B). At 7 days, the number of MMP-388 13-positive cells of the whole joint in the Compression group increased significantly 389 compared with the Intact and Non-Compression groups (Figure 6C). Moreover, the Non-390 Compression group also significantly increased the positive cell ratio compared to the Intact 391 group. Analysis of the medial tibial plateau showed similar results to the whole joint (Figure 392 6D), whereas for the lateral tibial plateau, the number of positive cells in the Compression 393 group increased significantly compared with the other groups (Figure. 6E). At 14 days, the 394 positive cells ratio of the whole joint in both ACL-R groups increased significantly compared 395 with the Intact group. Similar results were observed in the medial and lateral tibial plateau. 396 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 397 Figure 6. (A-B) Immunohistochemical images of MMP-13 in cartilage in the medial 398 and lateral tibial plateau. (C-D) In the whole joint level and medial tibial plateau, 399 Compression group increased the MMP-13 positive cell rate significantly compared with the 400 Intact and Non-Compression groups at 7 days. Furthermore, the Non-Compression group also 401 significantly increased the positive cell ratio compared to the Intact group. At 14 days, the 402 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 28 positive cells ratio in both ACL-R groups increased significantly compared with the Intact 403 group. (E) In the lateral tibial plateau at 7 days, the number of MMP-13 positive cells in the 404 Compression group increased significantly compared with the other groups. At 14 days, both 405 ACL-R groups increased the MMP-13 positive cell rate significantly compared with the Intact 406 group. Data are presented as the mean ± 95% CI. **P< 0.01; ***P< 0.001. Black scale bar, 407 100 µm. (F) Schematic summarizing the mechanism of OA development following ACL 408 injury with or without compression concomitant compression force. Compression-ACL 409 rupture in mice caused initial micro-damage on the cartilage surface and early cartilage 410 degeneration with MMP-13 production, leading to later bone formation in the subchondral 411 bone. 412 413

414 The mechanism of OA development following ACL injury involves two distinct 415 processes: initial direct micro-injury of the cartilage surface as a result of the compressive 416 force experienced during ACL injury, and secondary joint instability due to the deficiency of 417 the ACL. In this study, we compared the conventional Compression-induced ACL-R and 418 Non-Compression ACL-R models, with the aim of clarifying the individual effects of the 419 initial injury response and secondary joint instability on OA progression after ACL injury. 420 Although no significant differences in the joint laxity were observed between ACL-R groups, 421 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 29 the roughness of the cartilage immediately after injury was increased significantly in the 422 Compression ACL-R group but not the Non-Compression group. Similarly, MMP-induced 423 fluorescence intensity at 7 days was higher in the Compression group than in the Non-424 Compression group. Moreover, the MMP-13 positive cell ratio in the Compression group was 425 also increased significantly compared to the Non-Compression group at 7 days and micro-CT 426 analysis revealed that osteophyte volume in the Compression group was slightly higher than 427 in the Non-Compression group. Furthermore, sclerosis of tibial SCB in the Compression 428 group developed significantly more than in the Non-Compression group at 28 days, especially 429 in the medial tibial compartment. 430 The ACL contributes to joint stability and ACL deficiency results in an increase in 431 mechanical stress [27]. Our assessment of AP joint laxity immediately after injury revealed 432 that the Compression and Non-Compression groups had significantly increased joint laxity 433 compared to the Intact group. However, no differences were observed between the ACL-R 434 groups (Figure 1B). This result indicates that any differences in OA progression between the 435 ACL-R groups can be attributed to the direct effect of compressive stress experienced while 436 inducing ACL rupture. 437 Interestingly, microscopic analysis demonstrated that roughness on the surface of the 438 medial tibial plateau in the Compression group was significantly increased compared to the 439 Non-Compression group. However, no injuries such as cartilage erosion, meniscus injury, or 440 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 30 subchondral bone loss were observed histologically and morphologically immediately 441 following injury. A previous study using an intra-articular tibial plateau fracture mouse model 442 reported that compression force to the tibial plateau to a target load of 55 N at a rate of 60 N/s 443 caused massive soft tissue injuries, including to the synovium, meniscus, tendon, and 444 cartilage [28]. The mean compressive force applied with the Compression ACL-R model was 445 approximately 10 N at a loading rate of 1 mm/s [18]. Thus, the compressive force in the joint 446 during Compression ACL-R was remarkably lower than in the previous study using the tibial 447 plateau fracture model, therefore roughness on the cartilage surface indicated mild articular 448 damage in the Compression group rather than severe damage. 449 In addition to the initial micro-injury on the cartilage surface induced by compressive 450 loading, we also measured the biological response in the early stage using FRI with an MMP-451 activatable probe. MMP-3, MMP-9, and MMP-13 levels increased in animal models of 452 osteoarthritis [29-33], and MMP-2, MMP-3, MMP-9, and MMP-13 increased in in vitro 453 models of osteoarthritis [34], which can degrade collagens and proteoglycans of the articular 454 cartilage. Compressive loading during ACL injury increases the production of matrix-455 degrading enzymes and inflammatory cytokines in chondrocytes, which results in 456 chondrocyte apoptosis and cartilage degeneration [6, 35, 36]. A previous animal study 457 reported that even a single compressive load at 6 N without ACL rupture caused cartilage 458 degeneration with chondrocyte apoptosis [20]. However, in the current study no significant 459 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 31 differences were observed between ACL-R groups; the Compression group showed a high 460 MMP-induced fluorescence intensity using FRI at 7 days following injury. Furthermore, IHC 461 analysis demonstrated that MMP-13 expression in chondrocytes of the Compression group 462 increased significantly compared to the Non-Compression group. These molecular changes 463 may be due to the initial micro-injury detected as surface roughness, which is consistent with 464 early stage pathology of ACL injury in human patients. Interestingly, histological data also 465 showed that the OARSI score in the whole joint of the Compression group significantly 466 increased compared to the Intact group at 7 days. These results suggested that the 467 compressive force on the cartilage surface while inducing ACL rupture may have accelerated 468 cartilage degeneration in the early stage through the activation of MMPs. Alternatively, 469 considering that there was no difference in OARSI score at 14 and 28 days, it is also possible 470 that secondary joint instability is a more important factor in the progression of cartilage 471 degeneration after the initial stage. 472 Subchondral bone provides mechanical and nutritional support for cartilage, and 473 microenvironmental changes in subchondral bone might affect cartilage metabolism directly 474 or indirectly [37]. Generally, subchondral bone reacts to mechanical stress and is faster to 475 remodel than articular cartilage due to its robust innervation and blood supply that provide it 476 with a high capacity of turnover [38, 39]. We hypothesized that subchondral bone would react 477 to compressive force and induced structural changes during the early stage following ACL 478 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 32 injury. Indeed, we observed loss of subchondral bone volume in both ACL-R groups 479 following injury. However, µCT analysis showed that the Compression group induced 480 subchondral bone sclerosis in the medial tibial plateau compared to the Non-Compression 481 group at 28 days, and the lesion area was consistent with the cartilage degeneration. Some 482 reports have shown that cartilage degeneration precedes subchondral bone changes [40-42] 483 and osteoarthritic chondrocytes enhanced osteoblast differentiation in subchondral bone via 484 the ERK1/2 pathway [43]. Our results revealed that ACL injury with concomitant 485 compressive force initially caused micro-injury on the cartilage surface, then subchondral 486 bone remodeling and sclerosis in the medial compartment at an earlier time point than ACL 487 injury without compressive loading (Figure 6F). This earlier switch from bone loss to bone 488 sclerosis is consistent with an accelerated PTOA progression in the Compression ACL-R 489 group compared to the Non-Compression group. 490 There are two main limitations to this study. First, articular surface roughness 491 analysis was only performed immediately after the induction of each model. Microscopic 492 evaluation can determine the degree of cartilage degeneration as a surface rather than a line in 493 more detail. Roughness analysis was able to detect micro-injury on the cartilage surface at 0 494 days (Figure 2A), which was not possible to see with histological observation (Supplementary 495 Figure 5). Further roughness analysis at additional time points may enable us to reveal the 496 reaction of cartilage degeneration to compression force, especially in the early phase. Second, 497 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 33 we didn’t investigate the molecular biological mechanism of subchondral bone remodeling 498 and sclerosis during PTOA progression. It has been shown that the proliferation and 499 differentiation of osteoprogenitors are stimulated by platelet-derived growth factor-A, 500 transforming gene factor-β 1, and fibroblast growth factor-1, resulting in subchondral bone 501 formation [44]. We revealed that chondrocytes reacted to initial compression force and 502 increased MMP-13; therefore, additional experiments about the biological mechanism should 503 be performed to develop further understanding of these mechanisms. 504 In conclusion, concomitant joint contact while non-invasively inducing ACL rupture 505 in mice caused initial micro-damage on the cartilage surface and early cartilage degeneration 506 with MMP-13 production, leading to later bone formation in the subchondral bone. 507 Understanding the initial pathology of the ACL injury may be an important indicator of 508 disease etiology and represent a potential preventative approach for mitigating secondary OA 509 development. 510 511 Acknowledgments 512 The author(s) received no financial support for the research, authorship, and/or publication of 513 this article. We would like to thank Editage (www.editage.com) for English language editing. 514 The study design and summary scheme were created with BioRender.com. Research reported 515 in this publication was partially supported by the National Institute of Arthritis and 516 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 34 Musculoskeletal and Skin Diseases, part of the National Institutes of Health, under Award 517 Number R01 AR075013. 518 519 Author contributions 520 All authors approved the final submitted manuscript. 521 Study design: TK and BC 522 Making model and Data collection: KT and YL 523 Mechanical analysis: KT and BO 524 Fluorescent reflectance imaging: KT and YL 525 Morphological analysis: KT, YL, and KA 526 Histological analysis: KT, KA, and SE 527 Manuscript composition: KT, TK, and BC 528 529 Role of funding source 530 531 Competing interest statement 532 All authors have no conflicts of interest related to the manuscript. 533 534

535 1. Disease, G.B.D., I. Injury, and C. Prevalence, Global, regional, and national incidence, prevalence, and 536 years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990-2017: a 537 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 35 systematic analysis for the Global Burden of Disease Study 2017. Lancet, 2018. 392(10159): p. 1789-538 1858. 539 2. Thomas, A.C., et al., Epidemiology of Posttraumatic Osteoarthritis. J Athl Train, 2017. 52(6): p. 491-540 496. 541 3. Lohmander, L.S., et al., High prevalence of knee osteoarthritis, pain, and functional limitations in 542 female soccer players twelve years after anterior cruciate ligament injury. Arthritis Rheum, 2004. 543 50(10): p. 3145-52. 544 4. Sanders, T.L., et al., Incidence of Anterior Cruciate Ligament Tears and Reconstruction: A 21-Year 545 Population-Based Study. Am J Sports Med, 2016. 44(6): p. 1502-7. 546 5. Friel, N.A. and C.R. Chu, The role of ACL injury in the development of posttraumatic knee 547 osteoarthritis. Clin Sports Med, 2013. 32(1): p. 1-12. 548 6. Riordan, E.A., C. Little, and D. Hunter, Pathogenesis of post-traumatic OA with a view to intervention. 549 Best Pract Res Clin Rheumatol, 2014. 28(1): p. 17-30. 550 7. Whittaker, J.L., et al., Risk factors for knee osteoarthritis after traumatic knee injury: a systematic 551 review and meta-analysis of randomised controlled trials and cohort studies for the OPTIKNEE 552 Consensus. Br J Sports Med, 2022. 56(24): p. 1406-1421. 553 8. Yuan, X.L., et al., Bone-cartilage interface crosstalk in osteoarthritis: potential pathways and future 554 therapeutic strategies. Osteoarthritis Cartilage, 2014. 22(8): p. 1077-89. 555 9. Ajuied, A., et al., Anterior cruciate ligament injury and radiologic progression of knee osteoarthritis: a 556 systematic review and meta-analysis. Am J Sports Med, 2014. 42(9): p. 2242-52. 557 10. Thorstensson, C.A., et al., Choosing surgery: patients' preferences within a trial of treatments for 558 anterior cruciate ligament injury. A qualitative study. BMC Musculoskelet Disord, 2009. 10: p. 100. 559 11. Ferrero, S., et al., Impact of anterior cruciate ligament surgery on the development of knee 560 osteoarthritis: A systematic literature review and meta-analysis comparing non-surgical and surgical 561 treatments. Osteoarthr Cartil Open, 2023. 5(3): p. 100366. 562 12. Goldring, M.B. and M. Otero, Inflammation in osteoarthritis. Current Opinion in Rheumatology, 2011. 563 23(5): p. 471-478. 564 13. Liu-Bryan, R., Synovium and the Innate Inflammatory Network in Osteoarthritis Progression. Current 565 Rheumatology Reports, 2013. 15(5). 566 14. Mathiessen, A. and P.G. Conaghan, Synovitis in osteoarthritis: current understanding with therapeutic 567 implications. Arthritis Research & Therapy, 2017. 19(1). 568 15. Scanzello, C.R. and S.R. Goldring, The role of synovitis in osteoarthritis pathogenesis. Bone, 2012. 569 51(2): p. 249-257. 570 16. Ramme, A.J., et al., A novel rat model for subchondral microdamage in acute knee injury: a potential 571 mechanism in post-traumatic osteoarthritis. Osteoarthritis Cartilage, 2016. 24(10): p. 1776-1785. 572 17. Brown, S.B., et al., Characterization of Post-Traumatic Osteoarthritis in Rats Following Anterior 573 Cruciate Ligament Rupture by Non-Invasive Knee Injury (NIKI). J Orthop Res, 2020. 38(2): p. 356-367. 574 18. Christiansen, B.A., et al., Musculoskeletal changes following non-invasive knee injury using a novel 575 mouse model of post-traumatic osteoarthritis. Osteoarthritis Cartilage, 2012. 20(7): p. 773-82. 576 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 36 19. Maerz, T., et al., Articular cartilage degeneration following anterior cruciate ligament injury: a 577 comparison of surgical transection and noninvasive rupture as preclinical models of post-traumatic 578 osteoarthritis. Osteoarthritis Cartilage, 2016. 24(11): p. 1918-1927. 579 20. Wu, P., et al., Early response of mouse joint tissue to noninvasive knee injury suggests treatment 580 targets. Arthritis Rheumatol, 2014. 66(5): p. 1256-65. 581 21. Takahata, K., et al., Joint instability causes catabolic enzyme production in chondrocytes prior to 582 synovial cells in novel non-invasive ACL ruptured mouse model. Osteoarthritis Cartilage, 2023. 31(5): 583 p. 576-587. 584 22. Hsia, A.W., et al., Osteophyte formation after ACL rupture in mice is associated with joint 585 restabilization and loss of range of motion. J Orthop Res, 2017. 35(3): p. 466-473. 586 23. Takahata, K., et al., Joint instability causes catabolic enzyme production in chondrocytes prior to 587 synovial cells in novel non-invasive ACL ruptured mouse model. Osteoarthritis Cartilage, 2022. 588 24. Lockwood, K.A., et al., Comparison of loading rate-dependent injury modes in a murine model of post-589 traumatic osteoarthritis. J Orthop Res, 2014. 32(1): p. 79-88. 590 25. Satkunananthan, P.B., et al., In vivo fluorescence reflectance imaging of protease activity in a mouse 591 model of post-traumatic osteoarthritis. Osteoarthritis Cartilage, 2014. 22(10): p. 1461-9. 592 26. Glasson, S.S., et al., The OARSI histopathology initiative - recommendations for histological 593 assessments of osteoarthritis in the mouse. Osteoarthritis Cartilage, 2010. 18 Suppl 3: p. S17-23. 594 27. Kraeutler, M.J., et al., Anatomy and Biomechanics of the Native and Reconstructed Anterior Cruciate 595 Ligament: Surgical Implications. Journal of Bone and Joint Surgery, 2017. 99(5): p. 438-445. 596 28. Stiffel, V., et al., A Mouse Noninvasive Intraarticular Tibial Plateau Compression Loading-Induced 597 Injury Model of Posttraumatic Osteoarthritis. Calcif Tissue Int, 2020. 106(2): p. 158-171. 598 29. Hu, H., et al., Blocking of the P2X7 receptor inhibits the activation of the MMP-13 and NF-kappaB 599 pathways in the cartilage tissue of rats with osteoarthritis. Int J Mol Med, 2016. 38(6): p. 1922-1932. 600 30. Raghu, H., et al., CCL2/CCR2, but not CCL5/CCR5, mediates monocyte recruitment, inflammation and 601 cartilage destruction in osteoarthritis. Ann Rheum Dis, 2017. 76(5): p. 914-922. 602 31. Qu, R., et al., Ghrelin protects against osteoarthritis through interplay with Akt and NF-kappaB 603 signaling pathways. FASEB J, 2018. 32(2): p. 1044-1058. 604 32. Schmidli, M.R., et al., Inflammatory pattern of the infrapatellar fat pad in dogs with canine cruciate 605 ligament disease. BMC Vet Res, 2018. 14(1): p. 161. 606 33. Li, H., et al., TNF-alpha increases the expression of inflammatory factors in synovial fibroblasts by 607 inhibiting the PI3K/AKT pathway in a rat model of monosodium iodoacetate-induced osteoarthritis. 608 Exp Ther Med, 2018. 16(6): p. 4737-4744. 609 34. Jiang, Y., et al., BRD4 has dual effects on the HMGB1 and NF-kappaB signalling pathways and is a 610 potential therapeutic target for osteoarthritis. Biochim Biophys Acta Mol Basis Dis, 2017. 1863(12): p. 611 3001-3015. 612 35. Carbone, A. and S. Rodeo, Review of current understanding of post-traumatic osteoarthritis resulting 613 from sports injuries. J Orthop Res, 2017. 35(3): p. 397-405. 614 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint 37 36. Wang, L.J., et al., Post-traumatic osteoarthritis following ACL injury. Arthritis Res Ther, 2020. 22(1): 615 p. 57. 616 37. Hu, Y., et al., Subchondral bone microenvironment in osteoarthritis and pain. Bone Res, 2021. 9(1): p. 617 20. 618 38. Sniekers, Y.H., et al., A role for subchondral bone changes in the process of osteoarthritis; a micro-CT 619 study of two canine models. BMC Musculoskelet Disord, 2008. 9: p. 20. 620 39. Botter, S.M., et al., Cartilage damage pattern in relation to subchondral plate thickness in a 621 collagenase-induced model of osteoarthritis. Osteoarthritis Cartilage, 2008. 16(4): p. 506-14. 622 40. Panula, H.E., et al., Subchondral bone remodeling increases in early experimental osteoarthrosis in 623 young beagle dogs. Acta Orthop Scand, 1998. 69(6): p. 627-32. 624 41. Kwon, M., D. Nam, and J. Kim, Pathological Characteristics of Monosodium Iodoacetate-Induced 625 Osteoarthritis in Rats. Tissue Eng Regen Med, 2023. 20(3): p. 435-446. 626 42. Intema, F., et al., In early OA, thinning of the subchondral plate is directly related to cartilage damage: 627

from a canine ACLT-meniscectomy model. Osteoarthritis Cartilage, 2010. 18(5): p. 691-8. 628 43. Prasadam, I., et al., Osteoarthritic cartilage chondrocytes alter subchondral bone osteoblast 629 differentiation via MAPK signalling pathway involving ERK1/2. Bone, 2010. 46(1): p. 226-35. 630 44. Hu, W., et al., Microenvironment in subchondral bone: predominant regulator for the treatment of 631 osteoarthritis. Ann Rheum Dis, 2021. 80(4): p. 413-422. 632 633 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted May 10, 2024. ; https://doi.org/10.1101/2024.05.08.593114doi: bioRxiv preprint