Non-invasive ACL injury is a reliable and clinically relevant method for initiating post-traumatic osteoarthritis (PTOA) in mice. This injury method also allows for in vivo quantification of protease activity in the joint at early time points post-injury using protease-activatable near infrared probes and fluorescence reflectance imaging.
Traumatic joint injuries such as anterior cruciate ligament (ACL) rupture or meniscus tears commonly lead to post-traumatic osteoarthritis (PTOA) within 10-20 years following injury. Understanding the early biological processes initiated by joint injuries (e.g., inflammation, matrix metalloproteinases (MMPs), cathepsin proteases, bone resorption) is crucial for understanding the etiology of PTOA. However, there are few options for in vivo measurement of these biological processes, and the early biological responses may be confounded if invasive surgical techniques or injections are used to initiate OA. In our studies of PTOA, we have used commercially available near-infrared protease activatable probes combined with fluorescence reflectance imaging (FRI) to quantify protease activity in vivo following non-invasive compression-induced ACL injury in mice. This non-invasive ACL injury method closely recapitulates clinically relevant injury conditions and is completely aseptic since it does not involve disrupting the skin or the joint capsule. The combination of these injury and imaging methods allows us to study the time course of protease activity at multiple time points following a traumatic joint injury.
Osteoarthritis is a pervasive health issue that affects millions of people in the United States1. Post-traumatic osteoarthritis (PTOA) is a subset of OA that is initiated by a joint injury such as anterior cruciate ligament (ACL) rupture, meniscus injury, or intra-articular fracture2. The proportion of symptomatic OA patients that can be classified as PTOA is at least 12%3, and this etiology typically affects a younger population than idiopathic OA4. Mouse models of OA are crucial tools for investigating disease etiology and potential OA treatments on a much shorter timeline (4-12 weeks in mouse models compared to 10-20 years in humans). However, the methods to initiate OA in mice commonly involve invasive surgical techniques such as ACL transection5,6, removal or destabilization of the medial meniscus5,7,8,9,10,11,12,13,14,15,16, or a combination of the two17,18,19, which do not reproduce clinically relevant injury conditions. Surgical models also exacerbate inflammation in the joint due to disruption of the joint capsule, which could accelerate OA progression.
Non-invasive knee injury mouse models provide the opportunity to study biological and biomechanical changes at early time points post-injury and may yield more clinically relevant results20. Our lab has established a non-invasive injury model that uses a single externally applied tibial compression overload to induce anterior cruciate ligament (ACL) rupture in mice21,22,23,24. This non-invasive injury method is able to produce an aseptic joint injury without disrupting the skin or joint capsule.
Fluorescence reflectance imaging (FRI) is an optical imaging method that involves exciting a target with infrared light at a specific wavelength and quantifying the reflected light emitted at another wavelength. Commercially available protease-specific probes can be injected into animal models and FRI can then be used to quantify protease activity at specific sites such as the knee joint. This method has been widely used for in vivo detection of biological activities such as inflammation. The probes used for this application are fluorescently quenched until they encounter relevant proteases. Those proteases will then break an enzyme cleavage site on the probes, after which they will produce a near-infrared fluorescent signal. These probes and this imaging method have been extensively validated and used in studies of cancer25,26,27,28 and atherosclerosis29,30,31,32, and our group has used them for studies of the musculoskeletal system to measure markers of inflammation and matrix degradation23,24,33.
Together, non-invasive joint injury combined with in vivo FRI and protease activatable probes provide a unique ability to track inflammation and protease activity following a traumatic joint injury. This analysis can be done as early as hours or even minutes after injury, and the same animal can be assessed multiple times to study the time course of protease activity in the joint. Importantly, this imaging method may not be feasible when combined with surgical models of OA, since disruption of the skin and joint capsule results in a fluorescence signal that would confound the signal from within the joint.
All procedures described have been approved by the Institutional Animal Care and Use Committee at the University of California Davis. 3-month-old male C57BL/6J mice were used for the present study.
1. Non-invasive ACL injury
NOTE: ACL injury produced by an externally applied compressive load is a simple and reproducible method that closely recapitulates ACL injury conditions in humans. This protocol is written for a commercially available load frame instrument (see Table of Materials), but could be adapted for similar systems.
2. Animal preparation for FRI imaging
NOTE: For optical imaging, animal fur (particularly dark fur) is highly effective at blocking, absorbing, and scattering light, therefore, fur must be removed as much as possible from the area around the knee joints before imaging. A depilatory cream is typically more effective for fur removal than clippers. Nude or hairless mice don't require fur removal. However, fur removal is necessary for most commonly used mouse strains (e.g., C57BL/6). If possible, feed mice with low-fluorescence purified food before imaging. Standard mouse chow contains chlorophyll, which auto-fluoresces with a wavelength of around 700 nm and may affect data collection from the near-infrared FRI system.
3. Preparation of the probe solution
4. Retro-orbital injection
NOTE: Refer to Yardeni et al. regarding the details of this procedure37.
5. Fluorescence reflectance imaging
NOTE: The procedures in this section are specific to a commercially available optical imaging system (see Table of Materials) . Similar imaging can be performed with comparable systems.
After applying a single compressive force (1 mm/s until injury) on the lower legs of 3-month-old male C57BL/6J mice, ACL injury was consistently induced in all mice. The average compressive force at knee injury was approximately 10 N (Figure 1).
FRI analysis showed significantly greater protease activity in the injured joints of mice subjected to non-invasive ACL injury at 7 days following injury (Figure 2). FRI analysis of knee joints was also performed on mice that underwent surgical restabilization of the knee joint immediately after non-invasive ACL injury, similar to what has been previously described in rats35,36,38. This analysis showed considerably greater fluorescent signal in mice subjected to restabilization surgery than mice that did not have surgery at both 2 and 4 weeks post-injury. These data suggest that invasive surgical procedures can confound the analysis of protease activity in the joint.
Figure 1: Non-invasive ACL injury setup and a force-time plot during injury. (A,B) The mouse's lower leg is positioned vertically in the system, with the ankle placed in a notch of the top fixture and the knee joint placed in a shallow cup on the bottom fixture. The bottom fixture is locked into place with a set screw after manually applying a preload of 1-2 N. (C) Force-displacement plot, which shows ACL injury at approximately 9 N. Please click here to view a larger version of this figure.
Figure 2: Fluorescence reflectance imaging detecting protease activity in mouse knee joints. (A,B) Representative images of uninjured (A) and injured (B) mice following injury. (C) Average radiant efficiency of both knee joints for uninjured and injured mice one week following non-invasive ACL injury. Injured joints showed 43% greater average radiant efficiency compared to contralateral joints and joints from uninjured mice. (D) Normalized total radiant efficiency (R/L) for non-invasively injured mice and injured mice that were also subjected to joint restabilization surgery. A ~30%-80% greater radiant efficiency was observed in injured joints compared to contralateral joints at 1-4 weeks post-injury. In contrast, surgically operated joints exhibited ~300% greater radiant efficiency at week 4 compared to contralateral joints, suggesting a notable confounding effect of surgery. **P < 0.01. Please click here to view a larger version of this figure.
This protocol has established and rigorously described a reproducible, non-invasive method for inducing ACL injury in mice20,21,24,33. This simple and efficient injury method can be performed in just a few minutes, which facilitates high-throughput studies of PTOA. This injury method also closely recapitulates injury conditions relevant to human ACL injury. Surgical methods used to induce OA in mice may preclude the use of in vivo imaging methods to measure the time course and magnitude of protease activity in the joint following injury. In contrast, non-invasive OA mouse models (reviewed in20) combined with FRI provide a unique capability for in vivo imaging of protease activity in mouse knee joints following injury.
The inflammatory response following injury is critically important in OA progression. However, the methods used to analyze inflammation in the joint are typically expensive, time-consuming, and destructive. For example, techniques such as reverse transcription polymerase chain reaction (RT-PCR) or RNAseq can be used to quantify a wide array of genes in whole joints, individual tissues, or single cells. However, this method requires mice to be euthanized to acquire injured and uninjured knee joints. These mice cannot be analyzed at multiple time points, such as an early time point during the peak protease response (i.e., 3-14 days after injury) and a later time point when OA is more severe (i.e., 4-6 weeks after injury). In contrast, FRI combined with non-invasive joint injury provides the ability to analyze protease activity at multiple time points in the knee joints of mice in vivo39. This allows for longitudinal analysis of the same mice and makes FRI a relatively lower-cost outcome than RT-PCR or RNAseq. In addition, multiple probes or targets can be imaged simultaneously at different wavelengths, which can provide multiple results for different purposes. Measuring protease activity in the joint using FRI does not provide a rigorous quantification of all inflammatory processes that occur during OA progression, but the in vivo and longitudinal data provided by this method may still be useful for tracking the magnitude and time course of inflammatory protease activity following joint injury.
The fluorescence activatable probe solution used for FRI imaging of protease activity must be delivered intravenously (IV). The most common ways to perform IV injection in mice are tail vein injection and retro-orbital injection. Retro-orbital injection is often easier to perform and facilitates the necessary injection volume more easily than tail vein injection. Literature also indicates that retro-orbital delivery may cause less stress to mice with no difference in drug delivery or efficacy compared to the tail vein injection40. These findings suggest that retro-orbital injection is suitable for injecting the fluorescence activatable probe solution for FRI imaging.
The resolution of FRI is relatively low compared to some other imaging techniques, but the quantitative results can provide sufficient information on the time course and magnitude of the inflammatory protease response during OA progression. A limitation of this technique is that tissue autofluorescence could affect the results, but this problem can be solved with a thorough plan before the experiment (probe type, strain of mice, animal positioning, etc.). Unlike other preclinical imaging methods (e.g., microPET, microSPECT, microCT, MRI), FRI is not able to be directly translated to a clinical imaging modality due to the drastic differences in size between mice and humans since the depth of light penetration is limited. However, in preclinical studies using rodent models, the knee joint is close to the skin with minimal soft tissue coverage. Consequently, FRI is an effective tool for detecting protease activity in the knee joint of mice.
In conclusion, non-invasive ACL injury provides a simple and reproducible method to initiate PTOA in mice. This injury method also facilitates the use of protease-activatable probes and Fluorescence Reflectance Imaging for in vivo measurement of the time course and magnitude of inflammatory protease activity in mouse joints during OA progression. Future studies could use these techniques and the multiple commercially available near-infrared fluorescence activatable probes to investigate OA progression mechanisms in mice of different ages, sexes, and genetic backgrounds or to evaluate potential therapies for slowing or preventing OA progression after joint injury.
The authors have nothing to disclose.
Research reported in this publication was supported by the National Institute of Arthritis and Musculoskeletal and Skin Diseases, part of the National Institutes of Health, under Award Number R01 AR075013.
10x Phosphate-Buffered Saline | Tissue Protech | PBS01-32R | or equivalent |
Air Anesthetia System | Isoflurane vaporizor with induction chamber and nose cone | ||
Buprenorphine | Analgesic post-injury | ||
Depilatory Cream | Veet | B001KYPZ4G | or equivalent |
Fixtures | Custom-made knee fixture, ankle fixture, and platform | ||
IVIS Spectrum | Perkin Elmer | 124262 | Can also use comparable optical imaging system |
Kimwipes | Kimberly-Clark Corporation | 06-666 | or equivalent |
Living Image software | Perkin Elmer | ||
Materials testing systems | TA Instruments | Electroforce 3200 or equivalent | |
ProSense680 | Perkin Elmer | NEV10003 | Can also use other probes such as OsteoSense, MMPSense, Cat K, AngioSense, etc. |
Sterile Syringe with Needle | Spectrum Chemical Mfg. Corp. | 550-82231-CS | Covidien 1 mL TB Syringe with 28 G x 1/2 in. Needle, Sterile or equivalent |
Uniaxial load cell | TA Instruments | 20 N capacity | |
Vortex-Genie 2 | Scientific Industries, Inc. | SI-0236 | or equivalent |
WinTest software | TA Instruments | compatible with Electroforce 3200 |