Micro-Computed Tomography Analysis of the Knee in Aged Dunkin-Hartley Guinea Pigs after Intra Articular Injection

Summary

Dunkin-Hartley guinea pigs are an established animal model for osteoarthritis research. Such studies may benefit from intra-articular injections for various reasons, including investigating novel agents or treating disease. We describe a methodology for intra-articular knee injections in Guinea pigs and subsequent micro-computed tomography analysis assessing arthritis-associated knee changes.

Abstract

The purpose of this protocol is to guide researchers in performing a palpation-guided technique of intra-articular knee injection in guinea pigs and assessment using micro-computed tomography. Dunkin-Hartley guinea pigs are robust models for osteoarthritis research as they spontaneously develop osteoarthritis in their knees. Intra-articular drug delivery is a common method to study the effects of an investigational drug in vivo. In humans, therapeutic agents administered via intra-articular injection can offer pain relief and delay further progression of osteoarthritis. As with any species, the introduction of a needle into a joint space has the potential to cause injury, which can result in pain, lameness, or infection. Such adverse events can compromise animal welfare, confound study results, and necessitate additional animals to achieve study objectives. As such, it is imperative to develop proper injection techniques to prevent complications, especially in longitudinal studies that require multiple, repeated intra-articular injections. Using the presented methodology, five guinea pigs received bilateral knee injections under general anesthesia. Seven days after injection, animals were humanely euthanized for analysis of osteoarthritis severity. No adverse events occurred following anesthesia or knee injections, including limping, pain, or infection. X-ray micro-computed tomography analysis of the knee can detect pathologic changes associated with osteoarthritis. Micro-computed tomography data indicates osteoarthritis is more severe in older animals, as indicated by increased bone mineral density and trabecular thickness with age. These results are consistent with histologic changes and Modified Mankin scores, an established and widely used scoring system to assess arthritis severity in these same animals. This protocol can be utilized to refine intra-articular injections in guinea pigs.

Introduction

Osteoarthritis (OA) impacts 32.5 million US adults. It is caused by progressive loss of articular cartilage, mild inflammation of the tissues in and around the joints, and formation of osteophytes and bone cysts^1,2. Symptoms typically manifest in the later stages of the disease, with current treatments providing only palliative relief as well as having systemic side effects. The lack of disease-modifying drugs stems from a poor understanding of the underlying mechanisms of the disease³. As a result, there is a critical and ongoing medical need for improved agents to treat OA.

Several animal models of OA are available that examine different components of the disease processes⁴. While several surgical models exist, including transection of the anterior cruciate ligament and destabilization of the medial meniscus, these are invasive and require a high level of technical skill⁵. Chemically induced models are comparatively less invasive procedures typically used to study OA pain mechanisms⁶. One such widely used mouse model involves OA induction by an intra-articular knee injection of monosodium iodoacetate (MIA). This model generates a reproducible, robust, and rapid pain-like phenotype that can be graded by altering MIA dosage⁷. Technical details of inducing this model have been previously described⁷. Translation of this technique to larger rodents, like guinea pigs, is difficult due to their anatomical differences. Some differences include increased musculature surrounding the adjacent bones and joint space in the guinea pig and an articulating fibula and tibia compared to distal fusion seen in mice⁸. Dunkin-Hartley guinea pigs, a widely available guinea pig strain, are an established OA animal model as they naturally develop this disease, thereby offering a robust model for investigating the effects of novel therapeutics administered by intra-articular injection on disease progression⁹. Dunkin-Hartley guinea pigs start developing OA at three months, with males displaying an accelerated development and more severe phenotype¹⁰. In guinea pigs, OA progresses with age, and at 12 months, associated pathology is apparent on imaging¹¹. Spontaneous OA models, like the Dunkin-Hartley model, do not require any intervention to induce OA and thus recapitulate the development and progression of the disease phenotype in humans, thereby providing a powerful translational model¹⁰. Furthermore, the spontaneous development of OA allows for the internal control when novel therapeutics are administered unilaterally in a single knee of a given animal. This internal control minimizes the effects of inter-animal variabilities when analyzing data and may help reduce overall animal numbers.

X-ray Micro Computed Tomography (µCT) analysis is a powerful tool that allows for quantitative assessment of OA severity¹². µCT involves scanning multiple, high-resolution X-ray images, obtained from a rotating sample or rotating X-ray source and detector¹³. Then, three- dimensional (3D) volumetric data is reconstructed in the form of stacked image slices¹⁴. Because mineralized bone has excellent contrast on µCT, this modality can be used to assess 3D features and perform quantitative analyses of changes associated with OA^15,16,17. µCT offers several advantages over more widely used tools, including histopathology and gait analyses. In contrast to histologic assessment of one or few sections of tissues, µCT scans the entire joint and offers a more wholistic assessment of OA lesions¹⁸. While gait analysis can discern symptomatic changes in joint function over time, joint changes develop long before functional changes associated with OA. µCT can provide a more sensitive measure of OA development prior to the onset of lameness. Two particularly relevant quantitative measurements include bone mineral density and trabecular thickness as both increase throughout the progression of OA^19,20. It can be helpful to split the analysis into subchondral plate and trabecular bone, as they have different features, to achieve more robust measurements and comparisons.

The overall goal of this method is to help researchers successfully perform intra-articular injections on guinea pigs. The presented protocol utilized five-(n=2), nine- (n=1), and 12- (n=2) month-old intact, male Dunkin-Hartley guinea pigs; procedures can be extrapolated to other guinea pig strains and ages requiring intra-articular knee injections. In spontaneous models of OA, like the Dunkin-Hartley model, disease progression and response to serial treatment is often monitored over long periods of times, spanning weeks to months⁹. This extended protocol results in multiple intra-articular injections, and thus it is important to have proper injection technique to prevent adverse events, including pain, lameness, or infections, all of which can impact animal welfare and confound study results while necessitating additional animals on study. The presented protocol describes methodology of intra-articular injections in guinea pigs and subsequent analysis of µCT data.

Protocol

All methods described here have been approved by the Institutional Animal Care and Use Committee of the Medical University of South Carolina. The study followed the principle of 3R.

1. Intra-articular injection preparations

1. Allow Dunkin-Hartley guinea pigs to acclimate to facility for at least one week prior to starting the experiment.
   NOTE: 5- (n=2), 9- (n=1), and 12- (n=2) month-old male guinea pigs were used. Males display an accelerated development and more severe phenotype of OA.
2. Shave the knee area with an electric razor.
   NOTE: Be careful of the nipples medially.
3. Anesthetize guinea pig in an isoflurane chamber delivering 3-5% isoflurane in O₂ mixture (flow rate 2.5 L/min) and then transfer the guinea pig to a nose cone connected to a non-rebreathing anesthesia circuit. Adjust isoflurane to maintain surgical plane of anesthesia during the injection, typically with an oxygen flow rate of 0.5-1 L/min and 1-3% isoflurane.
   NOTE: Intra-articular injections cause mild, momentary pain. Animals are anesthetized during the procedure to prevent perception of painful stimuli and improve injection accuracy. In the presented study, the administration of analgesic agents, including nonsteroidal anti-inflammatory drugs, would interfere with OA progression²¹. Due to the momentary pain, provided anesthesia and potential of analgesics to confound the model, analgesics were not administered unless animals displayed side effects, including limping and signs of pain on joint palpation after injection. Investigators should consider the use of analgesics for routine injections. Analgesics are recommended when side effects occur. Analgesic regimens should be discussed with the institutional veterinarian and approved by the IACUC prior to initiating studies.
4. Ensure guinea pig is at appropriate anesthesia depth by lack of toe pinch response.
5. Place sterile ocular lubricant on both eyes to prevent desiccation and injury.
6. Dilute betadine with sterile water to 10%.
7. Dilute 200 proof ethanol with sterile water to 70% ethanol.
8. Prepare solutions for injection, in a biosafety cabinet to maintain sterility. In the presented protocol, a sterile vehicle (1x phosphate buffered saline) was utilized to inject both knees. Solutions can be changed based on research objectives.
   NOTE: Make sure to dilute fresh solutions for injection immediately before the injection session to ensure sterility. Any unused solutions should be discarded at the end of each injection session.
9. Fill sterile one-time-use insulin syringes with solutions for injection. Take care to utilize the smallest volume achievable to prevent overloading the joint space with volume. In the present study, 50 µL was used.
10. Place guinea pig and nose cone on a clean surface with a heating pad for thermal support and padding under the head to elevate it slightly.
11. Don surgical gown, hair net, sterile gloves, and mask while performing the injection procedure.
12. Pour 10% betadine onto a cotton ball and wipe both knee areas.
13. Pour 70% ethanol onto a cotton ball and wipe both knee areas.
14. Repeat 1.12 and 1.13 two more times.
    NOTE: For demonstration purposes, the corresponding video shows cleaning the knee and injection site once with 10% betadine and 70% ethanol. The injection site was subsequently cleaned using circular motions two more times, alternating these solutions. Three serial scrubs with alternating scrub solutions and alcohol are recommended to achieve aseptic technique.

2. Intra-articular injection

1. Place the guinea pig in supine position for entirety of the procedure.
2. Don a new pair of sterile gloves and palpate the knee joint.
   NOTE: In the presented protocol and video, autoclaved nitrile gloves were utilized. Sterile gloves, including either autoclaved nitrile gloves or surgical gloves, should be utilized for aseptic technique.
3. Manually flex the knee to 90°.
4. Move finger distal to the patella to locate the groove of the distal aspect of joint space by flexing and extending the hindlimb.
   NOTE: The patella can be palpated in this position as a small, firm structure located directly over the joint space. The tibia can be felt as a bony structure distal to the patella. Once the location of the tibia and patella are determined, the joint, felt as a groove, is between them, distal to the patella and proximal to the tibia.
5. Insert the insulin needle carefully on the midline distal to the patella within the joint space. The needle should be inserted 1-2 mm below the skin to enter the joint space.
   NOTE: The largest access window for the joint space while the knee is flexed is on the anterior aspect of the limb on midline, directly distal to the patella. Injecting on midline in the anterior-to-posterior direction will aid in accurately injecting into the joint space without penetrating boney structures. Accurately injecting into the joint space can be achieved using a lateral-to-medial approach, although the access window is narrower especially when the knee is flexed.
6. Inject 50 µL of the solution into the joint slowly. Ensure that the needle inserts easily, and contents are injected without resistance.
   NOTE: Make sure not to insert the needle too deep as it can cause joint or bone damage and result in unwanted inflammation and/or pain. If the groove corresponding to the joint space is not found, the needle could penetrate the femur, patella, or tibia. Therefore, it is beneficial to confidently palpate the groove corresponding to the joint space to prevent peri-articular injections or injuries associated with penetrating boney structures. If a bubble develops at the injection site under the skin, the injection was too shallow and the fluid has entered the subcutaneous space. Depending on the properties of the agent utilized, the drug may enter the joint space via diffusion, or another injection attempt may be needed.
7. Once done discard the needle into sharps bin.
   NOTE: For practice and training purposes, inject liquid containing a dye into the joint space of a cadaver in a similarly sized rodent or guinea pig. Then, dissect the joint to confirm the location of the injection.
8. Massage the knee by flexing and extending the joint a few times promote diffusion of the drug within the joint space.
9. Repeat steps 2.1-2.5 once on the contralateral limb with 1x PBS solution.

3. Recovery from intra-articular injection

1. Turn off isoflurane and maintain 100% flow by oxygen until the animal regains consciousness.
2. Place the animal on a heating pad for thermal support until ambulatory.
3. Apply an ice pack to the knee for 30 s with a paper towel as a barrier to help decrease swelling from the injection.
4. Assess animal gait when ambulatory prior to returning them to housing.
   NOTE: If any gait abnormalities are noted, analgesics and supportive care may be warranted. It is advisable to assess their gait again several hours after recovery from anesthesia to ensure normal mobility.

4. Micro computed tomography (µCT) scan

1. For the tissue harvest, establish a surgical plane of anesthesia with 100% oxygen and 5% isoflurane mixture.
2. Confirm a surgical plane of anesthesia with the lack of response to a toe-pinch stimulus. Humanely euthanize the animal via administration of ≥ 150 mg/kg of pentobarbital intravenously according to institutional policies and approved animal use protocol.
   NOTE: In the presented protocol, each of the five guinea pig received one injection in both knees. Animals were anesthetized and humanely euthanized one week after the injection.
3. Harvest both hindlimbs by dissecting the skin away from the surrounding musculature.
4. Next, disarticulate the hindlimb with Rongeurs at the mid-shaft of the femur and proximal to the ankle.
   NOTE: The scanning bed and specimen holder used was unable to accommodate the entire hindlimb of an adult guinea pig. Large specimen holders are commercially available for larger specimen sizes.
5. Place the tissues in neutral buffered formalin solution for 72 h for fixation before performing µCT.
6. Open µCT scan software and place sample with formalin in a compatible container that will fit into the µCT specimen folder while maintaining tissue in the field of vision.
7. Calibrate µCT machine for dark field and light field exposures according to manufacturer recommendations.
8. Scan the sample with Al+Cu filter at 18 µm. Use rotation step 0.7° for 360° with offset camera.
   NOTE: The scan automatically saves.

5. Image processing for evaluating bone microarchitectural parameters

1. Download and install µCT reconstruction software for the reconstruction of µCT images.
2. Select the software folder and double click to open the software.
3. Select one slice from the µCT images by clicking on an image slice.
4. Choose the reconstruction file destination. Select Browse and create a new folder named Recon. The selected file format should be BMP(8).
5. Check Misalignment Compensation.
   NOTE: Usually, the estimation is close to correct, but it can be manually adjusted to shift the overlapping images so that the right and left edges align as closely as possible.
6. Under Settings, apply Smoothing, Beam Hardening, CS Rotation, and Ring Artifacts algorithms.
   NOTE: It can be helpful to choose preview image to determine the clarity before reconstructing. The fine-tuning setting can also be helpful to determine which settings are the best.
7. Select Start to begin processing the reconstruction.

6. Collection of microarchitectural data from reconstructed images

1. Download and install Dataviewer.
2. Select VOI and orient the sample to align vertically for easier analysis at a later time.
3. Save edited VOI as a new folder.
4. Download and install CTAnalyser for the bone property analysis of microarchitectural parameters.
   NOTE: The free version of CTAnalyser is limited in functionality, so is recommended to obtain a full license.
5. Split the analysis to subchondral plate and trabecular bone by saving them as separate range of images.
   NOTE: Splitting the analysis is not necessary, but because subchondral plate and trabecular bone have different features, separate analyses can assist with robust measurements and comparison.
6. Select the range of images to analyze, starting with subchondral plate by clicking on the image slice that you want to start with.
7. Select the region of interest for each image to ensure that it is encompassing the bone by clicking on the region of interest tab.
8. Select the Binary Selection tab. Adjust the histogram so that the background and the bone is completely separate.
9. Select the Bone Mineral Density (BMD) tab. Save that data into a new analysis data folder.
10. Select Custom Processing and go to Internal tab.
11. First perform Thresholding and select Automatic Otsu, then Run.
12. Then select Despeckle and choose Remove black speckles, then Run.
13. Repeat Despeckle and choose Remove white speckles, then Run.
14. Choose 3D analysis and select Basic values and Additional values.
15. Repeat steps 6.2.2-6.4.5 to reset the image for trabecular bone analysis.
    NOTE: Make sure the output file is in a new folder with the same file as the BMD data.

Results

Before performing intra-articular injections on live animals, the above protocol was practiced on three rat cadavers to ensure correct injection location. During the practice sessions, 50 µL of 70% new methylene blue dye was injected into both knee joints using the methodology described above. This equates to six practice injections. After injections, the knee joint was dissected by incising through the cranial aspect of the joint space, distal to the patella and through the patellar ligament, to visualize the joint space and verify location of dye deposition. Because new methylene blue dye is a bright blue solution, the location of the dye deposition can be visualized grossly. Representative images from cadaver injections are included in Figure 1. Figure 1A demonstrates the presence of dye within the joint space, indicative of a correct location and technique of injection. During these practice sessions the knee joint was successfully injected at all six attempts (100% success rate). For the demonstration purposes, examples of incorrect injection location are included in Figure 1B-C. If the injection is administered in the subcutaneous space, fluid will accumulate causing a bleb under the skin as seen in Figure 1B. If the joint space is not injected, there will be no dye present within the knee joint, as seen in Figure 1C.

Each of the 5 guinea pigs underwent a single bilateral knee injection procedure with 1x phosphate buffered saline as part of a pilot experiment to ensure feasibility of injection and subsequent µCT analyses of age-related OA changes. Animals were visually assessed at least once daily following the procedure for overall health and adverse events. None (0%) of these animals experienced adverse events associated with injection, including pain, lameness, or infection. Seven days post injection, animals were placed under a surgical plane of anesthesia and humanely euthanized for knee harvest and subsequent µCT analyses.

As previously published, µCT analysis can be utilized to assess knee changes in guinea pigs, including the ability to measure quantitative changes associated with OA severity over time^22,23. In the presented study, µCT was utilized to verify OA changes in guinea pigs of differing ages. These results can be used as baseline data for future studies assessing novel treatments intended to delay OA progression. The methodology described above will allow for standardization of µCT analyses in future studies. The subchondral plate (Figure 2A) and trabecular bone (Figure 2B) have drastically different bone properties. As such, these regions are analyzed separately to yield a more robust result one week after the injection.

Bone mineral density (BMD) is greater in 12-month-old guinea pigs compared to 5- and 9- month olds (Figure 3). The mean subchondral plate BMD was 0.898 g/cm³, 0.952 g/cm³, and 0.588 g/cm³, in 12-, 9- and 5-month-old guinea pigs, respectively. This demonstrates 1.53 times increase in subchondral BMD for the 12-month old guinea pigs compared to 5-month old. The mean trabecular bone BMD was 0.825 g/cm³, 0.839 g/cm3, 0.427 g/cm³ in 12-, 9- and 5-month-old guinea pigs, respectively. This represents 1.93 times increase in trabecular BMD for the 12-month old guinea pigs compared to the 5-month old. Overall, BMD increases in Dunkin-Hartley guinea pigs as they age from 5 months to 12 months of age (Figure 3).

There were also differences in trabecular thickness between guinea pigs of various ages (Figure 4). For mean trabecular thickness, there is 1.38 times increase for the 12-month-olds (0.558 mm) compared to the 9-month old (0.403 mm), as well as a 2.48 times increase in the 12-month-old (0.558 mm) compared to the 5-month old (0.225 mm) guinea pigs.Therefore, the mean of the trabecular thickness is increased in Dunkin-Hartley guinea pigs as they age from 5- to 12-months.

Histologic changes and Modified Makin scores, a validated and widely utilized approach for assessing OA changes in 2D, support the µCT findings (Figure 5). Modified Mankin scores increased as the guinea pigs age (Figure 5A). Histologically, signs of OA including proteoglycan loss, hypocellularity, and fissures, increase in prevalence as guinea pigs age from 5- to 12-months of age (Figure 5B-D).

Figure 1: Representative images of correct and incorrect injection locations. (A) Dissected rat knee with the presence of new methylene blue dye within the joint space. Scale included to the left of the knee for reference. (B) Knee demonstrating a shallow injection, resulting in a bleb forming in the subcutaneous space. (C) Knee from panel B dissected confirming a lack of new methylene blue dye within the joint space. Scale included to the right of the knee for reference. In all images, cranial is at the top of the image and caudal is at the bottom. Please click here to view a larger version of this figure.

Figure 2: Segmentation for subchondral plate versus trabecular bone measurement fields. The subchondral plate has different properties from the trabecular bone. Analyzing these regions separately allows for comparisons between different regions throughout the course of OA progression. Region A (red) depicts the coronal view for the region of the subchondral plate segment used for analysis. Region B (green)depicts the coronal view for the trabecular bone segment used for analysis. Please click here to view a larger version of this figure.

Figure 3: Changes in Bone Mineral Density in aging Dunkin-Hartley guinea pigs. Bone mineral density (BMD) was assessed in guinea pigs at 5 (n=2), 9 (n=1), and 12 (n=2) months of age. Measurements were taken at the subchondral plate as well as the trabecular bone for two different measurements. Please click here to view a larger version of this figure.

Figure 4: Changes in Trabecular thickness in aging Dunkin-Hartley guinea pigs. Trabecular thickness measurements were taken from 5 (n=2), 9 (n=1), and 12 (n=2) month old guinea pigs. Please click here to view a larger version of this figure.

Figure 5: Histology analysis of OA changes in Dunkin-Hartley guinea pigs. (A) Modified Mankin scores for histologic samples from 5- (n=2), 9- (n=1), and 12- (n=2) month old guinea pigs. The Modified Mankin score was calculated by adding together individual scores of cartilage structure, cellularity, tidemark, and osteophyte formation. (B) Representative histology image stained with toluidine blue from a 5-month-old guinea pig. (C) Representative histology image stained with toluidine blue from a 9-month-old guinea pig. black *= proteoglycan loss. D. Representative histology image stained with toluidine blue from a 12-month-old guinea pig. black *= fissures; white *= hypocellularity. Please click here to view a larger version of this figure.

Discussion

Despite recent advancements in symptomatic treatment of OA, there is a complete lack of therapeutic agents that prevent onset or delay progression of OA²⁴. Currently, the only cure for severe OA is joint replacement, which is costly, invasive, and can result in patient morbidity and mortality²⁵. As a result, there is a dire need for continued research with animal models of OA and the sustained development of novel therapeutics. Several animal models are available to study different components of OA, including spontaneous and chemically or surgically induced models. The injection of chemicals, like in the MIA mouse model, are vital for studying pain but do not allow investigators to follow the progression of OA development and intervene at desired timepoints throughout the course of the disease²⁶. Surgical methods require invasive procedures and fail to faithfully mirror the initiation of OA in humans²⁷. Dunkin-Hartley guinea pigs naturally develop OA, making them a widely available spontaneous model of OA²⁸. This spontaneous model is ideal for translational research studying mechanistic OA development as it requires no interventions and features pathologic changes that mirror those in humans⁶. Previously published works have utilized intra-articular injections to test the effects of novel therapeutics on the progression of spontaneous OA in Dunkin-Hartley guinea pigs^29,30,31.

While intra-articular injections in mice have been previously described, anatomical differences pose additional challenges to this procedure in guinea pigs⁷. Although intra-articular injections are integral to guinea pig OA studies, this procedure has yet to be systematically detailed. With the described methodology, we highlight key considerations for refining intra-articular injections in guinea pigs, thereby reducing the potential for procedure related complications. Common side effects following intra-articular injections in humans include pain, swelling, septic arthritis, and neuropathy and these should be considered as potential complications when performing injections in guinea pigs³². In our experience, side effects, including limping and pain on joint palpation, can occur when the needle is inserted too deeply into the joint space and/or contacted a boney structure during injection. Having cadavers to practice our injections on before performing them on live animals helped ensure more accurate injections. In refining our technique to ensure a shallow injection depth, we observed no side effects following intra-articular knee injections. Therefore, proper injection technique is paramount to ensuring animal welfare, particularly for studies requiring multiple injections over many months. Key steps in the procedure include using aseptic technique; sufficient training prior to performing survival procedures; achieving a surgical plane of anesthesia to prevent pain perception and movement upon injection; and ensuring appropriate injection depth, to avoid contact with bone distally and peri-articular injections which may require administering the compound again.

To systematically assess OA related changes, we describe µCT analyses, including 3D image renditions, allowing for visualization, segmentation, labeling, and objective measurements of affected tissue. The methods described have been effective and efficient in systematically assessing OA joint and bone changes over time. Our µCT results show that 12-month-old Dunkin-Hartley guinea pigs have more severe OA than 5-month-old guinea pigs, as indicated by increased bone mineral density and increased trabecular thickness. This demonstrated worsening of OA over time aligns with previously published results in this OA model, including µCT analyses that assessed bone mineral density and trabecular thickness¹³. Previous studies, and the presented representative results, demonstrate and increase in BMD and trabecular thickness over time as measured by µCT^22,23. In addition, our histologic assessment of the knees in two-dimension supports the validity of the µCT data. Histologically, signs of OA are more pronounced in the 9- and 12- month-old guinea pigs compared to the 5-month-old animals. Histologic changes associated with OA include proteoglycan loss, fissures, and hypocellularity.¹⁰ Each of these features can be seen in the older guinea pigs. Additionally, older guinea pigs have higher Modified Mankin scores, a widely utilized and validated method for assessing OA severity¹⁰. While our histologic analyses demonstrate expected OA changes, µCT provides a more in-depth assessment of OA changes than histologic review as it scans an entire sample instead of one or few tissue sections¹⁸. Furthermore, it provides more objective measurements of the extent of bony lesions in both 2- and 3-dimensions¹⁸. Researchers should consult existing guidelines when scoring histologic OA changes, like the Osteoarthritis Research Society International histologic assessment guidelines¹⁰. Researchers should consider additional ancillary tests, including gait analyses, for in depth characterization of OA development³⁴. While gait analyses can reveal functional changes over time, OA lesions develop well before the onset of lameness, making µCT a more sensitive measure of OA related changes during early stages of disease.

While Dunkin-Hartley guinea pigs provide a robust model for studying OA progression and the effects of novel therapeutics, there are limitations to this model. Spontaneous models typically require a longer study period compared to the rapid development of changes following surgical or chemically induced OA³⁵. This can result in increased cost and study duration. As previously described, onset of OA varies between animals of the same age, which may require assessing OA related changes serially³⁶. Despite this variation, researchers can expect all Dunkin-Hartley guinea pigs to develop OA starting at 3 months, showing changes in subchondral and trabecular bone, followed by progressive OA changes as the animal ages^10,19. Researchers may reduce the effect of inter-animal variability by applying treatments to one limb in each animal, allowing the contralateral limb to serve as an internal control. Although beyond the scope of the presented methodology, ultrasound guided injections may assist in further improving injection accuracy and has been utilized in guinea pigs³⁷. The proposed study herein describes the feasibility of palpation-guided injection, without utilizing this imaging modality, a technique widely utilized in the field³⁷. While µCT is a powerful tool to analyze pathology and bone structure, the availability of software that supports extensive analytical features can introduce inter-user variability in analyses³⁸. In the present study, the diameter of the µCT scanning bed was 6.35 cm, thereby preventing imaging of the entire guinea pig or joint in situ. Larger bed attachments are commercially available that could facilitate whole animal imaging. Although the software is generally user friendly and offers several suggested settings, it may be beneficial to manually edit settings for higher resolution and accuracy. As automated techniques continue to evolve, including machine learning, these manual adjustments can be replaced with new, automatic software techniques to enhance standardization³⁹. While the presented manuscript describes step-by-step instructions for the utilized µCT and corresponding software, these same analyses can be performed with other brands and software; researchers should consult the corresponding user manuals to optimize and standardize image analyses when utilizing differing machines and/or software. To reduce variability, it is important to maintain the same settings across all samples. Subjectivity of image analyses can be further reduced by replicating analyses by multiple, blinded experimenters.

The primary goal of this protocol is to provide detailed guidance on performing a technically challenging procedure in guinea pigs and subsequent µCT analyses of associated OA changes. Standardization of technique and data analyses will help minimize cost and increase efficiency.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
200 Proof Ethanol	Decon Laboratories	2701	sterilizing agent
3D.SUITE software	Bruker		μ-CT analyzing software
Betadine Surgical Scrub	Avrio Health	67618-151-16	sterilizing agent
Insulin syringe with needle	Ulticare	91008	to perform injections
Isoflurane	Piramal	803249	anesthesize animal
Neutral Buffered Formalin	Fisher Scientific	23-427098	Fix tissue
Nrecon Software	Bruker		μ-CT reconstruction software
Phosphate Buffered Saline	Cytiva	SH30258.01	control and diluting agent
SkyScan 1176	Bruker		to scan samples