Tuina Manipulation to Reduce Inflammation and Cartilage Loss in Knee Osteoarthritis Rats

Summary

We present a protocol for Tuina manipulation performed on plaster-immobilized-induced knee osteoarthritis rats. Based on the preliminary results, it suggested that the efficacy of the method relied on the reduction of inflammation and cartilage loss.

Abstract

Clinical trials suggest that Tuina manipulation is effective in treating knee osteoarthritis (KOA), while further studies are required to discover its mechanism. Therefore, the manipulation of animal models of knee osteoarthritis is critical. This protocol provides a standard process for Tuina manipulation on KOA rats and a preliminary exploration of the mechanism of Tuina for KOA. The press and kneading manipulation method (a kind of Tuina manipulation that refers to pressing and kneading the specific area of the body surface) is applied on 5 acupoints around the knee joint of rats. The force and frequency of the manipulation were standardized by finger pressure recordings, and the position of the rat during manipulation is described in detail in the protocol. The effect of manipulation can be measured by pain behavior tests and microscopic findings in synovial and cartilage. KOA rats showed significant improvement in pain behavior. The synovial tissue inflammatory infiltration was reduced in the Tuina group, and the expression of tumor necrosis factor (TNF)-α was significantly lower. Compared to the control group, chondrocyte apoptosis was less in the Tuina group. This study provides a standardized protocol for Tuina manipulation on KOA rats and preliminary proof that the therapeutic effects of Tuina may be related to reducing synovial inflammation and delayed chondrocyte apoptosis.

Introduction

Knee osteoarthritis (KOA) is a degenerative disease mainly manifested in joint pain. Fibrosis, cracking, ulceration, and loss of articular cartilage are the main causes of this disease¹. KOA has a high prevalence and may result in a profound impact on the daily life of patients, causing disability in severe cases. Among people aged 45-84 years, the prevalence of KOA increases with age, and the prevalence among people aged 85 years and above is 15%, with a predominance in women^2,3. In addition, KOA might bring a serious economic burden on both the individual and the society. A survey showed that direct health care costs on KOA per capita reached up to $8,858 ± $5,120 per year⁴. With the aging of society, KOA has become a worldwide health problem and a major social issue, as well as a topical issue for scientific research.

Evidence-based studies have shown the effectiveness of Tuina manipulation in treating KOA⁵. Tuina manipulation could relieve pain and improve dysfunction in KOA patients, the mechanism of which is related to anti-inflammatory effects^6,7. Scholars found that Tuina manipulation effectively inhibited the expression of inflammatory factors interleukin (IL)-β and 5-hydroxytryptamine and slowed down the degeneration of articular cartilage in a rabbit KOA model⁸. It suggested that Tuina could promote blood circulation and metabolism at the lesion site, which helped clear inflammatory factors such as IL-1, IL-6 and tumor necrosis factor (TNF)-α, thereby alleviating the clinical symptoms of KOA⁹. In addition, the passive movement of the joint through Tuina manipulation can promote the penetration and diffusion of synovial fluid into the articular cartilage and improves tissue nutrient metabolism¹⁰. Other studies suggested that Tuina manipulation can effectively improve the biomechanical indexes in KOA patients¹¹. Manipulations applied on soft tissues can improve stress distribution over limbs and enhance balance function^12,13. At the same time, with some joint adjustment manipulations, the alignment of the lower limbs can also be adjusted to correct abnormal gaits^14,15.

The mechanism of action of Tuina manipulation in treating KOA remains to be explored, and therefore, an experimental study is necessary. The key to the application of Tuina in experimental animals is the standardization of modeling, animal fixation and intervention methods¹⁶. The modeling method determines whether the experimental animal can exhibit the characteristics of the disease. Meanwhile, appropriate fixation methods can facilitate the intervention of the Tuina manipulation and better reflect the effect of Tuina. The standardization of intervention methods is the most difficult part of Tuina manipulation. In 2010, the basic system of Chinese acupuncture standards mentioned the acupuncture point standards for experimental animals, providing the possibility for acupuncture and Tuina operations in animal experiments¹⁷. However, there are still difficulties in standardizing Tuina manipulation. There are multiple types of Tuina manipulation¹⁸. The choice of the specific manipulation mainly depends on the disease to be treated and the therapeutic theories the performer prefers. In the study of Tuina for KOA, more attention has been paid to the point-pressing manipulation (pressing the specific acupoints with the thumb or elbow), Yizhichan pushing manipulation (a pushing manipulation by wiggling the thumb), and press and kneading manipulation (which refers to pressing and kneading the specific area of the body surface by finger or palm)¹⁹. Press and kneading manipulation is one of the most widely used Tuina manipulations, which combines pressing and kneading to move the subcutaneous tissue²⁰. Press and kneading manipulation applied on acupoints can promote blood circulation and relieve pain and represents the therapeutic effect of Tuina on KOA¹⁹.

In this protocol, the operation of press and kneading manipulation on KOA rats will be described in detail, including selected acupoints, intensity and frequency of the manipulation and the body position of the rat, so as to provide a reference for future research.

Protocol

This study has passed the animal ethics review conducted by the experimental animal ethics committee of Yueyang Hospital of integrated traditional Chinese and western medicine affiliated with Shanghai University of Traditional Chinese Medicine (YYLAC-2022-166).

1. Experimental animal preparation and grouping

1. Animal Preparation
   1. Rear a total of 10 healthy SPF SD female rats of 200-220 g at room temperature (18-21 °C), humidity 40%-50%, 12 h: 12 h circadian rhythm alternations. Conduct pain-related animal experiments in strict compliance with the relevant provisions of animal ethical guidelines and guidelines.
2. Grouping of animals
   1. Randomly divide the rats into the Tuina group and control group. Treat rats of the Tuina group with press and kneading manipulation for 21 days after modeling. Put rats of the control group in the same Tuina room and place them in a black cloth bag simultaneously while the Tuina group is undergoing treatment.

2. Modeling of animals

1. Anesthetizing the animal
   1. Use isoflurane for gas anesthesia. Place the rat in the induction box with an induction concentration of 3%. Wobble the box after laying the rat down and confirm the anesthetization when the rat rollover with no attempt to return to the prone position.
   2. Remove the rat from the induction box and fix its nose in the anesthetic mask. Adjust the isoflurane concentration to 2% to maintain anesthesia. Confirm the anesthetization when the rat does not respond when pinching its paws. Apply eye ointment to the rats to prevent dryness when rats are anesthetized, as the eyelids cannot be closed.
2. Modeling method²¹​
   1. Use shaving machine to remove the hair of the right hind limb. Place a medical cotton pad between the right ankle and hip joint of the rat. Fix the right knee joint at 180° extension with 5-6 layers of wet plaster bandage evenly. Spiral wrap the plaster bandage starting from the ankle and covering 1/3 of the previous one. Use a hair dryer to dry and harden the plaster.
   2. External wrap the plaster with denture base material after the plaster bandage dries and hardens to fix the plaster and prevent it from gnawing.
   3. Mix the denture base materials to make it sticky and adhere the mixture to the outside of the plaster (the mixture should not exceed the edge of the bandage, Figure 1). After the mixture becomes hard, turn off the anesthesia machine and wait for the animal to wake up naturally. Supervise the rats to prevent anesthesia accidents before the rats wake up.
   4. Fix the plaster appropriately on the right hind limb of the rat, as a tight fixation restricts blood circulation, while a loose fixation tends to fall off. Observe the blood circulation in the terminal limb. If swelling of the terminal extremity or a purple complexion is detected, promptly cut off part of the plaster to help restore circulation. Remake the plaster if it is broken and fails to maintain the lower limb extension.
   5. Remove the plaster after 3 weeks of continuous immobilization. Use surgical scissors to cut off the denture base material outside and plaster bandage. Rinse the lower limb of the rat with saline and dry it with gauze. If there are local skin lesions, sterilize with iodophor.
3. Model verification²²
   1. X-ray-based verification
      1. Perform an X-ray examination of the right knee 1 day after the end of modeling. Take anteroposterior radiographs in the supine position with hip flexion at 30°, knee extension at 0°, and hip abduction at 15°. Keep the patella directly in front of the knee and put the radiator tube 110 mm from the knee joint.
      2. Take lateral radiographs in the right lateral decubitus position with right hip flexion at 30° and right knee extension at 0°. Make the left limb hip flexion at 70°, and knee flexion at 45°, and put the radiator tube 110 mm away from the knee joint. Set the detection parameters as exposure voltage 50 kV, current 250 mA, exposure dose 32 mAs, and exposure time 128 ms.
      3. Compare with normal rats' X-ray, check that the modeled knee X-ray shows narrower joint space with osteophyte hyperplasia at the edge.
   2. Osteoarthritis Research Society International (OARSI) scoring²³
      1. Place the rat in the euthanasia box and perfuse CO₂ at the rate of 30%-70% cage volume per minute. Stop perfusing CO₂ after detecting that the rat is immobile, not breathing, and that the pupil is dilated. Observe for another 2 min to confirm death.
         ​NOTE: Cervical dislocation can be performed after CO₂-based euthanasia as a secondary form to confirm death. Fix the rat on the table and grasp its tail with one hand. Press down on the head of the rat with the thumb and index finger of the other hand. Confirm death when hearing the sound of a crack, and the rat loses movement and heartbeat at the same time.
      2. Fix the rat in a supine position with a syringe needle on a foam board with the right hind limb flexed in abduction and external rotation. Pinch up the skin around the knee joint with surgical scissors. Expose the muscles around the knee joint by cutting the skin and then cutting the subcutaneous fascia.
      3. Cut off the femur and tibia diaphysis with bone scissors and remove the right knee joint. Gently remove the extra soft tissues, such as muscles and ligaments, outside the joint.
      4. Fix the joint in 4% paraformaldehyde for 24-48 h at 4 °C. Decalcify the joint in 10% formic acid solution for 3 days until the bone tissue can be easily poked with a needle.
      5. Place and trim the decalcified tissue in the fume hood and transfer it to a dehydration box in the dehydration machine. Add 75% ethanol for 4 h, then 90% ethanol for 2 h, followed by 95% ethanol for 1 h, absolute ethanol for 30 min, another round of fresh absolute ethanol for 30 min, alcohol benzene for 5-10 min, xylene for 5-10 min, another round of fresh xylene for 5-10 min, wax for 1 h, another round of fresh wax for 1 h, and final round of fresh wax for 1 h for dehydration and transparent wax immersion.
      6. Then, place the tissue into the machine for embedding. Cut the wax block into wax slices of 4 µm after the paraffin has set and flatten the slice in warm water. Place the slice on a glass slide and dry. Store it at room temperature.
      7. Observe the cartilage sample and score it according to the OA cartilage histopathology grade assessment (Table 1)²³. If the score of rats after modeling is significantly higher than that of the normal rats, modeling was successful.

Table 1. OA cartilage histopathology grade assessment. Grade is depth progression into cartilage. Total score = Grade x Staging. 0 for normal joints, 24 for severe arthritis. Please click here to download this Table.

Figure 1. Rats immobilized in plaster. After the rats were anesthetized, their right lower limbs were wrapped with plaster bandages, fixed in the hyperextended position, and covered with a layer of denture base materials outside. Please click here to view a larger version of this figure.

3. Tuina manipulation

1. Application area
   1. Select a total of 5 acupoints on the right hind limb of rats, including ST34, ST35, SP10, EX-LE4 and BL40 (Figure 2). Locate the acupoints according to the principles of acupoint positioning in ²⁴.
2. Position for application
   1. Cut a black cloth into a 9 cm x 15 cm bag with one side opening and tighten the opening with a rope. Before the Tuina manipulation, gently pull the rat's tail to make it burrow into the bag and expose its hind limbs outside the bag.
   2. Use one hand to keep the rat in the prone position, holding its tail and hind leg while the other hand applies press and kneading manipulation on a specific acupoint (Figure 3).
3. Tuina manipulation
   1. Operate press and kneading manipulation on the 5 acupoints on the right hind limb for 2 min each. Place the performers thumb on the selected acupuncture point to perform rhythmic press and kneading, driving the skin and subcutaneous tissue together in a circular motion.
   2. Use finger pressure recordings (units in Newton) to ensure consistent intensity and frequency of the manipulation. Keep the intensity between 3-5 N and the frequency at 2 Hz (Figure 4). Apply the manipulation once a day for 21 days.

Figure 2. Acupoints position. SP10 is located 5 mm above the inner knee joint in rats. ST34 is located 5 mm above the outer knee joint in rats. EX-LE4 is located in the medial side of the knee ligament in rats. ST35 is located in the lateral side of the knee ligament in rats. BL40 is located at the midpoint of the transverse popliteal stripe. Please click here to view a larger version of this figure.

Figure 3. Tuina manipulation applied on rats. The rats were kept in a black bag with their hind limbs exposed. The performer held the rat's tail with the left hand while the right hand performed the manipulation. Please click here to view a larger version of this figure.

Figure 4. Finger pressure recordings. A device that records the force and frequency of finger pressure is used for real-time feedback on the intensity and frequency in the process of Tuina manipulation. (A) Pressure sensor and transmission equipment. (B) Finger pressure recordings. (C) The force recorded during Tuina manipulation. Please click here to view a larger version of this figure.

4. Pain Behavior Tests

1. Test pain behavior before and after modeling, and at 1 day after (D1), 7 days after (D7), 14 days after (D14) and 21 days after (D21) the Tuina manipulation, including mechanical withdrawal threshold and paw withdrawal latency tests.
2. Mechanical withdrawal threshold (MWT)
   1. Place the rats in a 20 cm x 10 cm x 20 cm transparent tempered glass cubicle located on a 40 cm high stage which is made up of a wire lattice with a size of 10 mm x 10 mm aperture. Keep the room temperature at 23 °C ± 2 °C.
   2. Settle the rats in the behavioral laboratory for at least 2 h per day during the behavioral test phase to avoid interference with the test results due to the animals' lack of adaptation to the environment at the beginning of the formal test. Place the rats in the behavioral laboratory for 30 min before the start of the formal test to facilitate their adaptation to the environment and to reduce distracting factors.
   3. Use electronic Von Frey fibers to measure the MWT. Stimulate the rat with the fiber at the center of its foot and withdraw the fiber when the rat shows obvious movements such as raising the leg and avoiding. The machine can automatically record the maximum pressure value (N) at this moment.
   4. Start the next stimulation in the same rat at least 15 s after the current stimulation. Do not exceed 5 s during each stimulation to prevent sensitization to tactile stimuli in the paws of rats. Repeat the test 5x until the difference between the three consecutive measurements is insignificant (within 10 N).
   5. Delete the values with large differences (the maximum and the minimum values) and take the average of the remaining three values as the mechanical withdrawal threshold.
3. Paw withdrawal latency (PWL)
   1. Place the rats in a small compartment of transparent tempered glass with a size of 20 cm x 10 cm x 20 cm. Cover the top of the compartment with a transparent glass cover with ventilation holes. Keep the temperature of the transparent glass plate at 28-30 °C, on which the compartment is placed.
   2. Settle the rats in this environment for at least 30 min to acclimatize before the start of each formal test. If rats urinate or defecate in the compartment, clean it with absorbent paper in time to avoid affecting the subsequent light radiation heat transfer.
   3. Focus the spotlight on the center of the rat's foot and press the Start button. Press the Stop button when the rat shows obvious behaviors such as foot retraction or paw licking and record the time at this point. The spotlight irradiation time should not exceed 20 s to avoid damage to the rat's skin.
   4. Perform the next irradiation on the same rat after at least 10 min to prevent sensitization. Measure 5x on each rat.
   5. Remove the values with large differences (the maximum and the minimum value). Take the average of the remaining values as the PWL.

5. Sample preparation

1. Perform euthanasia on the mouse as described in 2.3.2.1.
2. Synovial membrane preparation
   1. Fix the rat in the supine position with a syringe needle on a foam board with the right hind limb flexed in the abduction and external rotation. Pinch up the skin around the knee joint with surgical scissors and expose the muscles around the knee joint of rats by cutting the skin and then cutting the subcutaneous fascia.
   2. Take the patellar ligament as a landmark to strip the muscle groups above the patellar ligament. Incise the ending of the patellar ligament (at the tibial tuberosity) carefully and find the synovial membrane when the ligament is pinched upwards from below.
   3. Carefully cut off the synovial tissue with ophthalmic scissors and wash off the blood and synovial fluid with pre-cooled saline. Fix the synovial membrane in 4% paraformaldehyde for at least 48 h after absorbing water from the tissue surface with clean gauze.
   4. Prepare the sample as in steps 2.3.2.5- 2.3.2.6. Perform scoring as in step 2.3.2.7.
3. Performing hematoxylin and eosin staining
   1. Dewax to water with xylene for 20 min, replace with another round of fresh xylene for 20 min, followed by treating with anhydrous ethanol for 5 min, replace with another round of fresh anhydrous ethanol for 5 min, then add 90% ethanol by volume for 5 min, 80% ethanol by volume for 5 min, 70% ethanol by volume for 5 min, and finally distilled water for 5 min.
   2. Immerse the slide in hematoxylin staining solution for 3-8 min. Remove the slide and rinse off the stain with distilled water. Move it into the differentiation fluid (1% hydrochloric acid alcohol) for almost 30 s, so that the slide fades to pale blue. Rinse it with distilled water, and place in eosin staining solution for 1-3 min.
   3. Dehydrate the slide with 95% ethanol volume fraction for 5 min, replace with another round of fresh 95% ethanol for 5 min, followed by anhydrous ethanol for 5 min, replace with another round of fresh anhydrous ethanol for 5 min.
   4. Make the slide transparent by adding xylene for 5 min, replace it with another round of fresh xylene for 5 min, and then seal it with neutral resin.
4. Terminal-deoxynucleotidyl transferase-mediated nick end labeling (TUNEL) on cartilage
   1. Dehydrate the cartilage sample in anhydrous ethanol, 90% ethanol, 85% ethanol and 75% ethanol until the dewaxing hydration is complete. Soak in PBS for 5 min. Add 3% H₂O₂ dropwise for 10 min.
   2. Add proteinase K working solution dropwise and digest at 37 °C for 10 min. Add 20 µL of labeling buffer per slice to keep it moist and shake off excess liquid after preparing the working solution. Add 20 µL of working solution to each slide and incubate for 2 h at 37 °C in a wet box.
   3. Add 50 µL of the closure solution drop by drop and close for 30 min. Then add 50 µL of diluted biotinylated anti-digoxin antibody (1:100 dilution) dropwise and incubate at 37 °C for 2 h in a wet box. Add 10 µL of SABC antibody diluent (1:100 dilution) dropwise and incubate at 37 °C for 2 h in a wet box.
   4. Add DAB color development solution (50 µL each of reagents A, B and C in 1000 µL of distilled water) dropwise for 10-15 min. The color development completes when it is brownish-yellow granular.
   5. Re-stain with hematoxylin for 3 s. After gradient dehydration and transparent treatment, dry at room temperature and seal the slide carefully with neutral gum. Pay attention to avoid leaving bubbles and overflowing glue.
5. Immunohistochemical analysis of IL-1β and TNF-α
   1. Dewax the slides routinely in xylene and hydrate them in gradient alcohol. Inactivate the endogenous peroxidase in the sections with 3% H₂O₂. Place the slide holder in 95 °C citrate buffer (pH 6.0) and incubate in a water bath above 95 °C for over 20 min. Take out the incubation box and leave it at room temperature for at least 20 min.
   2. Incubate 5% normal goat serum with PBS for 10 min at 37 °C and shake off excess liquid. Add 150 µL of antibody I drop by drop and let stand at 37 °C for 1 h, then store overnight at 4 °C. Next day, re-warm at 37 °C for 45 min.
   3. Wash 3x with PBS for 5 min each. Cover the tissue on the slide with 3% BSA and seal it at 37 °C for 30 min. Add 150 µL of antibody II drop by drop and let stand at room temperature for 1 h. Wash 3x with PBS for 5 min each and add 150 pL of DAB color development liquid dropwise. Observe the degree of staining under the microscope until the sample becomes brownish-yellow even to the naked eye.
   4. Immediately, rinse with PBS for 10 min. Stain again with hematoxylin, dehydrate in gradient alcohol, make slides transparent in xylene, and seal with neutral gum.
6. Statistical analysis
   1. Immunohistochemically stained sections with positive expression of the associated protein are yellow or brownish-yellow. Score the intensity of positive expression by Image J software for each group of immunohistochemical sections and use the evaluation criteria of average optical density (AOD), calculated by IOD divided by area.
   2. Use analysis software for statistical analysis. Use t-test if the data conformed to normal distribution and chi-square and use non-parametric test if they did not conform to normal distribution. Analyze repeated measurement data by generalized estimating equations.

Representative Results

Pain behavior tests
The MWT results showed that the MWT of the right hind limb after modeling was significantly lower than before (p<0.05). Compared with the control group, the MWT of rats was significantly elevated after Tuina (p<0.05; Figure 5 and Table 2).

Figure 5. Results of the mechanical withdrawal threshold test (, Newton). There were 5 rats each in the Tuina group and the control group. The MWT of the rats at different time points is shown in the figure. After the modeling, the MWT of the rats decreased significantly, suggesting that the pain of the rats was aggravated, and the KOA model was successfully prepared. Subsequently, the MWT gradually improved, suggesting pain relief. The generalized estimation equation was used for statistical calculation. The difference in MWT between the Tuina group and the control group was statistically significant at day 21 compared to that after modeling. The comparison between the two groups was statistically significant at D7, D14 and D21, *p<0.05. Moreover, rats in Tuina group have higher MWT than that in the control group. Please click here to view a larger version of this figure.

Table 2. Mechanical Withdrawal Threshold ( , N). Comparison before and after modeling, ^#p<0.05. Comparison between Tuina group and control group, *p<0.05. Please click here to download this Table.

The PWL results showed that the PWL of the right hind limb after modeling was significantly shorter than before (p<0.05). Compared with the control group, the MWT of rats was significantly prolonged after Tuina (p<0.001; Figure 6 and Table 3).

Figure 6. Results of the paw withdrawal latency test (, second). There were 5 rats each in the Tuina group and the control group. The PWL of the rats at different time points is shown in the figure. After the modeling, the PWL of the rats decreased significantly, suggesting that the pain of the rats was aggravated, and the KOA model was successfully prepared. At first, the improvement of PWT in the Tuina group was slower than that in the control group. After D7, rats in the Tuina group improved rapidly and surpassed the control group at D21. The generalized estimation equation was used for statistical calculation. The difference in MWT between the Tuina group and the control group was statistically significant at day 21 compared to that after modeling. The comparison between the two groups was statistically significant at D7, D14 and D21, ***p<0.001. Please click here to view a larger version of this figure.

Table 3. Paw Withdrawal Latency (, s). Comparison before and after modeling, ^#p<0.05. Comparison between the Tuina group and the control group, ***p<0.001. Please click here to download this Table.

Histomorphological assay
Upon analyzing the synovial membrane in the control group, inflammatory cell infiltration and fibrous tissue hyperplasia were seen in synovial tissue. The synovial cells were disorganized, and a small amount of capillary hyperplasia was seen around the synovial tissue (Figure 7).

Figure 7. Microscopic observation of the synovial tissue. (A) Synovial tissue in the control group. The disorganized arrangement of synovial cells is seen in the figure. The proliferating vessels are marked with red arrows, and inflammatory cells are marked with black arrows. (B) Synovial tissue in Tuina group. Synovial cells were more neatly arranged. Inflammatory cells are mainly distributed at the edges rather than infiltrating inwards. The proliferating vessels are marked with red arrows, and inflammatory cells are marked with black arrows. Please click here to view a larger version of this figure.

In the Tuina group, synovial cells were neatly arranged, with a small amount of inflammatory cell infiltration, fibrous tissue hyperplasia, and a small amount of capillary hyperplasia visible at the tissue margins.

Upon analyzing the cartilage, it was seen that the TUNEL staining positive area in the Tuina group was significantly smaller than that in the control group, indicating that there were fewer apoptotic chondrocytes in the Tuina group (Figure 8).

Figure 8. TUNEL staining of the cartilage. (A) Cartilage in the control group. The red part of the figure is the positive area of TUNEL staining. (B) Cartilage in Tuina group. The red part of the figure is the positive area of TUNEL staining. The positive area in the Tuina group is significantly smaller than that in the control group. Please click here to view a larger version of this figure.

Immunohistochemistry
For TNF-α, there was a significant difference in synovial TNF-α expression between the two groups, and the expression in the Tuina group was significantly lower than that in the control group (Table 4).

Table 4. TNF-α Expression in Synovium (, *10^-2). Two independent sample t-test was used for statistical analysis, *p<0.05. Please click here to download this Table.

For IL-1β, there was no significant difference in the amount of synovial IL-1β expression between the two groups, as measured by the statistics of Image J. However, the mean value of the Tuina group was subtly lower, indicating less expression (Table 5).

Table 5. IL-β Expression in Synovium (, *10^-2). Two independent sample t-test was used for statistical analysis. Please click here to download this Table.

Discussion

This study provides a protocol for Tuina manipulation on KOA rats. Through pain behavior tests and histomorphological findings, it suggested that such a series of Tuina manipulation applied to KOA rats could reduce synovial inflammation and cartilage apoptosis, which could be a reference of Tuina manipulation on animal models of KOA.

There are several critical procedures during the protocol. First, it's important to choose an appropriate method for inducing the KOA model. There are various methods to induce KOA model, including joint cavity injection²⁵, intra-articular surgery^26,27, joint braking method^28,29, etc. Since KOA models induced by invasive methods would leave wounds near the joints, which may interfere with the effect of Tuina manipulation, we chose the non-invasive method of joint immobilization. The joint immobilization method can be divided into hyper-extension and hyper-flexion fixation. Shang et al. compared the effects of these two fixation methods and found there was little difference in the effect of cartilage apoptosis²³. However, their experimental subjects were rabbits, which are larger in size and relatively easy to fix compared to rats. Fixed plaster in a flexion position is more likely to fall off under rats gnawing. Therefore, we chose the immobilization method of hyper-extension fixation to induce KOA. We referred to He et al. and used the plaster bandage with a tongue depressor as the fixation device³⁰. However, the gnawing ability of rats was stronger than expected, and the device could not serve as a fixation. Later we found a kind of denture base material, which can form a moldable but hard shell on the outside of the plaster, and effectively reduce the wear and tear of the fixation device by rats gnawing. A tight fixation would affect the blood circulation of the lower limbs, while too loose fixation would be easy to fall off. Therefore, the circulation of the lower limbs of the rats should be observed daily during the 3 weeks of modeling. The fixation should be released when the lower limbs become swollen and purplish. Re-install the fixation when it is loose, and the rats can flex their knees. During this study, the rats were not restricted from socializing during modeling, but it was found that they would chew each other's fixation to help each other break free. Perhaps isolating the rats would have resulted in better modeling. However, isolation might contribute to the rats' depression and stereotypical behavior. 

We consider that the key point of Tuina manipulation is to keep the consistency of the intensity and frequency. Previous research concluded that the optimal intensity for animal Tuina should be 80% of its maximum tolerable intensity³¹. At this point, the rats should show signs of receiving mechanical stimulation, but with no signs of pain or paw retraction. We tested the maximum tolerable intensity of rats, which is around 5-8 N. So, we set the pushing intensity to 3-5 N, which may lead to better efficacy. The frequency of the manipulation is 2 Hz, according to the textbook of Tuina. Previous studies have not set a standardization for Tuina manipulation, and the intensity and frequency might vary after operating different acupoints and altering the left and right hands during the experiment, which can affect the accuracy of the results. The finger pressure recordings used in this experiment provide real-time observation of the intensity and frequency of the manipulation during the experiment, which can keep the consistency of the manipulation. Some scholars use machines to simulate Tuina manipulation and apply it on animals, which has the advantage of standardization of the manipulation³².

For the selection of acupoints, we referred to the study of Wang and Liu and chose five acupoints around the knee joint to facilitate manipulation^33,34. We chose the plaster-immobilized method to induce KOA, which mostly ends up with muscle atrophy and joint stiffness³⁵. ST34 and ST35 belong to the foot Yangming stomach meridian which is critical in treating atrophy (mostly manifested as weakness and muscle atrophy) in traditional Chinese medicine theory and has good adaptability for KOA induced by the plaster-immobilized method. EX-LE4 is a common acupoint for the clinical treatment of KOA, mostly used together with ST35. Tan et al. found that acupuncture and moxibustion on ST35, ST36, and EX-LE4 could reduce inflammatory factor expression in the synovial membrane³⁶. SP10 and BL40 are the core acupoints of clinical Tuina treatment for KOA^37,38, and SP10 is most frequently used in Tuina prescriptions for KOA³⁹.

A difficult part of Tuina manipulation is the fixation of rats. Rats may struggle or avoid manipulation, which increases the difficulty of performing the experiment. Although rat immobilizers can immobilize rats and expose their hind limbs, the development of KOA limits hind limb flexion and extension. Therefore, settling rats with immobilizers may lead to pain and incomplete exposure of the hind limbs, which may affect Tuina manipulation. We designed a single-exit cloth bag based on the rat's preference for dark environments and its tendency to burrow. Close the cloth bag when the rat enters, which is, the head and forelimbs are located in the bag, and its hind limbs are exposed outside. This method can effectively reduce the struggle of rats and keep them quiet during Tuina.

The results of the pre-experiment suggest that Tuina manipulation may reduce the inflammatory response of synovial tissue and decrease the apoptosis of chondrocytes, thus acting as a treatment for KOA. However, more studies are needed to verify this.

There are several shortcomings of this protocol. First, the small size of rats makes it difficult to locate the acupoints precisely. The operator's fingers are too large compared to the acupoints on rats, which may affect the effect of Tuina manipulation. We had considered adding a rubber particle of about 2 mm in diameter on the fingertip to increase the accuracy of stimulating acupoints. But the particles might lead to higher intensity of pressure and increase the difficulty in controlling the consistency. In addition, it was more likely to damage the finger pressure recordings, so we did not choose this method. Second, our study ignored the change of inflammatory factors in the synovial fluid and chondrocyte apoptosis, which may weaken the credibility of the study. We will intensify research in this area in the future. Third, press and kneading manipulation is one of the Tuina manipulations, and cannot encompass the effects of Tuina in treating KOA. Other Tuina manipulations implemented in animals need to be further explored, including joint movement manipulations, and rubbing manipulation. Moreover, positive control would better show the effectiveness of the manipulation, but it was not designed in this study, and we will add this in a follow-up study.

Overall, this study provides a standardized protocol for Tuina manipulation on KOA rats, and the effectiveness of the manipulation has been verified by pain behavior tests and microscopical findings. The protocol preliminary proves that the therapeutic effects of Tuina may be related to reducing synovial inflammation and delayed chondrocyte apoptosis. More research is needed to explore the mechanism of Tuina for KOA.

Materials

Name	Company	Catalog Number	Comments
absolute ethanol	Supelco	PHR1070	For making specimen
ALMEMO admeasuring apparatus	ahlborn	2450-1	For Mechanical Withdrawal Threshold test
Anti-Digoxin antibody	Sigma-Aldrich	SAB4200669	For HE stain IHC or TUNEL
Anti-IL-1 beta	abcam	ab283818	For HE stain IHC or TUNEL
DAB Substrate kit	Solarbio	DA1010	For HE stain IHC or TUNEL
Denture base materials	Shanghai New Century	20000356	For model making
eosin	bioswamp	PAB180016	For HE stain IHC or TUNEL
Finger pressure recordings	Suzhou Changxian Optoelectronic Technology	CX1003w	For Tuina manipulation
formic acid solution	Sigma-Aldrich	695076	For decalcification
H2O2	Sigma-Aldrich	386790-M	For HE stain IHC or TUNEL
hematoxylin	bioswamp	PAB180015	For HE stain IHC or TUNEL
Isoflurane	Shanghai Yuyan Scientific Instrument Company	S10010533	For gas anesthesia
neutral resins	bioswamp	PAB180017	For HE stain IHC or TUNEL
Paraformaldehyde Fix Solution	Sigma-Aldrich	100496	For histology
PBS	Sigma-Aldrich	P3813	For HE stain IHC or TUNEL
Plantar Test Apparatus	IITC Life Science	/	For Paw Withdrawal Latency test
plaster of Paris bandage	WANDE	20150023	For model making
Proteinase K	Sigma-Aldrich	124568	For HE stain IHC or TUNEL
TNF Alpha Monoclonal antibody	Proteintech	60291-1-Ig	For HE stain IHC or TUNEL
TUNEL	Servicebio	GDP1042	For HE stain IHC or TUNEL
Wax	Sigma-Aldrich	327204	For making specimen
xylene	Shanghai Sinopharm Group	100092	For making specimen