This study presents an innovative running wheel-based animal mobility system to quantify an effective exercise activity in rats. A rat-friendly testbed is built, using a predefined adaptive acceleration curve, and a high correlation between the effective exercise rate and the infarct volume suggests the protocol's potential for stroke prevention experiments.
This study presents an animal mobility system, equipped with a positioning running wheel (PRW), as a way to quantify the efficacy of an exercise activity for reducing the severity of the effects of the stroke in rats. This system provides more effective animal exercise training than commercially available systems such as treadmills and motorized running wheels (MRWs). In contrast to an MRW that can only achieve speeds below 20 m/min, rats are permitted to run at a stable speed of 30 m/min on a more spacious and high-density rubber running track supported by a 15 cm wide acrylic wheel with a diameter of 55 cm in this work. Using a predefined adaptive acceleration curve, the system not only reduces the operator error but also trains the rats to run persistently until a specified intensity is reached. As a way to evaluate the exercise effectiveness, real-time position of a rat is detected by four pairs of infrared sensors deployed on the running wheel. Once an adaptive acceleration curve is initiated using a microcontroller, the data obtained by the infrared sensors are automatically recorded and analyzed in a computer. For comparison purposes, 3 week training is conducted on rats using a treadmill, an MRW and a PRW. After surgically inducing middle cerebral artery occlusion (MCAo), modified neurological severity scores (mNSS) and an inclined plane test were conducted to assess the neurological damages to the rats. PRW is experimentally validated as the most effective among such animal mobility systems. Furthermore, an exercise effectiveness measure, based on rat position analysis, showed that there is a high negative correlation between the effective exercise and the infarct volume, and can be employed to quantify a rat training in any type of brain damage reduction experiments.
Strokes exist continuously as a financial burden to countries globally, leaving countless patients physically and mentally disabled1,2. There is clinical evidence to suggest that regular exercise can improve nerve regeneration and strengthen neural connections3,4, and it is also shown that exercise can decrease the risk of suffering ischemic strokes5. With either a treadmill or a running wheel as an exercise training system, rodents, such as rats, serve as a proxy for humans for testing the effectiveness of exercises in a vast majority of clinical experiments6–8. A training system normally involves training a rat for a certain period of time, during which a rat runs at a certain speed. Therefore, the training intensity is generally calculated according to the exercise speed and duration6–8. The same approach is applied to estimate the amount of exercise required for neurophysiological protection. However, the experimental exercises are sometimes found to be ineffective, such as when a rat stumbles, falls, or grabs the rails once they are unable to catch up with the running wheel speed9–11. Needless to say, incidents of ineffective exercise significantly reduce the exercise benefit. Even though there is no any universally accepted approach currently to quantify the effective exercises for reducing brain damage, the level of effective exercises still stands as an objective appraisal for clinical researchers to illustrate the benefits of exercise in the discipline of neurophysiology.
There exist a number of limitations on commercially available animal mobility systems used in today's brain damage reduction experiments12. In a treadmill case, rats are forced to run by means of electric shocks, inducing tremendous psychological stress on the animals and thereby interference in the final neurophysiological test results8,13,14. Running wheels can be categorized into two types, namely voluntary and forced. Voluntary running wheels allow rats to run naturally, creating excessive variability due of the differences in the rats' physical traits and abilities15, while motorized running wheels (MRWs) employ a motor to turn the wheel, forcing rats to run. Despite also being a form of forced training, MRWs imposes less psychological stress on rats than treadmills13,16,17. However, experiments using MRWs have reported that rats sometimes interrupt the exercise by grabbing the rails on the wheel track and refusing to run at speeds exceeding 20 m/min9. These examples show that animal mobility systems currently available have an inherent disadvantage that inhibits effective exercising. For objective rat training purposes, the development of a highly effective training system but with low interference is therefore viewed as an urgent issue for neurophysiological exercise experiments.
This study presents a highly effective running wheel system for experiments on reducing the severity of the effects of the stroke11. In addition to a reduced number of interference factors during a training process, this system detects the running position of a rat using infrared sensors embedded in the wheel, thereby achieving a more reliable estimate of effective exercise activity. The psychological stress imposed by traditional treadmills and the frequent exercise interruptions in MRWs both skew the objectivity of the resulting exercise estimates. A positioning running wheel (PRW) system presented in this study is developed in an attempt to minimize the unwanted interference while providing a reliable training model for quantifying effective exercise.
Ethics Statement: The experimental procedures were approved by the animal ethics committee of Southern Taiwan University of Science and Technology Laboratory Animal Center, National Science Council, Republic of China (Tainan, Taiwan).
1. Constructing the Running Wheel Structure
NOTE: All acrylic should be transparent. Wash the disassembled wheel with water, then use alcohol to wipe the rubber track and acrylic sheets after each use.
2. Deploying the Infrared Sensors and Defining the Effective Exercise Area
NOTE: Take into account the running wheel size and the rat length in the design of an infrared system. A rat only triggers a single sensor at a time. In this experiment, rats are between 20 and 23 cm long.
3. Driving the Running Wheel
4. Constructing an Adaptive Acceleration Curve
5. Controlling the Software Program
NOTE: Exclusively develop a code for the microcontroller-based motor operation and for signal transmission from the infrared sensors to a computer for subsequent data analysis.
6. Operating the Positioning Running Wheel System
7. Training the Rats
8. Animal and Stroke Model
9. Assessing Neural Damage
This section is devoted to comparisons, made 1 week after surgery, on the mNSS scores, incline plane test results and brain infarct volumes among five groups. Figure 4A and 4B present the average mNSS scores and the average of incline plane test results, respectively. The PRW group appears as the best in terms of mNSS improvement. The significant differences between PRW and MRW and between treadmill and PRW clearly suggest that the PRW protects against stroke more effectively than other animal mobility systems currently available. Inclined plane tests are performed at significantly steeper inclination angles in all the exercise groups than in the control group over a time span of seven days after surgery, clearly demonstrating the benefits of exercise as a means of reducing the severity of the effects of the stroke. Particularly, the inclination angle in the PRW group was demonstrated as the steepest among all the exercise groups, and is even comparable with that in the sham group, showing a higher level of recovery than treadmill and MRW. Furthermore, Figure 4C shows that, after extracting the brain sections following 7 days of neural damage assessment, the PRW group not only exhibited a significantly smaller infarct volume than the control group but also exhibited the smallest infarct volume among all the exercise groups. It is hence clearly demonstrated that the rats trained using a PRW suffered significantly less amount of brain infarction damage than those using commercially available training systems, verifying the superiority of PRW in respect of brain damage reduction training.
This study presented a scientific approach for quantifying effective exercise activity in brain damage reduction training. During 3 weeks of training, there is a 98% effective exercise measure in PRW, whereas only 68% in MRW (Table 1). This significant difference in the effective exercise rate demonstrates that the superiority of the PRW training mechanism. The ineffective exercise measure, defined as 1 – the effective exercise measure and correlated with the mNSS score (Figure 4A), gives an 88% correlation with the mNSS score (Table 1). In addition, there exists an 85% correlation between the effective exercise measure and the incline plane angle (Table 1), and a 92% correlation between the ineffective exercise measure and the infarct volume (Table 1). Particularly, an effective exercise measure as high as 98% is correlated with an extreme low infarct volume in the PRW case. A significant correlation is hence demonstrated between the ineffective exercise and the extent of neurological damage.
Figure 1: PRW system. (A) Design drawing of a PRW. The running wheel is 55 cm in diameter and 15 cm in width. On the lower half of the running wheel, a hole has been drilled every 45° for infrared sensors installment. (B) Actual picture of the PRW. A layer of high friction rubber track is positioned on the inside of the acrylic wheel. A quarter-circle opening on one side of the running wheel acts as an entrance and an exit as well for the trained animals. An iron rod with bearings connects the running wheel to the triangular columns, supporting the running wheel. A motor is set on the outside of the running wheel track and is connected to the running wheel track by a 10 cm central-axis-mounted rubber disk. A microcontroller operates the motor and thereby commands the running wheel. A pair of semicircular, transparent acrylic sheets are attached to the triangular columns, and four pairs of infrared sensors are embedded in the acrylic sheets. Please click here to view a larger version of this figure.
Figure 2: Deployment of infrared sensors. (A) According to the size of the PRW and the body length of a rat, a pair of infrared sensors were deployed every 45° between 0° and 135° (producing a total of 8 sensors). Between 0° and 135°, rats exhibited a state of normal running, and therefore this area was defined as the effective exercise area. (B) In the MRW case, a pair of infrared sensors were deployed every 70° between 0° to 140°. Please click here to view a larger version of this figure.
Figure 3: Construction of an adaptive acceleration-training model for smooth speed-up exercise. The dashed lines represent the manually specified acceleration curves for the training of seven rats on Day 3, and can be characterized as an exponential function. Nonlinear curve fitting is then performed accordingly. The curve with circles represents the initial adaptive acceleration curve for Week 1, the Week 1 curve for short. The curve for Weeks 2 and 3 is an adjusted version of the Week 1 curve with a final speed to 30 m/min (Cfin = 30). Please click here to view a larger version of this figure.
Figure 4: Comparison on neurological damage assessment among groups over a time span of 7 days after surgery (each group with 9 rats). (A) Average mNSS scores (mean ± SD). There exists a significant variation among all the exercise and the control groups, evidence that exercise benefits brain damage reduction. The PRW group provides the lowest score among the exercise groups, demonstrating a superior neuroprotective mechanism to the other training systems. (B) Average hind-leg test angles (mean ± SD). A much steeper angle is demonstrated in the PRW than in the control group, and is demonstrated as the steepest among all the exercise groups. In addition, there was little difference between the PRW and the sham groups, indicating that PRW regenerated rats' hind-leg grip to a higher level. (C) Comparison on the infarct volume (mean ± SD). PRW acquires much smaller volume than the control group, and ranks the lowest among all the exercise groups, validating the prominent effect of PRW on brain damage reduction. Please click here to view a larger version of this figure.
Group | Effective exercise measure % (EEE) | mNSS | Inclined plane angle | Infarct volume |
PRW | 98.88 ± 1.11 | 23.54 ± 3.08 | 100 | 37.6 ± 1.08 |
MRW | 68.05 ± 5.39 | 70.7 ± 6.48 | 34.23 ± 4.48 | 72.76 ± 6.52 |
Control | 0 | 100 | 0 | 100 |
Correlation coefficient (R2) with EEE | -0.88 | 0.85 | -0.92 |
Table 1: Comparison on correlation between effective exercise activity and neurological damage. Effective exercise activity comparison among the PRW, MRW and the control groups. The PRW and MRW groups give a 98% and a 68% average effective exercise measure, respectively, after a 3 week training, meaning that PRW provides a larger amount of effective training. There exist a 0.88 correlation between mNSS and ineffective exercise measure, a 0.85 correlation between the effective exercise measure and the inclined plane angle, and a 0.92 correlation between the ineffective exercise measure and the infarct volume, respectively. In particular, an effective exercise rate up to 98% is correlated with an extremely small infarct volume in PSW. The data of mNSS, inclined plane angle, and infarct volumes are normalized.
Function | PRW (this study) | MRW | Treadmill |
Exercise training | Forced (laterally motorized) | Forced (centrally motorized) | Forced (electrical shock) |
Number of simultaneously training animals | Single | Single | Plurality |
Runway structure | Textured rubber belt | Bars | Rubber belt |
Trainable intensity | Low, intermediate, high | Low, intermediate | Low, intermediate, high |
Adaptive acceleration training | Yes | No | No |
Running position detection | Yes | No | No |
Deceleration training | Yes | No | No |
Effective exercise assessment | Yes | No | No |
Table 2: Comparison among animal mobility systems. PRW can be used at any level of training intensity. Combining a customized wheel with an adaptive training curve, PRW serves as a superior alternative to counterparts. Furthermore, infrared position detection technique is employed to quantify an effective exercise activity for brain damage reduction.
This protocol describes a highly effective running wheel system for reducing the severity of the effects of the stroke in animals. As a rat-friendly testbed, this platform is designed as well in such a way that a stable running speed can be maintained by rats throughout a running process by means of a predetermined adaptive acceleration curve. In typical training systems, preset training speeds and durations are set manually. Once an exercise commences, a preset speed is reached very shortly. In this context, it is very likely that rats are unable to reach higher speeds, making them tumble and fall and affecting the stability of their runs accordingly. The critical steps are, 1.1, 4.1 and 4.2, which are the key features in PRW as opposed to MRW. The integration between a spacious running track, as described in step. 1.1, and an adaptive acceleration-training model construction, referred to in step. 4.1 and 4.2, is presented as an improved version of a typical MRW. Such key features lead to a reduced infarct volume than MRW. The overall design of the presented system is designed as a rat-friendly testbed for a reduction in ineffective exercise. More specifically, 4 pairs of infrared sensors are deployed to detect the real-time position of a rat, providing a measure to quantify an effective exercise activity, defined in step. 6.6, for comparisons on correlation with the mNSS scores, inclined plane angle and brain infarction volume. This measure can be used to quantify any type of neurophysiological experiments, unrealized yet in conventional training platforms. However, it is very likely that an effective exercise cannot be detected for a small rat due to a sparse distribution of IR sensors. Furthermore, a major disadvantage relative to a treadmill is that only a single rat can be trained at a time on this platform. The system troubleshooting involves two parts. One is an accurate sensor alignment for signal transmission and reception due to the high directivity of IR, while the other is the running wheel rotating at specified revolutions per minute (rpm). An IR source/detector pair need to be aligned until a strong signal can be received by the detector. With regard to the running wheel, the 10 cm diameter rubber disk is worn out gradually when the wheel is rotated for a long period of time. Therefore, a spring need to be loosened as a way to compensate an inadequate rubber disk friction for normal wheel rotation. Table 2 gives a comparison on the forced animal mobility systems used in brain damage reduction experiments.
Tests give significantly better results in terms of the mNSS scores, incline angle and infarct volumes in the PRW group than in the control group (p <0.05). The PRW group was validated as the one providing the most amount of effective exercise training among all the exercise groups. In this study, when trained using the traditional MRW, rats are frequently found to hold onto the bars of the runway and refuse to run at a speed beyond 20 m/min, an agreement with a piece of prior work9. As a way to improve the rat training performance, the metallic runway is redesigned as a high-density rubber running track in this work. In a treadmill, psychological stress is inevitably imposed on the electric shock-driven rats, an unsolved problem in the discipline of physiology in the past. Therefore, a way must be found to reduce the tumble frequency and to ease the psychological stress imposed on rats during training. In this manner, test results can be interpreted more accurately, as a convincing way to demonstrate the exercise benefit to brain damage reduction. This is a major motivation behind this work.
This work successfully provides a quantitative measure of an effective exercise activity correlated with the infarct volume, the most direct evidence of stroke damage. Therefore, effective exercise in other types of animal-based tests can be qualified accordingly. As presented in6–8, both the exercise intensity and duration are user-specified in neurophysiological experiments, but not taking into account the effective amount of an exercise training. Effective exercise activity is validated as a key factor to stroke neuroprotection, using this rat-friendly and innovative animal mobility system.
It is believed that this platform can be applied to variable velocity training and related issues in the future. As pointed out in24,25, variable velocity training is viewed as a more effective training in the discipline of exercise physiology. Using infrared position detection technique as a basis, variable velocity training can be precisely conducted on athletes for a deep investigation into the neurophysiological protection mechanism.
The authors have nothing to disclose.
The authors would like to thank Dr. Jhi-Joung Wang, who is the Vice Superintendent of Education at Chi-Mei Medical Center, and Dr. Chih-Chan Lin from the Laboratory Animal Center, Department of Medical Research, Chi-Mei Medical Center, 901 Zhonghua, Yongkang Dist., Tainan City 701, Taiwan, for providing the shooting venue. They would also like to thank Miss Ling-Yu Tang and Mr. Chung-Ham Wang from the Department of Medical Research, Chi-Mei Medical Center, Tainan, Taiwan, for their valuable assistance in demonstrating the prototype system in real experiments with rats. The author gratefully acknowledges the support provided for this study by the Ministry of Science and Technology (MOST 104- 2218-E-167-001-) of Taiwan.
Brushless DC motor | Oriental Motor | BLEM512-GFS | |
Motor driver | Oriental Motor | BLED12A | |
Motor reducer | Oriental Motor | GFS5G20 | |
Speedometer | Oriental Motor | OPX-2A | |
Treadmill | Columbus Instruments | Exer-6M | |
Infrared transmitter | Seeed Studio | TSAL6200 | |
Infrared Receiver | Seeed Studio | TSOP382 | |
Microcontroller | Silicon Labs | C8051F330 | |
CCD camera | Canon Inc. | EOS 450D | |
Image processing software | Adobe Systems Incorporated | ADOBE Photoshop CS5 12.0 | |
Image analysis | Media Cybernetics | Pro Plus 4.50.29 | |
Sodium pentobarbital | Sigma-Aldrich (Saint Louis, MO, USA) | SIGMA P-3761 | |
Ketamine | Pfizer (Kent, UK) | 1867-66-9 | |
Atropine | Taiwan Biotech Co., Ltd. (Taoyuan, Taiwan) | A03BA01 | |
Xylazine | Sigma-Aldrich (Saint Louis, MO, USA) | SIGMA X1126 | |
Buprenorphine | Sigma-Aldrich (Saint Louis, MO, USA) | B9275 | |
Anesthesia | Sigma Chemical |