Core C - Behavioral & Electrophysiological Assessment
|MDR 539||Rat behavior|
|MDR 540||Mouse behavior|
|Core Director:||David Magnuson, Ph.D.|
|Personnel:||Darlene Burke, M.S.|
Currently, the Behavioral & Electrophysiological Assessment Core provides training, expertise and oversight for the following standardized assessments:
- The BBB Open Field Locomotor Scale for rats and the Basso Mouse Scale (BMS) for mice. These are the gold standard assessments of hindlimb function during open-field locomotion.
- The Louisville Swim Scale (LSS) for rats. This assessment was developed by Core C in collaboration with the Magnuson lab and is used to assess swimming after a spinal cord injury (Smith et al., 2006a).
- The horizontal ladder assessment for rats and mice. This assesses sensorimotor function and correlates well with white matter loss in the dorsal funiculi and lateral white matter. This assessment has replaced grid walking as the primary sensory-motor assessment following mild-moderate contusions incomplete laceration injuries.
- Hindlimb kinematics for rats: We use the MaxTraq and MaxMate software packages from Innovision Systems to assess hindlimb kinematics during swimming and walking. These techniques were established in the Cores over the last 2-3 years (Magnuson et al., 2009).
- Treadscan™ and digital footprint gait analysis for mice and rats. We use a commercially available Treadscan system to assess the gait of mice and a ventrally placed digital camera to make video of foot placement during walking for rats (e.g., Beare et al., 2009).
- Trans-cranial magnetic motor-evoked potentials or tcMMEP for rats. This technique assesses action potential conduction in descending axons located in the ventrolateral funiculus (Magnuson et al., 1998, Loy 2002, Cao 2005).
- The Magnetically-evoked inter-enlargement response or MIER for rats. This technique assesses action potential conduction in ascending axons located in the lateral funiculus. This technique was developed in the Cores B and C in collaboration with the Magnuson lab (Beaumont et al., 2006).
In addition, the proposed expansion of Core C will allow it to provide training, expertise and oversight for activity-based retraining of spinal cord injured rats. In collaboration with the laboratories of Drs. Harkema and/or Magnuson, retraining using the Robomedica™ body weight supported treadmill system and/or swimming will be available (Magnuson et al., 2009; Smith et al., 2006b).
Over the past 10 years the COBRE/KSCIRC Electrophysiology and Behavior Core has been involved in the functional assessments of thousands of rats and hundreds of mice. In addition, it has successfully developed and/or established novel and unique tools for assessing rodents with spinal cord or brain injuries. While each of the Core personnel has primary responsibilities, they have also been thoroughly trained in most, if not all, of the available assessments, giving the Core great flexibility in scheduling. Core personnel have taken on the two-fold responsibility of ensuring that quality assessments are done while also ensuring that members of the investigating laboratories are thoroughly trained in each of the assessment strategies. Thus, trainees and technicians are not passive observers during the assessment sessions, but go through a three stage process leading to proficiency which involves observation and training that includes blinded scoring, practice and correction as a redundant scorer and finally full participation with Core oversight. We routinely have members travel to Ohio State University for training or invite Ohio State faculty to Louisville for group refresher sessions. This approach ensures consistency for all User PIs across multiple studies and allows direct comparison of results from one study with those done previously. Given exact duplication of injury parameters/controls in Core C with the functional outcome measures of Core D ensures accurate interpretation of resultant data. This addresses one of the major barriers/weaknesses in the neurotrauma field, namely the modest level of reliability and repeatability of spinal cord injuries and levels of functional recovery.
Louisville Swim Scale (LSS).
In 2006, we began investigating different approaches to assessing rats and mice after spinal cord injury. The gold-standard BBB scale assesses overground locomotion and scores above an 8 or 9 depend on the capacity for weight support. We hypothesized, based on the widespread use of body-weight supported treadmill stepping in fully transected rats and cats, that removing the need for weight-support may allow the spinal cord below the level of a moderate to severe thoracic contusion injury to express greater amounts of hindlimb function. Thus, we developed the Louisville Swim Scale, an assessment tool, based loosely on the BBB, which assesses hindlimb movement while the weight of the animal is being supported by buoyancy. It assesses forelimb dependency and body position while swimming following spinal cord injury. It is an 18 point scale (0-17) that clearly distinguishes between mild, moderate and moderately-severe thoracic contusion injuries in rats (Smith et al., 2006a). Normal (uninjured) animals swim using their hindlimb only for forward motion. Immediately after a thoracic or lumbar contusion injury, they rely entirely on their forelimbs for forward motion. As they recover spontaneously, animals with mild contusion injuries regain some hindlimb movement and a near-normal kinematic pattern of kicking. Animals with moderate or moderately-severe injuries only recover a near-normal pattern of activity if they receive swim training as a rehabilitation strategy. Importantly, the normally bipedal nature of swimming provides a unique opportunity to assess a locomotor activity that is not dependent on forelimb-hindlimb coordination, nor is it an activity that animals are exposed to in their cages. All swimming activity is provided by the experimenter. Thus, swimming, as a model of activity-based retraining and as a functional assessment tool allows us a unique look into the plasticity of motor circuitry post-SCI.
Magnetically-evoked inter-enlargement response (MIER).
In 2006, we also developed a novel electrophysiological method, the MIER in rats (Beaumont et al., 2006). The utility of this assessment is two-fold. First of all, the tcMMEP that was the standard assessment being used at the time was found to correlate poorly with functional recovery in a variety of different injury models. Secondly, there was (and is) increasing interest in long and short propriospinal pathways within the spinal cord, in particular as targets for therapeutic intervention. Based on an earlier paper that demonstrated an EMG response in the masseter muscle following electrical stimulation of the sciatic nerve (Deriu et al., 2001) we were able to record EMG responses in the triceps and masseter muscles following magnetic stimulation of the sciatic nerve. This assessment is done on awake animals concurrent with tcMMEP recordings. The recording electrodes are moved to the triceps muscles in the forelimbs and a high-field 25 mm figure-8 magnetic transducer is used to stimulate the sciatic nerve. The response is distributed bilaterally in the spinal cord and results in moderate amplitude EMG responses in the triceps muscles with a latency of approximately 6 ms. It is carried in the lateral funiculus approximately at the level of the central canal (at T9). In contrast to the tcMMEP, the MIER is highly correlated with functional locomotor recovery as assessed by the BBB Open Field Locomotor Scale and the Louisville Swim Scale. The MIER is likely carried by the priopriospinal network of spinal cord interneurons that have become the focus of research efforts in numerous centers worldwide because of their apparent propensity for spontaneous and evoked plasticity, making them good potential targets for local regeneration and “bridging the gap” type repair strategies. The MIER represents a vital tool for examining mechanisms of pathophysiology and functional recovery that involve plasticity of the spinal intersegmental neural circuitry.
Fig. 1. The MIER raw EMG response is shown top left. Bottom left shows that the MIER recovers some amplitude after a moderate T9 contusion injury. The scatter plot on the right shows that, unlike the tcMMEP, the MIER correlates significantly with the BBB score.
Core C has been instrumental in the development of the kinematic and gait analysis tools that we use to assess overground locomotion and swimming in rats. The need for these assessments is primarily due to the perceived non-linear and subjective nature of the BBB Open Field Locomotor Scale. Kinematic and gait assessments are objective and are particularly sensitive to improvements in coordination and limb trajectory. These kinds of assessments also allow a better translation between our rat studies and the human studies being done in Dr. Harkema’s laboratory. A very basic assessment tool that we developed is the Plantar Stepping Index (PSI), which we use as a supplement to the Regularity Index originally developed by Frank Hamers for use with his Catwalk™ System (Koopmans et al., 2005). The PSI allows us to examine gait in spinal cord injured animals that are unable to perform well enough to be assessed by traditional footprint and gait analysis techniques. Unlike the Regularity Index, which only counts plantar steps made in one of four normal sequences, the Plantar Stepping Index allows every single plantar step made with a hindlimb to be taken into account. We have recently used the Plantar Stepping Index to show that step-training in shallow water brings about improvements in stepping in shallow water, but that the improvements do not transfer to overground stepping (Magnuson et al., in preparation). This finding has important implications for rehabilitation approaches by suggesting that early step-training is ineffective in rats because their functional recovery is already extensive.
Fig. 2. A shows that BBB scores for trained (in shallow water) and untrained rats with moderately-severe T9 contusion injuries are not different. In B, the Plantar Stepping Index (PSI) assessment shows that shallow water training resulted in improved stepping in shallow water but not overground. C shows an injured rat stepping in shallow water from the lateral and ventral views.
Fig. 3. The hindlimb swimming movement of a normal rat is show as a stick figure using the 3-segment, 2-angle model (A), described more fully in (B) with Iliac Crest (I), Hip (H), Ankle (A) and Toe (T). The angular excursion of the Iliac crest - hip – ankle angle (IHA) and the Hip – Ankle – Toe (HAT) angle during swimming are shown in (C). The angle-angle plot used for the Elliptical Fourier Analysis (EFAW) that provides the area and centroid of the plot as objective measures of the hindlimb movement during swimming is shown in (D).
Core C was instrumental in the development of a streamlined and relatively high-throughput kinematic assessment of rat swimming using an angle-angle plot and Elliptical Fourier Transformation as quantitative measures. For this assessment digital video of the hindlimb is used to create a stick figure using a 3 segment, 2 angle model where the segments are iliac crest – hip, hip – ankle and ankle – toe. This allows us to ignore the knee marker which is notoriously inaccurate due to skin slippage. More importantly, when the two angles created by these three segments are plotted against each other for a normal animal the angle-angle graph shows an ellipse. When this ellipse is analyzed mathematically using an Elliptical Fourier Transformation (Freeware program called EFAW), the area of the ellipse represents the out-of-phase excursion of the two angles and the centroid of the ellipse (position on the x-y plot) represents the overall limb position in terms of flexion and extension. Presently, it takes between 30 and 45 minutes per animal per time point, including recording time, for this analysis. Importantly, we have found that retraining after a moderate SCI can bring about a near normal angle-angle plot (area and centroid are not different from pre-injury levels) however, the toe velocity (relative to the hip) does not recover back to control levels. This suggests that the capacity for generating the pattern of hindlimb movement recovers with training while the force generated during the kicking does not
Fig. 4. Shown are the centroids for baseline, week 3 and week 6 post-injury swimming, and ellipse areas determined by EFAW analysis. These show that the pattern of hindlimb kicking continues to improve out to 6 weeks post-injury with swim training.
Gait Analysis using Treadscan. Core C has adopted a commercially available device, the Treadscan (by Cleversys, Inc.) as a highly sensitive gain-analysis system for mice and rats. This device is based on a clear treadmill, a mirror placed at 45° and a high-resolution digital video camera. The company Cleversys designed software that uses color and shape to identify the paws as mice walk on the treadmill. The software uses the paw placement position and timing to calculate a huge range of gait parameters including base of support, stance and swing times, coupling and coordination and toe spread. In a recently completed study, we thoroughly compared the sensitivity of the Treadscan with the BMS in mice following mild and moderate thoracic contusion injuries (Beare et al., 2009). We have since streamlined the analysis such that complete evaluation of a single animal requires only 20-30 minutes per time point and provides all the available gait parameters on a single spreadsheet.
Fig. 5 shows a still image taken from treadscan video. The software identifies the paws when placed on the treadmill and uses both timing and position to calculate more than 6 distinct gait parameters. The software is “trained” to identify the paws for each different strain of rat or mouse used. Once trained, the software works in an automated fashion allowing fairly high throughput.
Figure 1. Treadscan™ system for gait analysis of mice
Figure 2. Cadwell TCCMEP stimulator.