| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of 1Anatomy and Neurobiology, and 2Physical Therapy, Medical College of Virginia Campus, Virginia Commonwealth University, Richmond, Virginia
Submitted 30 May 2007 ; accepted in final form 23 August 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
extraocular; motoneuron; eye movement; oculomotor
Much work has been done on the physiology of EOM in animals such as the cat and squirrel monkey (17, 18, 29, 30, 40, 44). The evaluation of cat and squirrel monkey EOM motor units has revealed that they can be classified into five functional categories based on twitch capabilities, fusion frequency, and fatigue (18, 51, 52, 54). Although the EOM motor units in both these animals can be classified into similar groupings, each species displays different contractile characteristics (18, 50–52, 54). For example, the cat LR motor units produce a twitch tension nearly three times stronger than that of the squirrel monkey, whereas the monkey LR motor units are faster than the cat in both contraction time and fusion frequency (50, 51). Another potential difference between cat and squirrel monkey LR motor units is the presence of nontwitch motor units (24, 44, 51). Nontwitch motor units have been identified in the cat but not in the squirrel monkey to date. The nontwitch motor unit propagates an action potential only when stimulated with tetanic trains.
While contractile differences between species have been seen, another significant divergence between the studied species is that of predicted linear summation of forces (18–20, 50). The cat hindlimb whole muscle tensions have been shown by Burke et al. (8) to equal the predicted values from the summation of individual motor units. In contrast to hindlimb summation findings, EOMs display whole muscle tensions that are only a fraction of the predicted values obtained from single-unit data (18, 20). Specifically, the squirrel monkey actual LR whole muscle twitch force is only 5% of the predicted force, whereas the cat actual LR whole muscle twitch force is 37% of the estimated value (18, 20). The loss of actual contraction tension might be due to the EOM muscle fibers branching into neighboring fibers while other fibers insert in a serial manner (35) or the possibility that motoneurons within the EOM motor nuclei maintain polyneuronal "safeguard" innervation patterns that result in redundant motoneurons. This polyneuronal innervation would make the algebraic calculation of predicted tensions artificially high (36). Regardless of the reason for the "lost" actual EOM tension, the question remains as to why there is such a vast discrepancy between the cat and monkey using this force prediction method.
Because of the significant motor unit contractile differences seen between the cat and squirrel monkey, we wanted to expand the understanding of extraocular characteristics and anatomic parameters in another carnivore. The ferret is an animal that is widely used in biomedical research. Although some research has occurred in ferret multisensory and visual function (10, 25, 39, 41, 53), no previous work has focused on the LR motor units. We hypothesized that 1) ferret motor units could be classified into categories based on speed and fatigability, 2) nontwitch units would be found within the sample, and 3) the innervation ratio of LR muscle fibers to abducens motoneurons would be higher in the ferret than in previously tested animals.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
Nine adult male ferrets (Mustela putorius furo) weighing between 1.35 and 2.1 kg were preanesthetized with an intramuscular injection of Robinul-V (.2 ml/kg), ketamine (0.2 ml/kg), and xylazine (0.02 ml/kg) before each experiment. Following preanesthesia, the animal was given Isoflurane at 3.0% via a small mask for several minutes to make certain of deep anesthesia. The topical anesthesia Cetacaine was sprayed into the oral pharynx before intubation. An intubation tube was introduced to allow isoflurane to be continually administered throughout the duration of the experiment at a flow rate of 1.5–3.0% together with 
1.0 liter of oxygen per minute to keep the animal in a state of deep anesthesia (no withdrawal to paw pinch). The ferret vital signs were monitored at all times by way of a SurgiVet V9004 (Tulsa, OK) rectal probe that continually recorded heart rate (typical range 110–220 beats/min) and oxygen saturation (88–100%). Respiration rate (15–45 breaths/min) and expiratory carbon dioxide (2.7–4.7%) were recorded by the SurgiVet through an intubation tube. Body temperature was maintained by a heating pad under the animal. To maintain hydration, the animal was injected subcutaneously with 5 ml of normal saline after 3 or 4 h of experimental procedure and approximately every 2 h thereafter.
With the ferret's head secured in a Kopf stereotaxic frame using the Kopf squirrel monkey adaptor (model 1248, Kopf Instruments, Tujunga, CA), a midline incision was made from the nasion to the upper cervical region. The temporalis muscle on the right side as well as portions of the masseter muscle were removed to expose the periorbital structures on the lateral side of the eye. Because the lateral bony orbit is incomplete in the ferret, care was taken to ensure that the LR muscle was not damaged during the surgical procedure. The LR muscle was identified and a 5-0 silk suture passed behind and tied around the tendon at the point of insertion. This suture knot was secured with a drop of tissue glue before the tendon being released from the globe. As the LR tendon was being released from the globe, the underlying lateral retractor bulbi (RB) muscle slip was identified, separated from the global portion of the LR, and left attached to the globe. The suture loop from the LR muscle was placed around a hook attached to a semiconductor force strain gauge (Pixie model 8101, Endevco) adjusted to maintain correct anatomic alignment. This strain gauge has a natural frequency of 2 kHz and a compliance of 2 µm/g. The periodic application of warm mineral oil to the LR prevented the muscle from drying during the experimental procedure. Additional suture loops were placed through the anterior surface of the globe and connected to a second strain gauge (model MLT 050D, ADInstruments, Colorado Springs, CO). The second strain gauge was positioned directly anterior to the eye to record tension generated by the RB muscles.
A single craniotomy was performed through the right posterior parietal bone to the nuchal crest for the introduction of electrodes for stimulation of the abducens nerve and for probing the abducens nucleus. The midline sagittal crest was never disrupted to avoid the sagittal sinus. Through this craniotomy, a majority of the cerebellum was removed via suction to expose the floor of the fourth ventricle. A bipolar steel insulated electrode with 1.5-mm space between poles and 0.5-mm exposed tips was driven to the abducens nerve on the ipsilateral side (stereotaxic coordinates: posterior 6.2; lateral 2.9; and a variable depth depending on the size of the animal). The abducens nerve passes rostrally along the ventral surface of the cerebral peduncle. Cranial nerve VI was supramaximally stimulated via the bipolar electrode and the LR strain gauge was optimally positioned for maximal isometric tension in the LR. The RB strain gauge was also optimally positioned in the same manner.
All stimulation paradigms were delivered by an A.M.P.I. Master-8 programmable pulse generator (Jerusalem, Israel). Amplitude of the stimulation was regulated through a high-voltage World Precision Instruments (WPI) stimulus isolator (model A360, WPI, Sarasota, FL).
Insulated electromyographic (EMG) wires (0.05 mm, 59.4 
/m) were inserted through a 28-gauge needle into the LR muscle 3 mm from the detached tendon in the distal muscle and the second wire through the proximal muscle belly. These EMG wires were connected to a BioAmp (ADInstruments) interface with the computer recording program Chart (ADInstruments). The EMG was grounded with a silver wire through the contralateral temporalis muscle.
A tungsten electrode (WPI, Kapton-clad insulation, 5-µm shaft, tapered to a 1 µm tip, 4.8–5.2 M) was introduced at a 60° angle forward into the ipsilateral abducens nucleus. The ideal location for this tungsten electrode was determined by an antidromic field potential that was originated from the abducens nerve bipolar stimulating electrode and recorded via an ExCell3+ (FHC, Bowdoinham, ME) preamplifier. Once a strong abducens nucleus field potential was identified, the same tungsten electrode was used for the extracellular stimulation of individual motoneurons. The tungsten microelectrode was driven by a Kopf hydraulic microdrive.
When the tungsten electrode was driven close to a motoneuron, a small single pulse stimulation (10.2–76 µA) caused an all-or-nothing twitch response by the motor unit that was detected from the strain gauge on a Tektronix 5111A oscilloscope (Tektronix, Beaverton, OR). The stimulation amplitude and depth of electrode placement were adjusted to ensure only a single motor unit was activated. Following the technique of Macefield et al. (33), the confirmation of single-cell stimulation was made by reducing the amplitude of stimulation below the all-or-none contractile threshold of a single unit and the elimination of any EMG activity. As the stimulation intensity was raised, we noted a simultaneous return of contraction intensity and EMG response. The stimulation amplitude was then typically increased 30–50 µA above the single unit threshold before additional units were obviously recruited as evidenced by larger contractile forces and generally larger or more complex EMG shapes. Finally, the stimulation intensity was again lowered to just above the single-unit threshold but well below the level that recruited additional units. Within this range of acceptable single-unit stimulation intensity, the EMG was closely monitored to ensure that no alterations occurred during twitch or tetanic stimulation paradigms. If an amplitude adjustment did not ensure a single unit was being tested, the tungsten electrode was driven deeper in an attempt to better isolate the single motoneuron. If a motoneuron could not be isolated with absolute confidence, it was eliminated from further evaluation.
Each motor unit was evaluated by at least 10 single twitch stimulations (1 Hz at 0.2-ms duration) followed by nine tetanizing trains of 150 ms beginning at 50 Hz and increasing up to 220 Hz in increments of 20 Hz. The twitch evaluation was performed before any tetanic stimulation to avoid potentiation. Each twitch stimulation pulse was delivered once per second and the tetanic stimulation trains were separated by 5 s to allow the muscle force to completely return to baseline. To best evaluate whether the tetanic stimulation resulted in fusion, the rising slope of the contraction was observed for an absence of deflections in addition to a smooth plateau. The smoothness of a tetanic plateau varied with the overall strength of contraction and thus the signal-to-noise ratio.
Following this stimulation paradigm, a fatigue protocol was performed that consisted of 75-Hz trains of 500-ms duration per second for 2 full min. Occasionally while searching for motoneurons, a 50-Hz tetanic train was applied in an attempt to locate a distant motoneuron. These intermittent tetanic stimulations allowed the possibility of locating a nontwitch motor unit. A stimulated nontwitch unit may appear as a small fused contraction that minimally increases in tension with higher frequency stimulation. We confirmed the presence of a nontwitch motor unit located with tetanic trains by changing the tetanic stimulations back to single impulses that did not elicit any contractile response.
As many motor units as possible were evaluated from each animal. Following this individual motor unit testing on each animal, the same stimulation procedure was applied to the entire abducens nerve through the bipolar electrode to evaluate whole muscle twitch tension, tetanic forces, and fatigue.
Each force generated by the LR or RB was recorded through the transducers to the Chart program for later evaluation. The recorded twitches were analyzed for contraction time, maximal tension, and half relaxation time for 10 separate twitch recordings per motor unit. The contractile characteristics of each motor unit were expressed as mean values of each parameter. The tetanic contractions were evaluated for peak force. Individual tetanic trains were evaluated to determine the specific frequency that produced a characteristic fused tension. No effort was made to quantify the amount of tension from the RB slips still attached to the globe. The recording of RB activity was only used to document whether RB contraction occurred from the stimulation of the motoneuron. The fatigue index (FI) was determined using the Burke method as the ratio of the tension from the last of the fatigue stimulations to the tension from the first fatigue stimulation. The higher the FI, the more the motor unit is able to exert sustained tension throughout the fatiguing stimulation protocol.
Descriptive statistics were calculated for each contractile parameter, including twitch tension, maximum tetanic tension, twitch contraction time, twitch half relaxation time, and FI. An unpaired Student's t-test was performed to determine significance of differences between motor unit groupings at various stimulation frequencies.
Anatomic evaluation. In a separate series of experiments, three adult male ferrets (1.4–1.75 kg) were used for anatomic study of the LR muscle as well as histological evaluation of the abducens nucleus.
The ferret was anesthetized as described previously while maintaining a patent airway and being monitored for distress. While in this state, the lateral edge of the skin surrounding the right eye was carefully cut and the underlying connective tissue dissected to reveal the LR tendon as it inserts on the globe. After this tendon was carefully exposed, a small probe was passed under the muscle and pulled slightly forward to expose the LR muscle belly. A Hamilton microinjection 22-gauge needle was inserted 1 cm into the LR muscle belly maintaining the correct parallel alignment to the muscle. At this depth a bolus of 25–40 µl of horseradish peroxidase conjugated with the subunit B of cholera toxin (CTHRP) was injected. An additional 10–25 µl (50 µl total) was injected in the muscle along the needle track in areas closer to the distal tendon but still within the muscle belly. Following the CTHRP injection, gentle pressure was applied to the outer orbital edge of the muscle in an effort to prevent overflow of the CTHRP into other periorbital structures. The lateral eye connective tissue and skin was carefully sutured. Topical Xylocaine was applied to the surgical site. Following this procedure, the isoflurane was discontinued and the animal was monitored closely as it regained consciousness. As the animal became more alert, it was given water by mouth and returned to the animal care facility 1–2 h after the surgery. Twenty-four hours after the CTHRP injection, the animal was deeply anesthetized as described previously and then perfused, thereby killed, with a phosphate-buffered mixture of 1% paraformaldehyde and 1.25% gluteraldehyde over a 1-h period. Bilateral LR muscles were removed and frozen in embedding medium with liquid nitrogen-cooled isopentane. Following perfusion, the brain stem was removed and immersed in fixative for 8 h.
The brain stem was blocked and stored an additional 2 h in phosphate buffer. The blocked section was serially sectioned with a vibratome into 50-µm, transverse sections. At this point the sections were processed for HRP neurohistochemistry according to the modified Mesulam technique with tetramethylbenzidine (TMB) as the chromogen. Following the TMB incubation protocol, the sections were rinsed in distilled water and then mounted on glass slides. After 24–48 h of dehydration, the sections were counterstained with neutral red and coverslipped with Permount. The evaluation of the brain stem sections was conducted using the Image-Pro digital imaging software and an Olympus BH-2 microscope with an attached CoolSNAP-Pro digital camera (Media Cybernetics, Silver Spring, MD). Each section was analyzed for the presence of dark HRP reaction product in the motoneurons by focusing the microscope on the top of the slide and slowly focusing through the 50-µm section ensuring that each cell is identified and counted. By evaluating the top and bottom of each section as well as sequential sections, it was possible to identify any cell bodies that were cut and thus seen in two adjacent sections. Any cell that appeared to be cut was digitally photographed, and then the sequential section was analyzed in an attempt to locate the cut cell and ensure that it was only counted once.
The muscle samples were harvested and stored at –80°C until the time they were sectioned into 10-µm sections with a cryostat and mounted on glass slides. Muscle specimens were stained with hematoxylin and eosin and then coverslipped with Permount. The evaluation of three midbelly sections from each of eight different muscle specimens (3 samples from the anatomic study and 5 from nonstimulated muscles from the physiology study) was conducted using the same imaging system as for the brain stem sections. For each sample of LR muscle an average of three midbelly sections was counted. The muscle sample sections used to count muscle fibers were not adjacent sections but were not more than five sections away from each other. The individual muscle fibers were counted using Image-Pro software (Media Cybernetics) that allows a marker to be placed on each muscle fiber in the digital image. A correlation coefficient was performed between the average number of muscle fibers taken from each animal against the animal body weight. An unpaired Student's t-test was performed to determine significance of the difference between muscle fiber diameters in the global vs. orbital layers. The results are expressed as means ± SD. Statistical significance was set at P 
0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
Each of the motor units responded to a single electrical pulse as well as tetanic stimulation (Fig. 1A). The individual motor units in this study were located and initially evaluated using single pulses, which limited the possibility of locating nontwitch motor units as described previously (51). The LR motor units (n = 56) had an average twitch tension (Fig. 2A) of 30.2 ± 18.5 mg (range = 6.7–104 mg) and an average contraction time of 15.4 ± 5.79 ms (6.8–28 ms). Motor units with LR -RB split innervation (n = 6) had an average LR portion twitch tension of 27.8 ± 11.0 mg (17.3–42.9 mg) and an average LR portion contraction time of 23.7 ± 4.4 ms (17.9–29.5 ms).
|
|
The fusion frequencies of the LR motor units varied between 120 and 220 Hz with a mean frequency of 154.6 ± 20.7 Hz. To categorize the LR motor units into either fast or slow groupings, those motor units with a fusion frequency below the population mean are considered "slow" and those with a fusion frequency above the population mean are considered "fast" (18, 51, 54). All six of the LR-RB units had a fusion frequency at 140 Hz or below. In an effort to determine whether using the fusion frequency as a way of organizing the LR motor units into fast or slow classification produced distinctive groups, a frequency tension (F-T) curve was created from the tension for each frequency normalized against the tension produced at the highest frequency (220 Hz). The contractile differences between the groups were seen at each frequency as a percentage of the tension created at the highest frequency (Fig. 3). This method of using relative tensions allows the variability of absolute tensions within each group to be minimized. The fast units required a significantly higher stimulation frequency to achieve the comparable relative tension developed by the slow units at lower frequencies. Using an unpaired t-test, a significant difference was seen between relative tensions of the slow and fast units at 120 Hz (P = 0.013), 140 Hz (P = 0.017), 160 Hz (P = 0.034), and 180 Hz (P = 0.038).
|
Combining the fusion frequency and fatigue allowed the further classification into four groups (51). These groups were slow fatigue resistant (SR; n = 6), slow fatigable (SF; n = 15), fast fatigue resistant (FR; n = 16), and fast fatigable (FF; n = 19). Table 1 summarizes the specific contractile characteristics for the four categories above as well as the LR-RB units left together as one group.
|
|
650 µm rostral to caudal extension of the nucleus. An accessory abducens nucleus was seen ventrolateral to the primary abducens nucleus in one animal, indicating an overflow of CTHRP into one or more RB muscle slips.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
Motor unit contractile characteristics. Clamann (11) stated it well when he reported that "the muscles that move the eye are a world unto themselves." In comparing the EOMs to limb muscles, the fast contractile speed of eye muscles as well as a low innervation ratio that aids in the precision of controlled movement to provide clear binocular vision have been noted (3, 4, 11, 12, 49, 55).
The ferret LR twitch motor units were classified according to fusion frequency and FI. The distribution of fusion frequencies were continuous as was seen in the cat LR (51). We subdivided the motor units using the fusion frequency mean as the dividing point in an attempt to show a relationship to the five myosin heavy chain types isolated in ferret LR (53). The F-T curve (Fig. 3) clearly demonstrates that the slow units are a distinctive group that achieves maximal tension at notably lower frequencies than the fast units. The fast and slow motor unit groupings did not demonstrate significantly different relative tensions with frequencies below 120 Hz or above 200 Hz. At each of the slower stimulation frequencies, the motor units are all increasing nearly equally in tension as they summate from the twitch response. In the frequency range of 120–180 Hz, the slow motor unit group demonstrates a significantly greater relative tension than the fast group. Any differences between these groups are again lost at the frequencies at or above 200 Hz. At these higher frequencies, there is no statistical difference between fast and slow units because the slow motor units are past their optimal fusion frequency and have already achieved their maximal tension. Similar to hindlimb muscles (6), the extraocular slow units plateau and then decline in tension, whereas the fast units are reaching the optimal frequencies for them to achieve maximal tension. These different physiological reactions by fast and slow units to variable stimulation frequencies has also been seen in cat and human single unit evaluations (5, 56). The F-T curve (Fig. 3) also shows that the 75-Hz fatigue test used in this study was optimal in the ferret because no significant difference was seen between the motor unit groupings at that frequency (53). The 75-Hz fatigue testing frequency was well below the average fusion frequency thus preventing excessive stress on the motor unit. This F-T curve method of ensuring motor unit grouping differences has not been previously utilized in the EOM system.
The twitch-type motor units responded to single pulse stimulation that allowed the determination of twitch tension and then tetanic tension while progressing from lower to higher stimulation frequencies. The average ferret single motor unit twitch tension was very comparable to that seen previously in the cat but three times larger than in the squirrel monkey (Table 2) (17, 18, 20, 40, 50, 51, 54). When LR twitch contraction time is compared, the cat is two times and the squirrel monkey three times faster than the ferret. The fusion frequency in the ferret was much lower than in both the cat and squirrel monkey. In terms of tetanic contractions, the average tetanic tension produced in the ferret LR is much larger than both the cat and squirrel monkey (18–20, 50, 51).
|
The specific motor unit group characteristics outlined in Table 1 indicate that the FF motor unit group had the fastest contraction time, highest fusion frequency, and largest twitch tension, corresponding to that previously seen in the cat LR (51). The ferret SR motor unit group is surprising in that the average twitch tension is not the small fraction of the tension produced by the FF group as seen in the cat (92% in ferret vs. 21% in cat) (51). The tetanic tension generated from the ferret LR SR group is larger (111%) than the typically powerful FF motor units. The SR group in the cat had a much smaller tetanic tension that the FF group (23%) (51).
Whole muscle contractile characteristics. Although the ferret average single motor unit twitch is comparable to that of the cat, the whole muscle results were drastically different. The ferret whole muscle twitch (mean = 3.3 g) is more similar to the squirrel monkey than to the cat (18). The average cat LR maximal tetanic contraction (mean = 109 g) is 6.5 times greater than either the ferret (mean = 17 g) or squirrel monkey (mean = 13.7 g) although the weight of cat is only 3 times larger than these other animals (18, 51). Such a large disparity in whole muscle contractile capabilities between the cat and ferret LR muscles must be addressed in terms of a functional difference. Whereas the cat and ferret are both carnivore hunters, they appear to have completely differing hunting styles. Noticing a possible prey, the cat will make small eye and body adjustments to fixate its visual attention on the prey as it begins to quietly inch closer for an attempted capture (23, 45). In contrast to the cat, the ferret is a far-sighted nocturnal hunter that never stays still but is quickly exploring its environment with rapid up-and-down head movements (23, 32, 42). Vision does not seem to be a primary tool for ferret behavioral testing in that visual stimuli do not always attract the ferret's attention (46). When a ferret does watch a visual target, it seems to be most interested in targets moving at speeds of 25–45 cm/s (2), although the ferret is capable of tracking a light target up to 200°/s (23). Once the ferret has become interested in a potential prey, olfaction appears to become more important than vision in the hunting behavior (2).
The ferret oculomotor system is not well understood in terms of functional capabilities. There is a seeming contradiction with the presented results indicating the slow ferret LR response from extracellular stimulation to those of Hein et al. (23) that reveal that the potential speed of ferret oculomotor function to be comparable to cat that can produce a horizontal saccade over 250°/s (16). No literature exists that can help explain why the ferret utilizes rapid head and postural changes during play as well as hunting.
The ferret eyes are more laterally placed than either the cat or squirrel monkey, which increases the visual field to 270° compared with 180° for cats (42). Having a wider visual field and small laterally placed eyes reduces the ferret binocular visual field to as much as 80° (morphological estimation) or as little as 40° (electrophysiological mapping of cortical cell receptive fields) which is <15% of their visual field (22, 26). In contrast to the ferret, the cat has 120° of binocular vision comprising two-thirds of the entire visual field (22). In addition to having the eyes in a more lateral position, the ferret pupil is a horizontal slit coupled with a horizontal retinal visual streak extending nasally from the area centralis, suggesting a visual horizontal bias that might reduce the amount of required functional lateral globe movement (42).
Predicted muscle tension. Our laboratory has previously evaluated the actual lateral rectus tensions in cat and squirrel monkeys produced via stimulation of cranial nerve VI and found them to be smaller than predicted from multiplying the average single motor unit forces by the number of motoneurons (18, 20, 51). In the ferret, the average single LR motor units twitch force (30.2 mg) multiplied by the number of known motoneurons (n = 183) should produce a whole muscle twitch tension of 5.5 g while the actual value (3.3 g) is only 60% of the predicted. Actual ferret LR maximal tetanic tension produced only 28% of the predicted force (17 g actual vs. 60 g predicted). The loss of actual LR tension is even more drastic in the squirrel monkey in that the actual twitch tension is only 5% of the predicted value and the maximal tetanic tension is only 3.7% of estimated values (18).
The cause of this apparent "loss" of whole muscle force remains unclear. Extraocular muscle fiber serial arrangement and branching has often been postulated (1, 20, 28, 34, 38). The impact of this muscle architecture may result in tension being lost during whole nerve stimulation because the tension generated within one muscle unit cannot be completely transferred to intramuscular connective tissue because the adjoining muscle fibers are also contracting during the whole nerve stimulation (20).
Recent anatomic studies have suggested that extraocular pulleys coupled to the orbit may redirect some force in intact animals (14). The methods in this study removed the issue of periorbital tissues. It is useful to note that while the removal of the orbital pulley or check ligament in previous studies removed the restraint on the LR it did not appear to alter the force production in the cat or monkey (15). The loss of force identified in the EOM cannot be attributed to a pulley diverting tension away from the globe because stimulated muscle units isolated in the orbital layer produce tension on the LR tendon (52). Previous studies have shown that whole muscle and motor unit contractile properties remain consistent whether detached or attached to the globe (19, 31).
In addition to force dispersal and check ligament interactions, another potential source for the nonalgebraic summation of EOM motor units is that of polyneuronal innervation (27, 36, 37). Having several motoneurons innervating individual muscle fibers (polyneuronal) means that more motoneurons exist than are needed for the basic function of the EOM (36). The presence of "extra" motoneurons would explain why the simple algebraic prediction of whole muscle forces is drastically overestimated in the EOM as well as provide insight on the considerable differences in force losses between different animals. To illustrate, the squirrel monkey has been found to possess 2,000 motoneurons and the cat 1,100 motoneurons in the abducens nucleus (37). When this number of motoneurons is used as the multiplier, the resulting potential muscle tension is much larger than the animal can possibly achieve. The ferret on the other hand, with its relative few abducens nucleus motoneurons has a predictive tension much closer to the actual muscle force value. This interesting observation suggests that an animal such as the ferret that may be less reliant on vision for protection and hunting may have less redundancy in the motoneurons innervating the oculomotor system. The animals such as the squirrel monkey that use vision primarily over other senses might maintain a vast number of polyneuronally innervated muscle fibers as a safeguard mechanism to ensure visual function. This polyneuronal innervation may support the reason why the oculomotor system is protected during some motoneuron-wasting diseases (36, 43).
The findings in this study imply that the physiological characteristics of EOM motor units may vary depending on the functional requirements of the animal under investigation. The cat and squirrel monkey have fast LR contractile characteristics and a low LR innervation ratio that may aide in precise eye position adjustments. Contrast the cat and squirrel monkey that utilize these fine ocular corrections with the ferret that uses bodily adjustments to gain an approximate visual target and then focuses its attention with the use of olfaction in its social and hunting strategies. The ferret visual system may not require the speed and precision utilized by other animals and may correlate with the slower LR motor units and larger innervation ratio.
The impetus for this work was to evaluate LR motor units in another animal from the same order as the cat. Previous research on cat and squirrel monkey found vast differences in contractile tensions, speed of contraction, and innervation ratio. We have evaluated the ferret and found that its motor units have similar groupings as the cat using fusion frequency and fatigue. We now have a better understanding of the nature of the ferret LR muscle and motor units. This knowledge together with the observations of ferret head movements during social and hunting situations suggest that its focus on a visual target relies on both head and eye movements. Therefore, the ferret is an excellent animal model providing unique opportunities for research into the interaction between vestibular sensation and the neurodevelopment of the visual system.
|
GRANT |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. R. Anderson, J. Porrill, S. Sklavos, N. J. Gandhi, D. L. Sparks, and P. Dean Dynamics of Primate Oculomotor Plant Revealed by Effects of Abducens Microstimulation J Neurophysiol, June 1, 2009; 101(6): 2907 - 2923. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
P 
