INTRODUCTION

Top Abstract Introduction Methods Results & Discussion References

THE RODENT MODEL is the dominant model used in the study of skeletal muscle contractile function. In a Medline search of the period 1966-1997, rats and mice were employed in 43% of the skeletal muscle contractility studies, a percent >2.5-fold greater than that for any other single species. The main advantage of this model is the ability to measure contractile function in vivo, in situ, and/or in vitro. However, specific muscle training and chronic measurements of contractile function have been difficult in rodents. For example, it is difficult in rodents to study the functional adaptations to eccentric contraction-induced injury by using the more powerful within-animal study designs.

Direct electrical stimulation of the muscle has been used for muscle training and chronic contractile measurements (e.g., Refs. 2 and 9), but it is uncertain whether activation is equal throughout the muscle volume, and there is a good possibility of damage to the muscle fibers induced physically or electrochemically by the electrodes. Direct nerve stimulation would seem to be the best means of stimulation for chronic contractile measurements, but the construction and implantation of electrodes have been difficult in the rodent model. Relatively large, bulky electrodes (1, 6, 7) are hard to implant, especially in the mouse. These designs, coupled with inflexible lead wires, commonly result in nerve damage (Ref. 6; unpublished observations).

We have developed a relatively simple nerve cuff design for stimulation of the mouse common peroneal or tibial nerves. This design permits chronic in vivo contractile function measurements of the anterior and/or posterior crural muscles as well as training of these muscles and the tibial bone. The design should be easily adaptable to the rat model. In the following sections, the construction and implantation of the nerve cuff on the mouse common peroneal nerve will be described in detail, along with a discussion of the contractile function following surgery, success rates, and a comparison of contractile function with our established percutaneous stimulation model.

	METHODS

Top Abstract Introduction Methods Results & Discussion References

Animals

Experimental Procedures

Nerve cuff construction. The wire used most often in construction of the nerve cuff was Teflon-coated, multistranded 90% Pt-10% Ir wire (0.15-mm diameter; 10Ir9/49T, Medwire-Sigmund Cohn). This wire is highly flexible and is deemed critical to the success of the nerve cuff design. We also found 36-gauge, Teflon-coated, multistranded Pd wire (AS 787-36, Cooner Wire) provides good results at a lower cost, but supply of this wire from the manufacturer was uncertain. Silver wire was found unacceptable from an electrochemical standpoint, and we were unable to find small-diameter multistranded stainless steel wire with the flexibility of the Pt-Ir or Pd wire. The main disadvantage of the Pt-Ir-based nerve cuff is its cost [i.e., $5.40 to $10.00 (US dollars) per nerve cuff depending on whether the wire is bought in bulk].

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Construction of the nerve cuff: alignment of wires and suture knots before formation of wire loops. Protective cover on proximal end of wires is also shown.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Construction of the nerve cuff: forming the wire loops. Loops are formed by bending the multistranded wire around a 21-gauge needle.

Nerve cuff implantation. Before implantation, all components of the nerve cuff with the exception of the polyethylene tubing are autoclaved; gas sterilization is recommended for the polyethylene tubing. The nerve cuff implantation is conducted by using aseptic technique. The hair covering the dorsal cervical region and the lateral portion of the hindlimb is clipped, and the remaining hair removed by using a depilatory. The exposed skin is then aseptically prepared. For implantation of the nerve cuff on the left common peroneal nerve, the mouse is placed in right lateral recumbency. The left hindlimb is positioned so that the femur (which is visible through the skin) is perpendicular to the cranial-caudal axis of the mouse. The lower left leg is in turn held so that the tibia is perpendicular to the femur (Fig. 3). This positioning is maintained by taping the left foot on top of the right foot and both feet in turn to the underlying surgical table.

View larger version (78K): [in this window] [in a new window]   Fig. 3.   Nerve cuff implantation: mouse lateral left hindlimb showing incision sites for skin and muscle. Skin is shown as it appears after incision is made, with biceps femoris muscle being exposed. Skin incision is made on a line bisecting the 2 ends of the cut.

View larger version (91K): [in this window] [in a new window]   Fig. 4.   Nerve cuff implantation: position of the nerve cuff relative to common peroneal nerve, biceps femoris muscle flap, and gastrocnemius muscle.

Torque measurement. Torque production by the anterior crural muscles of the mouse was measured by using the apparatus previously described (e.g., Refs. 4 and 5). Nerve stimulation was done by using 75-µs biphasic pulses produced by a Grass S48 stimulator with a SIU-5 stimulus isolation unit set to capacity coupling mode. Isometric tetanic stimulations were elicited by 200-ms trains of pulses at 300 Hz.

Electromyography (EMG) of the tibialis anterior (TA) muscle. To test the compatibility of common peroneal nerve stimulation using our nerve cuff design with TA muscle EMG measurement, we chronically implanted EMG electrodes just beneath the fascial sheath covering the TA muscle (n = 12 mice). The electrodes consisted of 8.9-cm-long lengths of the same wire used for the nerve cuffs (i.e., Pt-Ir wire). The recording surfaces were prepared by deinsulating 3.0 mm of the wires' distal ends. The wires were routed underneath the fascial sheath at the TA muscle's midbelly by using a 23-gauge needle. The two wires ran parallel to each other at a spacing of 2.0 mm and were placed so that their longitudinal axes were perpendicular to that of the superficial TA muscle fibers. The wires were held in place by being sutured to the distal tendinous sheath of the biceps femoris muscle and by application of a small dab of Medbond cyanoacrylate adhesive to their distal tips. The proximal ends of the wires were routed to the dorsal cervical region where they were connected to a Grass P-15 amplifier. A wire acutely implanted beneath the skin in the abdominal region served as the reference electrode. The EMG signal was band-pass filtered (i.e., 10-3,000 Hz) and subsequently sampled by the same computer used for the torque measurements. EMG measurements were made on a given animal no earlier than 5 days after implantation of the EMG electrodes and 16 days after implantation of the nerve cuff.

	RESULTS AND DISCUSSION

Top Abstract Introduction Methods Results & Discussion References

Nerve cuffs were implanted on the common peroneal nerve in 112 mice. Criteria for a successful implantation were established from data collected by using our in vivo model with percutaneous stimulation and fentanyl-droperidol-diazepam anesthesia (Refs. 3, 4; unpublished observations); mean maximal isometric torque of the anterior crural muscles in those experiments (n = 300 mice) equaled 3.17 ± 0.42 N · mm, with 95% of the values being >2.54 N · mm. In the present study, we considered a nerve cuff implantation to be successful if torque equaled or exceeded 2.54 N · mm. In three-quarters of the experiments, torque was checked for the first time between 10 and 26 days after surgery. If the nerve had been damaged, complete recovery in torque was not evident until ~60 days after surgery. Generally, it was easy to distinguish a "successful" implantation from one that was not; torque averaged 2.9-fold higher in successful mice.

Overall, 75% of the implantations were successful; this percent includes implantations conducted during all stages of the technique development. More recently, the success rate has been 93% (i.e., 39 of 42). This improvement is attributed to 1) leaving a longer portion of the nerve cuff beneath the biceps femoris muscle flap, thus reducing the torque applied to the nerve by the cuff; 2) freeing up more nerve before placing in the nerve cuff; and 3) ensuring that the wire loops do not collapse on the nerve.

The mean maximal isometric torque of all successful mice equaled 3.38 ± 0.50 N · mm, 7% greater than the mean of the data for percutaneous stimulation (i.e., 3.17 N · mm); however, body mass of the mice used in the present study was on average 9% greater. These data indicate that successful nerve cuff implantation does not adversely affect the force-producing capability of the anterior crural muscles.

The allowance of sufficient recovery from the nerve cuff implantation is critical to the success of the procedure. If torque was measured before 5 days after the surgery, then erratic torque values were observed, presumably because connective tissue had not developed sufficiently to fix the nerve cuff in place. The erratic results were attributed to movement of the nerve cuff wire loops relative to the nerve, resulting in some stimulation of the adjacent gastrocnemius muscle. However, if torque was evaluated after 7 days, we observed good reliability for the measurement of isometric torque when using this stimulation technique; the within-subjects SD of replicate determinations measured six times over 2 mo equaled 5.5% (n = 6 mice). Furthermore, the nerve cuffs are easily viable out to 4 mo after surgery (n = 6 mice), and we have even followed two mice for 7 mo. When the nerve cuff does fail at these later times, it is usually due to damage of the nerve cuff proximal end inflicted by the mouse or breakage of the nerve cuff wires in the hip region.

We have found the nerve cuff stimulation procedure to be compatible with EMG measurements made on the TA muscle. We see no evidence for stimulation artifact in the EMG signal. Figure 5A shows a representative plot of EMG root mean square as a function of stimulation voltage. Above 1.4 V, there is no further increase in the EMG root mean square as one would expect for an artifact-free signal. The biceps femoris and triceps surae muscles effectively insulate the nerve cuff from the EMG electrodes. The relationship between EMG root mean square and isometric torque production is illustrated in Fig. 5B. Figure 6 shows the stability of both EMG root mean square and isometric torque over the 2-wk period after performance of 150 concentric contractions (n = 8 mice).

View larger version (17K): [in this window] [in a new window]   Fig. 5.   A: tibialis anterior muscle electromyographic (EMG) root mean square (RMS) as a function of stimulating voltage; data are from 1 mouse. B: relationship between peak isometric torque of anterior crural muscles and tibialis anterior muscle EMG RMS as determined by varying stimulation voltage; data are from the same mouse as in A. Solid lines, linear regression fit of data (r = 0.999). Tibialis anterior muscle contributes 89% of isometric torque produced by anterior crural muscles (5).

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Tibialis anterior muscle EMG and maximal isometric torque measured in the 14 days after performance of 150 concentric contractions (n = 8 mice). Two sets of points at day 0 reflect changes in the 2 measurements from immediately before to immediately after contraction bout. Measurements were made on a given animal no earlier than 5 days after implantation of EMG electrodes and 16 days after implantation of the nerve cuff.

In conclusion, we have demonstrated a means for chronic stimulation of the common peroneal nerve in the mouse by using a simple electrode design and implantation procedure. Success rates are high. Training and/or chronic contractile measurements can be initiated by ~2 wk after surgery with assurance that contractile function is stable. We established a minimum isometric torque criterion for categorizing an implantation as successful or not. There was a clear dichotomy in torque production between mice with damaged and undamaged nerves as long as the assessment was made in the first month (i.e., before recovery of damaged nerves had occurred). Finally, with minormodifications, this nerve cuff design could be adapted for use in other species.