Effects of the calcium antagonist nifedipine on the afferent impulse activity of isolated cat muscle spindles

Effects of the calcium antagonist nifedipine on the afferent impulse activity of isolated cat muscle spindles

Brain Research 954 (2002) 256–276 www.elsevier.com / locate / brainres Research report Effects of the calcium antagonist nifedipine on the afferent ...

1MB Sizes 1 Downloads 11 Views

Brain Research 954 (2002) 256–276 www.elsevier.com / locate / brainres

Research report

Effects of the calcium antagonist nifedipine on the afferent impulse activity of isolated cat muscle spindles ¨ M. Fischer*, S.S. Schafer Department of Neurophysiology ( Unit 4230), Hannover Medical School, Carl-Neuberg-Str. 1, D-30625 Hannover, Germany Accepted 12 July 2002

Abstract The impulse activity of muscle spindles isolated from the cat tenuissimus muscle was investigated under varying concentrations of the L-type calcium channel blocker nifedipine. At a concentration of 25 mM nifedipine impulse activity was clearly diminished in both primary and secondary endings. However, low concentrations of the drug (5–10 mM) exerted unexpected excitatory effects. The dynamic properties of primary endings in particular were augmented; those of secondary endings were also increased, although only slightly. A detailed analysis of the afferent discharge patterns obtained under ramp-and-hold stretches yielded the following effects of 10 mM nifedipine. (1) The initial burst at the beginning of the ramp phase of a stretch was increased in primary endings; (2) the peak dynamic discharge frequency at the end of the ramp phase was considerably increased in most primary endings; (3) the sensitivity of the peak dynamic discharge value to varying amplitudes and velocities of stretch was significantly enhanced in primary endings, and also increased, although only slightly, in secondary endings; (4) the rise in the discharge frequency during the ramp phase of a stretch was augmented in both types of ending, the effect being again stronger in primary endings; (5) the fast adaptive decay of the impulse frequency following the ramp phase of a ramp-and-hold stretch was significantly increased in primary endings, but remained unaffected in secondary endings. The enhanced dynamic properties of primary endings were also observed under small sinusoidal stretch stimuli (10 mm, 40 Hz), where nifedipine induced a significant shift in the position of the 1:1 driven action potentials toward smaller phase values. In view of an increase in tension in the isolated muscle spindle and an increased initial burst in primary endings in the presence of nifedipine, it is suggested that the drug facilitates the attachment of cross-bridges in the poles of the intrafusal muscle fibers. The increase in the dynamic properties of primary endings points to the possibility that the drug preferentially affects the nuclear bag 1 fiber. The inhibitory effect on the afferent discharge rate at high doses of the drug is interpreted as the consequence of a calcium channel block in the membranes of the sensory endings. The membrane potential of sensory endings appears to be highly dependent on sustained Ca 21 conductance.  2002 Elsevier Science B.V. All rights reserved. Theme: Sensory systems Topic: Somatic and visceral afferents Keywords: Isolated muscle spindle; Nifedipine; Primary and secondary endings; Ramp-and-hold stretch; Discharge pattern; Tension

1. Introduction The discharge patterns of primary and secondary sensory endings of the mammalian muscle spindle display a variety of dynamic and static properties. Primary endings respond very readily to the dynamic aspects of a mechanical stimulus that excites the receptor. They fire in a *Corresponding author. Tel.: 149-511-532-2773; fax: 149-511-5322776. E-mail address: [email protected] (M. Fischer).

1:1 fashion synchronously with sinusoidal stretches of very small amplitudes, and are largely sensitive to changes in the velocity of the dynamic ramp phase under a ramp-andhold stretch. Secondary endings, by contrast, are hardly affected by very small sinusoidal stretches. They respond most readily to the static components of a ramp-and-hold stretch. While the differences in the discharge patterns of primary and secondary endings are well known, their origin is poorly understood [22]. Different mechanical properties of the intrafusal muscle fibers might be involved, since primary endings and secondary endings

0006-8993 / 02 / $ – see front matter  2002 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 02 )03353-X

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

terminate on different intrafusal fibers: primary endings contact each type of intrafusal fiber, but secondary endings derive their physiological input mainly from the nuclear chain fibers [1,3]. However, it is the nuclear bag 1 fiber that shows the most extensive dynamic properties, and this is usually connected exclusively to the primary sensory ending. Nevertheless, the dynamic features of the primary ending’s response cannot be explained exclusively by the mechanical properties of the intrafusal bag 1 fiber. Giuox, Petit and Proske [14] and Scott [40] report that primary endings of muscle spindles which lack the bag 1 fiber display very similar dynamic properties during a passive stretch to those of spindles that possess both a bag 1 and a bag 2 fiber. Moreover, specific dynamic features such as the fast adaptive decay that follows the peak dynamic discharge at the beginning of the plateau phase of a rampand-hold stretch are certainly not simply mechanical in nature. Changes in the ionic conductance of the sensory ending membrane and of the encoder membrane respectively appear to be responsible for this characteristic feature of the discharge pattern [25,27,37]. As far as the ionic mechanisms are concerned, it is generally assumed that stretch activated cation channels contribute to the conductance changes occurring during a stretch of the sensory membrane. Hunt, Wilkinson and Fukami [25] demonstrate that the receptor potential obtained during brief stretches of the muscle spindle is mainly due to an increase in the conductance of sodium ions (Na 1 ). In this study we inquire into the relevance of calcium conductance to the process of mechanoelectric transduction and its contribution to the dynamic features of the muscle spindle response. Hunt, Wilkinson and Fukami report that D600, a derivative of the calcium antagonist verapamil, slightly reduced the receptor potential amplitude only if the muscle spindle was bathed in an Na 1 free Ringer’s solution, but not if it was bathed in a normal Ringer’s solution. However, Kruse and Poppele [27] mention a decrease in the firing rate of cat muscle spindle afferents even in a normal Ringer’s solution when they treated the isolated receptor with the calcium channel blockers nifedipine, diltiazem or CoCl 2 . Furthermore, Ito and collaborators [26] demonstrate a similar decrease in the impulse frequency of frog muscle spindle endings and a reduction in their receptor potential amplitude when they used the blockers verapamil and nifedipine. Thus it is reasonable to assume that the membrane depolarization occurring during a stretch might be supported by an increase in the conductance of calcium ions (Ca 21 ) as well as by that of sodium. The formerly described effects using dihydropyridines for calcium antagonism favor an inward current through voltage dependent L-type Ca 21 channels, a well-known class of Ca 21 channels that is found in almost all excitable tissues [43]. The existence of this channel has been shown for sensory neurons in dorsal root ganglion cells of chicken and rat where it is co-localized with N-type and T-type Ca 21 channels [31,42]. A single study

257

proves the voltage dependent Ca 21 channels (L-type and N-type) being located in the peripheral nerve fibers of the mouse knee joint by immunohistochemistry [17]. To our knowledge there is no study that similarly demonstrates the channel protein or mRNA of its subunits in the peripheral sensory endings of muscle spindles or in the membrane of intrafusal muscle fibers. However, it is likely that muscle spindle afferents also own that type of channel in their peripheral sensory endings, even though it is not proven so far. In this study, the role of the voltage gated L-type Ca 21 channel is investigated in detail by testing the effects of varying concentrations of the Ca 21 antagonist nifedipine on the discharge activity of primary and secondary endings of isolated cat muscle spindles. The data presented in this report will supply further evidence of the contribution of voltage gated Ca 21 channels to the transduction and encoding process in the mammalian muscle spindle. We observed a general depression of the discharge rate at high doses of nifedipine, and also excitatory effects, mainly enhancing the dynamic properties of primary endings, at low doses of nifedipine. By contrast, the dynamic properties of secondary endings remained nearly unchanged at low concentrations of the Ca 21 channel blocker.

2. Materials and methods

2.1. Spindle preparation and stretching of the isolated spindle The preparation of the isolated muscle spindle was ¨ similar to that previously described by Fischer and Schafer [12]. Muscle spindles were isolated from the tenuissimus muscles of cats that were anesthetized with sodium pentobarbital (45 mg / kg i.v.). The muscle and its nerve supply were excised from the hind limb and transferred into a modified Ringer’s solution [33]. The muscle was fixed in a dissection chamber in accordance with its in situ length when the angle of the femur / tibia joint was adjusted to 1358. Muscle spindles were isolated by removing all the attached extrafusal fibers, connective tissue and blood vessels. Usually, a small number of extrafusal muscle fibers were left in place around the poles of the spindle, in order to support the subsequent fixation of the isolated receptor to the holding rods of a stretching device, using a histoacryl glue (B. Braun-Melsungen). Again, the initial length (L 0 ) of the muscle spindle was adjusted to correspond to its in situ length. Stretching of the muscle spindle was effected by the movements of the two holding rods, which were connected to the membranes of two loudspeakers. These membranes were driven by D/A converted signals from a personal computer. The length changes were controlled with the aid of photocells that converted the intensity of a beam of light reflected from mirrors mounted on the holding rods of the

258

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

stretching device into a DC signal. For the measurement of the tension generated during a passive stretch of the isolated muscle spindle, a slightly different stretching device was used: in this case one pole of the muscle spindle was attached to a miniature mechanoelectrical transducer (SensoNor AE 801). The isolated spindle was repeatedly stimulated by rampand-hold stretches of varying amplitudes and ramp velocities. The stretch amplitudes used in this study were 2.5, 5 and 7.5% of the spindle’s initial length L 0 . The ramp velocities were 20, 40 and 60% of L 0 per second. The plateau phase of the stretch was held for 3 s and the pauses between individual stretches lasted 10 s. In some experiments the muscle spindles were stimulated by sinusoidal stretches of small amplitudes (5–20 mm peak to peak). The stimulation frequency was in the range of 10–100 Hz.

2.2. Recording technique and evaluation of afferent discharges During the preparation of the afferent nerve all the branches that did not supply the muscle spindle were cut. Then the spindle nerve was drawn into an oil-filled chamber and placed on an Ag electrode to provide for the extracellular recording of discharges. The reference electrode was placed close to the receptor in the bathing solution. The afferent discharges were amplified (Grass P511 AC pre-amplifier) and stored on tape for subsequent evaluation. In most experiments, the activity of one to three sensory endings was recorded simultaneously from one spindle nerve (Fig. 1a). The action potentials of different endings were distinguished by their varying amplitudes. Occasionally, action potentials of different endings superimposed and produced an apparently larger spike in the registration (see Fig. 1a). These spikes were handled as being discharges of the afferent fiber with long spikes. They were eliminated from the sequence of action potentials of the afferent fiber with small spikes. The resulting long interspike intervals were excluded from further evaluations, i.e. the calculation of the mean values of discharge frequencies. The differentiation between primary and secondary endings was based on various physiological criteria that have been described previously [12]. One of the most reliable criteria for differentiation was the property, displayed by primary but not by secondary endings, of discharging with each cycle of a sinusoidal stretch when the stretch amplitude was very small (5 mm) and the stimulation frequency was increased from 10 to 100 Hz. The responses of individual muscle spindle endings to ramp-and-hold stretches were evaluated by determining the instantaneous discharge frequencies from the sequence of action potentials and superimposing the responses to five successive stretches in one diagram, as shown for a primary and a secondary ending of a single muscle spindle in Fig. 1b and c respectively. Certain basic discharge

Fig. 1. Afferent activity of a primary and a secondary ending under a ramp-and-hold stretch in normal Ringer’s solution. (a): Oscillogram of action potentials recorded from the afferent spindle nerve at the beginning of a stretch (long spikes5primary ending; short spikes5secondary ending). (b,c): Discharge patterns of the same endings (b: primary ending; c: secondary ending) with five responses superimposed. IA: initial activity; IP: initial peak; PD: peak dynamic discharge; MST: maximum static discharge; FST: final static discharge. The graphs under the oscillogram and under the discharge patterns represent the changes in spindle length (L 0 : initial length; ramp velocity: 40% of L 0 / s).

frequencies were extracted from such a discharge pattern. IA is the initial activity that characterizes the background activity of the ending, and is calculated as the median value of the discharge frequencies during the last 500 ms before the stretch is started. A burst of discharges may occur at the beginning of the ramp phase and form the initial peak (IP) of the discharge pattern. The IP value was determined as the highest instantaneous discharge frequency that occurred during this initial burst. The peak

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

dynamic discharge (PD) is the value that chiefly reflects the dynamic properties of an ending, and is calculated as the median value of the discharge frequencies during the last 25 ms of the ramp phase. Two static values were obtained during the plateau phase of the stretch. The maximum static value (MST) is the highest discharge frequency during the plateau phase and is calculated as the median frequency over a period of 50 ms, usually starting 20 ms after the beginning of the plateau phase following the early fast adaptive decay. The successive slow adaptive component produces a further reduction in the discharge frequency until the final static value (FST) is reached at the end of the plateau phase. FST is the median frequency during the last 250 ms of the plateau phase.

259

acid; 1 g / l glucose [33]. The solution was continuously aerated with 95% O 2 and 5% CO 2 . The pH was adjusted to 7.4 and the temperature was kept constant at 35 8C. Nifedipine (Sigma) was added to the Ringer’s solution in final concentrations of between 5 and 25 mM. Since nifedipine is scarcely soluble in aqueous solutions, the substance was first dissolved in dimethylsulfoxide (DMSO). The most frequently used solution of 10 mM nifedipine contained 0.01% DMSO. In general, the solvent itself did not affect the activity of muscle spindle endings (see Fig. 2). Concentrations of up to 0.1% DMSO were tested. Only two endings seemed to respond to DMSO, generating a small increase in their firing rate. These endings were both eliminated from further evaluation.

2.3. Solutions 3. Results The experimental chamber (volume: 1 ml) containing the isolated muscle spindle was continuously perfused with a modified Ringer’s solution (perfusion rate: 3–5 ml / min). The ionic composition of the solution was: 118.6 mM NaCl; 4.75 mM KCl; 1.80 mM CaCl 2 ; 23.2 mM NaHCO 3 ; 1.19 mM KH 2 PO 4 ; 0.84 mM MgSO 4 ; 2.40 mM glutamine; 3.20 mM glycine; 0.97 mM histidine; 1.02 mM glutamic

The impulse activity of both primary and secondary endings of isolated cat muscle spindles was dramatically modified when the Ca 21 channel blocker nifedipine was applied to the bathing solution. Fig. 2 shows a representative example of the influence of 25 mM nifedipine on the four basic discharge frequencies IA (5initial activity), PD

Fig. 2. Representative example of the influence of 25 mM nifedipine on the basic discharge frequencies IA, IP, PD, MST and FST (see Fig. 1 for abbreviations) obtained from the responses of a primary (a) and a secondary (b) ending of one muscle spindle when the receptor was repeatedly stimulated by ramp-and-hold stretches. Horizontal bars at the top of the diagram represent the periods where nifedipine and its solvent DMSO were present in the bathing solution. DMSO did not affect the afferent activity. Nifedipine reversibly decreased the basic discharge frequencies IA, PD, MST and FST in both types of ending. IP (obtained from the primary ending only) increased under nifedipine. With the primary ending a transient increase in the discharge values occurred prior to the drug-induced inhibition, and the decrease in PD was slower than the decreases in MST, FST and IA.

260

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

(5peak dynamic discharge value), MST (5maximum static discharge value) and FST (5final static discharge value) of a primary ending (a) and a secondary ending (b) of one muscle spindle. Since the primary ending but not the secondary ending generated an initial burst, the initial peak discharge value (IP) is added to panel (a) only. The muscle spindle was repeatedly stimulated by 4–5 rampand-hold-stretches within a minute. The stretch amplitude was 5% of the spindle’s initial length L 0 , and the stretch velocity was 40% of L 0 per second. The horizontal bars at the top of the diagram mark the periods when DMSO (a solvent for nifedipine) and nifedipine were present in the bathing solution. The solvent DMSO did not affect the impulse activity of either type of muscle spindle ending. However, the application of nifedipine reversibly decreased the basic discharge frequencies with one exception, namely the IP of the primary ending, which increased. The static discharge values IA, MST and FST of both endings dropped to zero within a few minutes. Additionally, PD decreased. In the case of the primary ending, PD decreased more slowly than the static discharge frequencies (MST, FST and IA). In the case of the secondary ending, the four basic discharge frequencies declined in a similar manner, with IA decreasing to zero at first, being followed by FST and MST. Moreover, the PD additionally dropped to zero. However, in contrast to the primary ending the slope of PD decay did not seem to be different from the slope of decay of the static discharge frequencies. During the first 2 min or so, nifedipine induced a small and transient excitatory effect on the discharge rate of the primary ending before its depressant effect became dominant. This transient excitation did not occur in the secondary ending that is presented in this figure. However, some other secondary endings did also show a transient excitation. Since neither the site of action of nifedipine on the one hand nor its time of diffusion to the site of action on the other is known, this transient excitation might reflect either a time dependent or a concentration dependent effect of the drug. We therefore tested in some preliminary experiments various concentrations of nifedipine, delaying the measurement of the effects until 8–10 min after its application. If the excitatory effect were time dependent, it would cease during this period of delay and only the inhibitory effect would be seen. If the effect were concentration dependent, a sustained excitatory effect would be observed for a certain drug concentration even after a delay of about 10 min. Fig. 3 shows the results of these few experiments obtained from four primary endings (a) and six secondary endings (b). The mean values and the standard deviations of the basic discharge frequencies IP, IA, PD, MST and FST are plotted against the concentration of nifedipine in the bathing solution. For the sake of greater clarity, the symbols for IP, IA and MST are plotted with a slight horizontal displacement. It should be noted that the standard deviations in this figure do not only reflect the

Fig. 3. Mean values and standard deviations of the basic discharge frequencies IP, IA, PD, MST and FST plotted against the concentration of nifedipine. (a) Primary endings (n 5 4), (b) secondary endings (n 5 6). IA, PD, MST and FST decrease under 25 mM nifedipine in both types of ending; IP of primary endings increases. Low concentrations of the drug (5 mM) increase each of the basic discharge values in primary endings, with PD reaching its maximum at 10 mM nifedipine. With secondary endings IA is slightly increased at 5 mM nifedipine. The symbols for IP, IA and MST are plotted with a slight horizontal displacement.

variance of the nifedipine effects; they also represent the different levels of activity of the individual endings included in the evaluation. The effects of nifedipine were tested at concentrations of 5, 10 and 25 mM, and were compared to the basic discharge frequencies obtained in a normal Ringer’s solution (0 mM). Within 8–10 min after the application of low doses of the drug (5 and 10 mM) each ending appeared to have reached a nearly constant state of activity. However, we cannot exclude a very slowly continuing decrease in the remaining impulse activity (PD) under a nifedipine concentration of 25 mM. With primary endings we observed a weak increase in

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

each of the five basic discharge values under 5 mM nifedipine (Fig. 3a). IP displayed a further increase with increasing concentrations of nifedipine. However, this augmentation was not significant, since only two out of four afferents were clearly affected. Notably, mean PD value reached its maximum under 10 mM nifedipine. This was the only discharge value that was significantly increased when compared to the control values obtained in normal Ringer’s solution (paired Student’s t-test: P50.05). IA, MST and FST all declined to zero under 25 mM nifedipine. At this concentration one of the four primary endings became inactive apart from a short initial burst at the beginning of the stretch, but the other three endings continued firing during the ramp phase of the stimuli, so that the mean PD value was about 40 imp / s. With secondary endings the inhibitory effect of nifedipine was just as obvious as with primary endings. On average each of the four basic discharge frequencies decreased under 10 and 25 mM nifedipine (Fig. 3b). Four of the six secondary endings were completely inactive at 25 mM nifedipine. An excitatory effect of nifedipine could hardly be seen in the average results from these six endings. However, a slight increase in mean IA was observed under 5 mM nifedipine, due to the fact that three endings showed increased background activity, two remained unaffected, and only one ending’s IA decreased. Our findings clearly show that a high dose of nifedipine induces a dramatic fall in the activity of muscle spindle afferents. At low concentrations, however, the drug might induce excitatory and / or inhibitory effects. It is this state of activity that is most interesting when we inquire into the contribution of calcium channels to the dynamic and static components of the afferents’ responses, because at these concentrations, when only some of the calcium channels are blocked, we might be able to identify definite effects on specific components of the stretch response. At 10 mM nifedipine we observed the strongest augmentation for PD of primary endings that was even significant in these preliminary experiments. However, statistics on a small number of four endings are extremely unconfident. We therefore decided to investigate the effects of 10 mM nifedipine, as compared to 0 mM nifedipine in a normal Ringer’s solution, on a larger number of endings and under varying conditions of stretch.

3.1. Effects of nifedipine on the discharge patterns of primary and secondary endings under varying amplitudes and varying velocities of ramp-and-hold stretch The effects of nifedipine on the discharge frequencies of 20 primary and 22 secondary endings were investigated when the isolated muscle spindle was stimulated by rampand-hold stretches of varying amplitudes, while the stretch velocity was kept constant. Then the same collection of endings together with a further primary ending were tested

261

under changing ramp velocities, while the amplitude of the stretch was kept constant. The experiments were carried out in a normal Ringer’s solution and in a Ringer’s solution containing 10 mM nifedipine. The results are shown in Fig. 4, where the mean values of IA, PD, MST and FST are plotted against the amplitude of stretch in (a) and (b), and against the varying velocities of stretch in (c) and (d). Effects of nifedipine on IP will be described in a separate section (see Section 3.4). The broken lines in Fig. 4 represent the control values obtained in a normal Ringer’s solution; the solid lines connect the discharge values of the same endings obtained 10 min after the application of 10 mM nifedipine. For the sake of greater clarity it was omitted to display standard deviations in these diagrams. The standard deviations are large, reflecting predominantly the different levels of activity of the individual endings included in the evaluation. They range from 18 to 43 imp / s for the mean basic discharge frequencies of primary endings in Ringer’s solution and from 20 to 50 imp / s for the values obtained under nifedipine. For secondary endings the standard deviations are smaller: 16–24 imp / s in Ringer’s solution and 17–22 imp / s under 10 mM nifedipine. Fig. 4a shows that the PD, MST and FST of primary endings rose with an increasing amplitude of stretch, while IA stayed constant. This well known effect of a stretch stimulus was observed in a normal Ringer’s solution as well as in the solution containing the drug. The specific effect of nifedipine was an increase in the mean PD value, which was more obvious under large amplitudes of stretch (5–7.5% of L 0 ) than under the small stretch amplitude of 2.5% of L 0 . By contrast, the means of the static values MST and FST decreased in the drug solution. However, the third static value IA increased. Similar findings were observed for primary endings when the spindle was stimulated with stretches of varying velocities (Fig. 4c). Here, the excitatory effect of nifedipine on PD was most obvious under rapid stretches (40–60% of L 0 per second). And again, MST and FST declined but IA increased when nifedipine was applied to the bathing solution. In order to understand the contradiction in the behavior of the three static discharge values it is necessary to go into the details. Most of the primary endings did not fire spontaneously, i.e. 14 out of 21 primary endings did not generate an IA in a normal Ringer’s solution. However, two of these primary endings were excited by nifedipine in such a way that they produced an IA only in the solution containing the drug. These two endings strongly increased the mean value of IA under nifedipine (Fig. 4a). Thus the discrepancy between the effects of nifedipine on MST and FST on the one hand and IA on the other is misleading. Usually, the three static values increased or decreased in common. The PD value, however, was generally affected in a particular way: it often increased under nifedipine when the static values decreased. This extraordinary role of PD was apparent even when all the four basic discharge values increased:

262

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

Fig. 4. Mean values of the basic discharge frequencies IA, PD, MST and FST of primary endings (a: n 5 20; c: n 5 21) and secondary endings (b,d: n 5 22) plotted against varying amplitudes (a,b) and velocities (c,d) of a ramp-and-hold stretch in a normal Ringer’s solution (broken lines) and in a Ringer’s solution containing 10 mM nifedipine (solid lines). With primary endings the mean PD value increases in the presence of the drug (* paired Student’s t-test; P,0.05) and the mean static values MST and FST decrease. The static value IA, however, increases under nifedipine (see text for further explanation). With secondary endings the four basic discharge frequencies do not on average show any change.

the rise in PD was then usually larger than the rises in the static values. Conversely, when all the four basic discharge values declined the PD value fell less than the static values. According to these observations it is useful to distinguish between effects of nifedipine that influence the discharge frequency in general, therefore changing the ‘level of activity’ of an ending, and additional effects that exclusively influence a specific component of the response, e.g. the PD. The data indicate that effects on the three static discharge values (MST, FST and IA) reflect solely the influence of nifedipine on the level of activity of an ending, while changes in PD display the sum of effects on the level of activity and an additional excitatory effect. The

latter but not the former effects markedly changed the shape of the discharge pattern obtained under a stretch. To summarize the influence of 10 mM nifedipine on the level of activity: we observed that it was clearly enhanced in only five out of 21 primary endings, and was clearly reduced in six primary endings. The remaining ten primary endings displayed an activity level close to that in a normal Ringer’s solution. In line with this observation, there is little divergence between the broken and solid lines in the diagrams of Fig. 4a and c. Consequently this evaluation lacks significant effects of nifedipine apart from two exceptions: the mean PD value is significantly increased under nifedipine, when the spindle was treated with ramp-

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

and-hold stretches of large amplitude (7.5% of L 0 ) or high stretch velocity (60% of L 0 / s), respectively (paired Student’s t-test: P,0.05). In both cases, the specific excitatory effect of nifedipine on PD appears to surpass any effect of the drug on the level of activity. With secondary endings, there was no difference between the effect of nifedipine on the dynamic discharge value PD and that on the static discharge values. Fig. 4b and d demonstrate that on average the values of the four basic discharge frequencies did not change when the spindle was treated with 10 mM nifedipine, either when the amplitude or when the velocity of stretch was varied. However, it should be borne in mind that the mean values do not show the influence of nifedipine on the individual ending, as the level of activity increased in six of the 22 secondary endings and decreased in five. The results of these experiments indicate that nifedipine affects the muscle spindle endings in two different ways. Apart from the change in the level of activity that affects both types of ending, nifedipine seems to facilitate in particular the dynamic properties of primary endings. This effect will be elaborated in more detail using evaluations that extract the specific excitatory effects on the dynamics from the general effects of the drug on the level of activity.

3.2. Effect of nifedipine on the sensitivity of PD The slopes of the PD curves in Fig. 4 depict the variation of the sensitivity of the basic discharge value PD with the stretch amplitude, which will be designated the length sensitivity of PD, and with the stretch velocity, which will be designated the dynamic sensitivity of PD. In Fig. 5a the mean values and standard deviations of the length sensitivities of PD are displayed for 20 primary endings (Ia) and 22 secondary endings (II). Fig. 5b shows

263

the mean dynamic sensitivities of PD. Each pair of bars compares the mean sensitivity in a normal Ringer’s solution (dark bar) with that in a Ringer’s solution containing 10 mM nifedipine (light bar). Nifedipine significantly increased both the length sensitivity and the dynamic sensitivity of PD in primary endings (paired Student’s t-test: P,0.001 in Fig. 5a; P,0.05 in Fig. 5b). In a normal Ringer’s solution, PD rose by an average of about 4.5 impulses with each increase of 1% of L 0 in the stretch amplitude. However, under nifedipine PD increased by as much as 6.3 impulses with the same increase in the stretch amplitude. Thus the length sensitivity of PD was increased by about 40% of the control value. With seven out of 20 primary endings, the PD sensitivity was even doubled when the blocker was added to the bathing solution. The dynamic sensitivity of PD was somewhat less affected by nifedipine: an average increase of 20% of the control value was measured when it was applied to the muscle spindle. The sensitivity of the static discharge values MST, FST and IA to varying stretch amplitudes and velocities did not change under nifedipine (not shown). The increase in the length sensitivity and dynamic sensitivity of PD under 10 mM nifedipine was independent of the drug’s influence on the level of activity. As was mentioned in Section 3.1, the level of activity was clearly enhanced in a group of five out of 21 primary endings, and it was clearly reduced in a group of six primary endings. However, when nifedipine was applied, the mean PD sensitivities increased in both of these groups, and the same was observed for the group of the remaining primary endings that did not clearly change their levels of activity under 10 mM nifedipine. With secondary endings we observed a very small and insignificant increase in the mean length sensitivity of PD

Fig. 5. Effect of 10 mM nifedipine on the length sensitivity (a) and dynamic sensitivity (b) of PD in primary (Ia) and secondary (II) endings of isolated muscle spindles. Dark bars represent the mean control values and their standard deviations obtained in a normal Ringer’s solution; light bars represent the mean sensitivity values and standard deviations obtained under nifedipine. The length sensitivity and the dynamic sensitivity of primary endings are significantly enhanced by nifedipine (paired Student’s t-test: *P,0.05, ***P,0.001). With secondary endings the drug-induced augmentation is negligible.

264

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

(Fig. 5a) and no change in its mean dynamic sensitivity (Fig. 5b). Moreover, the sensitivities of the static discharge values did not change when nifedipine was added to the bath (not shown), except for a very small increase in the sensitivity of MST to varying stretch amplitudes.

3.3. Effect of nifedipine on the fast adaptive decay from PD to MST During the plateau phase of a ramp-and-hold stretch the adaptation of a muscle spindle ending is represented by the decline in the impulse frequency from PD to FST. This decay breaks down into fast adaptive decay from PD to MST and slow adaptive decay from MST to FST (Fig. 1b and c). The fast decay is large in primary endings and small in secondary endings. While fast adaptive decay has been reported to reflect the dynamic properties of an ending and is therefore mainly dependent on the velocity of a stretch, slow decay is exclusively dependent on the

amplitude of a stretch [7,37]. Our results obtained in Ringer’s solution confirm these findings for isolated muscle spindles (see broken lines in Fig. 4). The curves of PD and MST increase by nearly the same degree with increasing stretch amplitudes. Thus fast adaptive decay remains constant. However, the curves of MST and FST diverge, i.e. the slow decay increases with increasing stretch amplitudes (Fig. 4a and b). With increasing velocities of stretch only PD rises (Fig. 4c and d). Therefore fast adaptive decay is velocity dependent, but slow adaptive decay is not. How did nifedipine influence the amplitude of the fast and slow adaptive decay? Fig. 6 shows the mean values and standard deviations of the fast adaptive decay from the experiments with varying stretch amplitudes ((a) and (b)) and varying stretch velocities ((c) and (d)). Data from primary endings is displayed in (a) and (c), and from secondary endings in (b) and (d). Again, dark bars represent the control values obtained in a normal Ringer’s

Fig. 6. Influence of 10 mM nifedipine on fast adaptive decay (PD-MST) in primary endings (a,c) and secondary endings (b,d) obtained under varying amplitudes (a,b) and velocities (c,d) of a ramp-and-hold stretch. Dark bars represent the control values and their standard deviations obtained in a normal Ringer’s solution; light bars represent the mean values and standard deviations of the fast decay obtained under nifedipine. The number of endings that contribute data to the evaluation varies and is given in parentheses for each pair of bars. Nifedipine significantly increases fast adaptive decay under most of the varying stretch conditions (paired Student’s t-test: **P,0.01). By contrast, fast adaptive decay is not influenced by nifedipine in secondary endings.

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

solution and light bars those obtained in the drug solution containing 10 mM nifedipine. The number of endings that contribute data to the bars of each pair is constant (see numbers in parentheses above each pair of bars). However, the number of results varies between different pairs of bars, as we excluded from the evaluation those endings that did not produce an MST value under nifedipine. This occurred most frequently in primary endings where nifedipine lowered the three static discharge values and the amplitude of the stretch stimulus was small. If MST drops to zero, the difference between PD and MST is not a reliable value for the fast adaptation of an ending. The bar charts (a) and (c) show that nifedipine significantly increased the fast adaptive decay of primary endings under varying conditions of stretch (paired Student’s t-test: P,0.01). However, with a small amplitude of stretch (2.5% of L 0 ) the fast adaptive decay was nearly unchanged under nifedipine. It is worth mentioning that fast adaptive decay was independent of the amplitude of stretch in a normal Ringer’s solution but dependent on the stretch amplitude in the drug solution (Fig. 6a). This observation also held true when the data was reinvestigated with a constant number of 14 endings contributing data to each pair of bars. With secondary endings, fast adaptive decay was independent of nifedipine (Fig. 6b and d). The slow adaptive decay from MST to FST is not displayed in the figure. It did not show any significant change under nifedipine, either in primary endings or in secondary endings.

3.4. Effect of nifedipine on the initial burst of primary endings The initial burst that forms the IP of the discharge response to a brief stretch has been assumed to be due to an intensive lengthening of the equatorial region of the receptor at the start of the stimulus, because a number of myosin heads appear to be attached to actin in the polar ends of the intrafusal muscle fibers during the initial phase of a stretch, before detaching under a further extension [5,23,41]. Thus effects of nifedipine on IP might indicate a change in the stiffness of the intrafusal muscle fibers. In the experiment presented in Fig. 2a, the IP of the primary ending was increased as long as nifedipine was present in the bathing solution. The preliminary results displayed in Fig. 3 additionally indicate an augmentation of IP under nifedipine. However, only nine out of 21 primary endings produced a well-developed IP that could be easily compared in the normal Ringer’s solution and in the nifedipine solution. This is due to the fact that the formation of an IP is largely dependent on the history of the muscles spindle prior to the stretch stimulus [5,35,36]. The initial burst is best facilitated by fusimotor stimulation applied to the muscle spindle at its resting length prior to the test stretch. This procedure is thought to promote the

265

reorganization of cross bridges between actin and myosin. Conversely, an extensive lengthening of the receptor that often cannot be avoided during the isolation of the spindle weakens the development of an initial burst. Some spindles appear to recover spontaneously and display a well-formed IP in the primary ending’s discharge pattern, but others do not. Nevertheless, most of those endings that do not entirely recover display a steep initial increase at the start of the stretch before the impulse rate continues to ascend to the PD. An example is presented in the left-hand panels of Fig. 7a–c showing discharge patterns of one primary ending in a normal Ringer’s solution under varying stretch velocities (a: 20% L 0 / s; b: 40% L 0 / s; c: 60% L 0 / s). The stretch stimulus is depicted beneath each discharge pattern. Dotted lines represent zero imp / s. The amplitude of the initial increase is clearly enlarged with rising velocity of stretch but it remains constant when the amplitude of stretch is varied (not shown). Thus the initial increase shares the characteristics that are known for the initial burst. With some primary endings a well-developed initial burst arose from the initial increase when nifedipine was applied to the bathing solution. This effect is illustrated in the right-hand panels of Fig. 7a–c. The discharge patterns shown in these panels were obtained under 10 mM nifedipine from the same primary ending that lacks a well-developed initial burst in a normal Ringer’s solution (left-hand panels). According to these findings it is reasonable to combine the IP data with data of the initial increase so as to enlarge the number of endings for an evaluation of the nifedipine effects. Fig. 7d shows the results obtained from investigations with varying velocities of stretch. With the intention to eliminate from the evaluation unspecific excitatory or inhibitory effects of nifedipine on the level of activity the mean values of IP-MST were displayed. Even though each of the static discharge values mirrors the effect of nifedipine on the level of activity, a subtraction of MST was chosen because it is the static discharge frequency that rarely dropped to zero imp / s when nifedipine (10 mM) was applied to the muscle spindle. The dark bars in Fig. 7d represent mean control values of IP-MST obtained in a normal Ringer’s solution and the light bars represent those obtained in the drug solution. Numbers in parenthesis above each pair of bars denote the number of endings that contribute data to the evaluation. The nifedipine effect was an increase in IP-MST by more than 40% with regard to the control values. The effect was significant for a ramp velocity of 40% L 0 / s and 60% L 0 / s (paired Student’s t-test: P,0.05) but insignificant for the slow stretch (20% L 0 / s). It is worth mentioning that secondary endings usually produce a steep initial increase as well. Its amplitude also rises with increasing ramp velocities and remains constant with varying stretch amplitudes. With secondary endings, however, nifedipine tends to decrease the amplitude of the initial increase (not shown).

266

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

3.5. Effect of nifedipine on the rise in the discharge frequency during the ramp phase of a stretch

Fig. 7. Effect of nifedipine on the initial burst of primary endings obtained under varying velocities of a ramp-and-hold stretch. (a–c) Discharge patterns of one primary ending in a normal Ringer’s solution (left-hand panels) and in a solution containing 10 mM nifedipine (righthand panels). The graphs under the discharge patterns represent the changes in spindle length. A well-developed initial peak (IP) arises from the initial increase of the discharge patterns when nifedipine is present in the bathing solution. (d) Influence of 10 mM nifedipine on IP-MST in primary endings. Dark bars represent the control values and their standard deviations obtained in a normal Ringer’s solution; light bars represent the mean values and standard deviations of IP-MST obtained under nifedipine. The number of endings that contribute data to the evaluation varies and is given in parentheses for each pair of bars. Nifedipine significantly increases IP-MST under high velocities of stretch (paired Student’s t-test: *P,0.05).

In order to confirm the finding that nifedipine enhances the dynamic properties of primary endings, the rise in the discharge frequency during the ramp phase of a stretch was analyzed in more detail. From the impulse activity during the last section of the dynamic ramp phase of a stretch stimulus we calculated the linear regression line: the method is illustrated in Fig. 8a and b. The responses obtained during five stretches were superimposed in each diagram and the regression line calculated from the linear part of the impulse pattern. The regression coefficient (5the slope of the regression line) obtained from the ending’s discharge pattern in Ringer’s solution (a) was than compared with that obtained in a Ringer’s solution containing 10 mM nifedipine (b). The bar charts show the mean values of the regression coefficients of primary endings ((c) and (e)) and secondary endings ((d) and (f)) obtained in a normal Ringer’s solution (dark bars) and in 10 mM nifedipine (light bars). Results from experiments with varying stretch amplitudes are shown in (c) and (d), and from ones with varying stretch velocities in (e) and (f). If the impulse frequencies were widely scattered during the ramp phase, the calculation yielded an insufficient correlation coefficient (r , 0.6) and the regression coefficient was not reliable. In such a case the ending was excluded from the evaluation. Thus the number of endings represented by different pairs of bars in Fig. 8c–f varies. However, the bars of one pair always contain an equal number of endings (see number in parentheses above each pair of bars). Thus these diagrams should be used to compare the nifedipine effect under each condition of stretch separately, rather than the effects of changing the stretch amplitude or the stretch velocity. With primary endings the regression coefficients were larger under nifedipine than in a normal Ringer’s solution. This effect was observed for each of the different stimulus conditions (Fig. 8c and e). The increase was usually significant (paired Student’s t-test: P , 0.05). However, the nifedipine effect was not significant when the ramp phase of the stimulus was very short (stretch amplitude5 2.5% of L 0 ). Under this stretching condition it was hardly practicable to assess a reliable regression coefficient. Therefore only four primary endings were evaluated. However, the effect of nifedipine on these few endings showed the same trend as was observed under the other stretching conditions. Secondary endings also showed a small increase in the slope of the regression lines under nifedipine (Fig. 8d and f). Moreover, under some stretching conditions this increase was even significant. This result is surprising because on average neither IA nor PD changed under nifedipine (see Fig. 4). However, the shape of the impulse pattern between both values was slightly altered when nifedipine was applied to the bathing solution. Most of the

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

267

Fig. 8. Effect of nifedipine on the rise of the discharge rate during the ramp phase of a ramp-and-hold stretch. The slopes of the regression lines (5regression coefficients), calculated from the linear part of the afferent responses during the ramp phase, were measured in a normal Ringer’s solution (a) and in a Ringer’s solution containing 10 mM nifedipine (b). Mean values and standard deviations of the regression coefficients are plotted for different stretch amplitudes (c,d) and stretch velocities (e,f). Dark bars represent the mean control values obtained in a normal Ringer’s solution; light bars represent the mean values obtained under nifedipine. The number of endings that contribute data to the evaluation varies and is given in parentheses for each pair of bars. With primary endings (c,e) the mean regression coefficient is clearly enhanced under nifedipine; with secondary endings a minor increase is observed. (paired Student’s t-test: *P,0.05).

268

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

secondary endings show a rapid initial increase in the impulse rate at the beginning of the ramp phase of the stretch that is followed by a less steep increase ascending to PD (see Fig. 1c). It is this initial increase that was slightly diminished under nifedipine. Instead the slope of the subsequent impulse rate during the ramp phase was slightly enlarged. In summary, nifedipine enhanced the increase in the impulse frequency during the ramp phase of a stretch in both types of ending. However, the effect was smaller in secondary endings than in primary endings.

3.6. Effect of nifedipine on primary endings under sinusoidal stretches Small sinusoidal stretches with a high frequency of stimulation are suitable for testing the dynamic properties of muscle spindle endings [12,24]. Primary endings of isolated muscle spindles tend to fire during each cycle of a 10-mm sinusoidal stretch stimulus when the stimulation frequency varies between 10 and 100 Hz. Secondary endings, by contrast, not possessing dynamic properties equivalent to those of primary endings, do not usually show 1:1 driving under such a small amplitude of stretch unless the stimulation frequency is very close to the spontaneous discharge frequency of the individual ending. The high sensitivity of primary endings to the dynamic components of the stretch, i.e. velocity and acceleration, corresponds to the early phase position of their action potentials in relation to the sinusoidal increase in length. The better the dynamic properties are, the earlier firing may occur. We therefore tested whether nifedipine was able to enhance the dynamic properties of primary endings in such a way that it was possible to observe a phase shift toward smaller phase values within the stimulus cycle. Eleven primary endings were studied using sinusoidal stretch stimuli. While the muscle spindle remained under continuous stimulation, the normal Ringer’s solution of the bath was replaced by a solution containing 10 or 25 mM nifedipine. Fig. 9 shows two representative examples of the phase shift induced by nifedipine in primary endings. The two oscillograms in (a) show the action potentials of the afferent nerve of an isolated muscle spindle in a normal Ringer’s solution (upper trace) and in a solution containing 10 mM nifedipine (lower trace). The receptor was continuously stimulated by a sinusoidal stretch of 10 mm amplitude and a stimulation frequency of 40 Hz, which is depicted underneath. Responses to 30 sinusoidal stretches are superimposed in each oscillogram. Three different sensory endings can be distinguished by the sizes of their action potentials: the longest spikes belong to the primary ending of the muscle spindle, the medium and short spikes to two different secondary endings. In Ringer’s solution (upper trace) the three endings fire in a phase-locked manner in relation to the sinusoidal stretch stimulus. We

arbitrarily defined the most relaxed state of the muscle spindle as phase 08 of the sinusoidal cycle of stretch (see scale beneath the diagram). Thus the maximum stretch corresponds to a phase position of 1808, while the discharges of the primary ending occupy a phase position of about 628. The phase positions of the discharges of the two secondary endings are about 738 and 1008 respectively. When the bathing solution was replaced by a solution containing 10 mM nifedipine, the primary ending was the only one that continued firing in a phase-locked position relative to the stretch stimuli (lower trace); however, the action potentials shifted to an earlier phase position (ca. 308). By contrast, both secondary endings lost their phaselocked activity. This was in line with the observation that both secondary endings showed an increased level of activity under low doses of nifedipine (not shown in this figure). Their discharge frequencies rose from 40 imp / s in the normal Ringer’s solution to about 50 imp / s and about 53 imp / s respectively in the drug solution. The primary ending, however, continued firing at 40 imp / s. In Fig. 9b the shift in the phase position of the same primary ending discharge is plotted against time. The phase position of each individual action potential relative to the sinusoidal stretch is represented by a single gray dot. The change in the mean phase position that was calculated sequentially from small sections of the response of the primary ending (periods of 5 s) is represented by the curve. The horizontal bar at the top of the diagram marks the period during which nifedipine was present in the bathing solution surrounding the muscle spindle. Shortly after the bathing solution was exchanged, the phase position of the discharges shifted, as mentioned above, from about 628 to about 308. The early phase position remained nearly constant for the rest of the experiment. Fig. 9c shows a second example of a phase shift in the discharge activity of another primary ending in relation to a continuing sinusoidal stretch (10 mm, 40 Hz). In this case the normal Ringer’s solution was replaced by a solution containing 25 mM nifedipine, as is illustrated by the horizontal bar at the top of the diagram. In this ending, nifedipine induced an early transient phase shift towards smaller phase values and a late phase shift in the opposite direction, towards larger phase values. Thus in the first 2 min or so after the application of the drug, nifedipine induced the ending to discharge earlier in relation to the cycle of sinusoidal stretch, but subsequently the action potentials occurred at a point in time that was actually later than the time of discharge in a normal Ringer’s solution. These examples represent the two types of effect of nifedipine on the phase position that were usually observed in primary endings. The two different types of response were each observed under both 10 and 25 mM nifedipine. Furthermore, the type of response was not related to any definite stretch amplitude (5–20 mm) or stimulation frequency (40–100 Hz). Thus the variability in the efficacy of nifedipine during experiments using sinusoidal stretches

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

269

Fig. 9. Effect of nifedipine on the phase position of discharges in respect of small sinusoidal stretches (10 mm; 40 Hz). (a) Upper trace: oscillogram of the impulse activity recorded from the afferent spindle nerve during sinusoidal stretching of the spindle in a normal Ringer’s solution (superposition of 30 stretch cycles; long spikes5primary ending; medium and short spikes5two secondary endings). Lower trace: oscillogram of the same endings’ activity under 10 mM nifedipine. The graph underneath the oscillograms represents the changes in spindle length (L 0 5initial length). X-axes are scaled in phase degrees with respect to the cyclic sinusoidal stretch. Nifedipine induces a shift of the primary ending discharge to smaller phase values. The two secondary endings fail to discharge in a phase-locked manner under nifedipine. (b) Shift in the phase position of the primary ending discharge that is induced by the application of 10 mM nifedipine to the bathing solution (same ending as in a,b). (c) Shift in the phase position of another primary ending’s discharge induced by 25 mM nifedipine. For (b) and (c) horizontal bars at the top of each panel represent the periods where nifedipine is present in the bathing solution. Horizontal auxiliary lines represent the phase position measured shortly before nifedipine was applied. The phase position of each individual action potential is shown by a single gray dot. The change in the mean phase position is displayed by the curve. (d) Phase shift of the action potential position of eight primary endings (thin lines) in respect of small sinusoidal stretches (5–20 mm, 40 Hz) during the period of nifedipine presence (10–25 mM). Reference phase position in a Ringer’s solution is set to 08. Closed circles mark the maximum negative phase shift of an ending. Mean phase shifts and standard deviations of the collective of eight primary endings are displayed for different points in time (squares) and connected by thick lines. Nifedipine induced an early negative phase shift in each primary ending (paired Student’s t-test: *P , 0.05) and a subsequent delayed firing in half of the primary endings.

was similar to that observed under ramp-and-hold stretches. With 8 out of 11 primary endings the phase position of discharges was continuously acquired over a period of 8 min in a solution containing 10 mM (n55) or 25 mM (n53) nifedipine. The stretch amplitude was within a range of 5–20 mm and the stimulation frequency was 40 Hz. Results are collected in Fig. 9d where thin lines represent the phase shifts of each of these endings’ discharges relative to the spike position in a normal Ringer’s solution that is arbitrarily set to 08 for the sake of

comparison. The time scale (abscissa) starts with the beginning of the nifedipine application. Shifts to negative phase values represent earlier discharges; shifts to positive values represent delayed discharges. Each curve is composed of four measurements with each value representing an averaged phase position over a period of 5 s. The curves obtain (1) the reference phase position in normal Ringer’s solution (508); (2) the maximum negative phase shift at varying points in time after the start of nifedipine application (closed circles); (3) and (4) the phase shift measured 5 and 8 min after the start of nifedipine application. In

270

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

addition a thick line is inserted in Fig. 9d that connects the mean values of the phase shifts calculated from the collective of the eight primary endings. Obviously, each ending was affected by nifedipine and showed an early negative phase shift with varying delay following the application of nifedipine (range of delay: 60–282 s; mean: 139685 s). The degree of the negative phase shift was within a range of 2–408 (mean: 14.4615.68). Even though the variability was large the effect is significant (paired Student’s t-test: P,0.05). Four out of eight endings continued firing in the preceding phase position for the rest of the investigation. The remaining four endings showed a subsequent shift to delayed firing with regard to the position of discharge in a normal Ringer’s solution. Most of the endings did not reach a steady state in phase position during the 8 min of nifedipine presence. However, none of the primary endings lost its phase-locked activity like secondary endings did. This finding underlines the high sensitivity of primary endings to small sinusoidal stretch stimuli. The initial phase position was generally regained after a period of washing in a normal Ringer’s solution (not shown).

3.7. Effects of nifedipine on the tension of isolated muscle spindles In order to investigate the effects of nifedipine on the mechanical properties of muscle spindles, we measured the tension developed by the receptor during a passive stretch. In these experiments the normal Ringer’s solution of the bath was replaced by a Ringer’s solution containing 25 mM nifedipine, while the spindles were repeatedly stimulated by ramp-and-hold stretches (amplitude: 5% of L 0 , ramp velocity: 40% of L 0 per second). Thirteen muscle spindles were tested in this series of experiments. The isolated receptor was simultaneously observed under a microscope with 600-fold magnification. When the isolated muscle spindle was exposed to nifedipine, no visible length change of any part of the intrafusal fibers occurred. By contrast, contractions of nuclear bag fibers induced by adding succinylcholine to the bathing fluid were easy to recognize. Nevertheless, measurements using a mechanoelectrical transducer revealed a clear increase in tension in four muscle spindles resulting from the application of nifedipine. One such example is shown in Fig. 10. The tension, repeatedly altered by ramp-and-hold stretches, is plotted against time in Fig. 10a. The time scale is interrupted where long periods of wash-out (about 20 min) were included. For the sake of greater clarity a horizontal line corresponding to the initial tension of the spindle is inserted into the diagram; this was measured before the start of a stretch in a normal Ringer’s solution. When the normal Ringer’s solution was replaced by a Ringer’s solution containing 25 mM nifedipine (horizontal bar at the top of the diagram), the initial tension of the spindle

increased slightly. Furthermore, the peak tension, i.e. the maximum tension during each stretch, also increased. There may be confusion about the effect reaching its maximum some time after the application of nifedipine. Such delayed responses have been observed frequently and probably result from diffusion barriers hindering the onset of drug effects as well as the recovery of the spindle. In Fig. 9b averaged tension curves, each obtained from five successive responses to ramp-and-hold stretches, are plotted against an expanded time scale. Two curves are superimposed in the diagram: the lower curve represents the tension of the muscle spindle as measured in a normal Ringer’s solution (see bracket b in Fig. 10a), while the upper curve represents the tension affected by 25 mM nifedipine (see bracket b9 in Fig. 10a). The length change of the spindle is depicted in the curve underneath the diagram. The superimposition of the two tension curves in one diagram emphasizes the nifedipine effects. Initial tension (IT) increased by about 0.4 mg and final static tension (ST) at the end of the plateau phase of the stretch by about 0.6 mg, whereas peak tension (PT) increased by as much as 1.6 mg. The small increase in initial tension was not associated with any visible contraction of the intrafusal muscle fibers. By contrast, a single application of 5 ml of a solution containing 20 mg / ml succinylcholine induced (1) a clearly visible contraction and (2) a transient increase in the tension of the same muscle spindle. The latter is shown in Fig. 10a. Succinylcholine was added to the bathing solution close to the spindle after a 20-min period of washing in a normal Ringer’s solution. In Fig. 10c the averaged tension curves of the sections marked by c and c9 in Fig. 10a are superimposed. Again, the lower curve represents the tension developed in a normal Ringer’s solution. The upper curve represents the tension developed shortly after the single application of succinylcholine. The length change of the spindle during the ramp-and-hold stretches is depicted in the curve underneath the diagram. In a very similar way to what was observed with nifedipine, succinylcholine induced an increase in peak tension (2.3 mg) that was larger than the increases in initial tension (0.5 mg) or in final static tension (0.7 mg). With 10 out of 13 spindles the effect on tension of both nifedipine and succinylcholine could be compared. With the remaining three spindles no stable registration of tension was achieved, probably due to temperature effects influencing the measurements of the very sensitive mechanoelectrical transducer. Four out of ten spindles showed an increase in PT by more than 10% when 25 mM nifedipine was added to the bathing solution. The remaining six endings were hardly affected. The mean increases in the three tension indices were significant using the paired Student’s t-test. On average IT increased from 3.3561.16 to 3.5961.19 mg (P,0.01), PT increased from 7.4262.34 to 7.9162.64 mg (P,0.05), and ST increased from 6.9662.03 to 7.2162.13 mg (P,0.05). The effects

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

271

Fig. 10. Effects of nifedipine (25 mM) and succinylcholine (5 ml; 20 mg / ml) on the tension obtained from an isolated muscle spindle repeatedly stimulated by ramp-and-hold stretches (stretch amplitude: 5% of L 0 ; stretch velocity: 40% of L 0 / s). (a) Changes in the tension curve plotted against time. The time scale is interrupted where long periods of washing were included. (b,c) Mean tension curves, each obtained from five successive responses to stretch (see brackets b, b9, c and c9 in panel (a)), plotted against an enlarged time scale. IT5initial tension; PT5peak tension; ST5static tension. The changes in spindle length are depicted in the graphs beneath the tension curves in (b) and (c). Both nifedipine and succinylcholine increase the tension of the isolated muscle spindle. This causes a rise in IT, PT and ST, but it is PT that is most strongly augmented. When the spindle capsule is removed the tension of the spindle is markedly reduced (see panel (a)).

of succinylcholine (0.2 mg / ml) were generally stronger than those of nifedipine. Thus PT increased by more than 10% in seven out of ten muscle spindles. Mean IT increased from 3.4061.39 to 3.8161.38 mg (P,0.01), mean PT increased from 7.2762.29 to 8.1662.36 mg (P,0.001), and mean ST increased from 6.8362.05 to 7.3862.00 mg (P,0.01). At this point, we would like to point out that most of the tension measured during a ramp-and-hold stretch is developed by the outer capsule that surrounds the intrafusal muscle fibers in the equatorial region of the receptor. When the capsule was removed by microdissection, the initial tension and the changes in tension during a stretch (PT-IT) dropped to about one fifth of the values measured from the intact spindle (see right-hand section of Fig. 10a), since only the tension developed by the intrafusal fibers then remained. This observation will help to interpret the lack of strong effects of nifedipine on the mechanical properties of the majority of isolated muscle spindles (see Discussion).

4. Discussion The main effect of nifedipine on isolated cat muscle spindles was a decrease in the afferent discharge frequency (Fig. 2). The activity of both primary and secondary endings was markedly reduced, or even ceased, when the Ca 21 channel blocker was applied to the bathing solution at a concentration of 25 mM. However, with low concentrations of nifedipine (10 mM) some unexpected excitatory effects of the Ca 21 antagonist were observed, mainly in connection with the discharge patterns of primary endings. Despite a small increase in the level of activity that occurred in some endings of both types, the dynamic properties of primary endings in particular were potentiated. (1) The IP and PD of most primary endings increased (Figs. 7 and 4); (2) the sensitivity of PD to varying amplitudes and velocities of ramp-and-hold stretch rose (Fig. 5); (3) the increase in the discharge frequency during the ramp phase of a stretch was augmented (Fig. 8); (4) fast adaptive decay increased (Fig. 6); (5) the phase

272

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

position of the primary ending discharge shifted toward smaller values during sinusoidal stretches (Fig. 9). By contrast to these findings with primary endings, a low concentration of nifedipine hardly affected the dynamic properties of secondary endings (Figs. 4–6 and 8). The length sensitivity of PD and the rise in the discharge rate during the ramp phase of a stretch were only slightly increased and the fast adaptive decay remained strictly unchanged under 10 mM nifedipine. The phase shift was not tested, as secondary endings do not usually respond to small sinusoidal stretches in a phase locked manner with one action potential per stimulus.

4.1. The site of action of nifedipine The dual effect of nifedipine on the afferent discharge pattern with its inhibitory and excitatory components may be an indication of the existence of more than one site of action of the drug. In isolated muscle spindles, the intrafusal muscle fibers on the one hand and the afferent nerve fibers with their sensory terminals on the other might be targets for nifedipine. The outer spindle capsule, however, will be excluded from further consideration, since we found the same effects of nifedipine regardless of whether the spindle capsule was intact or had been removed. Effects on the cut ends of g -motoraxons appear to be unlikely as well. Two of our findings indicate that the intrafusal muscle fibers are involved in the excitatory effects of nifedipine. Firstly, the IP of the primary endings was potentiated when nifedipine was added to the bath (Fig. 7). Secondly, the tension of some isolated muscle spindles as recorded under ramp-and-hold stretches increased in the presence of nifedipine (Fig. 10). The IP of primary endings is formed by the brief initial burst of discharges at the start of a stretch. It is generally agreed that this initial burst is due to the short-range elasticity of the intrafusal muscle fibers resulting from cross-bridges that are formed between the actin and myosin filaments in the poles of the resting spindle prior to the stretch [5,10,23,34–36,41]. Consequently, stretching the spindle leads to a pronounced lengthening of its equatorial sensory region, which lacks contractile filaments, until at least some of the cross-bridges detach and the discharge frequency declines from the summit of the IP. Even though IP was not well developed in a number of primary endings, there was a clear tendency for nifedipine either to potentiate the IP in those endings that generally did produce an IP in a normal Ringer’s solution, or to evoke an IP in some primary endings where the initial burst was reduced to an initial increase in the normal Ringer’s solution. A reasonable interpretation of this result is that nifedipine facilitates the attachment of cross-bridges in the intrafusal muscle fibers, thus increasing the stiffness of the polar sections of the receptor and enhancing the IP.

This might also be the reason for the increase in tension that is observed in some muscle spindles under nifedipine when the receptor is passively stretched (Fig. 10). Again, the results can easily be explained in terms of increased stiffness of the spindle poles. The small rise in initial tension observed under nifedipine still points to there being a very small active contraction, possibly in only one type of intrafusal muscle fiber, even though we were not able to recognize any such contraction under the microscope. Further investigations may help to confirm our observations if it is possible to measure the effects of nifedipine on tension and on the sarcomere length of individual intrafusal muscle fibers under varying degrees of pre-stretch. It may be speculated that nifedipine preferentially affects the nuclear bag 1 fiber of the receptor, since stiffening the poles of this type of fiber will enhance the dynamic properties of primary endings, but not those of secondary endings, which derive their input mainly from the bag 2 and chain fibers. The suggestion that nifedipine acts on the nuclear bag 1 fiber is additionally supported by the fact that both nifedipine and succinylcholine were found to modify the averaged tension curves in a very similar way, when each curve was compared with the control in Ringer’s solution (Fig. 10b and c). Both nifedipine and succinylcholine enhanced the peak tension (PT) more strongly than the final static tension (ST) or the initial tension (IT). Gladden [15] demonstrates that acetylcholine, acting on the intrafusal muscle fibers in the same way that succinylcholine does [4], causes a contraction of bag 1 and bag 2 intrafusal muscle fibers of isolated muscle spindles, while nuclear chain fibers do not respond to the drug. Moreover, the bag 1 fiber is more sensitive to acetylcholine and succinylcholine than the bag 2 fiber. Diverse investigations on non-isolated muscle spindles confirm the succinylcholine effect on nuclear bag fibers [9,14,38,40]. We therefore suggest that nifedipine increases the stiffness of the nuclear bag fibers in the isolated muscle spindle in a manner resembling a weak stimulation by succinylcholine. This interpretation can in particular explain the effects of nifedipine on the dynamic properties of primary endings as long as the dynamic bag 1 fiber is involved. Before going into detail it is necessary to clarify why only a few muscle spindles showed a modified tension curve under nifedipine, while most spindles displayed minor changes in the tension curve but an effect on the afferent discharge rate. We believe that the weak effect of nifedipine on the intrafusal muscle fibers is just sufficient to induce changes in the very sensitive sensory endings, whereas the slight increase in stiffness is not very well reflected in the tension of the whole muscle spindle. As is demonstrated in Fig. 10a, the major part of the tension measured in our experiments is carried not by the intrafusal muscle fibers but by the spindle capsule. Thus the large changes in tension that occur in the capsule during a passive stretch may easily occlude the very minor changes

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

induced by an increase in the stiffness of intrafusal muscle fibers. So far, only the excitatory components of the nifedipine effect have been considered. However, at concentrations $25 mM it is usually the inhibitory action of nifedipine that is most prominent. Fig. 11 summarizes the effects of 25 mM nifedipine, and illustrates that both the excitatory and the inhibitory component are present simultaneously. Fig. 11a shows the effect of nifedipine over time on the basic discharge frequencies IA, PD, MST and FST, while Fig. 11b shows the simultaneously recorded changes in the tension of the whole receptor. A 25-mM sample of nifedipine was added to the bath for a short period of 3 min (see horizontal bar at the top of each diagram). The fine horizontal lines in (a) represent the mean values of IA and PD in a normal Ringer’s solution; those in (b) represent mean IT and mean PT in Ringer’s solution. These auxiliary lines make it easier to recognize small changes induced by nifedipine. The experiment illustrates that the inhibitory effect of nifedipine on the discharge frequency (IA and FST are reduced) goes in parallel with the excitatory effect on the spindle tension. Thus the

273

inhibitory effect of nifedipine cannot be explained in mechanical terms. Rather, a hyperpolarizing effect on the sensory nerve terminals is the most likely reason for the decrease in the discharge activity that occurs in all the basic discharge frequencies of both types of ending when the drug is applied for a longer time.

4.2. Effects of nifedipine on the level of activity of primary and secondary endings Since the muscular and neuronal effects of nifedipine counteract each other, it is not surprising that nifedipine causes heterogeneous effects on the level of activity in the muscle spindle afferents. The level of activity will rise if the muscular component is stronger than the neuronal component; it will fall if the neuronal component predominates. Our findings provide evidence that the inhibitory neuronal component of the effect on the discharge frequency is dominant at high concentrations of the drug (Figs. 2 and 3). However, Figs. 2 and 11 also show that the excitatory effect on the intrafusal muscle fibers develops faster than the inhibition, which overcomes that excitation later on. On average, the strengths of the two effects on the level of afferent activity seem to be almost equal at a concentration of 10 mM nifedipine. This circumstance made it possible to observe the effects of nifedipine on certain features of the afferent discharge pattern under various conditions of stretch.

4.3. Effects of nifedipine on the dynamic properties of muscle spindle endings

Fig. 11. Nifedipine effects over time on (a) the four basic discharge frequencies IA, PD, MST and FST of a primary ending and (b) the simultaneously recorded tension of its host spindle obtained under repetitive stimulation with ramp-and-hold stretches. Horizontal bars at the top of the diagrams represent the period when 25 mM nifedipine was present in the bathing solution. Fine horizontal lines make it easier to recognize small changes in the basic discharge values and tension. A decrease in IA and FST occurs simultaneously with a sustained increase in the spindle tension.

Our results demonstrate that nifedipine exerts a strong influence on the dynamic properties of primary endings. The dynamic properties of primary afferents are similarly increased during investigations with fusimotor stimulation of dynamic g -motorneurons that innervate the nuclear bag 1 fiber [7,11,21], and with an application of low concentrations of succinylcholine, which also preferentially stimulates this type of fiber [9,38]. However, an electrically or pharmacologically induced contraction of the nuclear bag 1 fiber usually also brings about an increase in the slow adaptive decay from MST to FST, which we did not find when the dynamic properties were augmented by nifedipine. Thus on the one hand our findings are consistent with the suggestion that nifedipine induces increased stiffness in the dynamic nuclear bag 1 fiber, but on the other hand the underlying mechanism might be slightly different. The effect of nifedipine on fast adaptive decay (Fig. 6) is worth discussing in more detail. This decay has been explained in terms of increased potassium conductance by the ending’s sensory membrane [25,37]. KCa channels may open due to an increased calcium influx during the ramp phase of a stretch. However, we cannot explain how nifedipine might influence this ional process, because a

274

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

blocking of the Ca 21 conductance during the ramp phase should additionally diminish the subsequent Ca 21 dependent potassium efflux, so that fast adaptive decay ought to be decreased rather than increased by nifedipine. We suggest that this ional mechanism is not influenced by nifedipine. Therefore, the Ca 21 influx during the ramp phase of a stretch appears to pass through channels that are insensitive to nifedipine. Nonetheless, it is known that fast adaptive decay is highly sensitive to changes in the velocity of the ramp stimulus [7,37]. With nifedipine, however, the lengthening of the sensory region is accelerated by mechanical means, since the postulated increase in stiffness of the intrafusal fibers results in a diminished degree of strain (i.e. length change of a unit length) in the polar section of the intrafusal fiber, whereas the strain in the equatorial sensory section is increased. Thus, even though the conditions of the stretch are not changed, nifedipine induces a redistribution of the lengthening. The lengthening is slowed down at the spindle poles, but accelerated in the equatorial part of the spindle. The increased velocity of the stretching of the sensory region will therefore explain the enhanced fast adaptive decay under nifedipine. However, it remains questionable why the fast adaptive decay becomes sensitive to varying amplitudes of stretch under nifedipine (Fig. 6a). This is not usually observed in a normal Ringer’s solution. We can only speculate that under ramp-and-hold stretches of small amplitude the stiffening of the spindle poles does not have the effect of speeding up the lengthening of the sensory region. This might depend on elastic components in series with the sensory region that have to be stretched first. On the other hand, this explanation does not seem to fit in with the observation that nifedipine was clearly effective in shifting the phase position of the action potential toward smaller phase values relative to the cycle of very small sinusoidal stretches, which means that the dynamic properties of primary endings are facilitated even under small elongations of the receptor. This might again reflect an increased stiffness in the spindle poles. About half of the spindles showed a late phase shift toward increasing phase values, subsequent to the early phase shift toward lower values. This late effect may reflect the more slowly developing inhibitory effect of nifedipine, hyperpolarizing the membrane of the sensory ending directly. To summarize, our findings support the notion that the dynamic properties of primary endings are largely dependent on their mechanical input from the intrafusal muscle fibers. The observation that nifedipine augments the dynamic properties of primary endings more strongly than those of secondary endings indicates that it might be the nuclear bag 1 fibers that are most sensitive to nifedipine. However, an increase in the stiffness of the other intrafusal muscle fibers cannot be excluded. In this case, the locations of the two types of sensory ending on the intrafusal fibers have to be considered to explain the differing

efficacies of nifedipine. The primary endings terminate in the equatorial region of the receptor, where intrafusal muscle fibers usually lack contractile filaments. Thus, the greatest effect of the stiffening of the poles by nifedipine is to enhance the elongation of that region. By contrast, secondary endings terminate juxtaequatorially in the positions S1–S5, where contractile filaments are common [1]. Thus, nifedipine will increase the number of attached cross-bridges not only in the spindle poles but also in the region where secondary endings contact the intrafusal fibers. However, an increased number of attached crossbridges in that sensory region will weaken the sensory strain, and therefore diminish the effects of the stiffening of the spindle poles. This may help to explain the limited excitatory effect of nifedipine on secondary endings as compared with the great effects on primary endings. The finding that nifedipine increases the initial burst in primary endings but reduces the initial increase in secondary endings supports this interpretation. Nevertheless, it should be borne in mind that different ionic mechanisms in primary and secondary endings might additionally contribute to their different dynamic properties.

4.4. Speculations on the mechanism of the nifedipine effects Nifedipine and other dihydropyridines are highly sensitive blockers of L-type Ca 21 channels [19,39,43]. Fleckenstein [13] was the first to describe the antagonist-like effects of dihydropyridines on heart muscle cells and smooth muscle cells. Numerous investigations have confirmed the antagonist-like effect of nifedipine, see review [32]. Thus, an inhibitory effect of the drug on the discharge activity of muscle spindle afferents was to be expected in our experiments. A reduced Ca 21 current through the afferents’ sensory membranes could easily explain the decrease in the activity level. Alternatively, the Ca 21 conductance of the encoder sites might have been reduced. These two sites of action cannot be distinguished within our investigation. Interestingly, the strong inhibitory effect under 25 mM nifedipine suggests that numerous Ca 21 channels are in an open state in the resting muscle spindle. Therefore, the membrane potential of the sensory endings appears to be highly dependent on a sustained Ca 21 conductance. By contrast, the mechanism of the excitatory effect of nifedipine is more difficult to understand, since a reduction in the Ca 21 influx would be expected to decrease rather than increase the contractility of intrafusal muscle fibers. However, there are some hints from other investigations that point to an agonist-like effect of nifedipine in different types of muscle fiber. Himori, Ono and Taira [20] described how low concentrations of nifedipine increased the basic tension of a papillary muscle; higher concentrations of nifedipine, however, had a depressing effect. Neuhaus and collaborators [29,30] measured the isometric force of

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

short toe muscle fibers of the frog and observed a shift in the threshold of force activation from an intracellular membrane potential of 250 mV under control conditions to about 275 mV under 1–3 mM nifedipine. Very similar effects were found in two further investigations on frog skeletal muscles testing the Ca 21 antagonist D600 [2,8]. The Ca 21 channel blocker diltiazem has been shown to decrease the Ca 21 influx and simultaneously to potentiate the twitch, tetanic and potassium contracture tensions in isolated muscle fibers [16]. These findings indicate that Ca 21 channel blockers may facilitate the contractility of skeletal muscle fibers. Even though the mechanisms of these effects are unknown, the findings favor the possibility that nifedipine also acts on intrafusal muscle fibers in an agonist-like manner, increasing the number of crossbridge attachments between actin and myosin. Furthermore, Hess, Lansman and Tsien [18] reported a mixed agonist-like and antagonist-like effect of dihydropyridines on the Ca 21 channel itself. The reasoning becomes even more complicated when the fact is taken into account that dihydropyridines are not perfectly selective Ca 21 channels blockers. Rather, they may affect various other types of voltage gated ion channel, as well as some ligand binding ion channels [6,28]. Further experiments are needed to explain the effects of nifedipine shown at the cellular level.

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

Acknowledgements [19]

We wish to acknowledge gratefully the technical assistance of Birgit Begemann during the experiments and the evaluation of data.

[20]

[21]

References [1] R.W. Banks, D. Barker, M.J. Stacey, Form and distribution of sensory terminals in cat hindlimb muscle spindles, Philos. Trans. R. Soc. Lond. B. Biol. Sci. 299 (1982) 329–364. ¨ [2] D. Berwe, G. Gottschalk, H.C. Luttgau, Effects of the calcium antagonist gallopamil (D600) upon excitation–contraction coupling in toe muscle fibres of the frog, J. Physiol. 385 (1987) 693–707. [3] I.A. Boyd, The isolated mammalian muscle spindle, Trends Neurosci. 3 (1980) 258–265. [4] I.A. Boyd, M.H. Gladden, J. Ward, The mechanical behaviour of the dynamic bag 1 fiber in isolated cat muscle spindles during rapid stretch, J. Physiol. 390 (1987) 138. [5] M.C. Brown, G.M. Goodwin, P.B. Matthews, After-effects of fusimotor stimulation on the response of muscle spindle primary afferent endings, J. Physiol. (Lond.) 205 (1969) 677–694. [6] D. Chesnoy-Marchais, L. Cathala, Modulation of glycine responses by dihydropyridines and verapamil in rat spinal neurons, Eur. J. Neurosci. 13 (2001) 2195–2204. [7] A. Crowe, P.B.C. Matthews, The effects of stimulation of static and dynamic fusimotor fibres on the response to stretching of the primary endings of muscle spindles, J. Physiol. 174 (1964) 109– 131. ¨ ¨ [8] M. Dorrscheidt-Kafer, The action of D600 on frog skeletal muscle:

[22] [23] [24]

[25]

[26]

[27]

[28]

[29]

275

¨ facilitation of excitation–contraction coupling, Pflugers Arch. 369 (1977) 259–267. M.B. Dutia, Activation of cat muscle spindle primary, secondary and intermediate sensory endings by suxamethonium, J. Physiol. (Lond.) 304 (1980) 315–330. B.B. Edin, The ‘initial burst’ of human primary muscle spindle afferents has at least two components, Acta Physiol. Scand. 143 (1991) 169–175. F. Emonet-Denand, Y. Laporte, P.B. Matthews, J. Petit, On the subdivision of static and dynamic fusimotor actions on the primary ending of the cat muscle spindle, J. Physiol. (Lond.) 268 (1977) 827–861. ¨ M. Fischer, S.S. Schafer, Temperature effects on the discharge frequency of primary and secondary endings of isolated cat muscle spindles recorded under a ramp-and-hold stretch, Brain Res. 840 (1999) 1–15. A. Fleckenstein, Specific pharmacology of calcium in myocardium, cardiac pacemakers, and vascular smooth muscle, Annu. Rev. Pharmacol. Toxicol. 17 (1977) 149–166. M. Gioux, J. Petit, U. Proske, Responses of cat muscle spindles which lack a dynamic fusimotor supply, J. Physiol. (Lond.) 432 (1991) 557–571. M.H. Gladden, Structural features relative to the function of intrafusal muscle fibres in the cat, Prog. Brain Res. 44 (1976) 51–59. H. Gonzalez-Serratos, R. Valle-Aguilera, D.A. Lathrop, M.C. Garcia, Slow inward calcium currents have no obvious role in muscle excitation–contraction coupling, Nature 298 (1982) 292–294. ¨ S. Just, C. Leipold-Buttner, B. Heppelmann, Histological demonstration of voltage dependent calcium channels on calcitonin generelated peptide-immunoreactive nerve fibers in the mouse knee joint, Neurosci. Lett. 312 (2001) 133–136. P. Hess, J.B. Lansman, R.W. Tsien, Different modes of Ca channel gating behaviour favoured by dihydropyridine Ca agonists and antagonists, Nature 311 (1984) 538–544. B. Hille, Ionic Channels of Excitable Membranes, 2nd Edition, Sinauer Associates, Sunderland, Massachusetts, 1992. N. Himori, H. Ono, N. Taira, Simultaneous assessment of effects of coronary vasodilators on the coronary blood flow and the myocardial contractility by using the blood-perfused canine papillary muscle, Jpn. J. Pharmacol. 26 (1976) 427–435. M. Hulliger, The mammalian muscle spindle and its central control, Rev. Physiol. Biochem. Pharmacol. 101 (1984) 1–110. C.C. Hunt, Mammalian muscle spindle: peripheral mechanisms, Physiol. Rev. 70 (1990) 643–663. C.C. Hunt, D. Ottoson, Initial burst of primary endings of isolated mammalian muscle spindles, J. Neurophysiol. 39 (1976) 324–330. C.C. Hunt, D. Ottoson, Responses of primary and secondary endings of isolated mammalian muscle spindles to sinusoidal length changes, J. Neurophysiol. 40 (1977) 1113–1120. C.C. Hunt, R.S. Wilkinson, Y. Fukami, Ionic basis of the receptor potential in primary endings of mammalian muscle spindles, J. Gen. Physiol. 71 (1978) 683–698. F. Ito, M. Sokabe, K. Nomura, K. Naruse, N. Fujitsuka, A. Yoshimura, Effects of ions and drugs on the responses of sensory axon terminals of decapsulated frog muscle spindles, Neurosci. Res. Suppl. 12 (1990) S15–S26. M.N. Kruse, R.E. Poppele, Components of the dynamic response of mammalian muscle spindles that originate in the sensory terminals, Exp. Brain Res. 86 (1991) 359–366. M.G. Lopez, R.I. Fonteriz, L. Gandia, M. de-la-Fuente, M. Villarroya, J. Garcia-Sancho, A.G. Garcia, The nicotinic acetylcholine receptor of the bovine chromaffin cell, a new target for dihydropyridines, Eur. J. Pharmacol. 247 (1993) 199–207. R. Neuhaus, The effect of nifedipine on slow Ca 21 inward current and force development in isolated skeletal muscle fibres of the frog, J. Physiol. 382 (1987) 122P.

276

¨ / Brain Research 954 (2002) 256–276 M. Fischer, S.S. Schafer

¨ [30] R. Neuhaus, R. Rosenthal, H.C. Luttgau, The effects of dihydropyridine derivatives on force and Ca 21 current in frog skeletal muscle fibres, J. Physiol. 427 (1990) 187–209. [31] M.C. Nowycky, A.P. Fox, R.W. Tsien, Three types of neuronal calcium channel with different calcium agonist sensitivity, Nature 316 (1985) 440–443. [32] L.H. Opie, Calcium antagonists. Mechanisms, therapeutic indications and reservations: a review, Q. J. Med. 53 (1984) 1–16. [33] R.E. Poppele, W.R. Kennedy, D.C. Quick, A determination of static mechanical properties of intrafusal muscle in isolated cat muscle spindles, Neuroscience 4 (1979) 401–411. [34] R.E. Poppele, D.C. Quick, Effect of intrafusal muscle mechanics on mammalian muscle spindle sensitivity, J. Neurosci. 5 (1985) 1881– 1885. [35] U. Proske, J.E. Gregory, The time-course of recovery of the initial burst of primary endings of muscle spindles, Brain Res. 121 (1977) 358–361. [36] U. Proske, D.L. Morgan, J.E. Gregory, Muscle history dependence of responses to stretch of primary and secondary endings of cat soleus muscle spindles, J. Physiol. (Lond.) 445 (1992) 81–95. ¨ [37] S.S. Schafer, The fast and slow components of receptor adaptation

[38]

[39] [40]

[41]

[42]

[43]

in the discharge frequency of the primary muscle spindles in the cat, EEG. EMG. Z. Elektroenzephalogr. Elektromyogr. Verwandte. Geb. 23 (1992) 12–19. ¨ S.S. Schafer, M.A. Schoppmeyer, Time-dependent fusimotor effects on the discharge of cat primary muscle spindle afferents induced by a long-lasting succinylcholine infusion, Brain Res. 709 (1996) 111–121. A. Schwartz, D.J. Triggle, Cellular action of calcium channel blocking drugs, Annu. Rev. Med. 35 (1984) 325–339. J.J. Scott, Responses of Ia afferent axons from muscle spindles lacking a bag 1 intrafusal muscle fibre, Brain Res. 543 (1991) 97–101. J.J. Scott, The ‘initial burst’ of muscle spindle afferents with or without terminals on the bag 1 intrafusal muscle fibre, Brain Res. 585 (1992) 327–329. R.S. Scroggs, A.P. Fox, Calcium current variation between acutely isolated adult rat dorsal root ganglion neurons of different size, J. Physiol. (Lond.) 445 (1992) 639–658. R.W. Tsien, P.T. Ellinor, W.A. Horne, Molecular diversity of voltagedependent Ca 21 channels, Trends. Pharmacol. Sci. 12 (1991) 349– 354.