The influence of basic fibroblast growth factor on acetylcholine receptors in cultured muscle cells

14

Neuroscience Letters, 144 (1992) 14 18 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00

NSL 08907

The influence of basic fibroblast growth factor on acetylcholine receptors in cultured muscle cells Z h e n g s h a n D a i a n d H. B e n j a m i n P e n g Department of ('ell Biology and Anatomy University of North Carolina, Chapel Hill, NC 27599 (USA) (Received 14 April 1992; Revised version received 8 June 1992; Accepted 9 June 1992)

Key words." Basic fibroblast growth factor; Acetycholine receptor; Neuromuscular junction; Tissue culture; Xenopus Acetylcholine receptors (AChRs) in Xenopus muscle cells undergo changes in channel kinetics during development in culture and these changes are

somehow related to innervation. Recently we showed that basic fibroblast growth factor (bFGF), when locally applied, can mimic the effect of nerve in inducing a postsynaptic-type development. In this study, we examined whether bFGF can influence the developmental changes of AChRs. Patch clamp method was employed to record single AChR channel currents from cultured Xenopus myotomal muscle cells and the kinetics of lowconductance AChR channels were analyzed. In cultures treated with 1/lg/ml bFGF at an early stage (stage 23), the burst duration of low-conductance AChR channels showed a 1.5-fold decrease between the first and second day in culture, while it underwent a remarkable 3-fold decrease during the same period in the control. Histogram analyses showed that the low-conductance channels were composed of a fast and a slow component and that the decrease in burst duration was due to a shift in the population from the slow to the fast. bFGF treatment appeared to slow down this shift by retaining the slow channels for a longer period of time. This effect is probably not due to channel modulation as the burst duration of short channel in older cells (stage 40) was not affected by bFGF. These data suggest that bFGF may enhance the metabolic stability of intrinsically short-lived AChRs. This effect seems to parallel the stabilization of junctional AChRs at the innervated endplate. Thus, bFGF, or a related polypeptide growth factor, may mediate this and other innervation-induced changes in the postsynaptic membrane.

M o t o r innervation exerts a p r o f o u n d influence on the skeletal muscle. A m o n g effects b r o u g h t about by innervation is the change in the physiological properties o f the acetylcholine receptors (AChRs). One type o f change is a transition from the low-conductance type to the highconductance type after innervation [6, 14]. In m a m m a l ian muscle, this is well correlated with the nerve-induced expression o f the adult type e-subunit and the suppression o f the fetal type y-subunit [16]. A n o t h e r developmental change is the decrease in the burst duration o f A C h R channels, which is best documented in the Xenopus m y o t o m a l muscle cells [1, 4, 5, 12]. A dramatic 3fold decrease in burst duration between the first and second day in culture is observed in the case o f low-conductance A C h R s . The mechanism underlying is not understood. A l t h o u g h this shortening in burst duration is independent o f innervation, Xenopus myocytes cocultured with neurons exhibited an increase in the channel open time in both subsynaptic and extrasynaptic areas [2]. Thus, this p h e n o m e n o n o f changes in A C h R channel kiCorrespondence." H.B. Peng, Department of Cell Biology and Anatomy, University of North Carolina, CB#7090, 108 Taylor Hall, Chapel Hill, NC 27599, USA. Fax: (1) (919)966 1856.

netics, similar to the y- to e-subunit transition, also appears to be under neural control. Recently we have shown that local application o f basic fibroblast growth factor ( b F G F ) mimics the effect o f nerve in inducing a postsynaptic type development in cultured Xenopus m y o t o m a l muscle cells [8]. In addition, a recent study has shown that this growth factor can modulate the behavior o f certain ion channels [10, l 1]. This p r o m p t e d us to examine its effect on developmental changes in the kinetics of A C h R channels. We concentrated on the low-conductance channel type, since its developmental change in burst duration is m u c h more pronounced. Here we report that b F G F causes a significant retardation in the decrease o f A C h R burst duration during development. This effect appears to parallel the previously reported influence o f nerve on the lengthening o f burst duration o f A C h R channels [2]. Muscle cells were isolated f r o m the m y o t o m e s of stage 17 t 8 Xenopus embryos [7] and cultured in a medium containing Steinberg solution (NaC1, 60 m M ; KC1, 0.67 m M ; Ca(NO3)2, 0.34 m M ; MgSO4, 0.83 m M ; H E P E S , 10 m M ; p H 7.4) supplemented with 10% L-15, 1% fetal bovine serum, and 0.1 mg/ml gentamycin at 22°C, according to a previously published m e t h o d [9]. The cell-at-

15

recording was done at 24°C. The single-channel current records were played back from the VCR through the PCM-1 adaptor and were digitized at 10 kHz with an IBM PC-AT computer. Before digitizing, the signal was filtered by an 8-pole Bessel filter (-3 dB at 2 kHz, Frequency Devices Model 902). Digital data were acquired by an event-triggered method and analyzed with Pclamp software (Axon Instruments Inc., Burlingame, CA). The burst duration of unitary channel current events longer than 200 ms were measured at their half-heights. Fig. 1A shows sample recordings and the burst duration histogram of low-conductance AChR channels from a 6-h culture. At this early stage, these low-conductance channels had long mean burst duration (11.0+0.5 ms, mean+S.E.M., n--2). However, after 1 day in culture, the burst duration of low-conductance AChR channels

tached patch clamp technique was employed to record single-channel currents from AChRs [3]. A patch clamp amplifier (List EPC-7) with a 50-GO feedback resistor was used. The current signal was recorded with a video cassette recorder (VCR) after digitizing through a PCM1 digital VCR-instrumentation recorder adaptor (44.6 kHz, 16 Bit, Medical Systems Corp., Greenvale, NY). Before electrophysiological recording, the culture medium was replaced with Ringer solution (NaC1, 120 mM; KC1, 2.5 mM; CaC12, 2 mM; HEPES, 10 mM; pH 7.4). The recording pipette was filled with Ringer solution containing 200 nM ACh with resistance between 5 to 10 MO. Holding potential was 80 mV hyperpolarized from resting membrane potential. At this holding potential, the single-channel current events of high- and low-conductance AChR channel could be well separated. The

N=(wl/tt)exp(-t/tt)+ (wz/tz)exp(-t/t z)

A

6 4

tt=2rns,

2 0

!,11

4o8°B6°_T I

I

1

I

re= 15ms

I

4o

t-'--

2O ~0

0

I

I

I

I

I

0

I

I

I

I

0

6O

4o 2O 0

I

Ih

,

I

.30-

I. . . .

I

I

I

I

I

40

60

80

G

20-

20 0

10I 0

t

'''i 20

j

i

I

40

60

80 ms

0

~ " .... 0

"1 ~

20

~'

m~ Fig 1. Single channel records and histograms o f burst duration of low-conductance AChR channels. Channel openings are shown as downward deflections. In both control (left column) and bFGF-treated (right column) muscle cells at different developmental stages, the histogram can be well fitted by a double-exponential equation as described here with indicated time constants. The low-conductance channels appear to be composed of a fast and a slow component. The decrease in burst duration is due to a shift in the population from the slow to the fast. b F G F slows down this shift. A was recorded from a 1-day culture; B and E from 2-day cultures; C and F from 4-day cultures; D and G from 6-day cultures.

16 showed a large decrease to a mean of 3.7+0.7 ms (n=8). Examples of these shortened channels and the burst duration histogram are shown in Fig. lB. Thus, consistent with previous results [4, 5, 12], there was a 3-fold decrease in burst duration during the first 2 days in culture. To test the effect of bFGF, recombinant human b F G F (a kind gift of Synergen, Boulder, CO) was applied to muscle cultures at stage 23 (about 6 h after plating) for 1 h at a concentration of 0.5-1 ¢tg/ml. Then the cultures were washed with bFGF-free medium and incubated for 1-5 days before recording. Alternatively, the cultures were continuously treated with b F G F at 0.5 ¢tg/ml after stage 23 until the time of recording. No significant difference was observed between these 2 experimental procedures. One day after b F G F treatment, the burst duration of low-conductance AChR channels still showed a decrease. But the rate of decrease was much reduced with respect to the control. Sample recordings and the burst duration histogram are shown in Fig. 1E. While cells in the control showed a 3-fold decrease in mean burst duration, only a 1.5-fold decrease was observed in b F G F treated cells between the first and second day in culture (Fig. 2). The mean burst duration in a 2-day-old, b F G F treated culture was 6.8+1.8 ms (n=13). This is significantly different from the control burst duration of 3.7 ms (P<0.001). In several subsequent days, the mean burst durations of low- conductance AChR channels in b F G F treated cultures were consistently longer than that in the control, although the difference gradually diminished. Sample recordings and burst duration histograms from control and bFGF-treated cells at 4 and 6 days in culture are shown in Fig. 1C,D,F,G. The mean burst duration as a function of development was plotted in Fig. 2. These data show that b F G F retarded the developmentally regulated shortening of low-conductance ACh channels. To study the mechanism of the prolongation of mean burst duration after b F G F treatment, we analyzed the burst duration histograms of low-conductance AChR channels in the control and bFGF-treated cultures at different developmental stages (Fig. 1). These histograms obtained from patchs that had the largest number of events at each developmental stage can be well fitted by the double exponential function:

N=(w,/t,) .exp(-t/t~)+ (w2/t2)exp(-t It2)

(1)

is the time constant of the fast decay phase and t2 is the time constant of the slow decay phase, w~ and w2 are experimentally determined parameters that indicate the relative contribution of the fast and the slow components (see below). For most of the histograms shown in Fig. 1B,C,E-G, a value of 2 ms for tl and 15 ms for t2 were given by the where

tI

•

~"Io o

Control bFGF-treated

8 6

"0

~

9

3 8

4

2

2 ©

~

o o

I

I

I

I

I

I

1

2

3

4

5

6

Age of culture, days Fig 2. Time course of the decrease in burst duration of low-conductance A C h R channels. In control, a 3-fold decrease in burst duration was observed between the first and the second day in culture. In contrast, cultures treated with b F G F at an early time showed only a 1.5-fold decrease in burst duration during the same period. Each point shows the mean burst duration (_+ S.E.M.). The n u m b e r of samples for each time point is also shown.

curve-fitting program in the Pclamp software for optimal fits, However, the population at day 1 contained few fast events (Fig. 1A) and the software did not give consistent fast decay time constant (h)- Likewise, few slow events were observed at day 6 in control cultures (Fig. 1D) and the estimate of the slow decay time constant (t2) was not consistent. Nevertheless, using the same t, and t2 values of 2 ms and 15 ms, we could obtain highly significant fits to all histograms, even to those at the extremes (Fig. 1A,D). (Thez-square values of these fits were close to the minimum and the goodness-of-fit values were always greater or equal to 2.) Thus, the same values of t, and t2 values were used to fit all histograms. The value (Wl/fi)/ ((w~/tO+(w2/t2)), which indicates the percentage of fast channels in the population, was calculated for each experiment. They were plotted in Fig. 3. These data allow us to draw 2 conclusions. First, the low-conductance AChR channels are composed of a fast and a slow component in all developmental stages examined and the decrease in burst duration with development is due to a shift in the population from the slow component to the fast one. Second, bFGF-treatment slows down this shifting process so that the slow component is more heavily represented in all stages as compared with the control. The effect of b F G F on the burst duration of low- conductance AChRs could conceivably be due to a change in the kinetics of the fast channels such that their opening state is prolonged. To examine this possibility, we studied the effect of b F G F on cells that already expressed predominantly fast channels. Cultures at 4 days old were treated with b F G F at 1 pg/ml for one day and then recorded on day 5. The mean burst duration of AChRs in

17 % A ~V

lOO I

tv-

8o

f f

~

~

~ .~

60

40

o

•

Control

V

bFGF-treated

I

I

I

I

I

I

1

2

3

4

5

6

Age of c u l t u r e , d a y s Fig 3. The time course of the increase in the percentage of fast lowconductance AChR channels. In control cultures, the transition from the slow to the fast channel type was completed within 2 days in culture. This transition was significantly retarded by bFGF treatment.

bFGF-treated cells was 3.7+1.6 ms (n=3). This was nearly identical to the value of 3.9+0.7 ms (n--4) in untreated, 5-day-old cells. Thus, the effect o f b F G F in prolonging the AChR channel burst duration as described above is not due to a change in the kinetics of the fast channels. In addition to bath application, we also studied the effect of local presentation of b F G F on the kinetic change of low-conductance AChR channels, b F G F coated 9 Hm polystyrene latex beads prepared with previously published methods [8] were applied to muscle cultures at stage 23. The bead-bearing cultures were incubated for 1-2 days before recording. To record AChR channels at the site of bead-muscle contact, we removed individual beads from the cell with a suction pipette having a tip diameter of 5 Hm and then applied the recording pipette to the site previously occupied by the bead. AChinduced single-channel currents were then recorded both at bead-muscle contacts and at bead-free sites on the cell. As shown in Table I, the mean burst duration of AChR TABLE I THE EFFECT OF BEAD-MEDIATED bFGF APPLICATION ON THE MEAN BURST DURATION OF LOW-CONDUCTANCE AChR CHANNELS Number of measurements are indicated in parenthesis. P<0.005 between bead-muscle contacts and control, and between bead-free sites and control (Student t-tests). Mean burst duration + S.E.M. (ms) Bead-muscle contacts Day2 8.6_+0.7 (3) Day3 6.1_+1.8 (2)

Bead-free sites

Control (no beads)

9.1_+2.3 (3) 7.4_+1.1 (4)

4.0_+0.7(3) 2.9_+1.7(2)

channels beneath the beads was prolonged relative to the control value at the same developmental stage. Surprisingly, the mean burst duration of channels away from beads in bead-treated cultures was also lengthened (Table I). Thus, local application of b F G F via beads appears to be equally effective as bath application in slowing down the low-conductance AChR channels. In addition, this local b F G F presentation appears to exert a global effect on channels away from sites of stimulation. We demonstrated a clear developmentally regulated shift in AChR channels toward the fast component. This is consistent with the finding of Rohrbough et al. [12] that Xenopus AChRs inserted into the membrane at later stages of development have shorter burst duration than the embryonic ones. Steele and Steinbach [15] also documented a class of low-conductance AChR channels in neonatal mouse skeletal muscle that have a mean burst duration 2 3 times longer than low-conductance channels recorded from denervated adult muscle fibers. Thus, the long burst duration is a characteristic of embryonic type low-conductance AChR channels. This type of channels is replaced by channels with the same conductance but shorter burst duration during development. The novel finding of this study is that the rapid shortening in burst duration of low-conductance AChRs during development is slowed down by bFGF. We showed that b F G F does not alter the mean burst duration of fast channels. Furthermore, our whole-cell recording has shown that b F G F affects neither the size nor the time course of desensitization of ACh currents (Dai and Peng, unpublished results). These data show that this growth factor exerts no deleterious effect on cells, nor does it alter the channel kinetics directly. As mentioned above, AChR channels inserted into the membrane of Xenopus muscle cells at a later stage tend to have short open time than those inserted at an earlier stage. As embryonic channels with long open time are degraded during development, they are replaced by channels with short open time. Thus, a simple mechanism to account for the action of b F G F is that b F G F stabilizes the intrinsically shortlived embryonic AChR channels against developmentally regulated degradation process. In this regard, the effect of b F G F bears similarity to that of innervation. During the formation of the neuromuscular junction, AChRs in the endplate membrane become metabolically stabilized under neural control [13]. The half-life of mature mammalian junctional AChRs is 8-10 days, whereas the half-life of AChRs in cultured myotubes or in denervated muscle is less than 24 h [13]. Our previous work has shown that local presentation of b F G F via beads to Xenopus muscle cells stimulates AChR clustering [8]. This study suggests that this local bead-mediated application of b F G F can also exert a global effect on the

18

kinetics of AChRs. Further studies of the role of this growth factor should promote our understanding of the signal-transduction pathways in synaptogenesis. We wish to thank Synergen for their generous gift of basic fibroblast growth factor. This study was supported by NIH Grant NS23583 and the Muscular Dystrophy Association. 1 Brehm, R, Kidokoro, Y. and Moddy-Corbett, F., Acetylcholine receptor channel properties during development of Xenopus muscle cells in culture, J. Physiol., 357 (1984) 203 217. 2 Brehm, R, Steinbach, J.H. and Kidokoro, Y., Channel open time of acetylcholine receptors on Xenopus muscle cells in dissociated cell culture, Dev. Biol., 91 (1982) 93-102. 3 Hamill, O.R, Marry, A., Neher, E., Sakmann, B. and Sigworth, F.J., Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches, Pflueger's Arch., 391 (1981) 85-100. 4 Leonard, R.J., Nakajima, S., Nakajima, Y. and Takahashi, T., Differential development of two classes of acetylcholine receptors in Xenopus muscle in culture, Science, 226 (1984) 55 58. 5 Leonard, R.J., Nakajima, S. and Carlson, C.G., Early development of two types of nicotinic acetylcholine receptors, J. Neurosci., 8 (1988) 4038~,048. 6 Mishina, M., Toshiyuki, T., Imoto, K., Noda, M., Takahasi, T., Numa, S., Methfessel, C. and Sakmann, B., Molecular distinction between fetal and adult forms of muscle acetylcholine receptor, Nature, 321 (1986) 40~411.

7 Nieuwkoop, RD. and Faber, J., Normal Table of Xenopus laevL~ (Daudin), North-Holland, Amsterdam, 1975. 8 Peng, H.B., Baker, L.P. and Chen, Q., Induction of synaptic development in cultured muscle cells by basic fibroblast growth factor, Neuron, 6 (1991) 237 246. 9 Peng, H.B., Baker, L.E and Chen, Q., Tissue culture of Xenopus neurons and muscle cells as a model for studying synaptic imduction. In B.K. Kay and H.B. Peng (Eds.), Xenopus laevis: Practical Uses in Cell and Molecular Biology, Methods in Cell Biology, Vol. 36, Academic Press, San Diego, CA, 1991, pp. 511 526. 10 Puro, D.G., A calcium-activated, calcium-permeable ion channel in human retinal glial cells: modu[ation by basic fibroblast growth factor, Brain Res., 548 (1991) 329 333. 11 Puro, D.G. and Mano, T., Modulation of calcium channels in human retinal glial cells by basic fibroblast growth factor: a possible role in retinal pathobiology, J. Neurosci., 11 (1991) 1873 1880. 12 Rohrbough, J. and Kidokoro, Y., Changes in kinetics of acetylcholine receptor channels after initial expression in Xenopus myocyte culture, J. Physiol., 425 (1990) 245-269. 13 Salpeter, M.M. and Loring, R.H., Nicotinic acetylcholine receptors in vertebrate muscle: properties, distribution and neural control, Prog. Neurobiol., 25 (1985) 297 325. 14 Schuetze, S.M. and Role, L.W., Developmental regulation of nicotinic acetylcholine receptors, Annu. Rev. Neurosci., 10 (1987) 403 457. 15 Steele, J.A. and Steinbach, J.H., Single channel studies reveal three classes of acetylcholine-activated channels in mouse skeletal muscle, Biophys. J., 49 (1986) 361. 16 Witzemann, V., Brenner, H.-R. and Sakmann, B., Neural factors regulate AChR subunit mRNAs at rat neuromuscular synapses, J. Cell Biol., 114 (1991) 125 141.