Neuron,
Vol. 6, 469-475, March, 1991, Copyright
0 1991 by Cell Press
Cyclic AMP Stabilizes the Degradation of O riginal Junctional Acetylcholine Receptors in Decervated Muscle S.-L. Shyng,* R. Xu, and M. M. Salpeter Section of Neurobiology and Behavior Division of Biological Sciences Cornell University Ithaca. New York 14853-2702
rapidly degrading (R,) AChRs (t1/2of ~1 day) synthesized in denervated muscleor expressed by myotubes in primary cell culture. Possible mechanisms for neural regulation of AChR degradation are discussed. Results
Summary We used mouse diaphragm muscle in organ culture to study the stabilization of acetylcholine receptor (AChR) degradation at denervated neuromuscular junctions. After denervation, the degradation rate of the AChRs present prior to denervation (slowly degrading, or Rs, AChRs) accelerates from the predenervation degradation half-life (t& of 4%10 days to a tljz of m2-3 days. We report that addition to the organ culture medium of pharmacological agents that elevate cytoplasmic CAMP levels (forskolin, dibutyryl CAMP, and 8-bromo-CAMP) reversed the change in t1/2 caused by denervation, whereas addition of l,%dideoxyforskolin, a forskolin analog that does not elevate cytoplasmic CAMP levels, did not reverse the effect of denervation. The degradation rate of AChRs in primary myotube cultures and that of the newly synthesized AChRs in denervated muscle were little affected by forskolin or dibutyryl CAMP. The possibility is raised that the modulation of Rs AChR degradation by innervation may be mediated by CAMP. Introduction Acetylcholine receptors (AChRs) at innervated neuromuscular junctions, variously called original, adult, or slowly degrading (Rs) AChRs, have a slow degradation half-life (t,,? of @-IO days). The Rs AChRs can exist in a stabilized or an accelerated state. A few days after denervation the degradation rate of the Rs AChRs accelerates to a t112of ~2-4 days (Levitt and Salpeter, 1981; Stanley and Drachman, 1981; Brett et al., 1982; Bevan and Steinbach, 1983; Wetzel and Salpeter, 1991); reinnervation causes it to revert to the stabilized, predenervation value (Salpeter et al., 1986). Since these changes in the degradation rate occur after the Rs AChRs have been inserted into the postsynaptic membrane, they do not involve de novo AChR synthesis, but probably arise either from posttranslational modifications of the AChRs or changes in their molecular environment. We show here that CAMP can mimic the effect of reinnervation in stabilizing the Rs AChRs in denervated muscle. Furthermore, while CAMP decreases the degradation rate of Rs AChRs at the denervated neuromuscular junction to predenervation levels, it has little or no effect on the degradation rate of the * Present address: Division of Biology, California Technology, Pasadena, California 91125.
Institute
of
In the present study we use the following definitions for the different AChRs, based on location, time of synthesis, or degradation properties. Original AChRs refer to AChRs synthesized in innervated muscle and present at the neuromuscular junction prior to denervation. New AChRs are those synthesized after denervation and present at both junctional and extrajunctional regions. Rs AChRs are slowly degrading AChRs whose degradation rate can be modulated by the nerve between a tin of 8-10 days in innervated muscle and one of 2-4 days after denervation (original AChRs are essentially of the RS type; Salpeter et al., 1986). R, AChRs are rapidly degrading AChRs, with a tin of ~1 day, for which neural modulation has not been demonstrated. (In mouse sternomastoid muscleand possibly in other muscles as well, new AChRs consist of -90% R, and -10% Rs; [Shyng and Salpeter, 19901.) Forskolin Stabilizes the Degradation Rate of Rs AChRs in Denervated Diaphragm Muscle Previous studies have shown that in diaphragm muscle the acceleration of the Rs AChR (to a t,n of ~~2-4 days) occurs in vivo by 4-6 days after denervation (Brett et al., 1982; Bevan and Steinbach, 1983; Wetzel and Salpeter, 1991). Thus by the time the diaphragm muscle is placed in organ culture (6 days after denervation), the degradation rate of the Rs AChRs that had been labeled in vivo at the time of denervation has already accelerated. In the present study, as in that of Wetzel and Salpeter (1991), the accelerated degradation rate (tin of ~2.5 + 0.5 days) was maintained for 12 days in organ culture (from day 6 to day 18 after denervation) with no spontaneous deceleration (see control curves of Figures 1,2, and 3). The maintenance of an accelerated degradation rate by the Rs AChR for a prolonged period in organ culture is like that seen in vivo and is unlike an earlier report by Bevan and Steinbach (1983). When forskolin (20 or 40 uM dissolved in 95% ethanol) was added to the organ culture medium, there was a small but significant decrease in the degradation rate of Rs AChRs. The t1/2 changed from 2.3 days in control to 3.9 days after addition of 40 PM forskolin (Figure 1). The difference between 20 and 40 uM forskolin was not significant (Figure 1). However, the combined data from these two concentrations of forskolin were significantly different from those of the control (F-test, p < 0.01). Ethanol alone (final concentration of 0.5%, i.e., the concentration of ethanol present in the medium of forskolin-treated muscles) did
Neuron 470
1’ ’ ’ ’ 012345678910
’
’
’
’
’
’
J
Days in Culture
1”“““““” 0 1 2
3
4
5
6
7
8
9 1011
1213
Days in Culture
Figure 1. Forskolin Stabilizes the Degradation of Junctional Rs AChRs in Denervated Mouse Diaphragm Muscles in Organ Culture
Figure 2. Degradation of Junctional Rs AChRs in Controls Is Not Significantly Different from That in 1,9,-Dideoxyforskolin-Treated Denervated Mouse Diaphragm Muscles in Organ Culture
Diaphragms were denervated and labeled with 1251-a-BgTx in vivo. Six days later, when the Rs AChRs had already accelerated, the muscles were placed in organ culture, and forskolin was added to the medium on the first day (day0). Degradation curves were fitted by linear regression and normalized to 100% on day 0 of the control curves (see Experimental Procedures). Error bars represent SEM of 3-22 muscles per time point. The tin is 2.3 days for control, untreated muscles and 3.5 and 3.9 days for 20 uM and 40 uM forskolin-treated muscles, respectively. The difference between the data from forskolin-treated and control muscles is highly significant by the F-test (p < 0.01). It appears from the data that there may be a delay of about 3-4 days before the forskolin effect can be seen. If this is the case, the difference in half-lives between forskolin-treated and control AChRs would be even greater.
Culturing, curve normalization, and curve fitting are described in Experimental Procedures. Error bars are SEM of 3-7 muscles per time point. The tliz value is 3 days for controls and 2.8 days for 40 uM 1,9,-dideoxyforskolin-treated muscles.
not affect AChR degradation (data not shown). Therefore, forskolin, like reinnervation (see Salpeter et al., 1986), causes the accelerated degradation rate of the Us AChRs to slow down.
The Effect of Forskolin Is via a CAMP-Dependent Pathway Since forskolin is known to activate adenylate cyclase (Seamon and Daly, 1981), we tested whether the effect of forskolin on Rs AChR degradation is mediated by CAMP. We first examined the effect of l,g-dideoxyforskolin, an analog of forskolin that does not activate adenylate cyclase, yet has been shown to resemble forskolin in affecting ion channel and other membrane protein properties by a CAMP-independent pathway (Laurenza et al., 1989). Unlike forskolin, 1,9dideoxyforskolin (40 uM) did not change AChR degradation (Figure 2). On the other hand, CAMP analogs had an even stronger effect than forskolin in stabilizing the Rs AChRs and slowed the degradation rate -3-fold. After the addition of 1 m M dibutyryl CAMP (DBcAMP), the degradation rate of the Rs AChRs switched from a t112of m2.9 days to one of ~8.1 days (Figure 3). Another CAMP analog, 8-bromo-CAMP (1 mM) showed a similar effect (see Table 1). In addition, 3-isobutyl-I-methylxanthine (1 mM), a phospho-
diesterase inhibitor (Beavo et al., 1970; Montague and Cook, 1971), also stabilized the degradation rate of the Rs AChRs (see Table 1). The finding that DBcAMP was more effective than forskolin in stabilizing the Rs AChRs could be due to concentration differences or the fact that DBcAMP has a higher resistance to phosphodiesterase (Kau kel and Hilz, 1972; Hsie et al., 1975). Our organ culture experiments show that elevation of the CAMP level stabilizes the accelerated degradation rate of Rs AChRs, just as was previously shown for reinnervation in vivo (Salpeter et al., 1986). These results suggest that the modulation of Rs AChRdegradation in the plasma membrane by innervation may be mediated by a CAMP-dependent mechanism.
Forskolin and DBcAMP Have Little or No Effect on the Degradation Rate of R, AChRs in Denervated Diaphragm Muscle or in Noninnervated Culture Myotubes In a previous publication (Shyng and Salpeter, 1990), we showed that the degradation rate of the Us AChRs, but not that of the R, AChRs, is modulated by innervation. The possibility is raised that the R, AChRs, which are synthesized predominantly in denervated or preinnervated embryonic muscle, lack the capacity to modulate their degradation rate in the plasma membrane or cannot do so during their short lifetime. Since DBcAMP can mimic the innervation effect, one test of the above suggestion is to determine whether the degradation rate of the R, AChRs in long-term denervated muscle or in cultured myotubes is affected by a change in CAMP levels. Figure 4 shows that daily doses of forskolin (40 uM) or DBcAMP (1 mM) had no significant effect (p >> 0.05) on the degradation rate of the AChRs on rat skeletal muscle cells in cell cul-
CAMP Modulates 471
Acetylcholine
II”“““““” 0 1 2 3 4 5
Receptor
Degradation
6 7 8 9 10111213 14
A-------0
Days in Culture
Labeling,cuIturing,andcurvefittingaredescribedinExperimental Procedures. Each data point is the mean of 3-6 animals f SEM (error bars for the DBcAMP-treated cells fall within the size of the symbols). The tr/2 value is Q.9 days for controls and -8.1 days for the DBcAMP-treated group. This difference is statistically highly significant (p 0.05). No time points later than 5 days were examined, since by day 3 in culture (i.e., day 4 after labeling), only ~6%-10% of the initially labeled R, AChRs are expected to remain undegraded. In fact,
Chemical
2
3
4
5
Days After Labeling
Figure 3. DBcAMP (1 mM) Is More Effective than Forskolin in Stabilizing Junctional Rs AChR Degradation in Denervated Mouse Diaphragm Muscle in Organ Culture
Table 1. Effect of Various Degradation of R, AChRs
1
Treatments
on the
Treatment
% Residual Label on Day 9 (Normalized to 100% on Day 0 of Culture)
Control Forskolin (20 pm) 1,9-Dideoxyforskolin (20 urn) SBromo-cAMP (ImM) Dibutyryl-cAMP (1 mM) 3-lsobutyl-I-methylxanthine (1 mM)
8.12 18.62 4.19 31.02 34.77 34.99
* f * f f f
0.59 1.41 0.57 4.40 2.15 2.32
Figure 4. DBcAMP (1 mM) and Forskolin (40 PM) Do Not Affect the Degradation Rateof R,AChR in Rat Myotubes in Cell Culture Each data point represents the mean of 4 samples f SEM. For each group the residual label was normalized to 100% at day 0 (or day4 in culture), which was the time of receptor labeling and the beginning of chemical treatment. (In this figure, the value at day0 is an experimental and not an extrapolated value.) Degradation curves obtained by linear regression gave tin values of 1.3 days, 1.2 days, and 1.3days forcontrols (middlecurve), DBcAMPtreated cells (top curve), and forskolin-treated cells (bottom curve), respectively. With all three treatments the observed t,n values of the AChRs were somewhat longer than expected as a result of the obvious deviation from a single exponential at later times. If only the first 3 days are considered, the t,R values in all cases are ~1 day. The longer than expected half-life probably reflects the presence of a small amount of a slow component (i.e., Rs in the accelerated state) in these noninnervated myotubes, as previously described for denervated muscle (Shyng and Salpeter, 1990). The curves were not extended far enough to make an assessment of the extent of this slow component.
when going to 5 days in culture, calculating the degradation half-life on the assumption of a single exponential can slightly overestimate the true half-life because some (~10%) of the new AChRs in denervated muscle may beslowlydegrading(or Rs in itsaccelerated form), which would then be further slowed by the DBcAMP (Shyng and Salpeter, 1990; see also Figure 4). We conclude from our results that, although DBCAMP can slow the degradation rate of the accelerated Rs AChRs to the predenervation level (t,n of Ql days; as seen in Figure 3), it has only a slight effect (if any) on the degradation rate of the R, AChRs, as seen in Figures 4 and 5. The slowing of the Rs AChR degradation rate is thus not a general, nonspecific effect of CAMP on slowing protein degradation, but depends on the AChR type. Discussion Studies on degradation of AChRs at the neuromuscular junction have led to the conclusion that there are two populations of AChRs (Rs and R,), differing in degradation properties, (Shyng and Salpeter, 1990). One of them, the RS form, can be in either a stabilized
NeLlrOn
472
1’
0
’
’
1
0
s
’
’
’
’
.
’
5
s
’
6
DBcAMP
lL-----0
I
2 3 4 Days in Culture
1
2 3 4 Days in Culture
5
6
Figure 5. DBcAMP Has Little Effect on the Degradation Rate of R, AChRs in Denervated Muscles at Either the Neuromuscular Junctions or the Extrajunctional Regions (A)AChRsattheneuromuscularjunctions;(B)AChRsattheextrajunctional regions. Diaphragm muscles were denervated and were labeled by injection of Y-a-BgTx in vivo 17 days later. By that time, the original Rs AChRs had degraded >90% and were replaced by R, AChRs. One day after labeling, muscles were placed in organ culture (day 0), and DBcAMP (1 mM) was added to theculture medium. Degradationwas followed for5days with or without DBcAMP. Each data point represents the mean of 5-6 animals f SEM. Linear regression curves give tin values for control and DBcAMP-treated muscles of 1.2 days and 1.5 days, respectively, at theendplate(A)and 1.4daysand 1.8days, respectively, at extrajunctional regions(B). Neither difference is statistically significant (p > 0.05; compare with data from Figure 3). Furthermore, as previously reported (Shyng and Salpeter, 1990), denervated muscle expresses some (-10%) slowly degrading (or Rs) AChRs, which would cause the trj2 values given here (as those in Figure 4), especially after DBcAMP treatment, to be a slight overestimate.
(observed t1j2 of -8-10 days) or an accelerated (tIiz of ~2-3 days) state. The stabilized state, which is seen at the endplates in adult innervated muscles, is converted to the accelerated state after denervation and back to the stabilized state upon reinnervation (Sal-
peter et al., 1986). In the present study we show that the neurally induced stabilization can be mimicked by agents that elevate CAMP levels in the muscle, suggesting that the stabilization of the RS degradation rate by innervation could involve a CAMP-dependent mechanism. Previous studies have reported that CAMP affects AChR desensitization (Huganir et al., 1986), synthesis(Betzand Changeux, 1979; Blosser and Appel, 1980; Fontaine et al., 1987), and subunit assembly (Ross et al., 1987). The present study reports that CAMP can affect AChR degradation, distinguishing between Rs and R, AChRs. Contrary to the situation with the RS AChRs, neither reinnervation (Shyng and Salpeter, 1990) nor the elevation of CAMP levels (present study) can decrease the degradation rate of the R, AChRs to the values seen at adult innervated neuromuscular junctions. The maximal tT12value of the R, AChR following DBcAMP treatment was less than 2 days. W e suggest that the development of metabolically stable AChRs at a neuro muscular junction after innervation cannot involve the stabilization of R, AChRs. The development of a slowly degrading, stable AChR population at the neuromuscular junction probably involves both a down regulation of the R, AChRs and their replacement by the Rs AChRs, as well as the stabilization of the R5 AChRs once they are in the plasma membrane. Electrical stimulation of denervated muscles causes stabilization of AChRs at the endplate band (Fumagalli et al., 1990; Rotzler and Brenner, 1990), but not those at the extrajunctional bands (Rotzler and Brenner, 1990). Rotzler and Brenner suggested that the different behavior of the junctional and extrajunctional AChRs in their study could reflect some, as yet unidentified, unique property of the neuromuscular junction. If such a unique property exists, it is not sufficient to stabilize the degradation rate of R, AChRs, since even at the endplate band, the R, AChRs did not slow their degradation rate in response to either reinnervation (Shyng and Salpeter, 1990) or CAMP (present study). Our results suggest that there may be an intrinsic difference in the responsiveness of R, and RS AChRs that is independent of localization. The differences in the responsiveness of Rs and R, AChRs to factors regulating degradation seem to be imparted during the synthesis of the AChRs, whether in innervated or denervated muscle. It is tempting to suggest that the Rs AChRs may be the E subunit-containing adult type and the R, AChRs may be the y subunit-containing fetal type of AChR (Cu et al., 1990; Shyng and Salpeter, 1990). One major problem with a suggestion that E and y subunits are involved in regulating AChR degradation properties is the fact that, in rat muscleat least, the decrease in AChRdegradation rate at the neuromuscular junction appears within a few days of synapse formation (Steinbach et al., 1979; Reiness and Weinberg, 1981; Rotzler and Brenner, 1990), whereas the E subunit at the junction is not expressed in significant amounts until several days later (Gu and Hall, 1988; Brenner et al., 1990). The
CAMP Modulates 473
Acetylcholine
Receptor
Degradation
timing of the appearance of the E subunit coincides with that of the high conductance AChR channels (e.g., Sakmann and Brenner, 1978; Fischbach and Schuetze, 1980) known to be associated with the E subunit (Mishina et al., 1986; Witzemann et al., 1987; Brehm and Henderson, 1988; Brenner and Rudin, 1989). Yet many of the characteristics of Rs and R, AChRsareconsistentwiththeirbeingsandysubunitcontaining, respectively. These include the following facts: First, Rs AChRs are synthesized mainly in innervated and R, AChRs in denervated muscle. Second, the small percentage of Rs (40%) in denervated muscle (Shyng and Salpeter, 1990) is consistent with the small percentage of E subunit present long after denervation (Henderson et al., 1987; Witzemann et al., 1987; Gu and Hall, 1988. Third, they subunit is down regulated by muscle activity (Goldman et al., 1988), as are the postdenervation extrajunctional AChRs (see Fambrough, 1979) and, as we suggest, the new R, junctional AChRs. Finally, E subunit-containing AChRs degrade more slowly than y subunit-containing AChRs in transfected COS cells (Gu et al., 1990). We also note, as it is possibly relevant to our present study, that the mammalian E subunit has a potential CAMP-dependent phosphorylation site,whereas they subunit does not (Huganir and Miles, 1989). However, unless the demonstrated difference in developmental time course between the appearanceof the E subunit and the slowing of AChR degradation can be resolved, some other property of the AChR imparted atthetimeof synthesis must be sought to explain the differences in degradation behavior between Rs and R, AChRs. In the search for factors that regulate AChR degradation, both structural changes in theAChR(Gu et al., 1990) and muscle activity (Avila et al., 1989; Brenner and Rudin, 1989; Rotzler and Brenner, 1990; Fumagalli et al., 1990) have been implicated. The present study shows that CAMP may be involved in one aspect of this regulation process, i.e., the modulation of the Rs AChRs at denervated neuromuscular junctions. Experimental
Procedures
Denervation of Diaphragm Muscle and labeling of AChRs Adult female mice (CD-I, Charles River Laboratories, Wilmington, MA) weighing 25-35 g were used. Under ether anesthesia, the left phrenic nerve was cut after being externalized with a glass hook through an incision at the level of the second rib. The AChRs were labeled by intrathoracic or intraperitoneal injection of 150 ul of 0.5 uM Y-a-bungarotoxin (a-BgTx), which can label ~60% of AChRs, as judged by a comparison with muscles bathed to saturation in 1251-a-BgTx in vitro (Wetzel and Salpeter, 1991). For studies on the degradation of original AChRs, the receptors were labeled at the time of denervation and muscles were removed 6 days later for culturing. For studies of new AChRs, synthesized after denervation, the receptors were labeled 17 days after denervation (at a time when >VO% of the original junctional AChRs had degraded and were replaced by new AChRs) and muscles were removed 1 day later for culturing. Culture of Diaphragm Animals were sacrificed by cervical translocation under ether anesthesia and perfused transcardially with oxygenated Kreb’s solution (4OC). The diaphragms were quickly dissected and im-
mersed in 4OC oxygenated Kreb’s solution. The denervated left diaphragm was separated from the right diaphragm, which we used as a control for calibrating AChR labeling efficiency (see below). Left diaphragms were then treated as previously described by Wetzel and Salpeter (1991). Briefly, they were pinned onto organ culture dishes and placed in a rocking incubator chamber in a medium consisting of Ml99 (Flow Labs, McLean, VA) plus 5% fetal calf serum and a number of nutritional supplements (given for optimal conditions [Wetzel and Salpeter, IVVI]). These included 2 m M B-hydroxybutyricacid (Sigma), L-glutamine (1.6 mM), ascorbic acid (IO ug/ml), 5-a-dihydrotestosterone (1 nM), pyruvate (0.9 mM), fructose (2.3 mM), and thiamine (35 PM). These optimal conditions were judged by the ability of the muscle to have normal fine structure, continued fibrillation, and an accelerated AChR degradation rate for more than 2 weeks in organ culture. The culture medium plus various pharmacological agents (see below) was replaced daily. Chemical Treatments The various pharmacological agents included forskolin and l,Vdideoxyforskolin from Calbiochem (San Diego, CA), CAMP analogsand 3-isobutyl-I-methylxanthinefrom Sigma(St. Louis,MO). Degradation of AChRs on Diaphragm Muscle in Organ Culture The degradation rate of junctional AChRs on the organ cultured muscles (and for the AChRs on myotubes in cell culture described below) was determined by the rate of loss of radioactivity from the endplate band (for justification, see reviews by Fambrough, 1979; Salpeterand Loring, 1985). Sincethe muscles were labeled by whole animal injection and the AChRs were not fully saturated, the label on the denervated muscle to be cultured was corrected for labeling efficiency by the relative label on the innervated right diaphragm determined at the time of removal from the animal. At different times during the culture period, muscles were fixed with 4% paraformaldehyde (in 0.05 M phosphate buffer [pH 7.41) for 2 hr and rinsed 3 times (10 min each) with 0.1 M phosphate buffer (pH 7.4). In each experiment the first control data point was taken 1 day after culturing began. The muscles were dissected into non-endplate bands and an endplate band(judged bytransilluminationorstainingforacetylcholinesterase). Both the endplate and the non-endplate bands were weighed, and radioactivity was counted using a Beckman gamma counter. The endplatespecific label was determined by subtracting the non-endplate band label from the endplate label on a per weight basis. Extrajunctional label was taken as the total muscle label minus the endplate-specific label. In each case the control data were first fitted by linear regression assuming a single exponential. The extrapolated value at day 0 was then set to 100% and used to normalize both the control and the experimental data on the assumption that muscles from both groups had the same label at the beginning of the culture period. Thedatapointsondayoin thefiguresoforgancultured muscles is thus an extrapolated and not an experimental value. Differences in degradation rates were analyzed statistically by the F-test as previously described (Shyng and Salpeter, 1990). Determining the Degradation Rate of Fetal AChRs on Rat Myotubes Primary cultures of rat skeletal muscle were prepared as described previously (Salpeter et al., 1982). Cells were plated on 24 multiwell plates at 2.5 x 104cell perwell in Dulbecco’s modified Eagle’s medium (DMEM; CIBCO) plus 10% fetal calf serum. Cytosine-I-B-arabinofuranoside (final concentration 10e5 M) was added 72 hr after plating. On day 4 after plating AChRs were labeled with ‘=I-a-BgTx (2 x 1O-8 M) in DMEM-HEPES buffer plus bovine serum albumin (1 mg/ml [pH 7.41) for 30 min at room temperature, followed by a series of washes in DMEM-bovine serum albumin for a total of 30 min. The original “conditioned” growth medium was then returned to the celLToobtain nonspecific binding, cells were incubated with nonradioactive a-BgTx (10-r M) for 15 min at room temperature, washed thoroughly, and then labeled with Y-a-BgTx (30 min), followed by thorough washing as described above. Chemical treatments began imme-
Netltoll 474
diately after AChR labeling. Culture medium and various chemicals were replaced daily. At different times after labeling, the amount of label remaining on the cells was determined bywashing off the medium, dissolving the cells in 2 N NaOH overnight, absorbing the NaOH solution on cotton plugs, and counting these in a gamma counter as described by Knaack and Podleski (1985). Half-life values were obtained by linear regression. Each curve was normalized to 100% for its own experimental value on day 0 (or day 4 in culture). For ail the groups in the present study we report the degradation half-life as the observed value (tr/& obtained from the loss of radioactivity. This can be corrected for a-BgTx unbinding (t,n..r, ~56 days; Cohen et al., 1990) and radioactive decay (trod, for lrz5 = 60 days) to give the true degradation rate (t,nd,,,) using the following equation: (fl/*degr)-’ = (tmd’
- (fli*““b)r’ - (tl/2dec)-’
Acknowledgments We thank Daniel Wetzel, Dorothy Bell, and Iris Greenberg for assistance with these experiments, Tom Podleski and Ron Harris-Warrick for helpful discussion, and Deborah Moslehi for preparing the manuscript. This work was supported by grants from the National Institutes of Health (NS09315) and from the Cornell University Biotechnology Program sponsored by the N.Y. State Science and Technology Foundation, by a consortium of industries, and by the United States Army Research Office and NSF. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisemem’in accordance with 18 USC Section 1734 solely to indicate this fact. Received
October
24, 1990; revised
January
8, 1991.
Fischbach, C. D., and Schuetze, S. M. (1980). A post-natal decrease in acetylcholine channel open time at rat end-plates. J. Physiol. 303, 125-137. Fontaine, B., Klarsfeld, A., and Changeux, 1. P. (1987). Calcitonin gene-related peptide and muscle activity regulate acetylcholine receptor a-subunit mRNA levels by distinct intracellular pathways. J. Cell Biol. 705, 1337-1342. Fumagalli, G., Balbi, S., Cangiano,A,and LBmo,T. (1990). Regulation of turnover and number of acetylcholine receptors at neuromuscular Junctions. Neuron 4, 563-569. Goldman, D., Brenner, H. R., and Heinemann, S. (1988). Acetylcholine receptor a-, B-, y-, and &subunit mRNA levels are regulated by muscle activity. Neuron 7, 329-333. Gu, Y., and Hall, Z. W. (1988). Immunological evidence for a change in subunits of the acetylcholine receptor in developing and denervated rat muscle. Neuron 7, 117-125. Gu, Y., France, A., Jr., Gardner, P. D., Lansman, J. B., Forsayeth, J. R., and Hall, Z. W. (1990). Properties of embryonic and adult muscle acetylcholine receptors transiently expressed in COS cells. Neuron 5, 147-157. Henderson, L. P., Lechleiter,J.,and Brehm, P. (1987). Singlechannel properties of newly synthesized acetylcholine receptors following denervation of mammalian skeletal muscle. J. Cen. Physiol. 89, 999-1014. Hsie, A. W., Kawashima, K., O’Neill, J. P., and Schroder, C. H. (1975). Possible role of adenosine cyclic 3’:5’-monophosphate phosphodiesterase in the morphological transformation of Chinese hamster ovary cells mediated by N6,0z’dibutyryl adenosine cyclic 3’5’-monophosphate. 1. Biol. Chem. 250, 984-989. Huganir, R. L., and Miles, K. (1989). Protein phosphorylation of nicotinic acetylcholine receptors. Crit. Rev. Biochem. Mol. Biol. 24, 183-215. Huganir, R. L., Delcour, A. H., Creengard, P., and Hess, C. P. (1986). Phosphorylation of the nicotinic acetylcholine receptor regulates its rate of desensitization. Nature 327, 774-776.
Avila, 0. L., Drachman, D. B., and Pestronk, A. (1989). Neurotransmission regulates stability of acetylcholine receptors at the neuromuscular junction. J. Neurosci. 9, 2902-2906. Beavo,J.A., Rogers, N. L., Crofford, 0. B., Hardman, J. C., Sutherland, E. W., and Newman, E. V. (1970). Effects of xanthine derivatives on lipolysis and on adenosine 3’-5’-monophosphate phosphodiesterase activity. Mol. Pharmacol. 6, 597-603. Betz, H., and Changeux, J. P. (1979). Regulation of muscle acetylcholine receptor synthesis in vitro by derivatives of cyclic nucleotides. Nature 278, 749-752. Bevan, S., and Steinbach, J. H. (1983). Denervation increases the degradation rate of acetylcholine receptors and end-plates in vivo and in vitro. J. Physiol. 336, 159-177. Blosser, J. C., and Appel, S. H. (1980). Regulation of acetylcholine receptor by cyclic AMP. J. Biol. Chem. 253, 3088-3093. Brehm, P., and Henderson, receptor channel function cle. Dev. Biol. 729, I-11.
L. (1988). Regulation during development
of acetylcholine of skeletal mus-
Brenner, H. R., and Rudin, W. (1989). On the effect of muscle activity on the end-plate membrane in denervated mouse muscle. J. Physiol. 470, 501-512. Brenner, H. R., Witzemann, V., and Sakmann, B. (1990). Imprinting of acetylcholine receptor messenger RNAaccumulation in mammalian neuromuscular synapses. Nature 344, 544-547. Brett, R. S., Younkin, 5. C., Konieczkowski, M., and Slugg, R. M. (1982). Accelerated degradation of junctional acetylcholine receptor-a-bungarotoxin complexes in denervated rat diaphragm. Brain Res. 233, 133-142. Cohen, M. W., Frair, P. F., Cantin, C., and Hebert, G. (1990). Developmental changes in the half-life of acetylcholine recep tors in the myotomal muscle of Xenopus laevis. J. Physiol. 426, 281-296. Fambrough, D. M. (1979). Control of acetylcholine skeletal muscle. Physiol. Rev. 59, 165-227.
receptors
in
Kaukel, E., and Hilz, H. (1972). Permeation ofdibutyryl CAMP into HeLa cells and its conversion to monobutyryl CAMP. Biochem. Biophys. Res. Commun. 46, 1011-1018. Knaack, D., and Podleski, T. R. (1985). Ascorbic acid mediates acetylcholine receptor increase induced by brain extract in L5 myogenic cells. Proc. Natl. Acad. Sci. USA 82, 575-579. Laurenza, A., Sutkowski, E. M., and Seamon, K. B. (1989). Forskolin: a specific stimulator of adenylyl cyclase or a diterpene with multiple sites of action? Trends Pharmacol. Sci. 70,442-447. Levitt, T. A., and Salpeter, M. M. (1981). Denervated have a dual population of junctional acetylcholine Nature 297, 239-241.
endplates receptors.
Mishina, M., Takai, T., Imoto, K., Noda, M., Takahashi, T., Numa, S., Methfessel, C., and Sakmann, B. (1986). Molecular distinction between fetal and adult forms of muscle acetylcholine receptor. Nature 327, 406-441. Montague, W., and Cook, 1. R. (1971). The role of adenosine 3’: 5’cyclic monophosphate in the regulation of insulin release by isolated rat islets of Langerhans. Biochem. J. 722, 115-120. Reiness, C. G., and Weinberg, C. B. (1981). Metabolic stabilization of acetylcholine receptors at newly formed neuromuscular junctions in rat. Dev. Biol. 84, 247-254. Ross, A. F., Rapuano, M., Schmidt, J. H., and Proves, J. M. (1987). Phosphorylation and assembly of nicotinic acetylcholine receptor subunits in cultured chick muscle cells. J. Biol. Chem. 262, 14640-14647. Rotzler, S., and Brenner, H. R. (1990). Metabolic stabilization acetylcholine receptors in vertebrate neuromuscular junction muscle activity. J. Cell Biol. 777, 655-661. Sakmann, B., and Brenner, H. R. (1978). Change nel gatingduring neuromusculardevelopment. 402. Salpeter,
M. M., and Loring,
of by
in synaptic chanNature276,401-
R. H. (1985). Nicotinic
acetylcholine
CAMP Modulates 47;
Acetylcholine
Receptor
Degradation
receptors in vertebrate muscle: properties, ral control. Prog. Neurobiol. 25, 297-325.
distribution
and neu-
Salpeter, M. M., Spanton, S., Holley, K., and Podleski, T. R. (1982). Brain extract causes acetylcholine receptor redistribution which mimics some early events at developing neuromuscular junctions. J. Cell Biol. 93, 417-425. Salpeter, M. M., Cooper, D. L., and Levitt-Cilmour, T. (1986). Degradation rates of acetylcholine receptors can be modified after they are inserted into the postjunctional plasma membrane o! the vertebrate neuromuscular junction. J. Cell Biol. 703,13991403. Soamon, K. B., and Daly, J. W. (1981). Forskolin: a unique diterpcne activator of CAMP-generating systems. J. Cycl. Nucl. Res. 7, 2(11-224. Shyng, S.-L., and Salpeter, M. M. (1990). Effect of reinnervation ori the degradation rate of junctional acetylcholine receptors synthesized in denervated skeletal muscles. J. Neurosci. 70, 3905-3915. Stanley, E. F., and Drachman, D. B. (1981). Denervation accelerates the degradation of junctional acetylcholine receptors. Exp. Neural. 73, 390-396. Steinbach, J. H., Merlie, J., Heinemann, S., and Bloch, R. (1979). Degradation of junctional and extrajunctional acetylcholine receptors by developing rat skeletal muscle. Proc. Natl. Acad. Sci. C SA 76, 3547-3551. Wetzel, D. M., and Salpeter,M. M. (1991). Fibrillation and accelerated AChR degradation in long term muscle organ culture. Muscle Nerve, in press. VVitzemann,V., Barg, B., Nishikawa, Y., Sakmann, B., and Numa, S. (1987). Differential regulation of muscleacetylcholine receptor y- and e-subunit mRNAs. FEBS Lett. 223, 104-112.
Vol. 6, 469-475, March, 1991, Copyright
0 1991 by Cell Press
Cyclic AMP Stabilizes the Degradation of O riginal Junctional Acetylcholine Receptors in Decervated Muscle S.-L. Shyng,* R. Xu, and M. M. Salpeter Section of Neurobiology and Behavior Division of Biological Sciences Cornell University Ithaca. New York 14853-2702
rapidly degrading (R,) AChRs (t1/2of ~1 day) synthesized in denervated muscleor expressed by myotubes in primary cell culture. Possible mechanisms for neural regulation of AChR degradation are discussed. Results
Summary We used mouse diaphragm muscle in organ culture to study the stabilization of acetylcholine receptor (AChR) degradation at denervated neuromuscular junctions. After denervation, the degradation rate of the AChRs present prior to denervation (slowly degrading, or Rs, AChRs) accelerates from the predenervation degradation half-life (t& of 4%10 days to a tljz of m2-3 days. We report that addition to the organ culture medium of pharmacological agents that elevate cytoplasmic CAMP levels (forskolin, dibutyryl CAMP, and 8-bromo-CAMP) reversed the change in t1/2 caused by denervation, whereas addition of l,%dideoxyforskolin, a forskolin analog that does not elevate cytoplasmic CAMP levels, did not reverse the effect of denervation. The degradation rate of AChRs in primary myotube cultures and that of the newly synthesized AChRs in denervated muscle were little affected by forskolin or dibutyryl CAMP. The possibility is raised that the modulation of Rs AChR degradation by innervation may be mediated by CAMP. Introduction Acetylcholine receptors (AChRs) at innervated neuromuscular junctions, variously called original, adult, or slowly degrading (Rs) AChRs, have a slow degradation half-life (t,,? of @-IO days). The Rs AChRs can exist in a stabilized or an accelerated state. A few days after denervation the degradation rate of the Rs AChRs accelerates to a t112of ~2-4 days (Levitt and Salpeter, 1981; Stanley and Drachman, 1981; Brett et al., 1982; Bevan and Steinbach, 1983; Wetzel and Salpeter, 1991); reinnervation causes it to revert to the stabilized, predenervation value (Salpeter et al., 1986). Since these changes in the degradation rate occur after the Rs AChRs have been inserted into the postsynaptic membrane, they do not involve de novo AChR synthesis, but probably arise either from posttranslational modifications of the AChRs or changes in their molecular environment. We show here that CAMP can mimic the effect of reinnervation in stabilizing the Rs AChRs in denervated muscle. Furthermore, while CAMP decreases the degradation rate of Rs AChRs at the denervated neuromuscular junction to predenervation levels, it has little or no effect on the degradation rate of the * Present address: Division of Biology, California Technology, Pasadena, California 91125.
Institute
of
In the present study we use the following definitions for the different AChRs, based on location, time of synthesis, or degradation properties. Original AChRs refer to AChRs synthesized in innervated muscle and present at the neuromuscular junction prior to denervation. New AChRs are those synthesized after denervation and present at both junctional and extrajunctional regions. Rs AChRs are slowly degrading AChRs whose degradation rate can be modulated by the nerve between a tin of 8-10 days in innervated muscle and one of 2-4 days after denervation (original AChRs are essentially of the RS type; Salpeter et al., 1986). R, AChRs are rapidly degrading AChRs, with a tin of ~1 day, for which neural modulation has not been demonstrated. (In mouse sternomastoid muscleand possibly in other muscles as well, new AChRs consist of -90% R, and -10% Rs; [Shyng and Salpeter, 19901.) Forskolin Stabilizes the Degradation Rate of Rs AChRs in Denervated Diaphragm Muscle Previous studies have shown that in diaphragm muscle the acceleration of the Rs AChR (to a t,n of ~~2-4 days) occurs in vivo by 4-6 days after denervation (Brett et al., 1982; Bevan and Steinbach, 1983; Wetzel and Salpeter, 1991). Thus by the time the diaphragm muscle is placed in organ culture (6 days after denervation), the degradation rate of the Rs AChRs that had been labeled in vivo at the time of denervation has already accelerated. In the present study, as in that of Wetzel and Salpeter (1991), the accelerated degradation rate (tin of ~2.5 + 0.5 days) was maintained for 12 days in organ culture (from day 6 to day 18 after denervation) with no spontaneous deceleration (see control curves of Figures 1,2, and 3). The maintenance of an accelerated degradation rate by the Rs AChR for a prolonged period in organ culture is like that seen in vivo and is unlike an earlier report by Bevan and Steinbach (1983). When forskolin (20 or 40 uM dissolved in 95% ethanol) was added to the organ culture medium, there was a small but significant decrease in the degradation rate of Rs AChRs. The t1/2 changed from 2.3 days in control to 3.9 days after addition of 40 PM forskolin (Figure 1). The difference between 20 and 40 uM forskolin was not significant (Figure 1). However, the combined data from these two concentrations of forskolin were significantly different from those of the control (F-test, p < 0.01). Ethanol alone (final concentration of 0.5%, i.e., the concentration of ethanol present in the medium of forskolin-treated muscles) did
Neuron 470
1’ ’ ’ ’ 012345678910
’
’
’
’
’
’
J
Days in Culture
1”“““““” 0 1 2
3
4
5
6
7
8
9 1011
1213
Days in Culture
Figure 1. Forskolin Stabilizes the Degradation of Junctional Rs AChRs in Denervated Mouse Diaphragm Muscles in Organ Culture
Figure 2. Degradation of Junctional Rs AChRs in Controls Is Not Significantly Different from That in 1,9,-Dideoxyforskolin-Treated Denervated Mouse Diaphragm Muscles in Organ Culture
Diaphragms were denervated and labeled with 1251-a-BgTx in vivo. Six days later, when the Rs AChRs had already accelerated, the muscles were placed in organ culture, and forskolin was added to the medium on the first day (day0). Degradation curves were fitted by linear regression and normalized to 100% on day 0 of the control curves (see Experimental Procedures). Error bars represent SEM of 3-22 muscles per time point. The tin is 2.3 days for control, untreated muscles and 3.5 and 3.9 days for 20 uM and 40 uM forskolin-treated muscles, respectively. The difference between the data from forskolin-treated and control muscles is highly significant by the F-test (p < 0.01). It appears from the data that there may be a delay of about 3-4 days before the forskolin effect can be seen. If this is the case, the difference in half-lives between forskolin-treated and control AChRs would be even greater.
Culturing, curve normalization, and curve fitting are described in Experimental Procedures. Error bars are SEM of 3-7 muscles per time point. The tliz value is 3 days for controls and 2.8 days for 40 uM 1,9,-dideoxyforskolin-treated muscles.
not affect AChR degradation (data not shown). Therefore, forskolin, like reinnervation (see Salpeter et al., 1986), causes the accelerated degradation rate of the Us AChRs to slow down.
The Effect of Forskolin Is via a CAMP-Dependent Pathway Since forskolin is known to activate adenylate cyclase (Seamon and Daly, 1981), we tested whether the effect of forskolin on Rs AChR degradation is mediated by CAMP. We first examined the effect of l,g-dideoxyforskolin, an analog of forskolin that does not activate adenylate cyclase, yet has been shown to resemble forskolin in affecting ion channel and other membrane protein properties by a CAMP-independent pathway (Laurenza et al., 1989). Unlike forskolin, 1,9dideoxyforskolin (40 uM) did not change AChR degradation (Figure 2). On the other hand, CAMP analogs had an even stronger effect than forskolin in stabilizing the Rs AChRs and slowed the degradation rate -3-fold. After the addition of 1 m M dibutyryl CAMP (DBcAMP), the degradation rate of the Rs AChRs switched from a t112of m2.9 days to one of ~8.1 days (Figure 3). Another CAMP analog, 8-bromo-CAMP (1 mM) showed a similar effect (see Table 1). In addition, 3-isobutyl-I-methylxanthine (1 mM), a phospho-
diesterase inhibitor (Beavo et al., 1970; Montague and Cook, 1971), also stabilized the degradation rate of the Rs AChRs (see Table 1). The finding that DBcAMP was more effective than forskolin in stabilizing the Rs AChRs could be due to concentration differences or the fact that DBcAMP has a higher resistance to phosphodiesterase (Kau kel and Hilz, 1972; Hsie et al., 1975). Our organ culture experiments show that elevation of the CAMP level stabilizes the accelerated degradation rate of Rs AChRs, just as was previously shown for reinnervation in vivo (Salpeter et al., 1986). These results suggest that the modulation of Rs AChRdegradation in the plasma membrane by innervation may be mediated by a CAMP-dependent mechanism.
Forskolin and DBcAMP Have Little or No Effect on the Degradation Rate of R, AChRs in Denervated Diaphragm Muscle or in Noninnervated Culture Myotubes In a previous publication (Shyng and Salpeter, 1990), we showed that the degradation rate of the Us AChRs, but not that of the R, AChRs, is modulated by innervation. The possibility is raised that the R, AChRs, which are synthesized predominantly in denervated or preinnervated embryonic muscle, lack the capacity to modulate their degradation rate in the plasma membrane or cannot do so during their short lifetime. Since DBcAMP can mimic the innervation effect, one test of the above suggestion is to determine whether the degradation rate of the R, AChRs in long-term denervated muscle or in cultured myotubes is affected by a change in CAMP levels. Figure 4 shows that daily doses of forskolin (40 uM) or DBcAMP (1 mM) had no significant effect (p >> 0.05) on the degradation rate of the AChRs on rat skeletal muscle cells in cell cul-
CAMP Modulates 471
Acetylcholine
II”“““““” 0 1 2 3 4 5
Receptor
Degradation
6 7 8 9 10111213 14
A-------0
Days in Culture
Labeling,cuIturing,andcurvefittingaredescribedinExperimental Procedures. Each data point is the mean of 3-6 animals f SEM (error bars for the DBcAMP-treated cells fall within the size of the symbols). The tr/2 value is Q.9 days for controls and -8.1 days for the DBcAMP-treated group. This difference is statistically highly significant (p 0.05). No time points later than 5 days were examined, since by day 3 in culture (i.e., day 4 after labeling), only ~6%-10% of the initially labeled R, AChRs are expected to remain undegraded. In fact,
Chemical
2
3
4
5
Days After Labeling
Figure 3. DBcAMP (1 mM) Is More Effective than Forskolin in Stabilizing Junctional Rs AChR Degradation in Denervated Mouse Diaphragm Muscle in Organ Culture
Table 1. Effect of Various Degradation of R, AChRs
1
Treatments
on the
Treatment
% Residual Label on Day 9 (Normalized to 100% on Day 0 of Culture)
Control Forskolin (20 pm) 1,9-Dideoxyforskolin (20 urn) SBromo-cAMP (ImM) Dibutyryl-cAMP (1 mM) 3-lsobutyl-I-methylxanthine (1 mM)
8.12 18.62 4.19 31.02 34.77 34.99
* f * f f f
0.59 1.41 0.57 4.40 2.15 2.32
Figure 4. DBcAMP (1 mM) and Forskolin (40 PM) Do Not Affect the Degradation Rateof R,AChR in Rat Myotubes in Cell Culture Each data point represents the mean of 4 samples f SEM. For each group the residual label was normalized to 100% at day 0 (or day4 in culture), which was the time of receptor labeling and the beginning of chemical treatment. (In this figure, the value at day0 is an experimental and not an extrapolated value.) Degradation curves obtained by linear regression gave tin values of 1.3 days, 1.2 days, and 1.3days forcontrols (middlecurve), DBcAMPtreated cells (top curve), and forskolin-treated cells (bottom curve), respectively. With all three treatments the observed t,n values of the AChRs were somewhat longer than expected as a result of the obvious deviation from a single exponential at later times. If only the first 3 days are considered, the t,R values in all cases are ~1 day. The longer than expected half-life probably reflects the presence of a small amount of a slow component (i.e., Rs in the accelerated state) in these noninnervated myotubes, as previously described for denervated muscle (Shyng and Salpeter, 1990). The curves were not extended far enough to make an assessment of the extent of this slow component.
when going to 5 days in culture, calculating the degradation half-life on the assumption of a single exponential can slightly overestimate the true half-life because some (~10%) of the new AChRs in denervated muscle may beslowlydegrading(or Rs in itsaccelerated form), which would then be further slowed by the DBcAMP (Shyng and Salpeter, 1990; see also Figure 4). We conclude from our results that, although DBCAMP can slow the degradation rate of the accelerated Rs AChRs to the predenervation level (t,n of Ql days; as seen in Figure 3), it has only a slight effect (if any) on the degradation rate of the R, AChRs, as seen in Figures 4 and 5. The slowing of the Rs AChR degradation rate is thus not a general, nonspecific effect of CAMP on slowing protein degradation, but depends on the AChR type. Discussion Studies on degradation of AChRs at the neuromuscular junction have led to the conclusion that there are two populations of AChRs (Rs and R,), differing in degradation properties, (Shyng and Salpeter, 1990). One of them, the RS form, can be in either a stabilized
NeLlrOn
472
1’
0
’
’
1
0
s
’
’
’
’
.
’
5
s
’
6
DBcAMP
lL-----0
I
2 3 4 Days in Culture
1
2 3 4 Days in Culture
5
6
Figure 5. DBcAMP Has Little Effect on the Degradation Rate of R, AChRs in Denervated Muscles at Either the Neuromuscular Junctions or the Extrajunctional Regions (A)AChRsattheneuromuscularjunctions;(B)AChRsattheextrajunctional regions. Diaphragm muscles were denervated and were labeled by injection of Y-a-BgTx in vivo 17 days later. By that time, the original Rs AChRs had degraded >90% and were replaced by R, AChRs. One day after labeling, muscles were placed in organ culture (day 0), and DBcAMP (1 mM) was added to theculture medium. Degradationwas followed for5days with or without DBcAMP. Each data point represents the mean of 5-6 animals f SEM. Linear regression curves give tin values for control and DBcAMP-treated muscles of 1.2 days and 1.5 days, respectively, at theendplate(A)and 1.4daysand 1.8days, respectively, at extrajunctional regions(B). Neither difference is statistically significant (p > 0.05; compare with data from Figure 3). Furthermore, as previously reported (Shyng and Salpeter, 1990), denervated muscle expresses some (-10%) slowly degrading (or Rs) AChRs, which would cause the trj2 values given here (as those in Figure 4), especially after DBcAMP treatment, to be a slight overestimate.
(observed t1j2 of -8-10 days) or an accelerated (tIiz of ~2-3 days) state. The stabilized state, which is seen at the endplates in adult innervated muscles, is converted to the accelerated state after denervation and back to the stabilized state upon reinnervation (Sal-
peter et al., 1986). In the present study we show that the neurally induced stabilization can be mimicked by agents that elevate CAMP levels in the muscle, suggesting that the stabilization of the RS degradation rate by innervation could involve a CAMP-dependent mechanism. Previous studies have reported that CAMP affects AChR desensitization (Huganir et al., 1986), synthesis(Betzand Changeux, 1979; Blosser and Appel, 1980; Fontaine et al., 1987), and subunit assembly (Ross et al., 1987). The present study reports that CAMP can affect AChR degradation, distinguishing between Rs and R, AChRs. Contrary to the situation with the RS AChRs, neither reinnervation (Shyng and Salpeter, 1990) nor the elevation of CAMP levels (present study) can decrease the degradation rate of the R, AChRs to the values seen at adult innervated neuromuscular junctions. The maximal tT12value of the R, AChR following DBcAMP treatment was less than 2 days. W e suggest that the development of metabolically stable AChRs at a neuro muscular junction after innervation cannot involve the stabilization of R, AChRs. The development of a slowly degrading, stable AChR population at the neuromuscular junction probably involves both a down regulation of the R, AChRs and their replacement by the Rs AChRs, as well as the stabilization of the R5 AChRs once they are in the plasma membrane. Electrical stimulation of denervated muscles causes stabilization of AChRs at the endplate band (Fumagalli et al., 1990; Rotzler and Brenner, 1990), but not those at the extrajunctional bands (Rotzler and Brenner, 1990). Rotzler and Brenner suggested that the different behavior of the junctional and extrajunctional AChRs in their study could reflect some, as yet unidentified, unique property of the neuromuscular junction. If such a unique property exists, it is not sufficient to stabilize the degradation rate of R, AChRs, since even at the endplate band, the R, AChRs did not slow their degradation rate in response to either reinnervation (Shyng and Salpeter, 1990) or CAMP (present study). Our results suggest that there may be an intrinsic difference in the responsiveness of R, and RS AChRs that is independent of localization. The differences in the responsiveness of Rs and R, AChRs to factors regulating degradation seem to be imparted during the synthesis of the AChRs, whether in innervated or denervated muscle. It is tempting to suggest that the Rs AChRs may be the E subunit-containing adult type and the R, AChRs may be the y subunit-containing fetal type of AChR (Cu et al., 1990; Shyng and Salpeter, 1990). One major problem with a suggestion that E and y subunits are involved in regulating AChR degradation properties is the fact that, in rat muscleat least, the decrease in AChRdegradation rate at the neuromuscular junction appears within a few days of synapse formation (Steinbach et al., 1979; Reiness and Weinberg, 1981; Rotzler and Brenner, 1990), whereas the E subunit at the junction is not expressed in significant amounts until several days later (Gu and Hall, 1988; Brenner et al., 1990). The
CAMP Modulates 473
Acetylcholine
Receptor
Degradation
timing of the appearance of the E subunit coincides with that of the high conductance AChR channels (e.g., Sakmann and Brenner, 1978; Fischbach and Schuetze, 1980) known to be associated with the E subunit (Mishina et al., 1986; Witzemann et al., 1987; Brehm and Henderson, 1988; Brenner and Rudin, 1989). Yet many of the characteristics of Rs and R, AChRsareconsistentwiththeirbeingsandysubunitcontaining, respectively. These include the following facts: First, Rs AChRs are synthesized mainly in innervated and R, AChRs in denervated muscle. Second, the small percentage of Rs (40%) in denervated muscle (Shyng and Salpeter, 1990) is consistent with the small percentage of E subunit present long after denervation (Henderson et al., 1987; Witzemann et al., 1987; Gu and Hall, 1988. Third, they subunit is down regulated by muscle activity (Goldman et al., 1988), as are the postdenervation extrajunctional AChRs (see Fambrough, 1979) and, as we suggest, the new R, junctional AChRs. Finally, E subunit-containing AChRs degrade more slowly than y subunit-containing AChRs in transfected COS cells (Gu et al., 1990). We also note, as it is possibly relevant to our present study, that the mammalian E subunit has a potential CAMP-dependent phosphorylation site,whereas they subunit does not (Huganir and Miles, 1989). However, unless the demonstrated difference in developmental time course between the appearanceof the E subunit and the slowing of AChR degradation can be resolved, some other property of the AChR imparted atthetimeof synthesis must be sought to explain the differences in degradation behavior between Rs and R, AChRs. In the search for factors that regulate AChR degradation, both structural changes in theAChR(Gu et al., 1990) and muscle activity (Avila et al., 1989; Brenner and Rudin, 1989; Rotzler and Brenner, 1990; Fumagalli et al., 1990) have been implicated. The present study shows that CAMP may be involved in one aspect of this regulation process, i.e., the modulation of the Rs AChRs at denervated neuromuscular junctions. Experimental
Procedures
Denervation of Diaphragm Muscle and labeling of AChRs Adult female mice (CD-I, Charles River Laboratories, Wilmington, MA) weighing 25-35 g were used. Under ether anesthesia, the left phrenic nerve was cut after being externalized with a glass hook through an incision at the level of the second rib. The AChRs were labeled by intrathoracic or intraperitoneal injection of 150 ul of 0.5 uM Y-a-bungarotoxin (a-BgTx), which can label ~60% of AChRs, as judged by a comparison with muscles bathed to saturation in 1251-a-BgTx in vitro (Wetzel and Salpeter, 1991). For studies on the degradation of original AChRs, the receptors were labeled at the time of denervation and muscles were removed 6 days later for culturing. For studies of new AChRs, synthesized after denervation, the receptors were labeled 17 days after denervation (at a time when >VO% of the original junctional AChRs had degraded and were replaced by new AChRs) and muscles were removed 1 day later for culturing. Culture of Diaphragm Animals were sacrificed by cervical translocation under ether anesthesia and perfused transcardially with oxygenated Kreb’s solution (4OC). The diaphragms were quickly dissected and im-
mersed in 4OC oxygenated Kreb’s solution. The denervated left diaphragm was separated from the right diaphragm, which we used as a control for calibrating AChR labeling efficiency (see below). Left diaphragms were then treated as previously described by Wetzel and Salpeter (1991). Briefly, they were pinned onto organ culture dishes and placed in a rocking incubator chamber in a medium consisting of Ml99 (Flow Labs, McLean, VA) plus 5% fetal calf serum and a number of nutritional supplements (given for optimal conditions [Wetzel and Salpeter, IVVI]). These included 2 m M B-hydroxybutyricacid (Sigma), L-glutamine (1.6 mM), ascorbic acid (IO ug/ml), 5-a-dihydrotestosterone (1 nM), pyruvate (0.9 mM), fructose (2.3 mM), and thiamine (35 PM). These optimal conditions were judged by the ability of the muscle to have normal fine structure, continued fibrillation, and an accelerated AChR degradation rate for more than 2 weeks in organ culture. The culture medium plus various pharmacological agents (see below) was replaced daily. Chemical Treatments The various pharmacological agents included forskolin and l,Vdideoxyforskolin from Calbiochem (San Diego, CA), CAMP analogsand 3-isobutyl-I-methylxanthinefrom Sigma(St. Louis,MO). Degradation of AChRs on Diaphragm Muscle in Organ Culture The degradation rate of junctional AChRs on the organ cultured muscles (and for the AChRs on myotubes in cell culture described below) was determined by the rate of loss of radioactivity from the endplate band (for justification, see reviews by Fambrough, 1979; Salpeterand Loring, 1985). Sincethe muscles were labeled by whole animal injection and the AChRs were not fully saturated, the label on the denervated muscle to be cultured was corrected for labeling efficiency by the relative label on the innervated right diaphragm determined at the time of removal from the animal. At different times during the culture period, muscles were fixed with 4% paraformaldehyde (in 0.05 M phosphate buffer [pH 7.41) for 2 hr and rinsed 3 times (10 min each) with 0.1 M phosphate buffer (pH 7.4). In each experiment the first control data point was taken 1 day after culturing began. The muscles were dissected into non-endplate bands and an endplate band(judged bytransilluminationorstainingforacetylcholinesterase). Both the endplate and the non-endplate bands were weighed, and radioactivity was counted using a Beckman gamma counter. The endplatespecific label was determined by subtracting the non-endplate band label from the endplate label on a per weight basis. Extrajunctional label was taken as the total muscle label minus the endplate-specific label. In each case the control data were first fitted by linear regression assuming a single exponential. The extrapolated value at day 0 was then set to 100% and used to normalize both the control and the experimental data on the assumption that muscles from both groups had the same label at the beginning of the culture period. Thedatapointsondayoin thefiguresoforgancultured muscles is thus an extrapolated and not an experimental value. Differences in degradation rates were analyzed statistically by the F-test as previously described (Shyng and Salpeter, 1990). Determining the Degradation Rate of Fetal AChRs on Rat Myotubes Primary cultures of rat skeletal muscle were prepared as described previously (Salpeter et al., 1982). Cells were plated on 24 multiwell plates at 2.5 x 104cell perwell in Dulbecco’s modified Eagle’s medium (DMEM; CIBCO) plus 10% fetal calf serum. Cytosine-I-B-arabinofuranoside (final concentration 10e5 M) was added 72 hr after plating. On day 4 after plating AChRs were labeled with ‘=I-a-BgTx (2 x 1O-8 M) in DMEM-HEPES buffer plus bovine serum albumin (1 mg/ml [pH 7.41) for 30 min at room temperature, followed by a series of washes in DMEM-bovine serum albumin for a total of 30 min. The original “conditioned” growth medium was then returned to the celLToobtain nonspecific binding, cells were incubated with nonradioactive a-BgTx (10-r M) for 15 min at room temperature, washed thoroughly, and then labeled with Y-a-BgTx (30 min), followed by thorough washing as described above. Chemical treatments began imme-
Netltoll 474
diately after AChR labeling. Culture medium and various chemicals were replaced daily. At different times after labeling, the amount of label remaining on the cells was determined bywashing off the medium, dissolving the cells in 2 N NaOH overnight, absorbing the NaOH solution on cotton plugs, and counting these in a gamma counter as described by Knaack and Podleski (1985). Half-life values were obtained by linear regression. Each curve was normalized to 100% for its own experimental value on day 0 (or day 4 in culture). For ail the groups in the present study we report the degradation half-life as the observed value (tr/& obtained from the loss of radioactivity. This can be corrected for a-BgTx unbinding (t,n..r, ~56 days; Cohen et al., 1990) and radioactive decay (trod, for lrz5 = 60 days) to give the true degradation rate (t,nd,,,) using the following equation: (fl/*degr)-’ = (tmd’
- (fli*““b)r’ - (tl/2dec)-’
Acknowledgments We thank Daniel Wetzel, Dorothy Bell, and Iris Greenberg for assistance with these experiments, Tom Podleski and Ron Harris-Warrick for helpful discussion, and Deborah Moslehi for preparing the manuscript. This work was supported by grants from the National Institutes of Health (NS09315) and from the Cornell University Biotechnology Program sponsored by the N.Y. State Science and Technology Foundation, by a consortium of industries, and by the United States Army Research Office and NSF. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisemem’in accordance with 18 USC Section 1734 solely to indicate this fact. Received
October
24, 1990; revised
January
8, 1991.
Fischbach, C. D., and Schuetze, S. M. (1980). A post-natal decrease in acetylcholine channel open time at rat end-plates. J. Physiol. 303, 125-137. Fontaine, B., Klarsfeld, A., and Changeux, 1. P. (1987). Calcitonin gene-related peptide and muscle activity regulate acetylcholine receptor a-subunit mRNA levels by distinct intracellular pathways. J. Cell Biol. 705, 1337-1342. Fumagalli, G., Balbi, S., Cangiano,A,and LBmo,T. (1990). Regulation of turnover and number of acetylcholine receptors at neuromuscular Junctions. Neuron 4, 563-569. Goldman, D., Brenner, H. R., and Heinemann, S. (1988). Acetylcholine receptor a-, B-, y-, and &subunit mRNA levels are regulated by muscle activity. Neuron 7, 329-333. Gu, Y., and Hall, Z. W. (1988). Immunological evidence for a change in subunits of the acetylcholine receptor in developing and denervated rat muscle. Neuron 7, 117-125. Gu, Y., France, A., Jr., Gardner, P. D., Lansman, J. B., Forsayeth, J. R., and Hall, Z. W. (1990). Properties of embryonic and adult muscle acetylcholine receptors transiently expressed in COS cells. Neuron 5, 147-157. Henderson, L. P., Lechleiter,J.,and Brehm, P. (1987). Singlechannel properties of newly synthesized acetylcholine receptors following denervation of mammalian skeletal muscle. J. Cen. Physiol. 89, 999-1014. Hsie, A. W., Kawashima, K., O’Neill, J. P., and Schroder, C. H. (1975). Possible role of adenosine cyclic 3’:5’-monophosphate phosphodiesterase in the morphological transformation of Chinese hamster ovary cells mediated by N6,0z’dibutyryl adenosine cyclic 3’5’-monophosphate. 1. Biol. Chem. 250, 984-989. Huganir, R. L., and Miles, K. (1989). Protein phosphorylation of nicotinic acetylcholine receptors. Crit. Rev. Biochem. Mol. Biol. 24, 183-215. Huganir, R. L., Delcour, A. H., Creengard, P., and Hess, C. P. (1986). Phosphorylation of the nicotinic acetylcholine receptor regulates its rate of desensitization. Nature 327, 774-776.
Avila, 0. L., Drachman, D. B., and Pestronk, A. (1989). Neurotransmission regulates stability of acetylcholine receptors at the neuromuscular junction. J. Neurosci. 9, 2902-2906. Beavo,J.A., Rogers, N. L., Crofford, 0. B., Hardman, J. C., Sutherland, E. W., and Newman, E. V. (1970). Effects of xanthine derivatives on lipolysis and on adenosine 3’-5’-monophosphate phosphodiesterase activity. Mol. Pharmacol. 6, 597-603. Betz, H., and Changeux, J. P. (1979). Regulation of muscle acetylcholine receptor synthesis in vitro by derivatives of cyclic nucleotides. Nature 278, 749-752. Bevan, S., and Steinbach, J. H. (1983). Denervation increases the degradation rate of acetylcholine receptors and end-plates in vivo and in vitro. J. Physiol. 336, 159-177. Blosser, J. C., and Appel, S. H. (1980). Regulation of acetylcholine receptor by cyclic AMP. J. Biol. Chem. 253, 3088-3093. Brehm, P., and Henderson, receptor channel function cle. Dev. Biol. 729, I-11.
L. (1988). Regulation during development
of acetylcholine of skeletal mus-
Brenner, H. R., and Rudin, W. (1989). On the effect of muscle activity on the end-plate membrane in denervated mouse muscle. J. Physiol. 470, 501-512. Brenner, H. R., Witzemann, V., and Sakmann, B. (1990). Imprinting of acetylcholine receptor messenger RNAaccumulation in mammalian neuromuscular synapses. Nature 344, 544-547. Brett, R. S., Younkin, 5. C., Konieczkowski, M., and Slugg, R. M. (1982). Accelerated degradation of junctional acetylcholine receptor-a-bungarotoxin complexes in denervated rat diaphragm. Brain Res. 233, 133-142. Cohen, M. W., Frair, P. F., Cantin, C., and Hebert, G. (1990). Developmental changes in the half-life of acetylcholine recep tors in the myotomal muscle of Xenopus laevis. J. Physiol. 426, 281-296. Fambrough, D. M. (1979). Control of acetylcholine skeletal muscle. Physiol. Rev. 59, 165-227.
receptors
in
Kaukel, E., and Hilz, H. (1972). Permeation ofdibutyryl CAMP into HeLa cells and its conversion to monobutyryl CAMP. Biochem. Biophys. Res. Commun. 46, 1011-1018. Knaack, D., and Podleski, T. R. (1985). Ascorbic acid mediates acetylcholine receptor increase induced by brain extract in L5 myogenic cells. Proc. Natl. Acad. Sci. USA 82, 575-579. Laurenza, A., Sutkowski, E. M., and Seamon, K. B. (1989). Forskolin: a specific stimulator of adenylyl cyclase or a diterpene with multiple sites of action? Trends Pharmacol. Sci. 70,442-447. Levitt, T. A., and Salpeter, M. M. (1981). Denervated have a dual population of junctional acetylcholine Nature 297, 239-241.
endplates receptors.
Mishina, M., Takai, T., Imoto, K., Noda, M., Takahashi, T., Numa, S., Methfessel, C., and Sakmann, B. (1986). Molecular distinction between fetal and adult forms of muscle acetylcholine receptor. Nature 327, 406-441. Montague, W., and Cook, 1. R. (1971). The role of adenosine 3’: 5’cyclic monophosphate in the regulation of insulin release by isolated rat islets of Langerhans. Biochem. J. 722, 115-120. Reiness, C. G., and Weinberg, C. B. (1981). Metabolic stabilization of acetylcholine receptors at newly formed neuromuscular junctions in rat. Dev. Biol. 84, 247-254. Ross, A. F., Rapuano, M., Schmidt, J. H., and Proves, J. M. (1987). Phosphorylation and assembly of nicotinic acetylcholine receptor subunits in cultured chick muscle cells. J. Biol. Chem. 262, 14640-14647. Rotzler, S., and Brenner, H. R. (1990). Metabolic stabilization acetylcholine receptors in vertebrate neuromuscular junction muscle activity. J. Cell Biol. 777, 655-661. Sakmann, B., and Brenner, H. R. (1978). Change nel gatingduring neuromusculardevelopment. 402. Salpeter,
M. M., and Loring,
of by
in synaptic chanNature276,401-
R. H. (1985). Nicotinic
acetylcholine
CAMP Modulates 47;
Acetylcholine
Receptor
Degradation
receptors in vertebrate muscle: properties, ral control. Prog. Neurobiol. 25, 297-325.
distribution
and neu-
Salpeter, M. M., Spanton, S., Holley, K., and Podleski, T. R. (1982). Brain extract causes acetylcholine receptor redistribution which mimics some early events at developing neuromuscular junctions. J. Cell Biol. 93, 417-425. Salpeter, M. M., Cooper, D. L., and Levitt-Cilmour, T. (1986). Degradation rates of acetylcholine receptors can be modified after they are inserted into the postjunctional plasma membrane o! the vertebrate neuromuscular junction. J. Cell Biol. 703,13991403. Soamon, K. B., and Daly, J. W. (1981). Forskolin: a unique diterpcne activator of CAMP-generating systems. J. Cycl. Nucl. Res. 7, 2(11-224. Shyng, S.-L., and Salpeter, M. M. (1990). Effect of reinnervation ori the degradation rate of junctional acetylcholine receptors synthesized in denervated skeletal muscles. J. Neurosci. 70, 3905-3915. Stanley, E. F., and Drachman, D. B. (1981). Denervation accelerates the degradation of junctional acetylcholine receptors. Exp. Neural. 73, 390-396. Steinbach, J. H., Merlie, J., Heinemann, S., and Bloch, R. (1979). Degradation of junctional and extrajunctional acetylcholine receptors by developing rat skeletal muscle. Proc. Natl. Acad. Sci. C SA 76, 3547-3551. Wetzel, D. M., and Salpeter,M. M. (1991). Fibrillation and accelerated AChR degradation in long term muscle organ culture. Muscle Nerve, in press. VVitzemann,V., Barg, B., Nishikawa, Y., Sakmann, B., and Numa, S. (1987). Differential regulation of muscleacetylcholine receptor y- and e-subunit mRNAs. FEBS Lett. 223, 104-112.
