Proc. Natl. Acad. Sci. USAVol. 84, pp. 6591-6595, September 1987Neurobiology

Regulation of nicotinic acetylcholine receptor phosphorylation inrat myotubes by forskolin and cAMP

(protein kinases/ion channel/desensitization)

KATHRYN MILES*t, DONNA T. ANTHONYt, LEE L. RUBINO, PAUL GREENGARD*, AND RICHARD L. HUGANIR**Laboratory of Molecular and Cellular Neuroscience and tLaboratory of Neurobiology, The Rockefeller University, 1230 York Avenue,New York, NY 10021-6399

Contributed by Paul Greengard, May 28, 1987

ABSTRACT The nicotinic acetylcholine receptor (Ac-ChoR) from rat myotubes prelabeled in culture with[32Plorthophosphate was isolated by acetylcholine affinitychromatography followed by immunoaffinity chromatogra-phy. Under basal conditions, the nicotinic AcChoR was shownto be phosphorylated in situ on the 1 and 6 subunits. Regulationof AcChoR phosphorylation by cAMP-dependent proteinkinase was explored by the addition of forskolin or cAMPanalogues to prelabeled cell cultures. Forskolin, an activator ofadenylate cyclase, stimulated the phosphorylation of the 6subunit 20-fold over basal phosphorylation and induced phos-phorylation of the a subunit. The effect of forskolin was dosedependent with a half-maximal response at 8 ,LM in thepresence of 35 ,uM Ro 20-1724, a phosphodiesterase inhibitor.Stimulation of 6 subunit phosphorylation was almost maximalwithin 5 min, whereas stimulation of a subunit phosphoryl-ation was not maximal until 45 min after forskolin treatment.Stimulation of AcChoR phosphorylation by 8-benzylthioade-nosine 3',5'-cyclic monophosphate was identical to that ob-tained by forskolin. Two-dimensional thermolytic phosphopep-tide maps of the 6 subunit revealed a single major phospho-peptide. These results correlate closely with the observedeffects of forskolin on AcChoR desensitization in muscle andsuggest that cAMP-dependent phosphorylation of the 6 subunitincreases the rate of AcChoR desensitization in rat myotubes.

Signal transduction at the postsynaptic membrane of theneuromuscular junction is mediated through the nicotinicacetylcholine receptor (AcChoR). This pentameric complex,composed of four types of protein subunits (a2f3y8), is botha neurotransmitter receptor and a membrane ion channel.The neurotransmitter acetylcholine binds to the AcChoRopening the ion channel, and the resulting influx of cationsdepolarizes the muscle cell (1). The neuromuscular junctionrepresents an excellent model system for investigating themolecular mechanisms involved in modulating intercellularsignal transduction processes.

Biochemical characterization indicates that the Torpedoelectric organ AcChoR is a phosphoprotein (2, 3). Theregulation of AcChoR phosphorylation in vitro has beenshown to involve at least three different protein kinasespresent in the postsynaptic membrane of the Torpedo electricorgan. The cAMP-dependent protein kinase phosphorylatesthe 'y and 8 subunits (4), protein kinase C phosphorylates thea and 8 subunits (5), and an endogenous protein tyrosinekinase phosphorylates the 83, y, and 8 subunits (6).

Protein phosphorylation is a major posttranslational mech-anism involved in regulating the function of proteins includ-ing ion channels and receptors (7-9). The effect of phospho-rylation on AcChoR function was directly examined in vitroby analyzing the ion-transport properties of purified Torpedo

AcChoR that was reconstituted into phospholipid vesicles.Phosphorylation of the purified AcChoR by cAMP-depen-dent protein kinase in vitro was shown to accelerate desen-sitization, the process by which the receptor is progressivelyinactivated in the presence of acetylcholine (10). In addition,electrophysiological studies have suggested that agents thatincrease the intracellular levels ofcAMP increase the rate ofAcChoR desensitization at the neuromuscular junction (11,12). Exposure of rat soleus muscle to forskolin, a potentactivator of adenylate cyclase, was shown to increase the rateof AcChoR desensitization. This effect of forskolin on therate of desensitization occurred over the same concentrationrange as that for activation of adenylate cyclase and waspotentiated by a phosphodiesterase inhibitor. The rate ofdesensitization was also increased by cAMP analogues (13).However, it is not known whether the AcChoR is phospho-rylated in muscle and whether changes in the state ofAcChoR phosphorylation could be responsible for acceler-ating AcChoR desensitization.To establish that the AcChoR in rat muscle is a phospho-

protein, we isolated the AcChoR following phosphate-label-ing of rat myotubes in culture. In this paper we demonstratethat the AcChoR is phosphorylated in situ in rat myotubesand that the state of phosphorylation of the a and 8 subunitsof the AcChoR increases after treatment of the myotubeswith forskolin and cAMP analogues. These findings suggestthat the AcChoR is phosphorylated by the cAMP-dependentprotein kinase in situ and that this phosphorylation increasesthe rate of AcChoR desensitization at the neuromuscularjunction.

MATERIALS AND METHODSMaterials. Pregnant rats (14-18 days of gestation) were

obtained from Taconic Farms (Germantown, NY). Tissueculture media and supplies were obtained from GIBCO,except for phosphate- and methionine-free media, whichwere purchased from Flow Laboratories. Radioisotopes andEn3Hance were obtained from New England Nuclear. Theaffinity resin was synthesized by incubating bromoacetyl-choline with reduced Affi-Gel 401 (Bio-Rad) (14). Hybridomacells secreting monoclonal antibody 88-B, a monoclonalantibody directed against the Torpedo AcChoR, were agenerous gift from Stanley Froehner (Dartmouth University,Hanover, NH) and rabbit anti-mouse IgG was purchasedfrom DAKO (Santa Barbara, CA). Reverse-phase HPLC wasconducted using a Waters ,Bondapac C18 column. Forskolinand 1,9-dideoxyforskolin were obtained from Calbiochemand 8-benzylthioadenosine 3',5'-cyclic monophosphate (8-benzylthio-cAMP) was obtained from ICN. Protein A-Seph-arose was purchased from Pharmacia. ortho-Na2VO4 waspurchased from Fisher. Trasylol was obtained from FBA

Abbreviation: AcChoR, acetylcholine receptor.tTo whom reprint requests should be addressed.

6591

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

Dec

embe

r 9,

202

0

Proc. Natl. Acad. Sci. USA 84 (1987)

Pharmaceuticals. Leupeptin, pepstatin, and antipain werepurchased from Chemicon (El Segundo, CA). Ro 20-1724 wasa gift from Hoffmann-LaRoche. All other chemicals wereobtained from Sigma.Muscle Cell Culture and Metabolic Labeling. Primary cul-

tures were established from the hind limb muscle of 20- to21-day rat embryos. Myoblasts were enzymatically dissoci-ated with 0.05% trypsin and cultured in Dulbecco's modifiedEagle's medium supplemented with 10% (vol/vol) horseserum, 2% (vol/vol) chicken embryo extract, and 33 mMglucose. Experiments were performed on 7- to 12-day-oldmyotube cultures. For metabolic labeling with phosphate ormethionine, cultures were washed either with phosphate-freeor methionine-free Eagle's minimal essential medium. Cellswere then incubated either with [32P]orthophosphate (0.5mCi; 1 Ci = 37 GBq) in 1 ml of phosphate-free medium per35-mm dish for 3.5 hr or [35S]methionine (0.5 mCi) in 1 ml ofmethionine-free medium per 35-mm dish for 3.5 hr or over-night as indicated.

Forskolin Treatment of Cultured Myotubes. Cultured cellswere treated with forskolin for the indicated times andconcentrations along with 35 ktM of the phosphodiesteraseinhibitor Ro 20-1724. Both compounds were dissolved in 95%(vol/vol) ethanol and were added directly to the culturemedium containing the radioactive label such that the cellswere not exposed to an ethanol concentration exceeding 1%.Ethanol alone (1%) was found to have no effect on phospho-rylation.

Acetylcholine Receptor Purification. Cultured myotubeswere rinsed with isotonic phosphate-buffered saline, and theAcChoR was isolated from each culture dish by solubilizingin 500 ,ul of a "lysis solution" containing 1% Triton X-100, 50mM Tris (pH 7.4), 150mM NaCI, 1 mM EDTA, 1 mM EGTA,10 mM N-ethylmaleimide, 100 ,uM NaVO3, 100 mM NaF, 10mM sodium pyrophosphate, leupeptin (10 ,ug/ml), pepstatin(10 ,g/ml), antipain (10 ,ug/ml), and Trasylol (10 units/ml).Insoluble material from each culture extract was removed bycentrifugation for 10 min at 10,000 x g, and the supernatantscontaining AcChoR were passed through individual 200-,lacetylcholine affinity columns. Each flow through was reap-plied to affinity resin, and the columns were rinsed twice with500 ,ul of the lysis solution. The specificity of AcChoRbinding to the columns was demonstrated by preincubatingparallel culture extracts with 25 mM carbamylcholine. TheAcChoR complex was eluted from the affinity resin with 500,ul of lysis solution containing 25 mM carbamylcholine. Theeluates from each acetylcholine affinity column were thenimmediately passed through individual 100-plA immunoaf-finity columns consisting of monoclonal antibody 88-B linkedto protein A-Sepharose by a rabbit anti-mouse IgG. Thecolumns were washed with 500 ,ul of lysis solution, and theAcChoR subunits were eluted from the immunoaffinitycolumns with sample buffer containing 125 mM Tris (pH 6.8),2% (wt/vol) NaDodSO4, 10% (vol/vol) glycerol, and 5%(vol/vol) 13-mercaptoethanol. The yield of AcChoR fromparallel cultures was consistently the same. The eluates fromimmunoaffinity columns were subjected to electrophoresison 9o NaDodSO4/polyacrylamide gels as described byLaemmli (15). Gels were dried and autoradiographed with theexception that those containing 35S were treated with En-3Hance prior to drying.Two-Dimensional Phosphopeptide Maps. Two-dimensional

thermolytic phosphopeptide mapping of proteins containedin gel pieces was performed as described (6).

Determination of [32PJATP Specific Activity. Cell cultureslabeled in situ with [32P]orthophosphate were washed inisotonic phosphate-buffered saline and then treated individ-ually with 1 ml of 0.4 M perchloric acid. The insolublematerial from each culture was removed by centrifugation for10 min at 10,000 x g, and the resulting supernatants were

neutralized with 0.1 ml of 4 M KOH. The KC104 precipitatesthat formed were removed by centrifugation for 10 min at10,000 x g. Nucleotides, extracted into the resulting super-natants, were purified by reverse-phase HPLC employingisocratic elution with 1 mM potassium phosphate buffer (pH6.0), pumped at 1 ml/min, and monitored by UV detection at254 nm (16). ATP from 100-pl aliquots ofeach cell extract wasidentified by retention time and quantitated by peak heightcompared to known ATP standards. Fractions eluting fromthe column were measured for radioactivity in a liquidscintillation counter.

RESULTSIn Situ [35S]Methionine Labeling and Isolation of Myotube

AcChoRi AcChoR was rapidly isolated, following solubiliza-tion of myotubes from cultures, by a two-column chroma-tography procedure consisting of acetylcholine affinity chro-matography followed by immunoafflinity chromatographyusing monoclonal antibody 88-B. Monoclonal antibody 88-B,raised against Torpedo AcChoR, has been shown to recog-nize specifically the y and ( subunits of the Torpedo AcChoRand to crossreact with cytoplasmic domains of mammalianAcChoR (17). With [35S]methionine labeling for 3.5 hr, threeproteins of molecular masses 42, 50, and 62 kDa were isolated(Fig. 1, lane 1). Labeling cultures with [35S]methionine for 12hr revealed an additional protein of 66 kDa (Fig. 1, lane 2).Confirmation that these proteins are subunits of the ratAcChoR was obtained by preincubating the cell homogenateswith 25 mM carbamylcholine before affinity chromatographythus preventing the AcChoR from binding to the acetylcho-line affinity column (Fig. 1 , lane 3). Alternatively, pretreatingthe immunoaffinity column with an excess of purified Tor-pedo AcChoR also resulted in the specific exclusion of these

1 2 3

66kDacr .*62kDa--m- i_50kDaD-_-- W

42kDa-- __

4 5 6 7

..- 6 -p-a

(5-0 -N

--* ol

Ro20-1724 - + + +

FORSKOLIN - - + +

FIG. 1. AcChoR isolated from myotube cultures prelabeled with[35Slmethionine or [32P]orthophosphate and regulation of AcChoRphosphorylation by forskolin and Ro 20-1724. Myotubes wereincubated with 0.5 MCi of (355]methionine for 3.5 hr (lane 1) or for12 hr (lanes 2 and 3) or with 0.5 mCi of [32P]orthophosphate for 3.5hr (lanes 4-7). In the presence of radioactive label, ntyotubes weretreated for 45 min with 20 ,uM forskolin and/or 35 ,uM Ro 20-1724 asindicated. AcChoR was solubilized, isolated by acetylcholine affinitychromatography followed by immunoaffinity chromatography, andanalyzed by electrophoresis and autoradiography. Cell homogenateswere preincubated with 25 mM carbamylcholine prior to acetylcho-line affinity chromatography to selectively inhibit AcChoR binding tothe column (lanes 3 and 7).

6592 Neurobiology: Miles et al.

am* wo

a&

don 4wm A6

Dow

nloa

ded

by g

uest

on

Dec

embe

r 9,

202

0

Proc. Natl. Acad. Sci. USA 84 (1987) 6593

proteins (data not shown). The 42-, 50-, and 62-kDa proteinsmost likely correspond to the a, (3, and 6 subunits of theAcChoR. AcChoR isolated from a clonal mouse cell line(BC3H1) has been reported to have subunits of42, 46, 48, and60 kDa, which have been suggested to be the a, X3, 'y, and 6subunits, respectively (18). The absence of the y subunit inour preparations may reflect its known sensitivity to prote-olysis (18, 19). Prolonged incubation with [35S]methioninerevealed trace labeling of a 66-kDa protein that appears to bedue to a delayed posttranslational processing of the 6 (62-kDa) subunit.

Phosphorylation of Myotube AcChoR in Situ. To measurethe extent of AcChoR phosphorylation in situ, muscle cellcultures were incubated with [32P]orthophosphate. 32p incor-poration into total cell protein reached a steady state after 3.5hr of metabolic labeling as measured by the 32p content oftrichloroacetic acid precipitated proteins. Rapid isolation ofthe AcChoR following labeling of cell cultures with [3IP]or-thophosphate demonstrated that the AcChoR is a phospho-protein in situ. Proteins of 50, 62, and 66 kDa were phos-phorylated (Fig. 1, lane 4). Preincubation of cell homogenatesfrom parallel cultures with 25 mM carbamylcholine prevent-ed the isolation of these phosphorylated proteins (Fig. 1, lane7). These 32P-labeled proteins comigrated with proteinsisolated after [35S]methionine labeling and were specificallycompeted by carbamylcholine during affinity adsorptionstrongly suggesting that the 50- and 62-kDa proteins corre-spond to the 8 and 6 subunits of the AcChoR. The specificincorporation of radioactive phosphate into AcChoR sub-units was also found to plateau at 3 hr. To determine whetherthe 66-kDa phosphoprotein was related to the 6 subunit,two-dimensional thermolytic phosphopeptide mapping wasperformed. Phosphopeptides generated by thermolysin di-gestion of the 62- and 66-kDa proteins have identical mobilityby two-dimensional thin-layer chromatography suggestingthat these proteins are variants of the same subunit (Fig. 2).The evidence that the 66-kDa protein is only apparent afterprolonged [35S]methionine labeling suggests that the 66-kDaprotein is related to the 62-kDa protein by posttranslationalprocessing rather than proteolytic digestion. We have, there-fore, designated the 66-kDa protein as the 6' subunit.

Regulation of Phosphorylation of Myotube AcChoR in Situ.The regulation of AcChoR phosphorylation by cAMP wasexplored by the addition of forskolin and the phosphodiester-ase inhibitor Ro 20-1724 or by addition of cAMP analogues.Addition of the phosphodiesterase inhibitor Ro 20-1724 for upto 1 hr did not influence the basal phosphorylation level of theAcChoR (Fig. 1, lane 5). However, treating myotube culturesfor 45 min with 20 ,iM forskolin in the presence of 35 ,uM Ro20-1724 increased the phosphorylation of the a, 6, and 6'subunits (Fig. 1, lane 6). A comparison of phosphate andmethionine labeling indicated that while phosphate incorpo-ration into the AcChoR increased in the presence offorskolin, methionine labeling of AcChoR subunits did notincrease, suggesting that forskolin increased the phosphoryl-ation state of the AcChoR rather than having an effect on theAcChoR levels under these circumstances. In addition, thisobservation is consistent with the assumption that our puri-fication procedure isolated both the phosphorylated andnonphosphorylated forms of the AcChoR. Two-dimensionalthermolytic phosphopeptide mapping indicated that phos-phorylation of the 6 subunit occurred on a single majorphosphopeptide (Fig. 2). Forskolin treatment during phos-phate labeling did not induce additional phosphopeptides(data not shown); therefore, it is likely that this majorphosphopeptide includes the phosphorylation site for thecAMP-dependent protein kinase. Addition of 20 AM 1,9-dideoxyforskolin, a derivative of forskolin that does notactivate adenylate cyclase, in the presence of 35 AM Ro

A

0)

0)

E0-Uo

S

+

Electrophoresis

B

-Cco

C)0

E0-CU

+Electrophoresis

FIG. 2. Two-dimensional thermolysin phosphopeptide maps ofthe 62-kDa (A) and 66-kDa (B) AcChoR subunits isolated from ratmuscle cell cultures. Myotube cultures were incubated with 2 mCi of[32P]orthophosphate for 4 hr, and AcChoR was isolated as describedin Fig. 1. Thermolytic phosphopeptide mapping of proteins containedin gel pieces was performed (circle designates origin).

20-1724 had no effect on AcChoR phosphorylation inmyotube cultures (data not shown).To investigate the possibility that forskolin caused an

increase in the specific activity of ATP in cultured myotubes,we purified ATP from prelabeled cultures by HPLC tomeasure its specific activity. Neither the yield nor specificactivity of ATP purified from forskolin-treated culturesdiffered significantly from control cultures (data not shown).The dose dependence of AcChoR phosphorylation in

response to forskolin was studied by treating myotubecultures for 1 hr with forskolin at concentrations ranging from0 to 50 ,uM (Fig. 3). An increase in phosphorylation of the a,8, and 6' subunits was detected that was dependent on thedose of forskolin and saturated at 20 gM forskolin. Quanti-tation of the autoradiogram optical density by scanningdensitometry indicated that forskolin stimulated the phos-phorylation of the 8 subunit 20-fold with a half-maximal effectoccurring at 8 ,M forskolin. The stimulation of phosphoryl-ation of the 8 subunit by forskolin varied among experimentsand ranged from 5- to 20-fold. Although the effect of forskolin

Neurobiology: Miles et al.

Dow

nloa

ded

by g

uest

on

Dec

embe

r 9,

202

0

Proc. Natl. Acad. Sci. USA 84 (1987)

A

o- .0% Aw aa-0-)

U--(

Forskolin (MM) 0 1 2 5 10 20 50

FIG. 3. Dose dependence of the effect of forskolin treatment ofrat myotubes on AcChoR phosphorylation. Myotube cultures wereincubated for 3.5 hr with 0.5 mCi of [32P]orthophosphate. Cells werethen treated with 35 ,uM Ro 20-1724 and the indicated concentrationsof forskolin for 1 hr. AcChoR was solubilized, isolated, and analyzedas described in Fig. 1.

on phosphorylation of the if subunit was highly variable, adecrease in phosphorylation was most consistently observed.The time course of the increase in AcChoR phosphoryl-

ation using 20 tLM forskolin showed that the rate of phos-phorylation of the a subunit was slower than that of the 6 and6' subunits (Fig. 4A). Phosphorylation ofthe 6 and 8' subunitsreached half-maximal response within 5 min, whereas phos-phorylation of the a subunit required 30 min to achievehalf-maximal phosphorylation. The time course of 8 plus 8'subunit phosphorylation after exposure to 20 ,uM forskolinwas quantitated by scanning densitometry (Fig. 4B).To test whether cAMP analogues had an effect similar to

that of forskolin on AcChoR phosphorylation, culturedmyotubes were treated with the cAMP analogue 8-benzyl-thio-cAMP (100 ,uM) for 60 min. 8-Benzylthio-cAMP stimu-lated the phosphorylation of the a, 6, and 8' subunits of theAcChoR, an effect similar to that observed with forskolintreatment (Fig. 5).

DISCUSSIONThe results presented in this paper demonstrate that thenicotinic AcChoR is phosphorylated in intact muscle cellsand that agents that increase intracellular cAMP levelsstimulate phosphorylation of the AcChoR. Treatment ofmuscle cells with forskolin or 8-benzylthio-cAMP increasedphosphorylation of the a, 8, and 8' subunits of the AcChoR.The concentrations of forskolin that were effective in stim-ulating phosphorylation of the AcChoR are in the dose-response range known to activate adenylate cyclase in othercell systems (20). These results suggest that the effect offorskolin on phosphorylation of the AcChoR is most likelymediated through activation of cAMP-dependent proteinkinase. The rapid time course of phosphorylation ofthe 8 and8' subunits ofthe AcChoR following treatment with forskolinis consistent with direct phosphorylation of the AcChoR bycAMP-dependent protein kinase. In contrast, phosphoryl-ation of the a subunit follows a much slower time course aftera considerable lag time and may reflect an indirect effect ofcAMP-dependent protein kinase. It is possible that anotherprotein kinase whose activity or synthesis is regulated bycAMP-dependent protein kinase phosphorylates the a sub-unit of the AcChoR.

Time (min) 0 1 2 5 10 20 40

B

z

0

-J

cc0I

3:

a-Cl)0mC--J

0

20

MINUTES

FIG. 4. Time course of the effect of forskolin treatment of ratmyotubes on AcChoR phosphorylation. (A) Myotube cultures wereincubated for 3.5 hr with 0.5 mCi of [32P]orthophosphate. Cells werethen treated with 35 ALM Ro 20-1724 and 20 ALM forskolin for theindicated times. AcChoR was subsequently solubilized, isolated, andanalyzed as described in Fig. 1. (B) Quantitation of AcChoR 8 plus8' subunit phosphorylation by soft laser densitometry scanning ofautoradiogram presented in A.

cAMP-dependent protein kinase has been shown tophosphorylate serine-353 of the -y subunit and serine-361 of

1 2

_

8BT-cAMP - +

-<- 3 FIG. 5. Regulation of AcChoRphosphorylation in rat myotubes

( by the cAMP analogue 8-benzyl-thio-cAMP (8BT-cAMP). Myo-tube cultures were incubated for3.5 hr with 0.5 mCi of [32P]ortho-phosphate. Cells were then treat-ed with 100 ,uM 8-benzylthio-cAMP for 30 min. AcChoR wassubsequently solubilized, isolat-ed, and analyzed as described inFig. 1.

6594 Neurobiology: Miles et al.

,. J*' o ,-

Dow

nloa

ded

by g

uest

on

Dec

embe

r 9,

202

0

Proc. Natl. Acad. Sci. USA 84 (1987) 6595

the 8 subunit of the Torpedo AcChoR (G. Yee and R.L.H.,unpublished results). The phosphorylation site for cAMP-dependent protein kinase has been conserved in the 8 subunitbut not in the y subunit of the mammalian AcChoR (22, 23).Interestingly, the E subunit, which has been reported tosubstitute for the y subunit in the mature form of the calfAcChoR, does contain the phosphorylation site for cAMP-dependent protein kinase (24). The experiments presentedhere confirm that the 8 subunit of the mammalian AcChoR isphosphorylated but no conclusions may be drawn concerningphosphorylation of the y subunit. It may be that the y subunitpresent in primary muscle cell cultures is not phosphorylatedby the cAMP-dependent protein kinase. Although the pri-mary sequence for the 8 subunit from rat is not available, itis interesting to note that the thermolysin phosphopeptide ofthe 8 subunit from rat has a mobility very similar to that of theTorpedo AcChoR 8 subunit phosphorylated by cAMP-de-pendent protein kinase. It is likely, therefore, that cAMP-dependent protein kinase regulates the phosphorylation of asite on the muscle AcChoR 8 subunit homologous to the sitedetermined on the Torpedo AcChoR 8 subunit.

Forskolin has been shown to increase the rate of AcChoRdesensitization in rat muscle cell cultures (13). Forskolinaccelerated AcChoR desensitization with a half-maximaleffect that occurred at 8 gM and was complete within 5 minafter forskolin treatment. This corresponds most closely tothe time course and dose dependency of 8 and 8' subunitphosphorylation observed after stimulating muscle cell cul-tures with forskolin. In contrast, since the increase inphosphorylation of the a subunit undergoes a longer timecourse, the stimulation of a-subunit phosphorylation byforskolin appears not to be correlated directly to AcChoRdesensitization rates. Physiological effects of forskolin onAcChoR receptor function that display a long time coursehave been observed. Miniature end-plate currents recordedextracellularly from rat soleus end plates exposed to forskolinfor at least 30 min were found to decay faster than nontreatedcontrols suggesting a shorter mean channel open time for theAcChoR in muscle treated with forskolin (11). However, ithas also been reported that single-channel currents recordedfrom chicken myotubes in the presence of forskolin for 30 minshowed a lengthening of mean channel open time of theAcChoR (25).The endogenous activator(s) of cAMP-dependent protein

kinase at the neuromuscular junction has not been identified.The neuropeptide calcitonin gene-related peptide has beenshown to increase cAMP levels (26) and to stimulate AcChoRsynthesis (27) in chicken myotubes. If calcitonin gene-relatedpeptide is released at the neuromuscular junction, where ithas been shown to be localized presynaptically in the mouse(28), it is an excellent candidate for modulating AcChoRfunction by stimulating cAMP-dependent phosphorylation ofthe AcChoR.

Protein kinase C may also be involved in regulatingAcChoR function in cultured myotubes. Muscle cells treatedwith phorbol esters, activators of protein kinase C, displayeda decreased sensitivity to acetylcholine and an increased rateof AcChoR desensitization (29). An endogenous proteintyrosine kinase, regulated by an unknown mechanism, mayalso phosphorylate the AcChoR at the neuromuscular junc-tion. In the Torpedo AcChoR the phosphorylation sites forcAMP-dependent protein kinase, protein kinase C, and aprotein tyrosine kinase are contained in one cytoplasmicdomain of the AcChoR subunits; This may reflect a conver-gence of various protein kinase systems toward a single

functional modification, an increased rate of AcChoR desen-sitization. It will be interesting to discern the functional roleas well as the activating signal for each protein kinase systemand to understand how they interact to modify AcChoRfunction.

Note Added in Proof. Phosphorylation of the nicotinic acetylcholinereceptor in mouse BC3HI myocytes has been demonstrated (21).

We thank Angus Nairn and Robert Nichols for a critical reading ofthis manuscript. This work was supported by a grant from TheCouncil for Tobacco Research USA, Inc. (R.L.H.) and a postdoc-toral fellowship from the Muscular Dystrophy Association (K.M.).

1. Stroud, R. M. & Finer-Moore, J. (1985) Annu. Rev. Cell. Biol.1, 317-351.

2. Gordon, A. S., Davis, C. G., Milfay, D. & Diamond, 1. (1977)Nature (London) 267, 539-540.

3. Teichberg, V. I., Sobel, A. & Changeux, J. P. (1977) Nature(London) 267, 540-542.

4. Huganir, R. L. & Greengard, P. (1983) Proc. Natl. Acad. Sci.USA 80, 1130-1134.

5. Huganir, R. L., Albert, K. A. & Greengard, P. (1983) Soc.Neurosci. Abstr. 9, 578.

6. Huganir, R. L., Miles, K. & Greengard, P. (1984) Proc. Natl.Acad. Sci. USA 81, 6968-6972.

7. Levitan, I. B. (1985) J. Membr. Biol. 87, 177-190.8. Sibley, D. R., Benovic, J. L., Caron, M. G. & Lefkowitz,

R. J. (1987) Cell 48, 913-922.9. Browning, M. D., Huganir, R. & Greengard, P. (1985) J.

Neurochem. 45, 11-23.10. Huganir, R. L., Delcour, A. H., Greengard, P. & Hess, G. P.

(1986) Nature (London) 321, 774-776.11. Middleton, P., Jaramillo, F. & Schuetze, S. M. (1986) Proc.

Natl. Acad. Sci. USA 83, 4967-4971.12. Albuquerque, E. X., Deshpande, S. S., Aracava, Y.,

Alkondon, M. & Daly, J. W. (1986) FEBS Lett. 199, 113-120.13. Middleton, P., Rubin, L. & Schuetze, S. (1986) Soc. Neurosci.

Abstr. 12, 148.14. Reynolds, J. & Karlin, A. (1978) Biochemistry 17, 2035-2038.15. Laemmli, U. K. (1970) Nature (London) 227, 680-685.16. Schweinsberg, P. D. & Loo, T. L. (1980) J. Chromatogr. 181,

103-107.17. Froehner, S. C., Douville, K., Klink, S. & Culp, W. J. (1983)

J. Biol. Chem. 258, 7112-7120.18. Merlie, J. P. & Sebanne, R. (1981) J. Biol. Chem. 256,

3605-3608.19. Huganir, R. & Racker, E. (1980) J. Supramol. Struct. 14,

13-19.20. Seaman, K. B. & Daly, J. W. (1986) Adv. Cyclic Nucleotide

Protein Phosphorylation Res. 20, 1-150.21. Smith, M. M., Merlie, J. P. & Lawrence, J. C. (1987) Proc.

Natl. Acad. Sci. USA 84, 6601-6605.22. Boulter, J., Evans, K., Martin, G., Mason, P., Stengelin, S.,

Goldman, D., Heinemann, S. & Patrick, J. (1986) J. Neurosci.Res. 16, 37-49.

23. Lapolla, R. J., Mayne, K. M. & Davidson, N. (1984) Proc.Natl. Acad. Sci. USA 81, 7970-7974.

24. Takai, T., Noda, M., Mishima, M., Shimizu, S., Furutani, Y.,Kayano, T., Ikeda, T., Kubo, T., Takahashi, H., Takahashi,T., Kuno, M. & Numa, S. (1985) Nature (London) 315,761-764.

25. Zani, B. M., Grassi, F., Molinaro, M., Monaco, L. & Eusebi,F. (1986) Biochem. Biophys. Res. Commun. 140, 243-249.

26. Laufer, R. & Changeux, J.-P. (1987) EMBO J. 6, 901-906.27. New, H. V. & Mudge, A. W. (1986) Nature (London) 323,

809-811.28. Takami, K., Kowai, Y., Shiosaka, S., Lee, Y., Girgis, S.,

Hillyard, C. J., Maclntyre, I., Emson, P. C. & Tohyama, M.(1985) Brain Res. 328, 386-389.

29. Eusebi, F., Molinaro, M. & Zani, B. M. (1985) J. Cell Biol.100, 1339-1342

Neurobiology: Miles et al.

Dow

nloa

ded

by g

uest

on

Dec

embe

r 9,

202

0