Identification of the Mouse Muscle 43,000-Dalton Acetylcholine Receptor-associated Protein (RAPsyn) by cDNA Cloning*

The nicotinic acetylcholine receptor and a receptor-associated protein of 43 kDa are the major proteins present in postsynaptic membranes isolated from Torpedo electric organ. Immunochemical analyses indi- cated that a protein sharing antigenic determinants with the receptor-associated protein is also present at receptor clusters of muscle cell lines and postsynaptic membranes of vertebrate neuromuscular junctions. We now provide definitive proof that a homolog of the 43- kDa protein exists in mammals. Complimentary DNA clones encoding the complete protein sequence have been isolated from the mouse muscle cell line, BC3H1. We heretofore refer to these proteins as nicotinic receptor-associated proteins at synapses or N-RAP-syns. The deduced sequence of mouse RAPsyn has 412 amino acids and a molecular mass of 46,392 daltons. The overall identity with Torpedo RAPsyn is 70%; some regions are extremely well conserved and are therefore postulated to be functionally important. Important domains, including the amino terminus and a CAMP-dependent protein kinase phosphorylation site, are conserved between species. Several structural fea- tures are consistent with the proposal that RAPsyn is a peripheral membrane protein that associates with membranes by virtue of covalently bound myristate. Although multiple mRNAs were identified in Torpedo electric RNA blot analysis reveals a single polyadenylated RAPsyn mRNA of ~ 2 . 0 kilo- bases in newborn and 4-week-old mouse muscle. Fi-nally, genomic DNA blot analysis indicates that a sin- gle N-RAPsyn


Identification of the Mouse Muscle 43,000-Dalton Acetylcholine
Receptor-associated Protein (RAPsyn) by cDNA Cloning* (Received for publication, May 6, 1988) Donald E. Frail, Laura L. McLaughlin, Jacqueline Mudd, and John P. MerlieS From the Department of Pharmacology, Washington University School of Medicine, St. Louis,Missouri 631 10 The nicotinic acetylcholine receptor and a receptorassociated protein of 43 kDa are the major proteins present in postsynaptic membranes isolated from Torpedo electric organ. Immunochemical analyses indicated that a protein sharing antigenic determinants with the receptor-associated protein is also present at receptor clusters of muscle cell lines and postsynaptic membranes of vertebrate neuromuscular junctions. We now provide definitive proof that a homolog of the 43-kDa protein exists in mammals. Complimentary DNA clones encoding the complete protein sequence have been isolated from the mouse muscle cell line, BC3H1.
We heretofore refer to these proteins as nicotinic receptor-associated proteins at synapses or N-RAPsyns. The deduced sequence of mouse RAPsyn has 412 amino acids and a molecular mass of 46,392 daltons. The overall identity with Torpedo RAPsyn is 70%; some regions are extremely well conserved and are therefore postulated to be functionally important. Important domains, including the amino terminus and a CAMP-dependent protein kinase phosphorylation site, are conserved between species. Several structural features are consistent with the proposal that RAPsyn is a peripheral membrane protein that associates with membranes by virtue of covalently bound myristate. Although multiple mRNAs were previously identified in Torpedo electric organ, RNA blot analysis reveals a single polyadenylated RAPsyn mRNA of ~2 . 0 kilobases in newborn and 4-week-old mouse muscle. Finally, genomic DNA blot analysis indicates that a single N-RAPsyn gene is present in the mouse genome.
The molecular mechanisms involved in the formation and maintenance of synapses are largely unknown. Postsynaptic specialization at the neuromuscular junction is characterized by both morphological and biochemical changes. These changes include a remodeling of the membrane surface such that junctional folds are developed, as well as the preferential localization of nicotinic acetylcholine receptor (AChR),' acetylcholinesterase, and a number of other less well character-~~ * This research was supported with funds from the Senator Jacob Javits Center of Excellence in the Neurosciences and Monsanto Company and with grants from the National Institutes of Health (to D. E. F. and J. P. M.) and the Muscular Dystrophy Association of America (to J. P. M.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EMBL Data Bank with accession number(s)

503962.
$ To whom correspondence should be addressed. The abbreviations used are: AChR, nicotinic acetylcholine receptor; N-RAPsyn, nicotinic receptor-associated protein at synapses. ized proteins (see Merlie and Sanes, 1986 for a review). Because the electric organ of the Torpedo electric fish is a rich source of AchR and postsynaptic membranes, it has facilitated the identification and biochemical characterization of synapse-specific proteins.
The AchR and a nonreceptor protein of 43 kDa are the most abundant proteins present in highly purified preparations of electric organ postsynaptic membranes (Sobel et al., 1977). This nonreceptor protein is of particular interest because it is a peripheral protein on the cytoplasmic surface (St. John et al., 1982;Bridgeman et al., 1987) whose ultrastructural localization has been shown to be coextensive with the AchR in the electric organ (Sealock et al., 1984). To date, this protein has been referred to as the 43-kDa protein based on its apparent molecular weight on sodium dodecyl sulfate-polyacrylamide gels. However, this reference, the 43-kDa protein, is ambiguous and creates confusion because two abundant proteins, creatine kinase and actin, are also present in the electric organ and have similar apparent molecular weights (Gysin et al., 1983). In order to avoid further confusion and to provide an identity for this nonreceptor protein, we refer to it as N-RAPsyn, nicotinic receptor-associated protein at synapses, to reflect its most characteristic features.
Speculation concerning the function of RAPsyn has centered on the colocalization of RAPsyn with AchR. RAPsyn does not play an obvious role in AchR function since the removal of RAPsyn from postsynaptic membrane vesicles does not alter receptor properties, including the kinetics of binding of acetylcholine, the binding of local anesthetic, and the stimulated efflux of "Na+ (Neubig et al., 1979). However, RAPsyn may play a role in the formation and maintenance of the postsynaptic membrane. It has been proposed that RAPsyn is a cytoskeletal protein that is involved in the localization of the AchR at the postsynaptic membrane (Froehner, 1986) since the rotational movement of the receptor is increased in the absence of RAPsyn (Rousselet et al., 1982). Although there is no evidence to suggest that RAPsyn and AchR are covalently linked, cross-linking studies indicate that RAPsyn is in close proximity to the 0 subunit of the AchR (Burden et al., 1983). RAPsyn has also been reported to be an actin-binding protein (Walker et al., 1984) and a protein kinase (Gorden et al., 1983;Gordon and Milfay, 1986).
Anti-Torpedo RAPsyn antibodies have been used to identify a cross-reacting epitope(s) that colocalizes with AchR at postsynaptic membranes of rat neuromuscular junctions (Froehner et al., 1981;Froehner, 1984). The colocalization of this RAPsyn epitope(s) with the acetylcholine receptor in both Torpedo electrocytes and vertebrate neuromuscular junctions indicates that RAPsyn-like proteins have an important function at cholinergic postsynaptic membranes. We recently isolated and characterized RAPsyn cDNAs from Torpedo electric organ (Frail et al., 1987) that encoded the chemically determined Torpedo RAPsyn protein sequence (Carr et al., Mouse Muscle ACh Receptor-associated Protein cDNAs 15603 1987). We have now identified and characterized mouse muscle cDNAs that encode a protein that is highly homologous with Torpedo RAPsyn, thereby providing definitive proof that RAPsyn exists in mammalian muscle. Important questions concerning the biochemistry of mouse RAPsyn are discussed with respect to predicted structural features of the protein.

EXPERIMENTAL PROCEDURES
Isolation of cDNA Clones-Total RNA was isolated from differentiated BC3H1 cells, a nonfusing mouse muscle cell line (Schubert et al., 1974), by the methods of Chirgwin et al. (1979) and poly(A+) mRNA was purified by chromatography on oligo(dT)-cellulose type 7 (Pharmacia LKB Biotechnologies Inc.) A cDNA libary was then prepared in the Xgtll bacteriophage as described previously (Buonnano et al., 1986). The probe used to screen the library was a 1.0kilobase fragment of the mouse RAPsyn gene. This fragment was isolated from a RAPsyn mouse genomic clone that was identified by low stringency hybridization with a Torpedo RAPsyn cDNA. The genomic fragment was labeled with [32P]dCTP by nick translation and used to screen approximately 250,000 recombinant plaques from the unamplified BC3H1 cDNA library. The filters were hybridized for 24 h at 42 "C in hybridization buffer (50% deionized formamide, 1 M NaCl, 50 mM Tris (pH = 7.5), 1% sodium dodecyl sulfate, 100 pg/ml denatured salmon sperm DNA, 0.1% each of bovine serum albumin, Ficoll, and polyvinylpyrrolidone) and washed at 65 "C in 0.2 X ssc (1 X ssc = 0.15 M NaC1, 0.015 M sodium citrate), 0.5% sodium dodecyl sulfate. A total of 37 putative positives were identified. The putative positive clones were purified through several rounds of plating, and DNA was prepared from five different isolates.
Sequencing of cDNA Clones-The cDNA inserts of the five isolates were excised with EcoRI restriction enzyme, subcloned into either the bacteriophage vector M13mp19 or the plasmid Bluescribe (+) (Stratagene), and used to transform JM109 bacteria. Sequence was obtained from recombinant M13mp19 single-stranded DNA by the dideoxy chain-termination method (Sanger et al., 1977). Alternatively, the dideoxy chain-termination method was used to obtain the sequence of Bluescribe subclones by the procedure of Chen and Seeburg (1985). In both cases, universal primers or synthetic oligonucleotides were used as sequencing primers for the enzyme Sequenase (US Biochemicals).
RNA (Northern) Blot Analysis-RNA ("Northern") blot analysis was performed as described previously (Buonanno et al., 1986). Briefly, total RNA was isolated from muscle tissue or cell lines by the method of Chirgwin et al. (1979), and poly(A+) mRNA was further purified by chromatography on oligo(dT)-cellulose. The mRNA was fractionated on 1.5% agarose gels containing formaldehyde and blotted onto Genescreen (Du Pont-New England Nuclear). A fragment of the mouse RAPsyn gene that encodes a segment of the amino terminus of the protein was labeled by nick translation and used for the hybridization. Hybridization and wash conditions were similar to those used to screen the cDNA library. DNA (Southern) Blot Analysis-High molecular weight mouse genomic DNA was isolated from BC3H1 cells by the method of Keene et al. (1981) and digested to completion with the indicated restriction enzymes (New England Biolabs) according to the manufacturers recommendations. The restricted DNA was then fractionated on 1.0% agarose gels and transferred to GeneScreen Plus (Du Pont-New England Nuclear). The blot was hybridized with nick-translated insert from either M19 cDNA or a mouse genomic clone that encodes a portion of the RAPsyn protein. Hybridization and wash conditions were similar to those used to screen the cDNA library.
Analysis of Sequence Data-Sequence analysis was performed using either the Microgenie Sequence Analysis Program (Beckman Instruments; Queen and Korn, 1984) or programs provided by the Washington University Medical School Biomedical Research Computing Facility. The algorithm of Kyte and Doolittle (1982), with a window of 7, was used to predict the hydropathy of the protein. The algorithm of Garnier et al. (1978) was used to predict protein secondary structure.

RESULTS
Characterization of RAPsyn cDNAs-Messenger RNA was isolated from a mouse muscle cell line, BC3H1 (Schubert et al., 1974), and used to construct a cDNA library in the bacteriophage hgtll. Approximately 240,000 recombinants from the unamplified library were screened with a 1.0-kilobase fragment of the mouse RAPsyn gene.
This fragment was previously identified and isolated from a mouse genomic library by low stringency hybridization with a Torpedo RAPsyn cDNA (data not shown). Thirty-seven putative RAPsyn cDNA clones were identified. Five of these clones were plaquepurified, and the inserts were subcloned into either M13mp19 or Bluescribe (Stratagene) vectors and sequenced. Structural maps of three of t h e inserts are shown in Fig. 1. One insert, M19, contained the complete coding sequence of mouse RAPsyn. We have previously characterized two RAPsyn mRNAs from Torpedo electric organ that encoded two forms of t h e RAPsyn protein, one form being 23 amino acids longer than the other at the carboxyl terminus (Frail et al., 1987). All five mouse RAPsyn cDNAs that we have sequenced encode a protein that is homologous with the long form of Torpedo RAPsyn.
The sequence of the RAPsyn mRNA is shown in Fig. 2. The mRNA contains 106 nucleotides of 5"noncoding sequence, 1236 coding nucleotides, and 214 nucleotides of 3'noncoding sequence. The sequence terminates in a short poly(A+) tail that is 15 nucleotides downstream of a polyadenylation signal (Proudfoot and Brownlee, 1976). The entire 5"noncoding sequence is also present in a mouse RAPsyn genomic clone thus indicating that this sequence does encode mouse RAPsyn mRNA sequences and is not the result of a cloning artifact (data not shown). The nucleotide conservation between Torpedo and mouse RAPsyn mRNA coding regions is 71%.
The Protein Sequence of Mouse RAPsyn-The deduced protein sequence of mouse RAPsyn, shown in Fig. 2, corresponds to the long form of Torpedo RAPsyn. The protein contains 412 amino acids, including the initiator methionine, and has a molecular mass of 46,392 daltons. The alignment of the mouse and Torpedo RAPsyn proteins, shown in Fig. 3, indicates that the RAPsyn protein sequence has been well conserved through evolution. The overall sequence identity of the two proteins is 70%. This high degree of homology between the mouse and Torpedo RAPsyn proteins is similar to the homologies between the mouse and Torpedo CY, p, y, a n d 6 subunits of the AchR (80, 59, 54, and 59%, respectively) (Boulter et al., 1985;Buonanno et al., 1986;Boulter et al., 1986;LaPolla et al., 1984). Some regions of the protein are extremely well conserved, indicating that they may be functionally important. In particular, the first 18 amino acids are identical and two large regions are highly conserved (starting The alignment of several cDNA inserts (thin lines) and the mouse RAPsyn mRNA (thick line) is shown. The numbers above the mRNA refer to nucleotides; the adenosine in the initiator methionine is +l. The numbers above the cDNA inserts refer to the first and last nucleotides of RAPsyn sequence present in the inserts. One insert has additional sequences ligated at one end that are unrelated to RAPsyn sequences (denoted b y w ) . The arrows denote nucleotide sequences obtained using RAPsyn-specific synthetic oligonucleotides (+) or a universal (vector-specific) oligonucleotide (+). Mouse Muscle ACh Receptor-associated Protein cDNAs

A T G G G G C A C I G A C C A C I A C A A A G C A A C A C I A I T W I A A A A G G A c r C , C A G C T G T A C C A G~C A A C C A G~G K ; A A c i G C A~
M o t Gly Gln Asp Gln Thr Lys Gln Gln 11e Glu Lys Gly Leu Gln Leu Tyr Gln Ser Asn G l n Thr o l u Lys Ala L.u

C A l i G T G T G G A n ; A A G G n ; C T G G A C l A A G G G C T A p l G T G A C C G T O G G C C O C~C O G G T A c r C , G G C T G C~O T A A U L G C T
Gln Val Trp M e t Lys Val Leu Glu Lys Gly See Asp Leu Val Gly Arg Phe Arg Val Leu Gly Cys Lou Val *r Ala

C R C T C C G A G A n ; G G C C G C T A C A A A G A G A n ; C r C l A A G m c c C G T G c n : C A l i A n G A T~~C O G G G A C l T j W Y ;~
His Ser Glu M e t Gly Arg Tyr Lys Glu M e t Leu Lys Phe Ala Val Val Gln I l e Asp Thr Ala Arg Gly Lou Glu Asp

G c p G A C T P C~c N 3 O A A A G C T A C c r C , A A C C l T j G C G C O C A G C A A T G A C l A A G A p l G A~G l y j~C A C A M A C C~T C C
Ala Asp Phe Leu Leu Glu Ser Tyr Leu Asn Leu Ala Arg Ser Asn Glu Lys Leu Cys Glu Phe His Lys Thr Ile Ser

T A C T G C A A G A C C T G C C T C O G C~C A p l G T G G C C A U ; G C T G G T O C C C A G C I T G G G G G T~~A C I C C l T j A G C A n ; G G C
Tyr Cys Lys Thr Cys Leu Gly Leu Pro Gly Thr Arg Ala Gly Ala Gln Leu Gly Gly Gln Val Ser Lou Ser X e t Gly

A T P G C C C P G C A G c A C G G T G A C C G C C C C A p l G A~G C A A p l G C T G T~C M : n ; C~G c C G A T A T C C A T C G G A G C Q i A G G O
Ile Ala Leu Gln His Gly Asp Arg Pro Leu Gln Ala Leu Cys Leu Leu Cys Phe Ala Asp Ile His Arg Ser Arg Gly

Gln Asp Leu Ala Glu Glu Val Gly Asn Lys Leu Ser Gln Leu Lys Leu
His Cys Leu Ser Glu Ser 1 1 . Tyr Arg Ser

A A A G G G c p G u \ c . c G T~C T G c G c A c c . c A c~A G T G A c I c I T P C c A c~T G C G T G G A G G K ;~~~T A C T G C o G c
Lys Gly Leu Gln Arg Glu Leu Arg Thr His Val Val Arg Phe His Glu Cys Val Glu Glu Thr Glu Leu Tyr Cy8 Gly

~C n ; T G G T G A G T C C A T C G G G G A C . A G G A A C A G C C G G C P G C A C I G C C~C C C T G C T C C C A C A T C m C A T C P C A G A T C i C
~e u cys Gly G l u See 1 1 . Gly Glu Arg Asn Ser Arg ~e u Gln Ala Leu Pro Cys Set His Ile Phe Ris Lou Arg Cys at Ledz7, 40144, 91%; starting at Ala225, 32/33, 96%). Also conserved is a consensus sequence for CAMP-dependant protein kinase phosphorylation near the carboxyl terminus (Fry  et al., 1986). The predicted secondary structure of this region (Gamier et al., 1978) is distinguished by a large amount of reverse turn structure (data not shown). Reverse turns are concentrated at the surfaces of proteins (Kuntz, 1972), and so the consensus sequence for CAMP-dependent phosphorylation would presumably be accessible to protein kinases and phosphatases. Finally, mouse RAPsyn, like Torpedo RAPsyn, has a very high cysteine content (21/412, 4.9%).

Leu Gln Asn Asn Gly Thr Arg Ser Cys Pro A m Cys Arg Arg Ser Ser bbt Lys Pro Gly Phe Val
We have used available algorithms to predict some structural features of the mouse RAPsyn protein. Purified Torpedo RAPsyn protein readily associates with lipid vesicles of var-ious compositions, and this association is disrupted by alkali (Porter and Froehner, 1985). We have, therefore, analyzed the hydropathic character of mouse RAPsyn using the hydropathy scale devised by Kyte and Doolittle (1982). This analysis, shown in Fig. 4, is useful for identifying membranespanning regions and short hydrophobic and hydrophilic regions of proteins. As expected, RAPsyn does not have any predicted membrane-spanning regions (segments of at least 19 residues with an average hydropathy value of greater than 1.6). In fact, the hydropathy profile is rather unremarkable in that there are no obvious hydrophobic regions that might be responsible for membrane interactions. We also used helical wheel diagrams (Schiffer and Edmundson, 1967) to detect possible amphipathic helices, but none were found. However,  I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I  mouse RAPsyn contains covalently bound myristate? and myristate has been shown to be involved in the membrane association of several other proteins, including pp6OSrc (see Towler et al., 1988 for review). Therefore, we predict that myristate is also involved in the association of RAPsyn with membranes in general and that other protein domains are involved in the specific association of RAPsyn with postsynaptic membranes. Hybridization Analysis-The expression of RAPsyn RNA in the mouse muscle cell line BC3H1 and in mouse muscle was analyzed by blot hybridization (Fig. 5). We previously determined that using a Torpedo RAPsyn probe, at least four RAPsyn mRNAs were present in electric organ; however, this probe detected a single RAPsyn mRNA of "2.0 kilobases in differentiated BC3H1 cells, a mouse muscle cell line (Frail et al., 1987). Blot hybridization analysis using a mouse RAPsyn probe also identified a single -2.0 kilobase mRNA in differentiated BC3Hl cells, thereby confirming our previous observations. We have now analyzed mouse muscle mRNA isolated at two different developmental stages, and the pattern of hybridization was the same as that in BC3Hl (Fig. 5). Therefore, whereas Torpedo electric organ has multiple RAPsyn mRNAs of sizes ranging from 1.6 to 6.0 kilobases, mouse muscle RAPsyn mRNA migrates as a single size of ~2 . 0 kilobases.

eC K A L F F P C K A E L V D V K W S L K R Y Y H Y A V V L L G L S A Y E C C E E S Y K I
The mouse RAPsyn gene was characterized by blot analysis of genomic DNA ("Southern" blotting) (Fig. 6). RAPsyn cDNA insert M19 (Fig. l ) , which encodes the RAPsyn protein, was hybridized to restricted mouse genomic DNA at high stringency; two (BamHI) or three (EcoRI and HindIII) genomic fragments were identified. However, a RAPsyn probe that encodes only the amino terminus of the protein hybridized to one of the genomic fragments in each of the three digests, indicating that the MI9 insert hybridized to multiple fragments derived from a single RAPsyn gene and not from two different genes. The restriction map of a RAPsyn mouse genomic clone is consistent with this conclusion (data not shown). We have not been able to identify additional hybridizing fragments under conditions of low stringency.

DISCUSSION
RAPsyn Is a Novel, Well Conserved Protein-RAPsyn was first described as a peripheral membrane protein that associated with nicotinic acetylcholine receptors at postsynaptic membranes of Torpedo electric organ (St. John et al., 1982; Sealock et al., 1984). The presence of RAPsyn a t vertebrate neuromuscular junctions has been inferred from immunofluorescence studies in which anti-Torpedo RAPsyn antibodies stained receptor-containing endplates of mouse muscle (Froehner et al., 1981). More recently, we detected a messenger RNA in both a mouse muscle cell line and chick muscle that hybridized a t low stringency with Torpedo RAPsyn cDNAs (Frail et al., 1987). We now report the isolation and characterization of mouse muscle RAPsyn cDNAs, thereby demonstrating that a protein that is highly homologous to Torpedo RAPsyn is synthesized by vertebrate muscle.
The protein sequences of mouse and Torpedo RAPsyn are highly homologous, sharing 70% sequence identity. Local homologies are even more impressive, including the first 18 amino acids at the amino terminus, L e~'~~-G l y '~~ and Ala225-Arg'"'. Since the function of RAPsyn is not known, we can only speculate that these highly conserved regions are intimately involved in the function of the protein. Mouse RAPsyn is myristoylated,2 but this alone cannot explain the high degree of conservation at the amino terminus because (i) although there are some primary sequence requirements for myristoylation, a consensus sequence does not exist and substitutions are allowed and (ii) apparently only the first 7-10 amino acids at the amino terminus specify the substrate for myristolyation (Towler et al., 1988).
RAPsyn appears to be a novel protein. We have searched GenBank (Bolt. Beranek, and Newman Laboratories, Cambridge, MA, Tape Release 52.0) and the protein databank of the National Biomedical Research Foundation (release 13.0), and no significant homologies with other genes or proteins were found. We have made an effort to compare RAPsyn protein sequences with protein kinases and G-proteins, as well as various known functional consensus sequences, in-*-associated Protein cDNAs cluding binding sites for ATP (see below) and GTP (Sullivan et al., 1986) and a repeated sequence present in Ca2+-dependent membrane-binding proteins (Geisow et al., 1986). RAPsyn was neither homologous with these proteins nor did it contain any of these consensus sequences. Mouse RAPsyn, like Torpedo RAPsyn, does have a consensus sequence for CAMPdependent protein kinase phosphorylation (Fry et al., 1986), although it is not known if it is used.
Is RAPsyn a Protein Kinase?-Two previous reports have indicated that RAPsyn is a protein kinase. In the first report, ATP analogs were shown to bind to Torpedo RAPsyn (Gordon et al., 1983). In the second report, immunopurified Torpedo RAPsyn was used to phosphorylate casein (Gordon and Milfay, 1986). Although our search of sequence databanks did not reveal homologies with other protein kinases, we have made a more direct analysis of possible sequence patterns shown to exist among protein kinases and the larger group of ATP-binding proteins. We have now attempted to identify these various patterns within the mouse RAPsyn protein.
Two different sequence patterns have been identified in ATP-binding proteins (Fry et al., 1986;Walker et al., 1982). One sequence pattern, the primary features of most all models of ATP-binding domains, includes, but is not restricted to, the primary sequence GXXXXG, where X denotes any amino acid. This motiff is apparently necessary for a conformational change in the protein after the binding of ATP (Fry et al., 1986). The second sequence pattern, usually found approximately 70 amino acids towards the carboxyl terminus of the first, includes several hydrophic residues followed by an aspartate that may be involved in hydrogen bonding to the ATP molecule (Fry et al., 1986). Bairoch and Clavierie (1988) derived model degenerate sequences for these two sequence patterns; these sequences, (LIV)GXGX(FY)GX(LIV) and cessfully identified protein kinases from sequence databanks. Others have considered crystallographic structures of known ATP-binding proteins in order to determine patterns of secondary structure around the ATP-binding domain. Models of secondary structure suggest that 1) the GXXXXG sequence is at or near a strand of B sheet and is immediately followed by a helix and 2) the aspartate terminates a strand of / 3 sheet (Fry et al., 1986;Bradley et al., 1987).
With these sequence considerations in mind, we have not been able to identify RAPsyn as a protein kinase. Mouse RAPsyn does not contain any sequences that satisfy the patterns of Bairoch and Claverie (1988). RAPsyn does have two GXXXXG sequences, and we have analyzed these further. The first sequence, GXXXXGSfi4', is not conserved between Torpedo and mouse and the predicted secondary structure preceding and following this sequence (data not shown) is not @ sheet or a helix, respectively, as suggested by Fry et al. Bradley et al. (1987). The predicted secondary structure surrounding the GXXXXG130-13S sequence (data not shown) does conform to these suggested models, but there are other primary sequence features shared among protein kinases that are not present in this region (Fry et al., 1986). Therefore, if mouse RAPsyn is a protein kinase, then it does not conform to the predicted primary and secondary sequence requirements established for most other protein kinases.
Concluding Remarks-The results of both RNA and DNA blot hybridization analysis are consistent with a single mouse N-RAPsyn gene. We do not yet know if the expression of this RAPsyn gene is restricted to muscle or if RAPsyn is associated with nicotinic cholinergic synapses in the brain. It is tempting to speculate that RAPsyn-like molecules are associated with neuronal synapses. Recently, a peripheral membrane protein

Mouse Muscle ACh Receptor
of M , 93,000 has been shown to be associated with the cytoplasmic domains of the postsynaptic glycine receptor complex (Schmitt et al., 1987). The structural relationship between this protein and RAPsyn is not known, although we have not been able to detect genes that are related to RAPsyn using low stringency hybridization conditions. The function of RAPsyn remains a mystery; however, we will now be able to use RAPsyn cDNAs to express the protein in environments where its biochemistry and function can be investigated.