Proc. Natl. Acad. Sci. USAVol. 92, pp. 3864-3868, April 1995Biophysics

Structural interpretation of the mutations in the (3-cardiac myosinthat have been implicated in familial hypertrophic cardiomyopathyIVAN RAYMENT*t, HAZEL M. HOLDEN*, JAMES R. SELLERSt, LAMEH FANANAPAZIR§, AND NEAL D. EPSTEIN§*Institute for Enzyme Research, Graduate School and Department of Biochemistry, 1710 University Avenue, University of Wisconsin, Madison, WI 53705-4098;and *Laboratory of Molecular Cardiology and §Inherited Cardiac Diseases Section, National Heart, Lung, and Blood Institute, National Institutes of Health,Bethesda, MD 20892

Communicated by Manuel F. Morales, University of the Pacific, San Francisco, CA, November 23, 1994

ABSTRACT In 10-30% of hypertrophic cardiomyopathykindreds, the disease is caused by >29 missense mutations inthe cardiac f8-myosin heavy chain (AYIH7) gene. The aminoacid sequence similarity between chicken skeletal muscle andhuman a-cardiac myosin and the three-dimensional structureof the chicken skeletal muscle myosin head have provided theopportunity to examine the structural consequences of thesenaturally occurring mutations in human ,8-cardiac myosin.This study demonstrates that the mutations are related todistinct structural and functional domains. Twenty-four areclustered around four specific locations in the myosin headthat are (i) associated with the actin binding interface, (ii)around the nucleotide binding site, (iii) adjacent to the regionthat connects the two reactive cysteine residues, and (iv) inclose proximity to the interface of the heavy chain with theessential light chain. The remaining five mutations are in themyosin rod. The locations of these mutations provide insightinto the way they impair the functioning of this molecularmotor and also into the mechanism of energy transduction.

The three-dimensional structure of a protein that has been impli-cated in a disease state is an important source of information forunderstanding the molecular basis of that state. This was firstdemonstrated by the pioneering studies on hemoglobin and sicklecell anemia (1-3). Although for many proteins site-directed mu-tagenesis is now the investigative method of choice, the study ofnatural genetic lesions continues to be of enormous value as thesource of information for understanding protein structure andfunction. (i) It provides one of the most powerful tools foridentifying the proteins and genes involved in the disease state, and(ii) it indicates residues or sections ofa molecule that are importantfor the normal function of the molecule.

Hypertrophic cardiomyopathy (HCM) is an autosomal dom-inant inherited cardiac disease, characterized by left ventric-ular hypertrophy and markedly variable phenotypic expression(4, 5). It is the most common cause of sudden death inotherwise healthy young individuals. The disease is caused bymissense mutations in the 13-myosin heavy chain (MHC) gene(MYH7) in 10-30% of HCM kindreds. More than 29 suchmutations have been identified (4-18). Most are associatedwith a high incidence of sudden death but a few have beenassociated with a more benign prognosis (4, 5). All have beenlocated in the globular head of myosin (subfragment 1, Si) orin the head-rod junction of the myosin molecule. The headregion contains all the necessary elements to generate move-ment of actin relative to myosin during ATP hydrolysis (19,20). At this time, the three-dimensional structure of the human3-cardiac myosin Si is unknown. However, much of the

primary sequence of myosin is highly conserved (21) such thatthe three-dimensional structures of all myosin molecules areexpected to be very similar (22). Thus, it is possible to use the

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

three-dimensional structure of chicken skeletal myosin Si tointerpret the effects induced by the substitutions in the MYI7gene (23, 24). Conversely, the location of the mutationshighlights important areas of the molecule and contributes tounderstanding the molecular basis of energy transduction.We describe here, relative to the structure of chicken skeletal

myosin Si, the location of 29 missense mutations in the ,3-cardiacmyosin (MIYH7) gene that have been shown to cause HCM.

METHODSThe study of the location of homologous residues in the chickenskeletal structure consisted of 29 amino acid residue substitutionsencoded by missense mutations in theMY7 gene. These HCM-causing mutations and their corresponding residues in the chickenfast skeletal muscle MHC are shown in Table 1. The locations ofthe amino acid mutations were mapped onto the chicken structureby aligning the sequences of the human cardiac and chickenskeletal muscle myosins with the GCG program package (25).There was 79% sequence identity between the two proteins in theirrespective myosin heads (26, 27).

RESULTSIn all cases, the mutations in the cardiac amino acid sequencemap to well-defined parts of the three-dimensional structure ofchicken skeletal myosin Si (Table 1). Throughout this paper,the mutations in the human MYI7 gene are indicated by thesingle-letter code for the normal and mutated amino acid,whereas the equivalent residue in the chicken amino acidsequence is indicated by its three-letter code. All of the dis-cussion of the locations of amino acids is relative to the chickenskeletal muscle myosin numbering.There is an extensive biochemical literature pertaining to the

chemical and physical properties of myosin (28). Much of this isdescribed with respect to the three major tryptic fragments, anN-terminal 25-kDa fragment, a central 50-kDa fragment, and aC-terminal 20-kDa fragment, generated by mild digestion ofmyosin S1 (29, 30). Although these peptides do not representindividual tertiary domains (23), they are convenient for describingthe three-dimensional location of the mutations.The three-dimensional locations of the point mutations in

the MYI7 gene as mapped onto the structure of chickenskeletal myosin S1 are shown in Fig. 1 and summarized inTable 1. The proposed interpretation of the biochemicalsignificance of these mutations is based in part on the modelfor the rigor actin-Sl complex (24) that was obtained bydocking the structure of chicken Si onto that of filamentousactin (32) within the constraints of the helical reconstructionof electron micrographs of rigor actin-Sl (33, 34). The muta-tions are clustered around specific locations in the head. In

Abbreviations: ELC, essential light chain; HCM, hypertrophic car-diomyopathy; MHC, myosin heavy chain; Si, subfragment 1; S2,subfragment 2.tTo whom reprint requests should be addressed.

3864

Dow

nloa

ded

by g

uest

on

Aug

ust 3

, 202

0

Proc. Natl. Acad. Sci USA 92 (1995) 3865

Table 1. Relation of the human f3-MHC mutations tostructural/functional domains in the chicken skeletalMHC molecule

Amino acid substitution in Equivalent residue inhuman sequence chicken sequence Ref(s).

Actin-myosin interfaceR403QR403LR403WK615NR663H

50-kDa crossoverV606M

Nucleotide binding pocketT124IY162CN187KN232SF244L

Outer end of ATP pocketR249QG256ER453C

Distal end of ATP pocketQ222K

Adjacent to SH1-SH2 helix and50-kDa to 20-kDa domaininterfaceF513CG584R

ELC binding interfaceG716RR719WR719QR723CG741RD778GS782N

Mutations in the rodR870HL908VE924KE930KE949K

Arg-405Arg-405Arg-405Lys-617Arg-665

Ile-608

Thr-125Tyr-163Asn-188Asn-234Phe-246

Arg-251Gly-258Arg-455

Ser-224

Phe-515Gly-586

Ala-718Lys-721Lys-721Arg-725Ala-743Asp-780Ala-784

Lys-872Leu-910Glu-926Glu-932Glu-951

4, 8, 12161614*

6, 7, 12

*

*

16*

767

R249QR453C0 K615NII Z V606M

St

9 FIG. 1. Two space-filling representations of the three-dimensional7 structure of the chicken skeletal myosin S1 showing the positions, in

yellow, equivalent to the substitutions in the 3-cardiac myosin gene.The mutations are labeled according to the sequence numbering and

9 changes in the ,8-cardiac myosin gene. The heavy chain is displayed in12 light green, violet, and beige to delineate the N-terminal, central, and13 C-terminal fragments. The regulatory and ELCs are displayed in blue17 and light green, respectively. All figures were prepared with MOLSCRIPT6 (31). (Upper) View of actin-binding face. (Lower) View into active site[0 pocket.*

*

4, 12, 1817127

*Unpublished mutations.

particular, one group is associated with the actin bindinginterface, a second is close to the active site, and a third groupis close to the interface of the heavy chain with the essentiallight chain (ELC). A fourth group is adjacent to the region thatconnects the two reactive cysteine residues, Cys-697 andCys-707 in the chicken sequence. The remaining mutations arefound in the myosin rod. The specific environments of themutations in Si are described below.

Substitutions in the Actin Binding Site. Three sites of substi-tutions have been identified that lie close to the actin bindinginterface, R403QLW, V606M, and K615N (Arg-405, Ile-608, andLys-617 in chicken skeletal MHC). There is an additional site,R663H (Arg-665), on the same face of the protein that might bealso involved in actin binding. The three sites are associated withthe flat face of the upper domain of the 50-kDa region ofthe MHC(Figs. 1 and 2). Arg-405 in chicken skeletal muscle myosin occursat the base of a loop that extends away from the bulk of themolecule. This loop has been predicted to be an integral part oftheactomyosin interface (24). The observed conformation of the loopis stabilized by an interaction with a symmetry-related molecule.Thus it is likely that this loop will adopt a different conformationin solution and during its interaction with actin. Arg-405 does notparticipate in any specific interaction with any other amino acid;

thus it is possible that this residue is only important when myosinis bound to actin and then could interact directly with actin orcontribute to the stability of the loop.

Biochemical studies on the R403Q mutation show thathuman 1-cardiac myosin containing this mutation exhibitsconsiderably reduced velocity in an in vitro motility assay anda reduced actin-activated MgATPase activity (35). In additiona rat a-myosin that contained this mutation at the equivalentposition also showed lower in vitro motility than wild type andhad a reduced Vmax and a higher Km for the actin-activatedMgATPase activity (36). These features are consistent with analtered actomyosin interaction.The V606M (chicken Ile-608) mutation occurs in the

chicken skeletal myosin structure shortly after the polypeptidechain crosses over from the lower to the upper segments of thecentral 50-kDa region. This residue is buried and lies againstthe upper 50-kDa domain. All of the residues in contact withIle-608 are identical to those in human 13-cardiac MHC.Insertion of a methionine side chain into this pocket mightdestabilize the molecule in this region or alter the structure inthe cross-over region.The equivalent amino acid to the K615N mutation (chicken

Lys-617) is a surface residue that is located at the top of thehelix-loop-helix motif that contains the V606M mutationdescribed above. It is not obvious what role this residue playsin either the actomyosin complex or protein stability. It isconceivable that it is involved in the actomyosin interactionsince it does face into the region between actin and myosin asit has been currently modeled (24).R663H (chicken Arg-665) is the fourth member of the muta-

tions located in the actomyosin interface. This residue is located at

Biophysics: Rayment et aL

;

Dow

nloa

ded

by g

uest

on

Aug

ust 3

, 202

0

Proc. Natl. Acad ScL USA 92 (1995)

FIG. 2. Stereo ribbon representation of myosin Si showing the location of mutations close to the actin binding site and the obvious cleft thatsplits the 50-kDa region into upper and lower domains. This cleft provides a communication pathway between the nucleotide and actin bindingsites (24). The heavy chain is displayed in green, red, and blue to delineate the 25-kDa, 50-kDa, and 20-kDa fragments.

the end of the first helix in the 20-kDa fragment and is directedtoward the proposed actin binding region. In chicken myosin Si,Arg-665 does not participate in a salt-bridge or hydrogen bondinginteraction and is solvent-exposed. It is difficult to provide astructural role for this mutation, although it does remove a positivecharge from the actomyosin interface.

Mutations near the Nucleotide Binding Site. The active siteof chicken skeletal myosin Si contains many completelyconserved residues (21). It would be expected that very fewmutations in this region would be compatible with life. Evenso, several of the genetic lesions associated with HCM arelocated directly in the nucleotide binding pocket. One group ofthese mutations includes T124L, N187K, N232S, and F244L(Thr-125, Asn-188, Asn-234, and Phe-246), all of which arelocated at the base of the pocket (Fig. 3).The side chains of Thr-125 and Asn-188 residues are in close

proximity in the structure of chicken skeletal myosin Si (Fig.3). The Oy of Thr-125 is not directly coordinated to any otherhydrogen bonding group. Furthermore, it is not within bondingdistance of ADP when it is bound to the high salt crystals (A.J. Fisher and I.R., unpublished results). However, these mu-

tations would be expected to alter the water structure in theactive site and thus affect the kinetic properties of the protein.

Likewise, Asn-234 and Phe-246 lie in close proximity to eachother but are on the opposite side of the phosphate binding looprelative to residues Thr-125 and Asn-188. The side chains for theseresidues are both buried and are brought together by a turn thatconnects the sixth strand of the (3-sheet motif in the myosin headto an a-helix (residues 218-233) that forms part of the base of theactive site pocket. This loop is adjacent to the phosphate bindingloop and contains Ser-243 that has been implicated in nucleotidebinding by chemical modification studies with vanadate (37). Thisloop forms the second major contact region with the phosphatesof the nucleotide (A. J. Fisher and I.R., unpublished results). Theside chains of Asn-234 and Phe-246 are not directly involved inphosphate binding. However, both of these residues are in aposition to influence the orientation of other residues in thephosphate binding pocket. In particular, Asn-234 forms a hydro-gen bond between its NS and the carbonyl oxygen of residue 242,whereas the side chain of Phe-246 forms a stacking interactionwith Phe-438. Thus it is expected that any changes in these sidechains would alter the position of critical residues in the nucle-

FIG. 3. Stereo ribbon representation of the location of mutations associated with the active site pocket. This also reveals the location of Gly-586(G584R) that lies close to the SH2 and the interface between the 50-kDa and 20-kDa fragments of the heavy chain.

3866 Biophysics: Rayment et at

Dow

nloa

ded

by g

uest

on

Aug

ust 3

, 202

0

Proc. Natl. Acad Sci. USA 92 (1995) 3867

otide binding pocket and compromise the catalytic function ofthemolecule.A second group of active site mutations is located at the end of

the nucleotide binding pocket at a distance of 29 A from thephosphate binding loop (Figs. 2 and 3). This group includesY162C, Q222K, R249Q, G256E, and R453C, which correspond toTyr-163, Ser-224, Arg-251, Gly-258, and Arg-455 in the chickenmyosin sequence. Arg-455 and Arg-251 are located close to eachother at the surface of the protein and face toward the putativeactin binding region of the molecule and thus might influence thisproperty of the molecule. This could occur by interacting directlywith actin or by influencing the closure of the narrow cleft thatsplits the 50-kDa tryptic fragment. This is because of their closeassociation with the first helix of the C-terminal 20-kDa trypticfragment that extends across the external interface of the upperand lower domains of the 50-kDa tryptic fragment. Similarly, theside chains for Tyr-163 and Gly-258 are opposite to each other butin this case are somewhat buried. Again the position of theseresidues suggests that they are located in a domain-domaininterface that might be affected when myosin cyclically interactswith actin and nucleotide. Ser-224 is located at the end of thenucleotide binding pocket 23 A from the phosphate binding loopand faces toward the solvent. It does not participate in anyhydrogen bonding interactions. This residue is a serine in thestructure of chicken skeletal myosin S1 whereas it is a glutaminein the -NMHC. In the structure ofchicken skeletal myosin Si, mostof the residues surrounding Ser-224 are identical to the equivalentresidues in the human ,3-MHC. At present it is difficult to assessthe importance of this location in the structure.

Mutations in the Lower Domain ofthe 50-kDa Fragment, Nearthe Reactive Sulihydryls. There are two mutations, F513C (chick-en Phe-515) and G584R (chicken Gly-586), that are associatedwith the lower domain of the 50-kDa tryptic fragment (Figs. 3 and4). In the native structure, both of these residues are close to thehelix that connects the two reactive cysteine residues (Cys-707 andCys-697). Phe-515 forms a stacking interaction with the side chainof residue 706, whereas Gly-586 is located in a turn that abuts thesecond reactive cysteine. It is expected that any other residue inthis latter location would alter the conformation of this segmentand change the domain-domain interactions. Chemical and struc-tural studies of myosin have indicated that there must be asignificant structural rearrangement associated with the reactivesulffihydryl groups during the contractile cycle (38-41). In thestructure of chicken skeletal myosin Si, these two residues areseparated by an a-helix where their side chains point in oppositedirections (23), yet they can be cross-linked in the presence ofnucleotide (42-44) and chemical modification of the cysteineschanges the kinetic properties of myosin (45, 46). It has been

suggested that domain movements around the reactive sulfhydrylpocket allows the molecule to bend during the contractile cyclewithout inducing major structural rearrangements within thedomains themselves (47). These two mutated residues lie at sucha domain interface and may modify the nature and timing ofconformational changes during the contractile cycle.

Mutations in the 20-kDa Tryptic Fragment. There are sevenmutations associated with the 20-kDa fragment (Fig. 4). These areparticularly interesting mutations since they highlight an area ofthe molecule whose importance was not obvious from the initialexamination of the x-ray structure (23, 24). These mutations fallinto two groups. The first group consists of G716R, R719WQ,R723C, and G741R, which correspond to residues Ala-718, Lys-721, Arg-725, and Ala-743. These are located in the small domainthat follows the reactive cysteines. The first three residues, Ala-718,Lys-721, and Arg-725, lie on the same side of a short a-helix thatforms an interface with the ELC. The remaining mutation(G741R) is closely associated with secondary structural elementsthat form the interface to the ELC.The second group of mutations in the 20-kDa fragment are

located in the long a-helix that forms the light chain bindingregion of the molecule. These are D778G and S782N (Asp-780and Ala-784 in chicken). This pair of mutations lies close to thetripartite interface between the N-terminal 25-kDa fragmentof the heavy chain, the ELC, and the long a-helix of the 20-kDafragment. Both sets of mutations suggest that the interfacebetween the heavy chain and the ELC is important forcoupling ATP hydrolysis with movement. To our knowledge,no previous biochemical study, prior to the determination ofthe structure of the myosin head, has suggested that theseresidues play any role in the transduction of energy from thehydrolysis of ATP into mechanical force. It will be importantto investigate in more detail the enzymatic and mechanicalproperties of these mutants.Mutations in the Rod. Five mutations have been observed

to occur in the a-helical segment (subfragment 2, S2) thatconnects the myosin head to the back bone of the thickfilament. These mutations cast light on the role of this part ofthe molecule in muscle contraction. It is expected that themyosin heads for these molecules are normal. Since the headalone is sufficient to generate force (19, 48), these mutationssuggest that their defect lies in the transmission of force to thethick filament array. This could occur in several ways. Theycould affect assembly of the thick filament or stability of theprotein. If these proteins are present at the same levels as wildtype then the effect may be due to a loss of tensile strength orstiffness (rigidity) in S2.

FIG. 4. Stereo ribbon representation of the distribution of mutations that lie close to the interface between the heavy chain and the ELC. Thisalso reveals the location of Phe-515 (F584C) that lies close to the reactive cysteine residues.

Biophysics: Rayment et al

Dow

nloa

ded

by g

uest

on

Aug

ust 3

, 202

0

Proc. Natl. Acad. Sci USA 92 (1995)

DISCUSSIONIn 10-30% of the HCM kindreds, the disease is caused by oneof >29 distinct ,3-MHC gene mutations located on the longarm of chromosome 14 (4-18,49). The mutations do not occurrandomly in the structure but, rather, cluster to four discretelocalizations in the head and to the proximal portion of S2.Many of the mutations lie at the interface between structuraldomains and may influence the transduction of chemicalenergy into movement. The mutations observed in the f3-MHCserve to indicate parts of the molecule that are important forfunction. They represent a class of molecules that are probablypartially impaired and as such highlight the more subtlefeatures of myosin. This implication has been confirmed inbiophysical studies of myosin and skinned muscle fibers fromhuman slow muscle that also express j3-MHC (35, 50).The observation that mutations lying in different functional

regions of the myosin molecule all lead to cardiac hypertrophysuggests that this phenotype is the consequence of the heart'sadaptation to impaired contractile function. This is consistentwith the recent description of two additional HCM diseasegenes encoding sarcomere proteins, a-tropomyosin and car-diac troponin T (51), and with one other chromosomal local-ization (52) whose corresponding gene is presently unknown.

Clinical expression of HCM is markedly heterogeneous.Some families are characterized by a high penetrance and earlyonset of the disease, whereas other families show a morebenign course (5). Additionally, within the same family, clin-ical expression in the heart and skeletal muscle may be quitevariable. These findings highlight the importance of modifyinggenes and environmental factors in determining the severity ofHCM caused by a distinct mutation in the individual patient.Further work may establish a relationship among the locationsof the mutations, its corresponding functional consequences,and the severity of the associated disease.

We thank the reviewers for their insightful comments on thismanuscript. This research was supported in part by grants from theNational Institutes of Health (AR35186 to I.R.). H.M.H. is anEstablished Investigator of the American Heart Association.

1. Ingram, V. M. (1957) Nature (London) 180, 326-328.2. Finch, J. T., Perutz, M. F., Bertles, J. F. & Dobler, J. (1973) Proc. Natl.

Acad. Sci. USA 70, 718-722.3. Perutz, M. F. (1992) Protein Structure: NewApproachesforDisease and

Therapy (Freeman, New York).4. Epstein, N. D., Cohn, G. M., Cyran, F. & Fananapazir, L. (1992)

Circulation 86, 345-352.5. Fananapazir, L. & Epstein, N. D. (1994) Circulation 89, 22-32.6. Fananapazir, L., Dalakas, M. C., Cyran, F., Cohn, G. & Epstein, N. D.

(1993) Proc. Natl. Acad. Sci. USA 90, 3993-3997.7. Watkins, H., Rosenzweig, A., Hwang, D.-S., Levi, T., McKenna, W. J.,

Seidman, C. E. & Seidman, J. G. (1992) N. Engl. J. Med. 326,1108-1114.

8. Geisterfer-Lowrance, A. A. T., Kass, S., Tanigawa, G., Vosberge,H. P., McKenna, W., Seidman, C. E. & Seidman, J. G. (1990) Cell 62,999-1006.

9. Greve, G., Anan, R., Watkins, H., Thierfelder, L., Shono, H., Spirito,P., Mares, A., Roberts, R. & Seidman C. E. (1993) Circulation 88,I-573 (abstr.).

10. Harada, H., Kimura, A., Nishi, A., Sasazuki, T. & Toshima, H. (1993)Biochem. Biophys. Res. Commun. 194, 791-798.

11. Hengstenberg, C., Komajda, M. & Schwartz, K. (1993) Trends Car-diovasc. Med. 3, 115-118.

12. Mares, A. J., Greve, G., Tapscott, T. & Roberts, R. (1993) Circulation88, I-572 (abstr.).

13. Consevage, M. W., Salada, G. C., Baylen, B. G., Ladda, R. L. &Rogan, P. K. (1994) Hum. Mol. Genet. 3, 1025-1026.

14. Nishi, H., Kimura, A., Harada, H., Toshima, H. & Sasazuki, T. (1992)Biochem. Biophys. Res. Commun. 188, 379-387.

15. Dausse, E., Dufour, C., Fetler, L., Dubourg, O., Vosberg, H.-P.,Komajda, M., Schwartz, K. & Guicheney, P. (1993) Circulation 88,I-572.

16. Dausse, E., Komajda, M., Fetler, L., Dubourg, O., Dufour, C., Carrier,L., Wisnewsky, C., Bercovici, J., Hengstenberg, C., Al-Mahdawi, S.,Isnard, R., Hagege, A., Bouhour, J. B., Desnos, M., Beckmann, J.,Weissenbach, J., Schwartz, K. & Guicheney, P. (1993) J. Clin. Invest.92, 2807-2813.

17. Watkins, H., Thierfelder, L., Hwang, D.-S., Seidman, J. G. & Sei-dman, C. E. (1992) J. Clin. Invest. 90, 1666-1671.

18. Al-Mahdawi, S., Chamberlain, S., Cleland, J., Nihoyanopoulos, P.,Gilligan, D., French, J., Choudhury, L., Williamson, R. & Oakley, C.(1993) Br. Heart J. 69, 136-141.

19. Toyoshima, Y. Y., Kron, S. J., McNally, E. M., Niebling, K. R., Toyo-shima, C. & Spudich, J. A. (1987) Nature (London) 328, 536-539.

20. Itakura, S., Yamakawa, H., Toyoshima, Y. Y., Ishijima, A., Kojima, T.,Harada, Y., Yamnagida, T., Wakabayashi, T. & Sutoh,K (1993) BiochemBiopys. Res. Conmmun 196, 1504-1510.

21. Warrick, H. M. & Spudich, J. A. (1987) Annu. Rev. Cell Bio. 3,379-421.

22. Chothia, C. & Lesk, A. M. (1986) EMBO J. 5, 823-826.23. Rayment, I., Rypniewski, W. R., Schmidt-Base, K., Smith, R., Tom-

chick, D. R., Benning, M. M., Winkelmann, D. A., Wesenberg, G. &Holden, H. M. (1993) Science 261, 50-58.

24. Rayment, I., Holden, H. M., Whittaker, M., Yohn, C. B., Lorenz, M.,Holmes, K. C. & Milligan, R. A. (1993) Science 261, 58-65.

25. Devereux, J., Haeberli, P.-S. & Smithies, 0. (1984) Nucleic Acids Res.12, 387-395.

26. Maita, T., Yajima, E., Nagata, S., Miyanishi, T., Nakayama, S. &Matsuda, G. (1991) J. Biochem. (Tokyo) 110, 75-87.

27. Jaenicke, T., Diederich, K W., Haas, H., Scheich, J., Lichter, P., Pfordt,M., Bach, A. & Vosberg, H.-P. (1990) Genomics 8, 194-206.

28. Cooke, R. (1986) CRC Crit. Rev. Biochem. 21, 53-118.29. Balint, M., Sreter, F. A., Wolf, I., Nagy, B. & Gergley, J. (1975)J. Bio.

Chem. 250, 6168-6177.30. Mornet, D., Pantel, P., Audemard, E. & Kassab, R. (1979) Biochem.

Biophys. Res. Commun. 89, 925-932.31. Kraulis, P. J. (1991) J. Appl. Crystallogr. 24, 946-950.32. Holmes, K. C., Popp, D., Gebhard, W. & Kabsch, W. (1990) Nature

(London) 347, 44-49.33. Milligan, R. A., Whitaker, M. & Safer, D. (1990) Nature (London)

348, 217-221.34. Milligan, R. A. & Flicker, P. F. (1987) J. Cell Biol. 105, 29-39.35. Cuda, G. F. L., Zhu, W.-S., Sellers, J. R. & Epstein, N. D. (1993) J.

Clin. Invest. 91, 2861-2865.36. Sweeney, H. L., Straceski, A. J., Leinwand, L. A., Tikunov, B. A. &

Faust, L. (1994) J. Biol. Chem. 269, 1603-1606.37. Yount, R. G., Cremo, C. R., Grammer, J. C. & Kerwin, B. A. (1992)

Philos. Trans. R. Soc. London B 336, 55-61.38. Cheung, H. C., Gonsoulin, F. & Garland, F. (1983) J. Biol. Chem. 258,

5775-5786.39. Dalbey, R. E., Weiel, J. & Yount, R. G. (1983) Biochemistry 22,

4696-4706.40. Huston, E. E., Grammer, J. C. & Yount, R. G. (1988) Biochemistry 27,

8945-8952.41. Botts, J., Thomason, J. F. & Morales, M. F. (1989) Proc. Natl. Acad.

Sci. USA 86, 2204-2208.42. Reisler, E., Burke, M., Himmelfarb, S. & Harrington, W. F. (1974)

Biochemistry 13, 3837-3840.43. Burke, M. & Reisler, E. (1977) Biochemistry 16, 5559-5563.44. Wells, J. A. & Yount, R. G. (1979) Proc. Natl. Acad. Sci. USA 76,

4966-4970.45. Sekine, T. & Kielley, W. W. (1964) Biochim. Biophys. Acta 81,

336-345.46. Sleep, J. A., Trybus, K M., Johnson, K. A. & Taylor, E. W. (1981) J.

Muscle Res. Cell Motil. 2, 373-399.47. Rayment, I. & Holden, H. M. (1994) Trends Biochem. Sci. 19,

129-134.48. Kishino, A. & Yanagida, T. (1988) Nature (London) 334, 74-76.49. Jarcho, J. A., McKenna, W., Pare, J. A. P., Solomon, S. D., Levi, T.,

Donis-Keller, H., Seidman, J. G. & Seidman, C. E. (1990) N. Engl. J.Med. 321, 1372-1378.

50. Sweeney, H. L., Lankford, E., Dantzig, J., Epstein, N. D. & Fanana-pazir, L. (1993) Circulation 88, I-526.

51. Thierfelder, L., Watkins, H., MacRae, C., Lamas, R., McKenna, W.,Vosberg, H.-P., Seidman, J. G. & Seidman, C. E. (1994) Cell 77,701-712.

52. Carrier, L., Hengstenberg, C., Beckmann, J. S., Guicheney, P., Duf-our, C., et al (1993) Nat. Genet. 4, 311-313.

3868 Biophysics: Rayment et aL

Dow

nloa

ded

by g

uest

on

Aug

ust 3

, 202

0