Rapid Publication

Skeletal Muscle Expression and Abnormal Function

of @-Myosin in Hypertrophic Cardiomyopathy
Giovanni Cuda, * Lameh Fananapazir, * Wen-Si Zhu, James R. Sellers, * and Neal D. EpsteinI
*Laboratory of Molecular Cardiology, tCardiology Branch, and 1Clinical Hematology Branch, National Heart,
Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892

Abstract There is, thus, allelic and nonallelic genetic heterogeneity

Hypertrophic cardiomyopathy is an important inherited dis- in HCM.
ease. The phenotype has been linked, in some kindreds, to the Evidence exists that f-myosin is also present in the slow
beta-myosin heavy chain (,8-MHC) gene. Missense and silent fibers of skeletal muscle of some animals (11-13). In this
mutations in the 0-MHC gene were used as markers to demon- study, we show that normal and mutant cardiac fl-MHC gene
strate the expression of mutant and normal cardiac fl-MHC message is also expressed in skeletal muscle of HCM patients
gene message in skeletal muscle of hypertrophic cardiomyopa- with identified missense mutations in the p3-MHC gene. The
thy patients. Mutant 8-myosin, also shown to be present in presence of the mutant cardiac p-MHC message in skeletal
skeletal muscle by Western blot analysis, translocated actin muscle indicates that the skeletal slow myosin heavy chain iso-
filaments slower than normal controls in an in vitro motility form is transcribed from the cardiac fl-MHC gene. We further
assay. Thus, single amino acid changes in 8-myosin result in show that mutant 3-myosin is present in skeletal muscle and
abnormal actomyosin interactions, confirming the primary role has abnormal function in an in vitro assay in which actin fila-
of missense mutations in ,-MHC gene in the etiology of hyper- ments are translocated by myosin bound to a coverslip surface.
trophic cardiomyopathy. (J. Clin. Invest. 1993. 91:2861- These results demonstrate that some j3-MHC gene mutations
2865.) Key words: cardiac hypertrophy- beta-myosin heavy in patients with HCM are associated with an abnormal acto-
chain * molecular genetics * mutation * muscle myosin interaction which may be the basis of their disease.

Introduction Methods
Hypertrophic cardiomyopathy (HCM)' is an autosomal domi-
nant inherited disease, and represents an important cause of Amplification of exon 23 off3-MHC. DNA was extracted from periph-
sudden cardiac death particularly in otherwise healthy young eral blood lymphocytes (14). PCR product for sequencing was per-
individuals such as athletes. The disease has been defined as an formed using intron 22 and intron 23 primers to avoid coamplification
increase in left ventricular wall thickness in the absence of an- of the highly homologous a-MHC gene. We have previously described
other cause of cardiac hypertrophy. Associated left ventricular primers and PCR parameters (7). Sequencing of total PCR product,
without subcloning, was performed with a Sequenase kit (U. S. Bio-
systolic and diastolic dysfunction, myocardial ischemia, and chemical Corp., Cleveland, OH) on single-stranded template generated
life-threatening arrhythmias frequently cause disabling symp- by alkaline denaturation of biotin labeled PCR products bound to
toms. Recently, the phenotype has been linked, in some streptavidin coated magnetic beads (Dynal, Inc., Great Neck, NY) (3).
kindreds, to the beta-myosin heavy chain (j3-MHC) gene lo- RNA was extracted (15) from fresh biopsies of skeletal muscle after
cated on the long arm of chromosome 14 (1-3). Several muta- homogenization in buffer with a Polytron (Brinkman Instruments,
tions resulting in substitutions of single evolutionary conserved Westbury, NY). Reverse transcriptase PCR was performed using the
amino acids in this gene have been described (4-8). These GeneAmp RNA PCR kit (Perkin Elmer Cetus, Norwalk, CT). Ran-
have all been detected in the head and head-rod junction re- dom hexamers were used to prime the reverse transcription reaction.
gions of the f3-MHC protein. The f-MHC gene, however, has Cycle lengths of 30 s and temperatures of 94, 61, and 72°C were used
for 30 cycles of denaturation, annealing, and extension, respectively.
been excluded as the disease locus in other kindreds (3, 9, 10). The 5'primer was 5'-GAGAAGAATGACCTGCAGCTCC-3'. The 3'
primer, 5'-AGCTTGGCAATGATCTCATCC-3', had a 40-bp gua-
This work was presented as an abstract at the 65th meeting of the nine-cytosine clamp affixed to the 5' end. The products digested by
American Heart Association, November 1992, and has appeared in PvuII (BRL, Gaithersburg, MD) was electrophoresed by 10% PAGE
abstract form (1992. Circulation 86:1229). and stained with ethidium bromide.
Address correspondence to Neal D. Epstein, Clinical Hematology Ampl/ification of exon 8 ofJ7-MHC. The primers were chosen from
Branch, Building 10, Room 7C-103, National Heart, Lung, and Blood intron 7 and intron 8. They were: 5'-CTGCCCTCCAAGGTCCTGTA-
Institute, National Institutes of Health, Bethesda, MD 20892. CC-3' and 5'-CCATTCCTCCACCAGTCCAAGTC-3', respectively.
Receivedfor publication 31 December 1992 and in revisedform I The cDNA library was constructed from skeletal muscle RNA using
March 1993. the Uni-ZAP vector (Stratagene Inc., La Jolla, CA). It was sequenced,
in part, using the Sequenase kit (Stratagene Inc.) according to the dou-
1. Abbreviations used in this paper: ,B-MHC, beta-myosin heavy chain; ble-stranded sequencing protocol of the manufacturer.
HCM, hypertrophic cardiomyopathy. Purification of proteins. All procedures were performed at 4°C.
Myosin was purified from human skeletal or cardiac muscle biopsies.
The Journal of Clinical Investigation, Inc. Briefly, 100- 150 mg of frozen tissue was pulverized and washed three
Volume 91, June 1993, 2861-2865 times in 5 vol of buffer A containing 20 mM Mops (pH 7.0), 40 mM

Skeletal Muscle Expression and Abnormal Function of f3-Myosin in Hypertrophic Cardiomyopathy 2861
KC1, 5 mM EGTA, 2 mM DTT, 0.01% NaN3, 0.1 mM PMSF, 10 A B
mg/liter pepstatin, 10 mg/liter tosyl-L-phenylalamine chloromethyl Genomic DNA cDNA Clone
ketone, 10 mg/liter leupeptin, 5 mg/liter chymostatin, and 10 mg/
liter tosyl-L-lysine chloromethyl ketone. To purify myosin, the washed
myofibrils were resuspended in buffer A and extracted for 15 min by -M. C T
T T
the addition of NaCi, ATP, and MgCl2 to final concentrations of 0.5 M, 4 C A
10 mM, and 10 mM, respectively. The sample was sedimented for 15 46 C
A Exon 9
min at 470,000 g using an ultracentrifuge (TLIOO; Beckman Instru- T
~* C A
ments, Inc., Palo Alto, CA). Rabbit skeletal muscle actin was prepared .:.

C Exon 8 G
as previously described ( 16). C G
G
Preparation ofantibodies. To specifically recognize the fl-MHC iso- **
G
c
form from skeletal muscle tissue, a polyclonal antiserum was raised T Codon
* T Codon
against a synthetic peptide corresponding to the unique carboxyl-ter- T
minus of the f,-MHC sequence (amino acids 1921-1935, reference 17) T/C) 244 C
* G J1

G Exon 8
and the specific IgG fraction was purified by affinity chromatography * T Intron 8 *
in which the antigen was coupled to Affi-Gel 15 (Bio-Rad Laborato- G
* C
ries, Richmond, CA) according to manufacturer's instructions. The ACGT T G C A
IgG did not cross-react with the fast skeletal MHC isoform, nor with
the cardiac a-MHC in a Western blot analysis. Figure 2. (A) A portion of a sequence from genomic DNA showing a
Another antibody was made to identify the expression of the mu- silent mutation in codon 2440 Phe of exon 8 in the f3-MHC gene of a
tated 403 An--Gln ,B-MHC in the skeletal muscle tissue of HCM patients. patient with the 908 Leu -* Val mutation. The arrow marks both a
This antibody was raised against a synthetic peptide (amino acids 392- T and a C which are present in this patient, who is a heterozygote for
402, reference 14) of the f3-MHC sequence (which does not include the the 2440 Ple mutation. (B) A portion of sequence of a clone obtained
mutated residue) and purified as described above. from a cDNA library constructed from skeletal muscle of the same
Identification of mutated f3-MHC from skeletal muscle tissue. Hu- individual. Only the C is present, demonstrating the segregation of
man soleus muscle myosin was electrophoresed on 5% SDS-PAGE one of the two alleles shown in genomic sequence of the f,-MHC gene
containing a 30:0.4 ratio of acrylamide:bis acrylamide, which allows in A.
the separation of the fast from the (3/slow MHC isoform. The fl-MHC
band was excised from the gel, electroeluted using a micro-electroelu- of 50-100 ,1 and extensively digested with 2 mg/ml endoproteinase
tor (Amicon Corp., Danvers, MA), concentrated to a final volume Arg-C (Calbiochem-Behring Corp., San Diego, CA). The digest was
concentrated to 50-Mi final vol, electrophoresed on a 20% SDS-PAGE,
and blotted onto a PVDF membrane (Millipore Corp., Bedford, MA)
Co
0 using a Milliblot-SDE apparatus (Millipore). The blot was probed with
the rabbit polyclonal IgG raised against a synthetic peptide extending
El. from residues 392-402 in the f3-MHC sequence.
In vitro motility assay. The in vitro motility assay was performed in
two different ways. In some instances either pure cardiac ,B-myosin or
soleus skeletal muscle myosin was bound to the nitrocellulose-coated
96 bp- surface in monomeric form as described in Umemoto and Sellers ( 18 ).
75 bp- To selectively study the motility ofB-myosin from soleus muscle with-
out interference from the contaminating fast myosin, the anti-,B-MHC
1 180 base pairs COOH-terminal-specific antibody was used to affinity select the (-
Exon 22 Exon 23 myosin. In this case the f3-myosin-specific IgG was first introduced into
the motility assay flow cell at a concentration of 0.1-0.5 mg/ml in
buffer B containing 0.5 M NaCI, 10 mM Mops (pH 7.0), 0.1 mM
EGTA, 1 mM DTT, and allowed to incubate for 20 min. The unbound
antibody was washed out and the surface was coated with 0.5 mg/ml
BSA in buffer B. After 1 min, soleus muscle myosin was added in buffer
B. The same buffer was used, after 5 min, to remove the unbound
material, which contains the fast myosin isoform. The in vitro motility
assay was otherwise performed and the results quantified essentially
75 base pairs .2 according to Homsher et al. ( 19). The composition of the assay buffer
96 base pairs was 20 mM KCI, 10 mM Mops (pH 7.2), 5 mM MgCl2, 1 mM ATP,
274 base pairs 0.1 mM EGTA, 10 mM DTT, 0.7% methylcellulose, 2.5 mg/ml glu-
0 base
b
cose, 0.1 mg/ ml glucose oxidase, and 0.02 mg/ ml catalase.
370 pairs Specificity of binding of ,B-myosin to the coverslip was determined
by SDS-PAGE analysis of the unbound protein after washout and of
Figure 1. A PvuII digest of reverse transcriptase PCR fragments, am- the protein bound to the nitrocellulose surface following extraction of
plified from skeletal muscle RNA, encompassing the 908 amino acid the surface with 2% SDS sample buffer.
residue mutation in the ,B-MHC gene of the same affected individual
and two controls. Below is a genomic map ofthe region showing the
location of the two PCR primers, marked with horizontal arrows, Results
which generated a 370-bp fragment from RNA, easily distinguished
from a 1,1 80-bp fragment obtained from DNA. One of the PvuII Expression of the f3-MHC gene message in skeletal muscle.
sites in exon 23 is destroyed by the mutation and upon digestion Skeletal muscle RNA was obtained from members of two
yields an aberrant 96-bp fragment present only in the affected indi- kindreds in which we previously linked missense mutations in
vidual together with the predicted normal 75-bp fragment. P, PvuII the genomic sequence of the cardiac i3-MHC gene to HCM (3,
site; *, PvuII site destroyed by the 908 Leu -o Val mutation. 7). A C -- G transversion was detected in a PCR fragment,

2862 G. Cuda, L. Fananapazir, W.-Si. Zhu, J. R. Sellers, and N. D. Epstein

 A
Arg-C Proteolytic Fragment of Wild Type Sequence - 3.5 kDa:

-EEQAEPDGTEEADKSAYLMGLNSADLLGGLCHPR-
T Tg
Arg-C Arg-C

B
Figure 3. Schematic showing the wild-type
Arg-C Proteolytic Fragment of 403Arg9Gln Mutant Sequence - 6.6 kDa: (A) and mutant (B) sequences following
enodoproteinase Arg-C cleavage. Cleavage
sites are shown by arrows. The Arg -- Gln
-EEQAEPDGTEEADKSAYLMGLNSADLLGGLCHPQVKVGNEYVTKGQNVQQVIYATGALAKAVYER- substitution at amino acid residue 403 de-
I T stroys a cleavage site, resulting in longer
Arg-C Ar,rg-C (6.6 kD vs 3.5 kD) sequence.

amplified from genomic DNA of an affected member of myosin, while an extra peptide of 6.6 kD, together with the
kindred 2755 (7). The mutation results in a Leu -* Val substi- wild-type peptide, was observed in the digested #-myosin from
tution in codon 908 and destroys a PvuII site. Hence, a reverse HCM patients due to the replacement of Gln for Arg at residue
transcriptase PCR fragment, amplified from skeletal muscle 403 (Fig. 4).
RNA of an affected member, yielded an aberrant band when In vitro motility assay studies. Skeletal muscle biopsies
digested with PvuII (Fig. 1). Comparison of the normal and from patients with identified mutations of the fl-MHC gene
aberrant bands shows that the ratio of mutant to normal mes- were used to study the pathophysiology of the mutant myosin
sage in pectoralis muscle is > 50%. Multiple cDNA clones from in an in vitro motility assay (21 ). This assay measures the rate
a library constructed from the same skeletal muscle RNA were of sliding of rhodamine phalloidin-labeled actin filaments
sequenced in part, and matched the previously published car- translocated by myosin molecules bound to a nitrocellulose-
diac fl-MHC gene sequence obtained from cardiac cDNA ( 17) coated surface. To study the function of 3-myosin from individ-
and a genomic clone (20), (data not shown). Sequence of a uals with the 908 L'u'vaI and 403 AgGln mutations, this isoform
PCR fragment encompassing codon 244 Phe amplified from ge- had to be separated from the fast-myosin isoform, also present
nomic DNA of the same individual shows the two alleles (C in skeletal muscle. Beta-myosin was first enriched through the
and T) of a polymorphism reflecting a silent mutation (Fig. 2 choice of the soleus muscle, as it contains 75-90% slow fibers.
A). A cDNA clone from this individual contained the An antibody raised to the unique carboxyl-terminal end of ,B-
908'Leu-'va mutation which cosegregated with the C in codon
244Phe (Fig. 2 B), further confirming that the j3-MHC message
in human skeletal muscle is transcribed from the cardiac /3-
MHC gene on chromosome 1 4q. a b C Figure 4. The 403 Arg -
Gln mutant fl-MHC is
Expression of normal and mutant #-myosin in the slow, present in human soleus
type I, skeletal muscle fibers in a patient with the 403 Arg -o _1 46 K muscle. The Arg -- Gln
Gln mutation. Loss of arginine in this mutation allowed dis- substitution at residue
crimination of mutant from normal ,B-MHC by digestion with 403 allows for detection
endoproteinase Arg-C, which specifically cleaves at the car- - 30 K of mutant protein when
boxyl-terminal site of arginine residues. An antibody was the ,B-MHC is digested
raised against a peptide corresponding to amino acid residues with arginine specific
392-402. In the control unmutated myosin the Arg-C peptide endoproteinase Arg-C.
recognized by this antibody should have a molecular mass of - 21.5 K Anantibodytoasyn-
thetic peptide flanking
-
3.5 kD, whereas in the 403A" GI" mutation, the loss of Arg the mutation recognizes
would result in a new, larger peptide with a molecular mass of in Western blot the pre-
6.6 kD (Fig. 3). UV - 6.5 K dicted 3.5-kD peptide in
Beta-myosin was extracted from soleus muscle biopsies of Bf-MHC from a normal
HCM patients with the 403M"G"n mutation and normal con- individual (lane B). In
trols. The p-myosin heavy chain was separated from the fast- the affected heterozy-
myosin heavy chain isoform by 5% SDS-PAGE. The band -. 3.4 KK gous individual an
corresponding to ,B-MHC was excised from the gel and exten- 3*4 aberrant 6.6-kD peptide
sively digested with endoproteinase Arg-C. The sample was resulting from the loss
of Arg at residue 403 as
then applied on a 20% SDS-PAGE, which allows the separation well as the normal 3.5-
of very low molecular weight peptides, and was blotted to a kD peptide is recognized (lane A). Prestained low molecular mass
PVDF membrane and probed with the antibody. A single pep- standards (Amersham Corp., Arlington Heights, IL) are shown in
tide at the expected size of 3.5 kD was detected in control lane C.

Skeletal Muscle Expression and Abnormal Function off-Myosin in Hypertrophic Cardiomyopathy 2863
A B C

ANTI P-MHC Aby

a

f.
b d

3-MYOSIN
.......4;.
.......__
7- y
'V.y
a b

FWST* =
- P-MHC FAST MYOSIN o-MHC- - FAST MHC

F-ACTvN
F-ACTIN
FAST MYOSIN

Figure 5. Use of anti-fl-MHC carboxyl-terminal specific IgG to immunoabsorb (3-myosin for in vitro motility assays. (A) 5% SDS-PAGE (a, b)
and Western blot (c, d) probed with an antibody prepared against a synthetic peptide (RAKSRDIGTKGLNEE) derived from the unique car-
boxyl-terminal region of f3-MHC. Lanes a and c contain myosin from soleus muscle; lanes b and d contain myosin from human ventricle. (B)
Schematic diagram showing the strategy for using the carboxyl-terminal fl-MHC specific antibody for immunoabsorbing the ,B-myosin isoform
from myosin purified from soleus muscle which is a mixture of fast skeletal and #-myosin isoforms. (C) 5% SDS-PAGE showing the myosin
that is bound to the coverslip via the carboxyl-terminal #-MHC specific antibody (lane a) and the flow through containing the fast skeletal
muscle myosin isoform.

MHC sequence was used to immunoabsorb the ,B-myosin to Discussion

the nitrocellulose-coated surface. This antibody specifically de-
tects, in a Western blot analysis of myosin from skeletal and We have determined that a mutant 13-cardiac myosin is ex-
cardiac muscle, only the f3-MHC (Fig. 5, A and B). SDS-PAGE pressed in skeletal muscle of HCM patients with identified 13-
analysis of both the flow through and immunoabsorbed frac- MHC mutations and has abnormal function. This strengthens
tions confirmed that only f3-myosin was bound to the anti- the hypothesis that fl-MHC mutations are indeed responsible
body-coated surface (Fig. 5 C). Immunoabsorbed 1-myosin for the disease in some kindreds. The consistency of the find-
from normal individuals translocated actin filaments at a rate ings is underscored by the fact that two unrelated kindreds
of 0.49±0.04 ,um/s. The actin sliding velocity in individuals (Western European versus Asian origin) with the same
with the 908 Iu-val or the 403ASg-Gln mutations were only 403A`-GIn mutation had the same degree of abnormal myosin
0.17±0.01 and 0.09±0.02 gm/s, respectively (Fig. 6). Since function. The cardiac findings in the two kindreds with the
both mutant and normal beta myosins were present, the effect 403 A-Gln and 908 ''I mutations have been similar. How-
of the mutation may be even larger than observed. The anti- ever, the natural histories are markedly different: the
body did not affect the rate of actin filament translocation by 403 AOg~lln mutation is characterized by 100% penetrance in
pure ,-myosin obtained from cardiac ventricle (Fig. 6). adults, early onset, and a high frequency of sudden death (7).
In contrast, in the 908Leu-va' mutation, HCM is not apparent
in childhood, the penetrance is 63% in adults, and there is a low
frequency of sudden death (7). Despite these differences, myo-
A B sin from patients with either mutation translocates actin fila-
IJ No Ab ments with similar velocities in the in vitro motility assay. This
E
0.8
Plot Ab

0.5
T may be due, in part, to the nature of the motility assay in its
measurement of an unloaded velocity ( 19). The quantities of
pure 13-myosin isolated precluded other biochemical analyses
- 0.4

such as ATPase activity or actin interaction. Measurement of

~ ~ a 0.3
these parameters and others such as loaded tension or ATP
consumption may further discriminate between the functions
of these two mutant ,-myosins compared to normal myosin.
Our experiments suggest that, in individuals with myosin-asso-
Pwctoralis Soleus Ventricle Control (L9f->V) (R403->Q) ciated HCM, the disease may be explained by abnormal acto-
Figure 6. Mutations at residue 403 and 908 decrease the rate of actin myosin interactions resulting in decreased myosin function
filament sliding. (A) Velocity of actin filaments sliding over human and that the hypertrophy may be compensatory.
pectoralis and soleus muscle myosin and purified human ventricular
myosin, assayed in the absence (white bars) and in the presence
(black bars) of the carboxyl-terminal ,l-MHC specific antibody. The We would like to thank Drs. Robert Adelstein and Arthur W. Nienhuis
data are expressed as the mean±SD of the rate of 20-100 actin fila- for their encouragement and support.
ments in a single representative experiment. (B) Rate of translocation
of actin filaments over myosins with distinct point mutations. Myosin References
preparations from four individuals with the 403 "'ln mutation and 1. Jarcho, J. A., W. McKenna, J. A. P. Pare, S. D. Solomon, R. F. Holcombe,
five individuals with the 908 L"u-Val mutations were analyzed. Myosin S. Dickie, T. Levi, H. Donis-Keller, J. G. Seidman, and C. E. Seidman. 1989.
from five normal subjects were used as controls. Mapping a gene for familial chromosome 14q 1. N. Engl. J. Med. 321:1372-1378.

2864 G. Cuda, L. Fananapazir, W.-Si. Zhu, J. R. Sellers, and N. D. Epstein

 2. Hejtmancik, J. F., P. A. Brink, J. Towbin, R. Hill, L. Brink, T. Tapscott, A. sion of cardiac myosin heavy chain and actin gene involvement in hypertrophic
Trakhtenbroit, and R. Roberts. 1991. Localization of gene for familial hypertro- cardiomyopathy of several french families. Circ. Res. 71:3-8.
phic cardiomyopathy to chromosome 14q I in a diverse U.S. population. Circula- 1 1. Lompre, A. M., B. Nadal-Ginard, and V. Mahdavi. 1984. Expression of
tion. 83:1592-1957. the cardiac ventricular a- and #-myosin heavy chain genes is developmentally
3. Epstein, N. D., L. Fananapazir, H. J. Lin, J. Mulvihill, R. White, J.-M. and hormonally regulated. J. Biol. Chem. 259:6437-6446.
Lalouel, R. P. Lifton, A. W. Nienhuis, and M. Leppert. 1992. Evidence ofgenetic 12. Mahdavi, V., A. Chambers, and B. Nadal-Ginard. 1984. Cardiac a- and
heterogeneity in five kindreds with familial hypertrophic cardiomyopathy. Circu- #-myosin heavy chain genes are organized in tandem. Proc. Nail. Acad. Sci. USA.
lation. 85:635-647. 8 1:2626-2630.
4. Tanigawa, G., J. A. Jarcho, S. Kass, S. D. Solomon, H. P. Vosberg, J. G. 13. Sinha, A. M., D. J. Friedman, J. M. Nigro, S. Jakovcic, M. Rabinowitz,
Seidman, and C. E. Seidman. 1990. A molecular basis for familial hypertrophic and P. K. Umeda. 1984. Expression of rabbit ventricular a-myosin heavy chain
cardiomyopathy: an a/# cardiac myosin heavy chain hybrid gene. Cell. 62:991 - messenger RNA sequences in atrial muscle. J Biol. Chem. 259:6674-6680.
998. 14. Bell, G. I., J. H. Karam, and W. J. Rutter. 1981. Polymorphic DNA region
5. Geisterfer-Lowrance, A. A. T., S. Kass, G. Tanigawa, H. P. Vosberg, W. adjacent to the 5' end of human insulin. Proc. Nall. Acad. Sci. USA. 78:5759-
McKenna, C. E. Seidman, and J. G. Seidman. 1990. A molecular basis for famil- 5763.
ial hypertrophic cardiomyopathy: a beta-cardiac myosin heavy chain gene mis- 15. Chomczynski, P., and N. Sacchi. 1987. Single-step method of RNA isola-
sense mutation. Cell. 62:999-1006. tion by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Bio-
6. Watkins, H., A. Rosenzweig, D. S. Hwang, T. Levi, W. McKenna, C. E. chem. 162:156-159.
16. Eisenberg, E., and W. W. Keilley. 1974. Troponin-tropomyosin complex.
Seidman, and J. G. Seidman. 1992. Characteristics and prognostic implications J. Biol. Chem. 249:4742-4748.
of myosin missense mutation in familial hypertrophic cardiomyopathy. N. Engl. 17. Jaenicke, T., K. W. Diederich, W. Haas, J. Schleich, P. Lichter, M. Pfordt,
J. Med. 321:1108-1114. A. Bach, and H.-P. Vosberg. 1990. The complete sequence of the human fl-myo-
7. Epstein, N. D., G. M. Cohn, F. Cyran, and L. Fananapazir. 1992. Differ- sin heavy chain gene and a comparative analysis ofits product. Genomics. 8:194-
ences in clinical expression of hypertrophic cardiomyopathy associated with two 206.
distinct mutations in the #-myosin heavy chain gene: a 908 I 'av mutation and a 18. Umemoto, S., and J. R. Sellers. 1990. Characterization of in vitro motility
403ArO' mutation. Circulation. 86:345-352. assays using smooth muscle and cytoplasmic myosins. J. Biol. Chem.
8. Nishi, H., A. Kimura, H. Harada, H. Toshima, and T. Sasazuki. 1992. 265:14864-14869.
Novel missense mutation in cardiac beta myosin heavy chain gene found in a 19. Homsher, E., F. Wang, and J. R. Sellers. 1992. Factors affecting move-
Japanese patient with hypertrophic cardiomyopathy. Biochem. Biophys. Res. ment of F-actin filaments propelled by skeletal muscle heavy meromyosin. Am. J.
Commun. 188:379-387. Physiol. 262 (Cell Physiol. 31):C714-723.
9. Solomon, S. C., J. A. Jarcho, W. McKenna, A. Geisterfer-Lowrance, R. 20. Liew, C. C., M. J. Sole, K. Yamauchi-Takihara, B. Kellam, D. H. Ander-
Germain, R. Salerni, J. G. Seidman, and C. E. Seidman. 1990. Familial hypertro- son, L. Lin, and J. C. Liew. 1990. Complete sequence and organization of the
phic cardiomyopathy is a genetically heterogeneous disease. J. Clin. Invest. human cardiac beta-myosin heavy chain gene. NucleicAcids Res. 18:3647-3651.
86:993-999. 21. Kron, S. J., Y. Y. Toyoshima, T. Q. P. Uyeda, and J. A. Spudich. 1991.
10. Schwartz, K., J. Beckmann, C. Dufour, L. Faure, F. Fougerousse, L. Assays for actin sliding movement over myosin-coated surfaces. Methods Enzy-
Carrier, C. Hengstenberg, D. Cohen, H.-P. Vosberg, A. Sacrez, et al. 1992. Exclu- mol. 196:399-416.

Skeletal Muscle Expression and Abnormal Function of #-Myosin in Hypertrophic Cardiomyopathy 2865