SUBSTRUCTURE OF THE MYOSIN MOLECULE AS VISUALIZED BYELECTRON MICROSCOPY*

BY HENRY S. SLAYTER AND SUSAN LOWEY

THE CHILDREN S CANCER RESEARCH FOUNDATION, AND THE DEPARTMENTS OF PATHOLOGY AND

BIOLOGICAL CHEMISTRY, HARVARD MEDICAL SCHOOL, BOSTON

Communicated by John T. Edsall, August 7, 1967

The general topography of the myosin molecule has been elucidated by electronmicroscopy of shadow-cast preparations.'-3 The model obtained from these datais that of a globular head, about 200 A in diameter by 50 A high, attached to arodlike tail 1300 to 1400 A long by 20 A wide. This picture of the myosin mole-cule had been anticipated on the basis of physicochemical and X-ray diffractionstudies on myosin and its proteolytic fragments.4 From light-scattering meas-urements of the length and mass of the highly a-helical portions of myosin6 andX-ray diffraction investigations,7 it was concluded that the rod portion of myosinconsists largely or entirely of a two-stranded coiled coil.5 However, other physical-chemical studies of myosin have led to the proposal that myosin consists of threeidentical subunits.8-'0 Some substructure has been seen in the globular head ofthe myosin molecule by electron microscopy,2' 11 but these observations have beeninadequate to permit the determination of the number of subunits.

In the present study the use of improved techniques in specimen preparationhas made it possible to detect two subunits in the globular region of myosin. In-dividual subunits of similar size have also been isolated from preparations of myo-sin which were exposed to low concentrations of insoluble papain. These findingssuggest that the major portion of the myosin molecule consists of two polypeptidechains.

Protein Preparations. M\Ayosin was prepared as previously described.5 Theproteolytic fragments of myosin, heavy meromyosin and heavy meromyosin sub-fragment 1, were isolated from myosin which had been digested with insolublepapain. The papain suspension was prepared by coupling the enzyme withdiazotized p-amino benzyl cellulose.'2 Details of this procedure will be describedelsewhere. These protein preparations, at a concentration of about 1 to 2 mg/ml,were dialyzed exhaustively against 1 Ml ammonium acetate, pH 7.2, and then dilutedfor spraying onto mica.Shadow Casting.-Preparations of myosin and subfragments in 1 ill ammonium

acetate, pH 7.2, at a concentration of 0.05 mg/ml, were sprayed through a high-pressure spray gun at freshly cleaved mica. Specimens were evacuated to a pres-sure in the 10-7 Torr range and shadow-cast with platinum evaporated from atungsten filament, at a shadow to height ratio of 10: 1. Amounts of metal evapo-rated were minimized to permit optimal resolving power, consistent with adequatecontrast enhancement. Larger amounts of evaporated metal led to increased grainsize and resultant poorer resolution. The specimen was either held stationary orrotated through ten or more revolutions during the brief (10-sec) shadow-castingprocess. During evaporation the pressure rose only briefly to the middle of the10-6 Torr scale. A thin layer of carbon was evaporated on top of the platinumreplica. The preparation was removed from the shadow-casting apparatus and

1611

Dow

nloa

ded

by g

uest

on

Dec

embe

r 30

, 202

0

1612 BIOCHEMISTRY: SLAYTER AND LOWEY PROC. N. A. S.8 | | | - l 1 l | | || l I l WpF | | ||Nlt_ gS ! M s | s s s s g Zs fi | r R R R =d M _d | § p S S S s - l R - X X!X il _ ^ X B_ E i._ ! D_^ S S w g X X lS - ^ N NN | gSED S F _E N s ^_§ |gg gg________

qi ,jrwFf.sW..i_ea-g5iI;_0iF.A_____

-___R__k===___|__*||||__E_________________l

___l____

Y _______ s__.__E__ffi_wr.onso__ -__

_t.Sa_W/-_|_____________

I__

|w--R

FIG. 1.-SIYOSin molecules rotary shadow-cast with platillum. AIagnification 1052000X.

backed with 0.25 per cent collodion. Fragments of the dried replica films weremounted on 200-mesh copper grids.

Electron Microscopy. Electron microscopy was carried out on a Siemens 1A

Dow

nloa

ded

by g

uest

on

Dec

embe

r 30

, 202

0

VOL. 58, 1967 BIOCHEMISTRY: SLAYTER AND LOWEY 1613

FIG. 2.-Myosin molecules shadow-cast unidirectionally with platinum. Magnification 105,000X.

electron microscope, at a magnification of 27,000 X. M\icrographs were recordedsufficiently close to focus that metal grain at the 20 A level was clearly resolved onthe original plates. Plates were enlarged four to six times photographically using

Dow

nloa

ded

by g

uest

on

Dec

embe

r 30

, 202

0

1614 BIOCHEMISTRY: SLAYTER .4AD LOWEY PROC. N. A. S.

FIG. 3.-Top row: A composite of selected unidirectionally shadowed myosin. When shadowsare perpendicular to the line connecting the two heads, the cleft can often be seen. At otherangles the shadows can obscure structure. Second row: Rotary shadowed molecules with theaxis of the doublet perpendicular to the rod. Third and fourth rows: Rotary shadowed moleculeswith the axis of the doublet parallel to the rod. 'Magnification 175,000X.

a contact intermediate to reverse contrast, so that the shadowing metal wouldappear light. A 70-,4 objective aperture was used to enhance contrast.

Results. Myosin substructure: When myosin was prepared for electron micros-copy by the rotary shadowing technique, it was found that 60 per cent of themyosin molecules had bipartite heads, each lobe of which was approximately halfthe size of the single heads found in the remaining 40 per cent of the molecules

Dow

nloa

ded

by g

uest

on

Dec

embe

r 30

, 202

0

\OL. 58, 1967 BIOCHEMISTRY: SLAYTER AND LOWIEY 1601.

FIG. 4.-Sedimentation patterns of the water-soluible fraction ofthe papain digests. Solvent = 0.05 Al KCl, 0.005 AI phosphate,pH 7. Both runs are at 59,780 rpm, temperature = 20'C, barangle = 55'. Sedimentation proceeds to the left. Top: Proteinconcentration -0.55, prepared from 3% myosin digested 5 minmwith 0.0025 mg/ml insoluble papain; bottom: same sample digestedfurther with 0.3 mg/ml insoluble papain.

(Fig. 1). Measurements of the large, single heads indicated a diameter of about200 A, in agreement, with previously reported values.3' 11 Obviously, if heads withtwo lobes are shadowed unidirectionally, the one lobe could easily overshadow theother, with the net result of a large percentage of apparently single heads (Fig. 2).

Occasionally, there are hints of a cleft between lobes even in the unidirectionallyshadowed preparations, but the thicker, one-sided metal deposit tends to obscuresuch fine structure (Fig. 3, top row). The globular, bipartite head of myosinappears to possess considerable flexibility: Mlany of the molecules show the pairof lobes arranged parallel to the axis of the rod or, alternatively, perpendicular tothe axis (Fig. 3). It seems as if each of the small lobes may be attached by a singlestrand to the more rigid rod portion of myosin, and by virtue of this structure canassume a wide variety of positions. Thus, the subunits can lie adjacent to eachother or be separated by as much as 150 A. Previous estimates of head size haveprobably been confused by the diversity of arrangements of the two subunits, whichmay lead to a range of lengths or widths depending on the shadowing direction.It is more meaningful to consider the size of the individual lobes, rather than toattempt to define a discrete size for the globular region of myosin.

It is noteworthy that when the rod portion of mVosin is shadow-cast by therotary method, small kinks and bends are more apparent than when it is shadowedunidirectionally. Thus, there are obvious advantages to the rotary method forthe measurement of contour length in myosin.Papain fragmiients of Ryosin: The use of insoluble papain in the digestion of

myosin has provided a simple means for obtaining relatively undegraded fragmentsfrom myosin.'2 When myosin is digested briefly with extremely low concentrationsof papain (0.002 mg/ml), the water-soluble fraction of the digest shows two peaksin the ultracentrifuge: The leading peak has an intrinsic sedimentation constantof 7.2S, which is the value determined for heavy meromyosin (HMIMI). (Referto Fig. 7 for a pictorial explanation of the nomenclature used in this section.) Theslower peak has a value of 5.9S, which is the value commonly accepted for heavymeromyosin subfragment 1 (HAIM\J S-1), the enzymic portion of HME13' 14 (Fig.4, top). Electron micrographs of this preparation show that about one third ofthe particles have heads similar in appearance to those seen in myosin. The tailsof many of these HM\IA molecules are not observed, possibly due to breakage dur-ing preparation for electron microscopy. The remaining two thirds of the particles,

Dow

nloa

ded

by g

uest

on

Dec

embe

r 30

, 202

0

1616 BIOCHEMISTRY: SLAYTER AND LOWEY PROC. N. A. S.

FIG. 5. Top row: Early digest with papain showing HM1\ circled and HMMi S 1.Bottom row: Advanced digest with papain showing HM1\L S 1.

Magnification 150,000X.

Dow

nloa

ded

by g

uest

on

Dec

embe

r 30

, 202

0

\OL. 58, 1967 BIOCHEMISTRY: SLAYTER AND LOWEY 1617

FIG. 6.-Distributions of widths A Bof HMM S-1 and half heads of myo-sin. (A) Distribution of HMIM S-1 100particles present in the early papaindigest, excluding HMM particles(see Fig. 5, top). (B) Distributionof HMM S-1 particles in advanced 50 - 30digest (see Fig. 5, bottom). Virtually IIall the particles in the fields meas-ured are included in the histogram. I(C) Distribution of single lobes asmeasured in native myosin mole-cules. 20 100 180 20 100 100 180

WIDTH IN A

identified as HMM\Ii S-1, have the appearance of a single lobe of the pair character-istic of HMAI and myosin molecules (Fig. 5, top row). Distribution of diameters forthe HAIM S-1 particles and, for comparison, the individual lobes of bipartite headsof myosin molecules are shown in Figure 6A and C. The skewing seen in thesehistograms may be due to asymmetry of the lobes, or may simply be ascribed tolimitations in the assessment of particle size.

If more papain is added to the early digest, the HAIMI is completely convertedto the smaller fragment, HMM S-1 (Fig. 4, bottom). (The trace of slow-sediment-ing protein in the ultracentrifuge pattern probably represents the tail of HMMI, orHMMI S-2.) The corresponding electron micrographs show only single lobes(Fig. 5, bottom row). These particles are more uniform in size than those fromshorter digests, as shown by the histogram in Figure 6B.Conclusions.-By rotary shadowing myosin molecules, it has been possible to

detect two subunits in their globular regions. Each of these subunits has approxi-mately the same diameter as that found for the isolated HAIM S-1 molecule.There can be little doubt that the small, nearly spherical HMM S-1 fragments con-taining both the ATPase and actin combining sites of myosin are located in thehead of the myosin molecule.11 13, 14 The question of whether these two HMA\I S-1subunits are identical remains moot. On the basis of the electron micrographspresented here, as well as the earlier hydrodynamic and X-ray diffraction studies,

GHMM S-1---(120,000)

[ F--~~~~~~~HMMS-2-LMM (150,000) (60,000)

i HMM(340,000)

4 - MYOSIN (500,000)

FIG. 7.-Schematic representation of the myosin molecule.References to these molecular weights can be found in ref. 12.

Dow

nloa

ded

by g

uest

on

Dec

embe

r 30

, 202

0

1618 BIOCHEMISTRY: SLAYTER AND LOWEY PROC. N. A. S.

the myosin molecule can be depicted as consisting of two polypeptide chains. Thetwo supercoiled a-helical segments of the molecule stabilize each other by strongside-chain interactions along most of their length,7 while each chain terminates in aseparate globular conformation. Such a model is illustrated schematically in Figure7. This drawing is designed to identify the proteolytic fragments, and to show howthese fragments are related to the native myosin molecule. The molecular weightsgiven should only be considered as approximate values, since the size is dependent oilthe conditions of the proteolytic degradation.'2 The mass of the globular lobe seenin electron micrographs may be calculated from the most frequent diameter of

0 0

about 90 A, assuming that a correction for 20 A of shadowing metal is applicable.The value of about 150,000 so obtained is consistent with the molecular weight of120,000 which has been reported for HAIMI S-1.1lX-ray diffraction studies of living frog muscle'5 as well as X-ray diffraction and

electron-microscope studies of glycerinated insect flight muscle" have shown thatthe cross bridges between actin and myosin filaments undergo movement oil con-traction or in the transition from a relaxed state to rigor. These observations onwhole muscle imply that the portion of the myosin molecule involved in crossbridges must be capable of movement. In this study, we have demonstrated thatthe two globular subunits of myosin have the potential for a high degree of flexibility.Rotarv shadowing also serves to emphasize the bending in the rod region of myosin.Whether this apparent flexibility has relevance for the interaction between myosinand actin remains to be elucidated.

* This investigation was supported in part by research grants from the National Institutes ofHealth, U.S. Public Health Service, no. AMI-04762 National Institute of Arthritis and Metabolic1)iseases (to S. L.), no. GM1-14237 National Institute of General Medical Sciences (to H. S. S.),and from no. Fl1-05526, Division of Research Facilities and Resources, National Cancer Instituteto the Children's Cancer Research Foundation.

1 Rice, R. V., Biochim. Biophys. Acta, 52, 602 (1961); ibid., 53, 29 (1961).2Zobel, C. R., and F. D. Carlson, J. Mol. Biol., 7, 78 (1963).Huxley, H. E., J. Mol. Biol., 7, 281 (1963).Cohen, C., J. Polymer Sci., 49, 144 (1961).

5Lowey, S., and C. Cohen, J. Mol. Biol., 4, 293 (1962).6 lloltzer, A., S. Lowey, and T. Schuster, in The AMolecutlar Basis of .\eoplasia (Austin: Uni-

versitv of Texas Press, 1962), p. 239.*Cohen, C., and K. Holmes, J. Mlol. Biol., 6, 423 (1963).8 Small, P. A., W. F. Harrington, and W. W. Kielley, Biochimn. Biophys. Acta, 49, 462 (1961).9 Woods, E. F., S. IHimmelfarb, and W. F. Harrington, J. Biol. Chem., 238, 2374 (1963).10 Young, D. MI., S. Himmelfarb, and W. F. Harrington, J. Biol. Chem., 240, 2428 (1963).11 Rice, R. V., A. S. Brady, R. H. D)ePue, and R. E. Kelly, Biochemi. Z., 345, 370 (1966).12 Lowey, S., in Symposionn on Fibrous Proteins (Australia, 1967), in preparation.13 eller, H., and S. V. Perry, Biochenm. J., 80, 217 (1961).14 Mueller, H., J. Biol. Chem., 240, 3816 (1965).15 Huxley, H. E., W. Brown, and K. C. Holmes, Nature, 206, 1338 (1963).16 Reedy, AI. K., K. C. Holmes, and R. T. Tregear, Natufre, 207, 1276 (1963).

Dow

nloa

ded

by g

uest

on

Dec

embe

r 30

, 202

0