Differences between cardiac and skeletal muscle actins

Journal of Molecular and Cellular Cardiology ( 19%1) 13, 108 1- 1086 RAPID Differences between COMMUNICATION Cardiac and Skeletal Muscle Actins (Received 14 September 1981, accefited in revised form 7 October 1981) Actin is present in practically every living ceil and at the same time appears to be a highly conserved protein [8]. At present three categories of actin, cc, p and y, have been positively identified [9]. Of these, the tc category comprises actins from striated muscle fibres (skeletal and cardiac) as well as constituting a major component of aortic smooth muscle actin. Striated muscles possess actins of apparently identical molecular weight [JO], molecular charge [14] and other properties [II]. Cardiac and skeletal actins are therefore very similar but not identical. One per cent of the amino acid sequences differ between cardiac and skeletal muscles but these differences (a single residue inversion and two substitutions) involve no change in the overall charge of the actin monomer [14]. Furthermore, small differences were recently detected using the sophisticated radioimmunoassay technique [ 131. Recently, we reported that the dimensions of the lanthanide-induced skeletal actin tubes [7] vary with the radius of the lanthanide ion used [.5]. If the actin tube dimensions are sensitive to these small changes in cationic radii (101.5 to 114.3 pm) then tube dimensions may also be sensitive to other changes such as the amino acid changes described above. Accordingly, we measured the dimensions of tubes formed from cardiac and skeletal muscle actins and found significant differences. Furthermore, the data suggest that there may be differences between actins prepared separately from the atria and ventricles. Actin tubes were prepared as described previously [S, 71 from freshly purified (P raseodymium) G-actin by adding sufficient Pr(II1) to achieve a Pr:actin ratio of five or greater [7]. Figure 1 illustrates electron micrographs of actin tubes made from skeletal [Figure l(a)], atria1 [Figure l(b)] and ventricular [Figure 1 (c)] sources. These structures were cylindrial in solution [S] and were flattened by drying following the application of the negative stain (1 o/o uranyl acetate). By viewing the tubes at angles of 80” and 100” to the axis of the tube, two sets of repeating rows may be observed. The separation between these rows (5.5 nm) represents the long axis of the actin monomer [I, 61. Similarly, when the actin tubes are viewed along their long axis, finer striations, 3.3 nm apart, can be observed which represent the smaller of the projected monomer dimensions [I, 61. The degree of resolution in these structures (1.65 nm [I, 6]> is considerably higher than that observed in F-actin [15] and its aggregates [12]. Three measurements were made from electron micrographs of the actin tubes: (i) the angle, 8, subtended between the front and rear 5.5 nm wide rows; (ii) the number, N, of 5.5 nm wide rows/helical repeat; and (iii) the flattened width, D, of the actin tubes. Table 1 compares the skeletal muscle actin tube parameters (published elsewhere) [5] with those prepared from bovine atria and ventricles. The 022-2828/81/121081+06 $02.00/O 0 1981 Academic Press Inc. (London) Limited 3~2 1082 Rapid Communication: P. J. Finlayson FIGURE 1. and C. G. dos Remedios Card&c and Skeletal Muscle Actins 1083 comparison was made because the entire amino acid sequence is known for actin from both sources and because they are available in convenient quantities. A clear and significant difference was detected between the mean measurements of skeletal actin tubes compared with those prepared from the atria and ventricles. However, the difference between the actins from atria and ventricles was smaller and two of the parameters (0 and D) were not significantly different. As observed in a previous report [5] the number of rows/helical repeat was always even for all tubes. Measurement of the flattened tube width (D) has the disadvantage that it is dependent on the unavoidable variations in magnification inherent in routine electron microscopy. The relationship : D = 5.5 x N/2 (sin 8/2) is valid provided that the actin monomer dimension which determines the separation between the measured rows (i.e. 5.5 nm), is constant for both skeletal and cardiac actins. This assumption seems reasonable since Acanthamoeba (y type) has recently been shown to have the same projected dimensions as skeletal muscle actin [I]. Accordingly, the projected dimensions of the actin monomer were determined from optical transforms of the actin monomer lattice for actins isolated from skeletal, atria1 and ventricular muscle fibres. Measurements were made of the reciprocal lattice dimensions, a* and b*, and the angle between them, y*. The optical transforms were calibrated using a standard grating. The results, summarized in Table 1, demonstrate that there is no significant difference in the projected dimensions of the actin monomer for the three types of striated muscle fibres. Thus, we conclude that there are significant differences between skeletal muscle actin and ventricular actin, and between skeletal muscle actin and atria1 actin. Because the actin monomer dimensions appear to be constant, these differences probably reflect the way the monomers are arranged to form the actin tubes. Differences between atria1 and ventricular actins are more blurred and in the case of the actin tube pitch (0) and tube diameter (D), they are not significant. However, the lanthanide ions themselves do not apparently alter the conformation of the actin monomer [3]. Furthermore, the Pr(II1) can be removed and replaced by a Ca(I1) or Mg(I1) and the actin will polymerize normally [4]. An important outcome of these results is that very small differences in actin monomers can be easily and conveniently detected. These differences may reflect variations in the primary structure of these actins. For example, there arc four known [14] differences (one inversion and two substitutions) between skeletal and cardiac actins. The methods used in actin tube analyses involve routine electron microscopy of the negatively stained actin tubes. These procedures are therefore considerably less specialized than the radioimmunoassay [13], which can also detect FIGURE 1. Electxm micrographs of: (a) rabbit skeletal muscle actin tubes showing the parameters that were measured; (b) bovine atria1 tubes; (c) bovine ventricular actin tubes. Actin monomer was prepared from acetone powders of muscle as described by dos Remedios and Barden [I]. SDS PAGE using 7.5% acrylamide slab gels demonstrated that actin samples were at least 95% pure and that they migrated as a single band at protein loadings up to 50 pg. The ability of atria1 and ventricular actin to polymerize (~~a = 1.0) was equal to skeletal actin (qre,, = 1.0). Cardiac and skeletal actin have a similar ability to activate myosin ATPase activity. We observed that the maximum activation was achieved at an actin:myosin molar ratio of - 2. Actin tubes were formed by adding 10 mol Pr (in excess of the free ATP concentration) [4] /mol of actin. Actin tubes were stained using 1 o/0 uranyl acetate as described previously [5J and examined in a Philips EM 301 operated at 60 kV. Scale bar = 0.1 pm. TABLE 1. Actine tube parameters (N, 0, D) and unit cell dimensions N (number of rows per helical repeat) Skeletal actin Atria1 actin Ventricular actin 6 (angle in degrees between front and back rows) 13.0 f 0.2 (113) 19.7 f 0.2 17.3 f 0.3 (32) 25.4 5 0.5 (31) 16.2 -& 0.3 (40) 25.4 -& 0.6 (41) (116) (a *, b*, y*) D (flattened tube width in nm) obtained a* (nm) 211 f 0.4 (93) 6.5 f 194 * 1.2 (45) 195 f 1.1 (41) 6.4 f 0.04 (20) 0.06 from optical transforms b* (nm) 5.5 f 0.03 a*/b* Y” (7 1.19 -j= 0.006 (20) 5.4 f 0.05 (12) (12) 6.6 & 0.05 (11) 5.5 f 0.03 (11) 86.8 f 0.2. (20) 1.17 f 0.02 86.2 f 0.4 (12) 1.20 f 0.02 86.6 f 0.5 (11) All of the tube parameters were significantly different from each other at 1yO confidence levels by t tests, except in the case of cardiac tube widths and angles. a* and b* are the reciprocal projected monomer dimensions and y* is the angle between the two axes. t Tests of the optical transform data showed all differences to be not significant at 5% confidence levels. Mean of measurements + S.E.M. (number of measurements made). Cardiac and Skeletal Muscle Actins 1085 differences between skeletal and cardiac actins. Thus, the technique should have wide appeal. Although it was not stated, it is probable that the cardiac actin used in sequencing [la] was mostly derived from the ventricles since they form the bulk of the myocardium. It would therefore be of interest to sequence atria1 and ventricular actin. Currently, these studies are being e’xtended to other sources of actin whose sequences are known [14]. The classification [9] of actin isomers into CL, 8 and y categories was originally based on the separations obtained using two-dimensional gel electrophoresis and is now widely recognized. However, because of the differences detected by sequencing, radioimmunoassay methods as well as those described in this report, it is becoming increasingly clear that at least the OLcategory contains different isomeric forms. We propose to subdivide the actins into skeletal (ask) and cardiac (c&a) classes and it is possible that the cardiac category may also be subdivided into ventricular actin (c&v) and atria1 (aCa) types. Analyses of the 8 and y categories may reveal similar subclasses. Acknorerledgements We thank Dr Julian Barden and other members of the Muscle Research Unit for their helpful discussions. We are grateful to Mr Angelo Valois for technical assistance. This research was supported by grants from the National Health and Medical Research Council of Australia. Peter J. Finlayson was the recipient of a scholarship from the National Heart Foundation of Australia. Peter J. Finlayson and Cristobal G. dos Remedies* Muscle Research Unit, Department of Anatomy, The University of Sydney, Sydney 2006 New South Wales, Australia KEY~WORDS: Lanthanides; Cardiac actin; Skeletal actin; Actin tubes. REFERENCES I. 2. 3. 4. 5. 6. AEBI, U., SMITH, P. R., ISENBERG, G. & POLLARD, T. D. Structure of crystalline actin sheets. Nature 288, 296-298 (1980). BARDEN, J. A. & DOS REMEDIOS, C. G. Crystalline actin tubes. I. Is the conformation of the lanthanide-induced actin tube monomer more like F-actin than G-actin? Biochimica et biophysics acta 624, 163-173 (1980). CURMI, P. M. G. & BARDEN, J. A. Conformational studies of G-actin containing bound lanthanide. Proceedings of the Australian Biochemical Society 14, 36 (1981). DOS REMEDIOS, C. G. & BARDEN, J. A. Effects of Gd(II1) on G-actin: inhibition of polymerization of G-actin and activation of myosin ATPase activity by Gd-actin. Biochemical and Biophysical Research Communications 77, 1339- 1346 ( 1977). actin tubes. II. The DOS REMEDIOS, C. G., BARDEN, J. A. & VALOIS, A. A. Crystalline effect of various lanthanide ions on actin tube formation. Biochimica et biophysics acta 624, 174-186 (1980). DOS REMEDIOS,~. G.,BARDEN, J. A., VALOIS, A. A. & TULLOCH, P. A. Actin tube structures formed by Gd3+ and analy?ed by electron microscopy and electron diffraction. In Fibrous Proteins: Scientijic, Industrial and Medical Aspects, Volume 2, D. A. D. Parry & L. K. Creamer, Eds, pp. 23-32. London: Academic Press (1980). * Correspondence to: Associate Professor C. G. dos Remedios. 1086 7. 8. 9. 10. 11. 12. 13. 14. 15. Rapid Communication: P. J. Finlayson and C. G. dos Remedios DOS REMEDIOS, C. G. & DICKENS, M. J. Actin microcrystals and tubes formed in the presence of gadolinium ions. Nutare 276, 73 l-733 (1978). ELZINGA, M. & Lu, R. C. Comparative amino acid sequence studies on actins. In Contractile Systems in .Non-Muscle Tissues, S. V. Perry, A. Margreth & R. Adelstein, Eds, pp. 29-37. Amsterdam: Elsevier (1976). GARRELS, J. I. & GIBSON, W. Identification and characterization of multiple forms of actin. Cell 9, 793-805 (1976). KATAGIRI, T. Gel electrophoretic analysis of structural proteins of the normal canine and human hearts. Japanese Heart Journal 18, 705-710 (1977). MCCUBBIN, W. D. & KAY, C. M. Physicochemical studies on bovine cardiac globular actin. Biochimica et biophysics acta 214, 272-281 (1970). MOORE, P. B., HUXLEY, H. E. & DEROSIER, D. J. Three dimensional reconstruction of F-actin, thin filaments and decorated thin filaments. Journal of Molecular Biology 50, 279-295 (1970). MORGAN, J. L., HOLLADAY, C. R. & SPOONER, B. S. Immunological differences between actins from cardiac muscle, skeletal muscle and brain. Proceedings of the National Academy of Sciences,U.S.A. 77, 2069-2073 (1980). VANDEKERCKHOVE, J. & WEBER, K. The complete amino acid sequence of actins from bovine aorta, bovine heart, bovine fast skeletal muscle, and rabbit slow skeletal muscle. Dz&entiution 14, 123-133 A protein-chemical analysis of muscle actin differentiation. (1979). WAKABAYASHI, T., H&LEY, H. E., AMOS, L. A. & KLUG, A. Three dimensional image reconstruction of actin-tropomyosin complex and actin-tropomyosin-troponin T-troponin I. Complex. Journul of Molecular Biology 93, 477497 (1975).