Heterogeneity of beta-type myosin isozymes in the human heart and regulational mechanisms in their expression. Immunohistochemical study using monoclonal antibodies

A single serine in the carboxyl terminus of cardiac essential myosin light chain-1 controls cardiomyocyte contractility in vivo

Although it is well known that mutations in the cardiac essential myosin light chain-1 (cmlc-1) gene can cause hypertrophic cardiomyopathy, the precise in vivo structural and functional roles of cMLC-1 in the heart are only poorly understood. We have isolated the zebrafish mutant lazy susan (laz), which displays severely reduced contractility of both heart chambers. By positional cloning, we identified a nonsense mutation within the zebrafish cmlc-1 gene to be responsible for the laz phenotype, leading to expression of a carboxyl-terminally truncated cMLC-1. Whereas complete loss of cMLC-1 leads to cardiac acontractility attributable to impaired cardiac sarcomerogenesis, expression of a carboxyl-terminally truncated cMLC-1 in laz mutant hearts is sufficient for normal cardiac sarcomerogenesis but severely impairs cardiac contractility in a cell-autonomous fashion. Whereas overexpression of wild-type cMLC-1 restores contractility of laz mutant cardiomyocytes, overexpression of phosphorylation site serine 195-deficient cMLC-1 (cMLC-1(S195A)) does not reconstitute cardiac contractility in laz mutant cardiomyocytes. By contrast, introduction of a phosphomimetic amino acid on position 195 (cMLC-1(S195D)) rescues cardiomyocyte contractility, demonstrating for the first time an essential role of the carboxyl terminus and especially of serine 195 of cMLC-1 in the regulation of cardiac contractility.

Separation of large mammalian ventricular myosin differing in ATPase activity

To investigate a possible heterogeneity of human ventricular myosin, papillary muscles of patients with valvular dysfunction were examined using a modified native gel electrophoresis. Myosin was separated into 2 components termed VA and VB, whereby the VA to VB proportion appeared to depend on the ventricular load. The proportion of the faster migrating band VA was correlated (P<0.05) with end-diastolic pressure and the aortic pressure-cardiac index product. The regression based on these variables accounted for 67% of the variation in VA (R2=0.67). The VA proportion was, however, not significantly correlated with cardiac norepinephrine concentration. The ATPase activity of the 2 components of myosin was assessed from the Ca3(PO4)2 precipitation by incubating the gel in the presence of ATP and CaCl2. The ATPase activity of VA was 60% of that of VB. The VA and VB forms were observed also in the cat (31.4% VA), dog (32.1% VA), pig (28.5% VA), wild pig (33.7% VA), and roe deer (30.5% VA). VA and VB were not detected in the rat exhibiting the 3 isoforms V1, V2, and V3, rabbit (100% V3), and hare (86% V1). The data demonstrate a heterogeneity of large mammalian ventricular myosin, whereby an increased cardiac load appeared to be associated with a higher myosin VA proportion that exhibited a reduced ATPase activity.

Structural and functional diversity of human ventricular myosin

The role of subcellular alterations in the process of heart failure remains ill-defined. Because contractile performance of failing heart muscle is depressed, possible alterations in the myosin molecule could be of particular relevance. There is increasing evidence that myofibrillar ATPase activity is reduced in congestive heart failure, whereas the findings on myosin ATPase are still controversial. The molecular causes of the reduced activity are currently not known. Because alpha-MHC is present only in small amounts in normal ventricles, a shift in favor of beta-MHC is of minor importance. Also immunohistochemical data on subspecies of beta-MHC seem not to provide an explanation. A new type of myosin heterogeneity was found by optimizing native polyacrylamide gel electrophoresis in the presence of pyrophosphate. Two bands (VA and VB) were observed in ventricles of patients with valvular disease. Because the two bands were detected also in normal hearts of large mammals, the existence of VA/VB cannot be diagnostic of diseased heart. However, the VA/VB ratio was influenced by the hemodynamic load, whereby the fast migrating band (VA) increased with the diastolic and systolic load. Because a relationship with the hemodynamic load was observed only in surgical muscle specimens, it appears that this heterogeneity is prone to post mortem modification. Further work is required to identify the molecular nature of this heterogeneity and to examine the therapeutic potential of a pharmacological modification of the VA/VB ratio.

Contractile protein alterations in heart failure

After nearly three decades of intense investigation, the precise cellular mechanisms underlying impaired contractility in heart failure remain to be defined. Nevertheless, growing use of the tools of molecular biology promises new insights into how alterations of contractile proteins mediate the functional derangements of failure.

Specificity of autoantibodies to histone H1 in SLE: relationship to DNA-binding domains

Autoantibodies in the sera of patients with systemic lupus erythematosus were examined with respect to their specificity for proteolytic fragments of histone H1 that retain, or do not retain, DNA-binding domains. 16 of 31 sera contained IgG and IgM antibodies to histone H1. IgM antibodies to H1 in 8 sera (50%) were directed at 18 kD and 20 kD alpha-chymotrypic H1 fragments that bore binding sites for DNA, as identified by staining immunoblots containing the fragments with ssDNA plus 6/0, a mouse monoclonal antibody against ssDNA, IgM with this type of histone H1 specificity did not react with comparably-sized V8 protease fragments of H1. IgM antibodies to H1 in the other patients were directed against entirely different epitopes which were preserved in V8 protease digests of H1. In serial studies of three patients during different phase of their SLE, the level of antibodies against the 18 kD and 20 kD histone H1 fragments varied in parallel with the level of anti-ssDNA antibodies in one and varied inversely in the other two. The data suggest that a significant proportion of autoantibodies to histone H1 are directed at a limited number of epitopes localized to H1 fragments containing DNA-binding sites.

The complete sequence of the human beta-myosin heavy chain gene and a comparative analysis of its product

We have isolated and sequenced the gene and the cDNA coding for the human cardiac beta-myosin heavy chain (designated MYH7). The gene is 22,883 bp long. The 1935 amino acids of this protein (Mr223,111) are encoded by 38 exons. The 5' untranslated region (86 bp) is split by two introns. The 3' untranslated region is 114 bp long. Three Alu repeats were identified within the gene and a fourth one in the 3' flanking intergenic region. The molecular organization of this gene reflects the conservative pattern with respect to size, coding ratio, and number or position of introns characteristic of vertebrate sarcomeric myosin heavy chain genes. The protein sequence of the human beta-heavy chain was compared with corresponding (homologous) sequences of rabbit, rat, and hamster as well as with the (heterologous) embryonic heavy chain sequences of rat, chicken, and man. The results show that protein subregions responsible for basic functions of myosin heavy chains (nucleotide binding and actin binding) are very similar in homologous and heterologous heavy chains. Regions that differ in their primary sequences in heterologous heavy chains appear to be highly conserved within mammalian beta-myosin heavy chains. Constant and variable subregions of heavy chains are discussed in terms of functional significance and evolutionary relatedness.

Infarcted heart uptake and biodistribution of radiolabelled anti-myosin monoclonal antibody in rat and dog myocardial infarct models

A new mouse monoclonal antibody that recognizes alpha- and beta-heavy chains of human atrial and ventricular myosin and beta-heavy chain of human slow skeletal muscle myosin was obtained. The 125I- and 111In-labelled antibody, and its F(ab')2 and Fab fragments localize in isoproterenol induced infarcted rat heart, with the F(ab')2 fragment showing the highest uptake. Comparison with 99Tc-pyrophosphate uptake in infarcted dog heart, induced by selective obstruction of a coronary artery, suggest that the 111In-labelled F(ab')2 localizes specifically in infarcted myocardium only.

Shortening velocity and myosin and myofibrillar ATPase activity related to myosin isoenzyme composition during postnatal development in rat myocardium

The relation between functional properties of the contractile apparatus, such as shortening velocity and ATPase activity, and myosin isoenzyme composition was studied in ventricular myocardium of adult (60-90-day-old) rats and of newborn (3-day-old) and young (10- and 20-day-old) rats. In adult animals, variations of isomyosin pattern were produced by reducing food intake and by changing the thyroid state. Hyperthyroidism was induced with triiodothyronine daily injection for 15 days; hypothyroidism was induced with iodine-free diet and KClO4 in drinking water for 50-60 days. The following parameters were studied: 1) calcium-magnesium-activated and magnesium-activated ATPase activity of washed and purified myofibrils, 2) calcium-activated ATPase activity of purified myosin, 3) isomyosin composition and relative content of alpha-myosin heavy chains (alpha-MHCs), and 4) force-velocity curve of left and right ventricle papillary muscles. To take into account the difference in excitation-contraction coupling between newborn and adult myocardium, the determination of the force-velocity curve was repeated in Krebs' solution with normal [CaCl2] (2.5 mM) and in Krebs' solution with high [CaCl2] (10 mM). During postnatal growth, the relative content of alpha-MHC increased and reached a maximum at about 20 days. Pronounced increases of myofibrillar and myosin ATPase activity and in shortening velocity occurred during the same period. In adult hyperthyroid rats, alpha-MHC content as well as enzymatic activity and shortening velocity were higher than in control adult animals. Hypothyroidism and food deprivation caused a decrease of alpha-MHC content and a reduction of both enzymatic activities and shortening velocity. The study of the relations between alpha-MHC relative content and functional parameters showed that 1) in ventricular myocardium of adult rats a linear relation existed between alpha-MHC content and myosin and myofibrillar ATPase activity and shortening velocity, and 2) in newborn and young rat ventricular myocardium, both enzymatic activities and shortening velocity were lower than would have been expected on the basis of the linear relation described above. This latter observation could be accounted for by a variation in specific activity of myosin during postnatal development or by the presence of peculiar isomyosins that cannot be detected with usual electrophoretic techniques.

Cardiac beta myosin heavy chain diversity in normal and chronically hypertensive baboons

We have identified two distinct beta-myosin heavy chains (MHCs) present in baboon myocardium by electrophoresis in gradient pore gels and by Western blots with anti-MHC MAb. The two beta-MHCs have molecular masses of 210 and 200 kD and share several antigenic determinants including an epitope recognized by a beta-MHC-specific MAb. A fivefold increase in the level of the 200-kD beta-MHC was observed in the hypertrophied left ventricles of baboons with chronic (5.3 +/- 0.7 yr) renal hypertension. A 60% increase (P less than 0.01) in BP and a 100% increase (P less than 0.001) in left ventricular mass to body weight ratio occurred in hypertensive baboons compared with normotensive animals. The Ca2+-activated myosin ATPase activity in hypertrophied left ventricles was decreased by 35% (P less than 0.05) compared with controls. Normal levels of the 200-kD MHC were detected in the right ventricles and intraventricular septa of the hypertensive animals. These data suggest that cardiac MHCs of primates may exist in alternative molecular forms that are indistinguishable by nondenaturing gel electrophoresis and that increased concentration of a second beta-MHC is associated with ventricular hypertrophy (r = 0.55). The functional significance and mechanisms that control the concentration of beta-MHC subspecies remain to be determined.

Isolation and characterization of the complete human beta-myosin heavy chain gene

The entire gene coding for the human beta-myosin heavy chain has been isolated from genomic EMBL3A phage libraries by chromosomal walking starting from clone gMHC-1, reported earlier (Appelhans and Vosberg 1983). gMHC-1 has been shown to carry coding information for the C-terminal two-thirds of beta-myosin heavy chain, which is expressed in cardiac muscle and in slow skeletal muscle fibers (Lichter et al. 1986). Three DNA clones were identified as overlapping with gMHC-1 by restriction mapping and DNA sequencing. They span a 30-kb region in the genome. About 22 kb extend from the initiation codon ATG to the poly(A) addition site. The clones include about 4 kb of 5' flanking sequences upstream of the promoter. Comparisons of beta- and alpha-myosin heavy chain sequences indicate that gene duplication of the cardiac myosin heavy chain isogenes preceded the mammalian species differentiation.

Expression of myosin in atrial areas of the bovine myocardium

1. A comparison of myosins from defined areas of the bovine atrial myocardium was performed by measuring Ca2+-ATPase activity and electrophoretic separation of myosin light chains. 2. Some areas of atrial myocardium contained myosin with slightly higher ATPase activity than others. 3. There were also clear differences in the amount of one ventricular light chain of myosin in defined regions of atrial myocardium. 4. No close relationship existed between the expression of ventricular and atrial myosin light chains and myosin ATPase activity.

Distribution pattern of alpha and beta myosin in normal and diseased human ventricular myocardium

All fibers in three normal, four dilated, and two ischemic human ventricles were classified according to their myosin content using three sets of monoclonal antibodies each specific for one myosin heavy chain isoform (alpha, beta and beta'). Numerous fibers contained only beta myosin heavy chain (denoted as beta fibers), others contained either alpha and beta, or beta and beta' myosin heavy chain (denoted as alpha beta and beta beta' fibers, respectively). The percentages of alpha beta fibers were systematically determined along the walls of seven homologous regions of the ventricular myocardium. In all ventricles, there was an alpha beta-fiber transmural gradient, with less alpha beta fiber in the subendocardium than in the subepicardium. More alpha beta fibers were found in the right than in the left ventricular wall but there was no difference between the mid-portion and the apex of the free wall of each ventricle. The diseased ventricles contained a lower alpha beta fiber percentage than the normal hearts. beta beta' fibers were very rare in the normal ventricles (less than 5%) and almost inexistent in pathological hearts. The correlation between the mean alpha beta fiber percentages of the diseased hearts and their cardiac indices (r = 0.88, P less than 0.05) suggests that the small amount of alpha myosin distributed in a large number of ventricular fibers could play a role in the contractile performance of the heart. In conclusion, this study provides evidence for 1) an alpha beta fiber transmural gradient, and 2) a lower alpha myosin ratio in diseased than in normal human ventricle.