Creatine kinase isozyme expression in embryonic ... - Springer Link

Anatomy and Embryology

Anat Embryo1 (1989) 179:387 393

9 Springer-Verlag 1989

Creatine kinase isozyme expression in embryonic chicken heart Wouter H. Lamers 1, Willie J.C. Geerts I, Antoon F.M. Moorman 1, and Robert P. Dottin 2 1 Department of Anatomy and Embryology, University of Amsterdam, Meibergdreef t 5, 1105 AZ Amsterdam, The Netherlands 2 Department of Biological Sciences, Hunter College, CUNY, 695 Park Avenue, New York, NY 10211, USA

Summary. The distribution pattern of creatine kinase (EC 2.7.3.2) isozymes in developing chicken heart was studied by immunohistochemistry. Creatine kinase M, which is absent from adult heart, is transiently expressed between 4 and 11 days of incubation. During that period, numerous muscular cells in the roof and septum of the atrium, in the interventricular septurn and on top of the trabeculae cordis and at the rim of the outflow tract stain strongly with a polyclonal antibody that is specific for the M subunit. In the ventricle and outflow tract, the M-positive cells are found mainly subendocardially and in the right half, at the transition of conducting and working myocytes. Creatine kinase B, which is the predominant adult isozyme, is initially expressed to a high concentration in a small group of disperse myocardial cells in the upstream part of the inflow tract. When compared to the expression pattern of cardiac myosin heavy chains, the observed creatine kinase expression pattern suggests that M-positive cells are mainly found in areas that participate in the formation of cardiac conductive tissue, whereas B-positive cells are first found in areas that are involved in the generation of cardiac rhythm. Key words: Distribution pattern - Creatine kinase isozymes - Embryonic chicken heart - Immunohistochemistry

Indroduetion

A very economical mechanism for energy transfer may be required for several energy consuming processes in excitable tissues, such as muscle and neural tissue (for a discussion: see Kammermeier 1987b). Recent experiments and calculations show that the phosphocreatine shuttle between mitochondria and cytosolic microcompartments represents an intracellular mechanism to assure a high efficiency of energy transduction via ATP. The shuttle mechanism requires the presence of creatine kinases (CK, E.C. 2.7.3.2) at the site of ATP production (" mitochondrial" CK) and at the site of consumption ("cytosolic" CK) (for a review, see Bessman 1985, 1987). In all species two different but related cytosolic gene products, designated '~ (for muscle) and " B " (for brain) and one unrelated, mitochondrial gene product (designated " M i " ) are found (for a review, see Babbitt et al. 1986). The enzyme exists as a dimer. In the cytosol 3 different isozyme are found, MM, BB and the Offprint requests to: W.H. Lamers

hybrid MB. The MM isozyme is prevalent in mature skeletal muscle and mammalian myocardium, whereas the BB isozyme is prevalent in neural tissue and avian myocardium (Eppenberger et al. 1967). Only adult mammalian myocardium contains appreciable amounts of the MB isozyme (Roberts et al. 1975). Creatine kinase enzyme levels increase relatively late in organ development (Ziter 1974; Hall and De Luca 1973; Ingwall etal. 1981; Eppenberger etal. 1964, 1983; Robinson 1987; Perriard et al. 1987). Initially all organs that contain CK, express low levels of the BB isozyme. In avian and mammalian skeletal muscle and in mammalian heart, CK BB is subsequently replaced by the CK MM isozyme. The hybrid enzyme CK MB is formed during the transitional period (Ziter 1974; Hall and DeLuca 1973; Perriard et al. 1978). Despite pronounced changes in isozyme expression with development, very few studies have focussed on the spatial distribution of the expression patterns of the respective isozymes during the transitional period. We have recently studied the distribution patterns of atrial and ventricular myosin heavy chains in developing chicken heart (Sanders et al. 1986; de Groot et al./987; de Jong et al. 1987). These studies showed that prior to its morphological differentiation, the developing conductive tissue is characterized by a coexpression of atrial and ventricular myosin heavy chains. In view of our interest in the origin of the cardiac conductive tissue, it appeared of interest to look at the development of the distribution pattern of CK isozymes in embryonic chicken hearts.

Materials and methods Animals

White Leghorn chicken embryos of 3-13 days of incubation were staged according to Hamburger and Hamilton (1951) and fixed at room temperature for 2 h in a mixture of methanol, acetone, acetic acid and water (9:9:2:4 by vol.). After dehydration in dimethoxypropane and embedding in paraplast (Paraplast Plus, Lancer, Oxford), serial sections of the embryos were prepared for immunohistochemistry. Antibodies

The isolation of the creatine kinase BB and MM isozymes from chicken heart (BB) and pectoralis muscle (MM), respectively, and the preparation and characterization of the polyclonal antibodies have been described (Schweinfest

388

389 et al. 1982; Kwiatkowsky et al. 1985). In cultures of developing skeletal myocytes, the CK-B antibody recognizes, in addition to the CK-B enzyme, a protein of approx. 35 Kd. To exclude spurious staining patterns in the heart due to this contaminant, sections were incubated with antibody that was absorbed with a serial dilution of purified heart CK BB protein (Dawson and Eppenberger 1970). Since only minor contaminations are observed in the purified CK BB preparation, one may assume a difference in the ratio of the concentration of CK-B and contaminant and in the ratio of the concentration of their respective antibodies. Hence, one would expect extinction of the CK-B staining pattern at lower concentrations of added CK BB proteins than one would expect extinction of the staining pattern of the contaminant. Such a change in the staining pattern was not observed. Immunohistochemistry

Antibody reactions were performed according to the indirect, unconjugated peroxidase-antiperoxidase method (Sternberger 1986) with 3,3'-diaminobenzidine as a substrate on 7 gm serial sections that were affixed to chrome aluminum-treated, gelatin-coated slides and that were pretreated with 3% H2Oz in methanol. Serial sections were incubated with 1:1000 and 1:2000 dilution of the primary antibodies and 1:1000 of the corresponding pre-immune serum. The stained sections were illuminated through a green filter (Zeiss) and registered on Agfa-Pan (25 ASA) film. This procedure strongly enhanced contrast in the stained sections but, as a consequence, sacrificed to some extent the differential staining intensities that were observed with the CK-B antiserum. Results

In the youngest stages examined the CK antibodies mainly stain neural and muscular tissue. This is illustrated in Fig. 1 for an embryo of 4 days of incubation (Hamburger and Hamilton stage 24). To demonstrate the specificity of the CK-B and CK-M antibodies, serial sections were incubated. The CK-B antibody stains neural and muscular tissue, e.g. the neural tube (Fig. 1 A), the neural retina (Fig. 1 C), the myotomes (Fig. I A ) and parts of the heart, whereas the CK-M antibody stains muscular tissue only, i.e. the myotomes (Fig. 1 B) and parts of the heart. Unfortunately, the CK-B antibody gives an appreciably higher background than the CK-M antibody, especially in dense tissue. In the heart, the anterior (cranial) part of the outflow tract (Fig. 1 G), the top of the trabeculae cordis (Fig. 1 H) and the atrial roof (Fig. 1 G, H and N) contain numerous muscular cells that stain strongly with the CK-M antibody. By contrast, the CK-B antibody shows a strong reaction in a small group of cells at the inflow tract of the heart, but stains only weakly in the outflow tract and the trabecu-

lae (Fig. 1 F, I). This small group of cells (Fig. 1 I-L) may represent the sinoatrial node: they stain positive with antibody against atrial myosin heavy chain (Fig. 1 K) but, in contrast to the rest of the inflow tract, not with antibody against ventricular myosin heavy chain (Fig. 1 L) (see also : Discussion). Erythrocytes stain rather intensely with the anti-B antibody. However, no CK enzyme activity could be detected in isolated embryonic erythrocytes. The staining pattern that is found in 4-day old chicken embryos does not change during the rest of the embryonic period. In embryos of 7 days incubation (Hamburger and Hamilton stage 31), intensive staining with the CK-M antibody is found at the rim of the anterior (pulmonary) and to a much lesser extent around the posterior (aortic) part of the muscular outflow tract (Fig. 2 B), in the ventricular septum and trabeculae (Fig. 2D) and in the atrium (Fig. 2B, D). In addition, a few isolated cells stain at the inflow tract (Fig. 2B, D, H). Staining with the CK-B antibody is most intense at the inflow tract, around the venous entrance (Fig. 2A, C, E). Typically, these cells are not organized in the coherent muscular architecture of the rest of the myocardium, but lay dispersed, embedded in interstitial tissue. Less intense staining with CK-B antibody is observed at the rim of the outflow tract and in the interventricutar septum and trabeculae, i.e. in areas that are also strongly positive for the CK-M subunit. Interestingly, such double staining is only observed in the left half of the ventricle and the outflow tract, but not in the right half of the ventricle and the atrium (cf. Fig. 2 F, I, and 2 G + J). At the upstream part of the inflow tract only cells are found that are positive for CK-B (cf. Fig. 2E, H). Further downstream, near the venous entrance into the atrium, the number of double staining cells increases (cf. Fig. 2C, D). The relation of the CK-M positive cells to the developing ventricular conduction system can be most clearly illustrated in the 9.5 days embryo (Hamilton and Hamburger stage 36, Fig. 3). In Fig. 3A, B, a section through the atrioventricular bundle is shown. Whereas the bundle itself stains minimally, myocardial cells around the club-shaped end of the atrioventricular bundle stain strongly positive with the CK-M antibody (Fig. 3 E, F). The positive staining cells are interlaced with CK-M negative myocardial cells. Towards the apex, the positively staining cells are found almost exclusively in the subendocardial myocytes of the right ventricle (Fig. 3D). The CK-M positive cells loose their stainability upon the completion of the embryonic period (i.e. Hamburger and Hamilton stage 37-38; cf. Butler and Juurlink, 1987). Discussion Specificity o f the CK-isozyme staining reactions

CK-B and CK-M proteins are homologous, especially in the carboxyterminal 80% of the protein (see for a review

Fig. 1A-N. Creatine kinase isozyme expression in the 4-day chicken embryo (Hamburger and Hamilton stage 24). Serial sections were incubated with antibodies either against creatine kinase BB isozyme (A~ C, E, F, I, M) or against the creatine kinase MM isozyme (B, D, G, H, J, N). Specificity of staining with the respective antibodies is shown for the neural tube (A, B), the myotomes (A, B), the neural retina (C, D), the outflow tract of the heart (E, G), the ventricle an atrium (F, H). M, N show details of the roof of the atrium and I, J show details of a group of cells at the upstream end of the inflow tract. The staining pattern of these cells with anti-atrial myosin heavy chain (K) and anti-ventricular myosin heavy chain (L) are also shown to adstruct the special characteristics of these cells. Abbreviations. nt neural tube; rn, myotome; nr, neural retina; ao, aorta; l, lens; oft, outflow tract; a, atrium; v, ventricle; tr, trabeculae (arrows); san, sinoatrial node. Bar =0.2 mm

390

391

Fig. 3A-F. Creatine kinase isozyme expression in the 9.5 day chicken embryo (Hamburger and Hamilton stage 36). Serial sections were incubated with antibodies either against the creatine kinase BB isozyme (A, C, E) or against the creatine kinase MM isozyme (B, D, F). Specificity of staining with the respective antibodies is shown for the ventricular conductive system only: the CK-M subunit is expressed in the subendocardial myocytes of the right ventricle (D, arrows) from the basis of the atrioventricular bundle (B, F) downward. Abbreviations. avb, atrioventricular bundle; my, mitral valve; rv, right ventricle; lv left ventricle; ivs interventricular septum. Bar = 0.2 mm

Fig. 2A-J. Creatine kinase isozyme expression in the 7-day chicken embryo (Hamburger and Hamilton stage 31). Serial sections were incubated with antibodies either against the creatine kinase BB isozyme (A, C, E, F, G) or against the creatine kinase MM isozyme (B, D, H, I, J). Specificity of staining with the respective antibodies is shown for the outflow tract (A, B), the atrium (A-F, H, I), the upstream part of the inflow tract (A, B, E, H), the downstream part of the inflow tract (C, D) and the ventricle (C, D, G, J). Abbreviations: oft, outflow tract; sh, sinus horn; a atrium; ao, aorta; p semilunar valves of pulmonary trunc; sv sinoatrial valves; av, atrioventricular valves; ias, interatrial septum; ivs interventricular septum. Bar = 0.2 mm

392 of sequences: Babbitt et al. 1986). It is therefore not surprising that the CK-B and CK-M antibodies are only specific in immunodiffusion, immunoadsorption or immunoprecipitation assays (Perriard eta1. 1978; Kwiatkowski etal. 1985), i.e for the native isozymes. As can be observed in the photographs, there is only a partial overlap between the distribution of CK-B and CK-M subunits in the embryo, as identified with the respective antibodies. Thus, the CK-M antibody does not stain the neural tube (Fig. 1 B) nor the majority of the CK-B containing myocardial cells in the upstream part of the inflow tract (Figs. 1J, 2H). The CK-B antibody does not stain the atrial myocytes (Figs. I M ; 2E, F) and the subendocardial CK-M containing myocytes of the right ventricle (Figs. 2 G, 3 C, E). The most likely explanation of these observations is that the conformation of the CK isozymes is sufficiently maintained to assure specific staining. Based on these arguments, those areas that stain exclusively with one antibody should contain the homodimer isozyme (BB or MM), whereas those that stain with both antibodies can contain combinations of BB and MM and/or the heterodimer MB.

Relation of the CK-isozyme distribution to that of myosin heavy chain isozymes Various energy-driven processes have been found to be directly dependent on the free energy of ATP hydrolysis rather than on the ATP concentration (Kammermeier 1987a). A primary benefit of the colocalization of a phosphocreatine-associated ATP-regenerating system with ATP-consuming processes in functionally dedicated microcompartmerits may therefore be the limitation of the free energy change due to diffusion problems (Kammermeier 1987b). Whereas a heterogeneous intracellular distribution of creatine kinase appears to be a direct consequence of its postulated function, the presently observed intercellular heterogeneous distribution of creatine kinase isozymes in developing heart may identify functionally different myocardial cells. In particular, the presence of the M-isozyme is unexpected in view of its virtual absence in adult chicken heart (Eppenberger et al. 1967). When comparing the 3-dimensional distribution patterns of CK isozymes (present study) with that of myosin heavy chain isozymes (Sanders et al. 1986; De Groot et al. 1987; De Jong et al. 1987), a number of interesting similarities emerge. Especially (the developmental changes in) the distribution patterns of CK-M and the areas of double expression of atrial and ventricular isomyosins initially resemble and subsequently are contiguous with each other. Furthermore, an interesting combination of CK BB isozyme expression and atrial myosin mono-expression is found in a group of cells at the inflow tract.

CK-M. Both CK-M and atrial plus ventricular isomyosin co-expression are found in the anterior (cranial) part of the outflow tract at 4 days of incubation (Fig. I ; cf. Fig. 5 of Sanders et al. 1986), but not in the posterior (caudal) part. Although at 7 days CK-M expression is also seen in a small ring at the posterior (aortic) part of the outflow tract, its extension remains smaller than that in the anterior (pulmonary) part, and does not extend into the lesser curvature, as isomyosin co-expression does (Sanders et al. 1986). Similarly CK-M (Fig. 1) and both isomyosins are expressed in the trabeculae and the interventricular septum.

However, already in the 7-day chicken embryo double expression of both isomyosins is located closer to the top (free margin) of the interventricular septum (cf. Fig. 4 of De Groot et al. 1987) than the expression of CK-M (Fig. 2). The reason for this difference in localization becomes clear in the 9.5-day embryo: whereas isomyosin coexpression is confined to the atrioventricular bundle (Fig. 3 of Sanders et al. 1986), CK-M expression is found in myocardial cells around the end of the atrioventricular bundle and continues around the lumen of mainly the right ventricle (Fig. 3). To a certain extent, a contiguity in the distribution pattern of CK-M and isomyosin coexpression can also be found in the atrium: whereas isomyosin coexpression is initially confined to the sinoatrial and atrioventricular junction (cf. Fig. 2 of De Groot et al. 1987), CK-M is expressed in the atrial roof (Fig. 1). Subsequently, CK-M expression becomes concentrated in the interatrial septum (Fig. 2), whereas isomyosin coexpression is still found at the sinoatrial and atrioventricular junctions and at the junction of the interatrial septum and the endocardial cushion (cf. Fig. 4 of De Groot et al. 1987). Our previous studies of the developmental changes of isomyosin expression have made it plausible that isomyosin co-expression identifies the developing ventricular (Sanders et al. 1986; De Jong et al. 1987) and possibly also the atrial (De Groot et al. 1987; De Jong et al. 1987) conducting system. CK-M therefore appears to be transiently expressed in ceils at the transition of "conducting" and "working" myocardial cells (cf. Vassall-Adams 1978). The functional significance of this observation, especially the right sided dominance in both ventricle and outflow tract, remains to be established.

CK-B. Strong mono-expression of the CK-B subunit, i.e. expression of the CK BB isozyme, is only found in a group of disperse myocardial cells in the upstream part of the inflow tract (Figs. 1, 2). In contrast to the adjacent cells of the sinoatrial junction, these cells were found to express only the atrial isomyosin (Fig. 1K, L, cf. De Groot et al. 1987). Atrial myosin monoexpression may be a characteristic of pacemaker cells (for a discussion, see De Jong et al. 1987). Since the conduction system may initially have both neurogenic and myogenic properties (Lamers et al. 1987), it is tempting to speculate that the simultaneous expression of only the atrial isomyosin and the CK BB isozyme identifies pacemaking cells. In addition to these typical structural properties, the cells also have the correct location and distribution for chicken nodal cells (Gossrau 1969). Acknowledgements. This research was supported in part by NIH grant AM 34479 to RPD. We thank Dr. R. Charles and Mr. F. de Jong for carefully reading the manuscript. We are indebted to Mr. C. Hersbach and Mr. A.A. van Horssen for art work and to Mr. R. ten Hagen for typing the manuscript. References

Babbitt PC, Kenyon GL, Kuntz LD, Cohen FE, Baxter JD, Benfield PA, Buskin JD, Gilbert W, Hauschka SD, Hossle JP, Ordahl CD, Pearson ML, Perriard JC, Pickering L, Putney S, West BL, Zivin RA (1986) Comparison of nine creatine kinase primary structures : implications for structure-activity relationships. J Protein Chem 5:1-14 Bessman SP (1985) The creatine-creatine phosphate energy shuttle. Ann Rev Biochem 54:831-862

393 Bessman SP (1987) The creatinephosphate energy shuttle - The molecular asymmetry of a "pool". Anal Biochem 161 : 519-523 Butler H, Juurlink BHJ (1987) An atlas for staging mammalian and chicken embryos. CRC Press, Boca Raton, FL Dawson DM, Eppenberger HM (1970) Creatine kinase. Methods Enzymol 17:995-1002 De Groot IJM, Sanders E, Visser SD, Lamers WH, De Jong F, Los JA, Moorman AFM (1987) Isomyosin expression in developing chicken atria: a marker for the development of conductive tissue? Anat Embryol 176:515-523 De Jong F, Geerts WJC, Lamers WH, Los JA, Moorman AFM (1987) Isomyosin expression patterns in tubular stages of chicken heart development: a 3-D immunohistochemical analysis. Anat Embryol 177: 81-90 Eppenberger HM, Eppenberger M, Richterich R, Aebi H (1964) The ontogeny of creatine kinase isozymes. Dev Biol 10:1-10 Eppenberger HM, Dawson DM, Kaplan NO (1967) The comparative enzymology of creatine kinases. J Biol Chem 242:204-209 Eppenberger HM, Perriard JC, Wallimann T (1983) Analysis of creatine kinase isozymes during muscle differentiation. In: Rattazi MC, Scandalios JG, Whitt GS (eds) Isozymes: Current topics in biological and medical research, vol 7. New York, Alan R Liss, pp 19-38 Gossrau R (1969) Topographische und histologische Untersuchungen am Reitzleitungssystem yon V6geln. Z Anat Entw Gesch 128:163-184 Hal1 N, DeLuca M (1975) Developmental changes in creatine phosphokinase isoenzymes in neonatal mouse hearts, Biochem. Biophys Res Commun 66:988-994 Hamburger V, Hamilton HL (1951) A series of normal stages in the development of the chicken embryo. J Morphol 88:49-92 Ingwall JS, Kramers MF, Friedman WF (1981) Developmental changes in heart creatine kinase. In: Jacobus WE, Ingwall JS (eds) Heart creatine kinase. William and Wilkins, London, pp 9-17 Kammermeier H (1987a) High energy phosphate of the myocardium: concentration versus free energy change. Basic Res Cardiol 82 [suppl 2] : 31-36 Kammermeier H (1987b) Why do cells need phosphocreatine and a phosphocreatine shuttle ? J Mol Cell Cardiol 19 : 115-118

Kwiatkowski RW, Ehrismann R, Schweinfest CW, Dottin RP (1985) Accumulation of creatine kinase mRNA during myogenesis : molecular cloning of a B-creatine kinase cDNA. Dev Biol 112:84-88 Lamers WH, te Kortschot A, Los JA, Moorman AFM (1987b) Acetylcholinesterase as a marker for the early development of rat cardiac conductive tissue. Anat Rec 217:361-370 Perriard JC, Caravatti M, Perriard ER, Eppenberger HM (1978) Quantitation of creatine kinase isoenzyme transitions in differentiating chicken embryonic breast muscle and myogenic cell cultures by immunoadsorption. Arch Biochem Biophys 191:90-100 Perriard JC, Eppenberger HM, Hossle JP, Sch/ifer B (1987) Genes and proteins of chicken creatine kinase isozymes : developmental regulation and functional significance. In: Rattazi MC, Scandalios JG, Whitt GS (eds) Isozymes: current topics in biological and medical research, vol 14, New York, Alan R Liss, pp 83-101 Roberts R, Henry PD, Sobel BE (1975) An improved basis for enzymic estimation of infarct size. Circulation 52 : 743-754 Robinson MM (1987) Fixation and immunofluorescent analysis of creatine kinase isozymes in embryonic skeletal muscle. J Histochem Cytochem 35:71~722 Sanders E, de Groot HM, Geerts WJC, de Jong F, van Horssen AA, Los JA, Moorman AFM (1986) The local expression of adult chicken heart myosin during development. II. Ventricular conducting tissue. Anat Embryol 174:187-193 Schweinfest CW, Kwiatkowski RW, Dottin RP (1982) Molecular cloning of a DNA sequence complementary to creatine kinase M mRNA from chickens. Proc Natl Acad Sci USA 79 :4997-5000 Sternberger LA (1986) Immunohistochemistry, 3rd ed. John Wiley and Sons, New York Vassall-Adams PR (1982) The development of the atrioventricular bundle and its branches in the avian heart. J Anat 134:169-183 Ziter FA (1974) Creatine kinase in developing skeletal and cardiac muscle of the rat. Exp Neurol 43 : 539-546

Accepted September 22,1988