J Mol Cell Cardio122,

Histochemical

361-370

(1990)

and Biochemical Expression During Lauren

Analysis Cardiogenesis

J. Sweeney

and Susan

of Myosin Heavy in the Rat

Chain

W. Kelley

Department of Anatomy, Loyola University, Stritch School of Medicine,

Maywood,

IL 60153, USA

(Received 9 May 1989, accepted in revised form 6 December 1989) L. J. SWEENEY AND S. W. KELLEY. Histochemical and Biochemical Analysis During Cardiogenesis in the Rat. journal of Molecular and Cellular Cardiology

of Myosin Heavy Chain Expression (1990) 22, 361-370. Little is known

about the tissue-specific expression of contractile proteins during cardiogenesis in the mammalian heart. Since the myosin heavy chain (HC) isoform expressed in the adult correlates with myocardial functional capacity, we undertook an analysis of myosin HC expression in atria1 and ventricular myocardia during fetal cardiogenesis in the rat heart. Cardiac HCs were separated by electrophoresis under denaturing conditions. The expression of the predominant adult isoform HCa was localized within developing fetal cardiac chambers by immunohistochemistry with a specific monoclonal antibody (R 37). Results demonstrated that myosin HC isoform expression followed tissue-specific patterns during cardiogenesis in the rat. Atria1 myocytes expressed HCa throughout development. The ventricles expressed exclusively HCa in the adult, but HC/I expression predominated in fetal ventricles. Fetal ventricles also expressed minor amounts of HCa, whose amount and distribution varied with developmental stage. HCa was initially confined to tracts of cells in the trabeculae, suggestive of future conduction system cells. A more extensive population of H&-expressing cells appeared several days before birth in a pattern which could represent the prenatal initiation of HCa expression in working myocardial cells. These

results indicate that there is tissue-specific developmental modulation of myosin isoform expression during fetal development. tally distinct KEY

Results also demonstrated myosin HC in fetal hearts.

Atria; Cardiogenesis; Myosin; Ventricle.

WORDS:

cardium;

that this modulation

Electrophoresis:

Introduction An impressive number of studies on the mammalian heart have documented both the functional significance of expression of myosin heavy chain (HC) isoforms, and the different patterns of isoform expression of this contractile protein in atria1 and ventricular myocytes. However, these data come primarily from examination of the neonatal and adult heart. There have been no systematic studies of the content and distribution of myosin HC isoforms in the mammalian heart during prenatal development. Many studies have reported ventricular HC isoform (and mRNA) content, some at several stages during fetal development (Lompre et al., 1981; Schwartz et al., 1982; Chizzonite et al., 1984a; Mahdavi et al., 1987). However, none of these have examined the distribution of these isoforms or

Immunohistochemistry;

+ 10 $03.00/O

expression

of a third

Monoclonal

electrophoreti-

antibodies;

Myo-

their mRNAs. Only two studies have examined myosin HC content in fetal atria1 myocytes, each at a single time point late in development (Chizzonite et al., 1984b; Samuel et al., 1986). Thus, an understanding of the factors controlling developmental regulation of myosin gene expression, and their significance to myocardial physiology, is limited by the paucity of information available for the prenatal mammalian heart. In all mammalian hearts, atria and ventricles express two myosin HCs, HCa and HCj3 (Hoh et al., 1979; Lompre et al., 1981; Schwartz et al., 1982; Chizzonite et al., 1984b; Samuel et al., 1986) which are encoded by different genes (Mahdavi et al., 1982). Myosin HCa is capable of generating a 2-3-fold greater speed of contraction than HC/?, but at the expense of a correspondingly higher utiliz-

Please address all correspondence to Lauren J. Sweeney, Dept. Medicine, 2160 South First Ave., Maywood, IL 60153, USA. 0022-2828/90/030361

may include

of Anatomy,

Loyola 0

University, 1990 Academic

Stritch

School

Press Limited

of

L. J. Sweeney

362

ation of ATP (Carey et al., 1979; Schwartz et al., 1981; Ebbrecht et al., 1982). Atria and ventricles show different intrinsic patterns of expression of these isoforms from late fetal development through postnatal life (Chizzonite et al., 1984a; Dechesne et al., 1985a, 1987; Samuel et al., 1986)) and different capacities to respond to functional and hormonal chalin isoform expression lenges by changes (Lompre et al., 1981; Chizzonite et al., 1984a, b; Samuel et al., 1986; Mahdavi et al., 1987). In addition, the impulse conduction system has a distinct myosin phenotype from that of the working myocardium in all mammalian species examined, expressing both HCa and HCB (Forsgren el al., 1982; Gorza et al., 1986; Kuro-o et al., 1986; Dechesne et al., 1987; Komuro et al., 1987), and a third HC immunologically related to the embryonic skeletal myosin HC (Gorza et al., 1988a, b). In this study we utilized electrophoretic separation of myosin HC isoforms and immunohistochemical localization with a specific monoclonal antibody to analyze the regulation of myosin HC gene expression during cardiogenesis in the rat myocardium. We report here that this regulation is both tissue and region-specific, and may include the expression-of a unique HC species in the working myocardium of both atria and ventricles during fetal development. Materials

and

Methods

Animal model Pregnant Sprague-Dawley rats were sacrificed by COz inhalation, and fetuses removed. Developmental stages (in days postcoitum; fertilization plug = day 0) follow Theiler staging as amplified in Butler and Juurlink, 1987. Adult myocardial status was assessed in the 4 week male, which has been shown to demonstrate exclusive expression of HCcl in both atria1 and ventricular cardiac chambers (Dechesne et al., 1985a, 1987; Samuel et al., 1986). Comparison of immunohistochemical localization and biochemical analysis of myosin was made in fetuses from the same pregnancy. Immunohistochemical

localization isoforms

of myosin HC

Previously published protocols for tissue preparation and immunofluorescence were em-

and S. W. Kelley ployed (Sweeney, et al., 19871, with the following specifications applicable to rat hearts. Hearts from 18-2 1 day fetuses and adults were frozen at - 160°C in aluminum boats containing OCT embedding medium (Milesr. Younger fetuses were first infiltrated with 15”,) sucrose in PBS for 2-4 h. Ten pm thick frozen sections were incubated first with primary anti-myosin antibodies (dilutions established empirically) and then with the appropriate rhodamine-conjugated IgG secondary antibody (1:50, ICN Immunobiologicals) or peroxidase-conjugated IgG secondary antibody ( 1: 10, Cappel). The peroxidase reaction, which yielded a dark reaction product, was developed by incubation in Hz04 and 4chloro-1 -napthol in methanol. Slides were examined with a Leitz epifluorescence microscope and photographed on Kodak Plus-X film. Monoclonal

antibody characterization

The monoclonal antibodies used in this study have been previously characterized for their specificity for myosin HCs by immunoblot immunofluorescence local(Western blot), ization, and RIA or ELISA. Antibody R 37 is specific for HCc( in the rabbit (Chizzonite et al., 1982; Eisenberg et al., 1985) and rat (Chizz0nit.e et al., 1982). Ab CCM 52 reacts with all mammalian cardiac myosin HCs, but shows a greater affinity for HC/I (Chizzonite et al., 1982, Clark et al., 1982). The specificity of Ab HPM 7 for myosin HC has been previously documented with regard to avian skeletal (Kennedy et al., 1988) and cardiac muscle isoforms (Sweeney et al., 1989). Identijcation

and quantitation of myosin isoforms by SDS-PAGE

Myosin HC isoforms were separated by denaturing polyacrylamide gel electrophoresis (SDS-PAGE) following previously published protocols (Sweeney et al., 1989). Low percentage gels (5-6O/b), low loads (nanograms of myosin HCs) and a mini-gel apparatus were important to both maximal separation and sharpness of bands. Myosin extraction: Minced muscle was incubated at 100 mg/ml in a modified Guba-Straub high salt extraction solution (0.3 M NaCl, 0.1 M NaH2P04.H20, 0.05 M

Fetal

Cardiac

Na2HP04, 1 mM EDTA, 0.2 mM ATP, pH 6.5) for 1 h on ice. The supernatant was stored at - 20°C. Myosin extracts were denatured to separate HC and LC subunits by 1:l addition of 2X SDS loading buffer (final concentration 0.125 M Tris/Cl, pH 6.8, 4% SDS, 10% mercaptoethanol, 20% glycerol), and boiled for 3 mins. Electrophoresis: Denatured extracts were loadrd on 0.75 mm thick SDS polyacrylamide slab gels at 250-300 ng myosin HC/lane (determined from comparison with known loads of myosin HC molecular weight marker, Sigma). We utilized a 5.6% separating gel 133.5?, acrylamide/0.3~/, bis stock; 0.38 M Tris/Cl, pH 8.8; 0.1% SDS) and 4% stacking gel (3096 acrylamide/0.440/b bis stock, 0.125 M Tris/Cl, pH 6.8; 0.1% SDS) (Dreyfuss et al., 1984). The discontinuous Tris/Cl buffer system employed was based on the Laemmli system (1970) with the addition of 0.1 o/o SDS (Ausubel et al., 1988). Electrophoresis was carried out at constant voltage (200 V) on a mini-gel apparatus (Mini Protean II, BioRad) until the myosin HCs had run 2/3 of the way through the gel (l$ h). Detection of protein bands by silver staining the gels followed protocols described in Sweeney et al., 1987. Results

Electrophoretic

separation of myosin HC isoforms

In order to directly compare the myosin HCs expressed in atria and ventricles, myosins were electrophoretically separated under denaturing conditions. This dissociated the HCs from their light chains (LCs), which differ in atria1 and ventricular myosins, and can impart different mobilities to myosins which contain identical HCs. Under denaturing conditions, HCa has a slower electrophoretic mobility than HCfl (Esser et al., 1988). We found that the 4 week adult atria and ventricles expressed only a single HC with this slower mobility [Fig. l(A)]. Fetal ventricles, on the other hand, expressed predominantly the faster migrating HC, HCfi [Fig. 1 (B)]. These findings are consistent with previous results utilizing non-denaturing electrophoresis (Chizzonite et al., 1982; Schwartz et al., 1982; Dechesne et al., 1987). Systematic examination of a number of prenatal stages revealed that isomyosin con-

363

Isomyosins

tent was not static during fetal development. The minor amounts of HCa which fetal ventricles expressed oscillated with stage of development. The level of HCcr observed until 1.5 days [Fig. l(B), lane 71 decreased to a minimum at 18 days (lane 5), and then increased again (maximal at 21 days, lanes 2-3 1. Thr fetal atria at all stages of development CYpressed HCa, as did the adult atria [Fig. 1 (A, B)]. Atria expressed minor, barely detrctable levels of HCfi only through 15 days ot fetal development [Fig. 1 (B), lane 61. The separation of myosin HC isofbrms under denaturing conditions led to the unanticipated finding of a HC band in fiatal atria and ventricles which migrated between HCa and HCP. We have arbitrarily designated this band as HCa’, since it was only present in tissues which were also expressing HCa SW Discussion). This third band was most clearl) seen in the 15 and 18 day fetal ventricles and atria [Fig. 1 (B), lanes 4-7; enlar,qcd \irw and diagram, Fig. 1 (C-D)]. Immunohistochemical

localization isoforms

oj‘qusin

I-IO’

Immunohistochemical analysis provided information on the localization of the myosin HC isoforms detected by electrophoretic separation. We were able to delineate all myocardial cells by utilizing Ab HPM 7, which detected all myosin HC isoforms expressed in the heart at all stages [Figs 2(A), 3(A), 4(12/l. The specificity of Ab R 37 for HCa was determined in previous studies with respect to the postnatal HCa isoform (Chizzonite et al., 1982). All adult atria1 and ventricular myocardia (including the conduction system) were strongly and uniformly reactive with Ab R 37 [Fig. 2(B)] at 4 weeks. In addition, Ab CCM 52 [Fig. 2(C, D)], which is preferential for HCB (Clark et al., 1982) reacted with myocytes distributed in patterns indicative of conduction system cells in the adult ventricles. The pattern of co-expression of HCa and HC/? which this suggests would be consistent with previous findings in conduction tissue (Forsgren et al., 1982; Gorza et al., 1986; Kuro-o et al., 1986; Dechesne et al., 1987; Komuro et al., 1987). The patterns of atria1 and ventricular myosin HCa expression changed during fetal heart

364

L. J. Sweeney

B4

+

21d A

VA

VB

and

S. W. Kelley

.-18dA

-15dV

A

V

FIGURE 1. Electrophoretic separation of myosin HCs. Myosin HCs (250-300 ng) from rat atria (A) and ventricles (V) separated by denaturing gel electrophoresis, and detected by silver staining. (A): Comparison of4 week adult (Ad.) and I8 day fetal (18 d F.) isomyosins. Adult A and V (Lanes 2 and 3) contained a single HC (HCa) with the slowest electrophoretic mobility. The predominant HC species in fetal V (HC$) had the fastest mobility (18 day shown, Lane 5). A previously undetected HC of intermediate mobility (designated HCa’) was seen in fetal A (Lane 4) and V (Lane 5). Lane 1: Reference myosin HC molecular weight (MW) marker, ,250 ng (200 000 Da). Lane 6: mix of HCa + HCB (18 d. fetal V + adult V). (B): Developmental modulation of fetal isomyosin HC expression. HCa and HCa’ are expressed in both fetal A and V: 21 d. A (Lane 1) and V (Lanes 2 and 3), 18 d. A (Lane 4) and V (L ane 5), and 15 d. A (Lane 6) and V (Lane 7). HCB is the predominant isoform in the fetal V at all stages. The amount of HCa/HCa’ in fetal V is less at 18 d. than at either 15 or 21 d. Same conditions as in A, but run slightly further through the gel to enhance separation ofHCa and HCa’. (C): Diagram of relative mobilities of HCa, HCa’, and HCB. (D): Enlarged view of 15 day fetal atria1 and ventricular myosin HCs seen in B (lanes 6 and 7) to clearly demonstrate relative mobilities of HCa, HCa’, and HCB. Note that only at 15 days do the A show any detectable levels of HCfi expression.

development in a manner which provided insight into (and was qualitatively consistent with) electrophoretic results. Initially, the embryonic atria were uniformly and strongly reactive with Ab R 37. This uniform reactivity continued through the initial stages of cardiac septation [shown at 15 days, Fig. 3(B) 3. Thereafter, the fetal atria showed a clear localization of Ab R 37 reactivity in some portions of the atria1 appendages. Reactivity

was confined to the subendocardial myocardium in some regions [shown at 18 days, Fig. 4(B)]. In other regions, there was a more heterogenous, apparently random distribution of reactivity with Ab R 37, in which positive and negative cells were adjacent to each other [Fig. 4(C)]. This heterogeneous pattern became the predominant one by 21 days (not shown). The absence of Ab R 37 reactivity in portions of the 18-21 day fetal

FIGURE 2. Adult heart. Distribution of myosin HC expression in the 4 week adult rat heart. Reactivity of ant)myosin Abs was developed with fluorescently-conjugated secondary antibodies unless specified otherwise. (A) Myocurdial archlecture is delineated by Ab HPM 7, which is reactive with all myosin HCs expressed in cardiac muscle. Outline shows regions contained in B and C. (B) Ab R 37 shows strong reactivity with all atria1 and ventricular myocardium at a stage when HCa is thr onI\ isoform detected in the heart by electrophoresis. (C) Ab CCM 52 (which is strongly reactive with HCB) also showed weak cross-reactivity with HCa in the adult cardiac myocytes. (D) Ab CCM-52 showed strong reactivity with conduction tissue in the ventricles (arrows), suggesting co-expression of HC/? with HCa in these cells. -

atria was not due to expression of HCfi, since there was no detectable expression of HC/? in the atria during this period [Fig. 1 (B), lanes 1, 41. However, it was consistent with expression of a third HC [Fig. l(B), lane 43 with which AB R 37 does not react. In the fetal ventricles, Ab R 37 reactivity also varied in localization with development. During the initial stages of cardiac septation i through 15 days), reactivity was confined to a small number of ventricular myocytes. Tracts of myocytes in the extensive trabecular network were strongly reactive with Ab R 37 [Fig. 3 (B-D)]. In addition, a sleeve of reactive cells continuous with the atria1 myocardium extended along the developing atrioventricular valves into the ventricles [Fig. 3(B, C)]. By day 18 of fetal development, ventricular mass and morphology had changed dramatically to assume a thick, compact configuration close to that of the mature heart [Fig. 4(A)]. Tracts of cells strongly reactive with Ab R 37 were still present, but comprised a small percentage of the total muscle mass. These cells were almost exclusively in sub-endocardial locations indicative of conduction system components [Fig. 4(D)]. In addition, a separate population of HCcc-expressing cells was now detected in a small region of the right ventricular free wall. They were only weakly

reactive with Ab R 37 (Fig. 4(B, E)]. The paucity of reactive cells in either category was consistent with the small amount of total HCc( expression detected by electrophoresis [Fig. 1

(-4 WI.

By the day before birth (21 days), electrophoresis detected a large increase in HCcl expression in the ventricles. Immunofluorescence localization with Ab R 37 revealed a substantial increase in the area of the right ventricular wall which was reactive. In addition, reactivity now extended in from the epicardial surface at the interventricular septum. The levels of reactivity were still low, however. Consistent with this, similar increases in HCa content occurred in both apical and basal regions [Fig. 1 (B), lanes 2-31. The tracts of strongly reactive cells (putative conduction cells) were still infrequent. The question of HCfl distribution in the fetal ventricles could only be addressed indirectly utilizing Ab CCM 52, since it showed low levels of cross-reactivity with HCa as well on immunofluorescence [Fig. 2 (C)l. It is most likely that HC$ was expressed in all fetal ventricular myocytes, since strong, uniform levels of reactivity with Ab CCM 52 were detected throughout the fetal ventricles at all stages (results not shown). Fetal atria demonstrated the same low level of cross-reactivity as the adult atria and ventricles [Fig. P(C)].

366

L. J. Sweeney

and

S. W. Kelley

FIGURE 3. 15 day fetal heart. Distribution of myosin HC isoform expression is representative of the developing bteart architecture is delineated by Ab HPM 7 reactivity with all cardiac myosin HCs. The to this point. (A) Myocardial ventricular wall is comprised largely of a loose trabecular network (Tr; arrows). Ventricular and atria1 septa (V’S, AS) with are forming. LA, RA: left, right atria; LV, RV: left, right ventricles. SM = skeletal muscle myosin (intrrcostals) which HPM 7 also reacts. (B) Distribution of HCU expression (as detected by reactivity with Ab R 37) is uniform throughout the atria. 11L the ventricle it is confined to isolated trabecular myocytes more clearly seen in magnifications C and D. Note absent :e of reactivity in the outer ventricular wall and developing ventricular septum. (C) Higher magnification of HCa-expressing myocytes in portion of heart outlined in B. Strands of ventric ular trabecular myocytes are strongly reactive with Ab R 37 ( arrowheads). In addition, a sleeve of reactive atria1 myoc :ytes extends a short distance into the ventricular chamber along the surface of the atrioventricular valve (arrows). (D) Trabecular myocytes in the ventricular apex (region outlined in B) also express HCa (arrowheads). Note abs, encr of reactivity in the outer (epicardial surface) ventricular wall (arrow).

FIGURE 4. 18 day fetal heart. Distribution of myosin HC isoform expression after septation is complete has changed from that observed earlier in development. (A) Myocardial architecture delineated by Ab HPM 7 reactivity (developed with the peroxidase reaction). Section shown is through RV cavity, anterior to LV cavity. Thick, compact ventricular walls now have only a thin trabeculated layer at the endocardial surface. Cardiac septation is complete, as is the separation of Pulmonary Artery (PA) and Aorta (AO). Coronary arteries (CA) are well developed. (B) Distribution ofAb R 37 reactivity in right half of adjacent serial section. Reactivity in atria is heterogeneous: in thicker portions of the atria, it is strong only in the endocardial-surface myocytes. The remaining atria1 myocardium in these regions shows little or no reactivity. In the ventricles, only occasional endocardial-surface myocytes are mildly reactive, as well as small portions of the right ventricular free wall. This ventricular reactivity is seen more clearly in magnifications of outlined regions in D and E. respectively.

367

CC)-I ‘hin walled atria1 appendage (region outlined on LA in A) demonstrates the other major pattern of distribution ofAb R 37 reactivity. Adjacent myocytes show very different levels ofreactivity with no apparent pattern (arrows): this transi tic ms into regions (generally the thin endocardial-surface pectinate muscles, arrowheads) of uniformly strong reacti vi1 ‘Y. myocytes on the endocardial surface of both chambers are moderately or stron,qly rractive P: II solated ventricular with J4b ) R 37 (arrows). free wall myocardium have begun to show low levels of Ab R 37 rravtivity. (E) Slmall portions of the right ventricular These rt tgions am usually adjacent to coronary artery branches i CA!.

368

L. J. Sweeney

and S. W. Kelley

. Discussion

This study was undertaken to provide a systematic examination of myosin HC isoform expression in the mammalian myocardium during prenatal development. Previous studieshave documented the predominance of HCfi expression (and its mRNA) in the prenatal ventricle of all mammalsexamined, with most focus on the rat heart. These previous reports have recorded the amount of HCa and HC/3 via the native myosin forms which contain them: F’s, the HC$ homodimer; Vi, the HCa homodimer; and Vz, the HCj?HCa heterodimer. However, the percentages reported have varied widely between studies. In the rat heart ventricle, reports have recorded 80-90% F’s at 15, 19, and 2 1 fetal days (Chizzonite et al., 1984a), 70-75% F’s at 16 and 20 fetal days (Lompre et al., 1981), and F’s except for a “trace” of Vi/V* at 18 fetal days (Schwartz et al.; 1982). Some of these apparent discrepanciescould be due to differences in developmental maturity on a given day of fetal development, even within the same strain; they could be due as well to slight differences between the rat strains employed (Sprague-Dawley versus Wistar). On the other hand, the dip we observed in ventricular HCa expression between early fetal (15 days) and late fetal (21 days) development could represent a substantive basis for these differences. More importantly, our results demonstrate that the amount of a given isoform expressedis given meaning by the localization of that expression. Our results demonstrate that the upregulation of HCa expression occurs to a modest extent in specific portions of the working myocardium in the late fetal ventricle. Experimental studies would be required to determine whether this localized upregulation is a responseto changing metabolic or functional conditions. It is interesting that it occurs in the absenceof the elevated thyroid hormone levels which are known to be required for the HCB to HCa transition which occurs rapidly after birth (Lompre et al., 1981, 1984; Chizzonite et al., 1982, 1984a). While it has been shown previously that the fetal rat ventricular myocyte is capable of HCcr expression, it has been demonstrated only by experimentally introducing elevated thyroid

hormone levels (Chizzonite et al., 1984a; Gustafson et al., 1987). The differences detailed in this study between myosin gene programs of atria and ventricles throughout prenatal development are consistentwith the postnatal differences in responsivenessof atria1 and ventricular myocardia to stimuli. The atria showed virtually no expression of HCB after 15 days in our study. The one previous study on prenatal rat atria (Samuel et al., 1986) showed a small amount of HCB at 19days, the only fetal stage examined. The late fetal rabbit atria also showed only a small amount of HCB (Chizzonite et al., 1984b). The atria of many neonatal and adult mammals, including the rat, have been shown to expresseither low levels of HCB (Lompre et al., 1984; Dechesneet al., 1985), or no HCB (Chizzonite et al., 1984b; Samuel et al., 1986). Adult atria show much lessmodulation of their HC phenotype in response to hormonal stimuli than ventricular myocytes (Chizzonite et at., 1984b; Dechesne et al., 1985a;Samuel et al., 1986). Indeed, significant transition in atria1 myosin HC expression has been shown only in the human atria, and only in responseto mechanical overload (Yazaki et al., 1984; Mercadier et al., 1987). The third cardiac myogenic cell type, conduction tissue, has been shown to express a number of gene programs distinct from those of the “working” myocardium of the surrounding atria and ventricles. Recent studies have followed this distinction back to early stages of embryonic development. It is of interest that the distribution of HCa (R 37 reactivity) we observed was virtually identical at early fetal stagesto distribution of several proteins which are expressed in developing conduction tissue (Viragh and Challice, 1977). These include acetylcholinesterase (Lamers et al., 1987), atria1 natriuretic peptide (Thompson et al., 1986; Scott and Jennes, 1987; Toshimori et al., 1987; Zeller et al., 1987) and a myosin HC with embryonic skeletal myosin HC characteristics (Gorza et al., 1988). In addition to changing patterns of expressionof the known isoformsHCa and HCB, our results suggest the possibility that fetal cardiac development may include the expression of a previously unrecognized HC isoform. This band, which we have designated

Fetal Cardhc as HCa’, is expressed in both atria and ventricles at all observed stages of fetal development. Since we observed this band at different levels at each of these stages, we do not feel it is a degradation product or artifact. We felt it was important to give a specific designation to this third band to distinguish it from an isomyosin HC with HC$-like immunologic characteristics whose expression has recently been proposed in the adult and fetal atria of large mammals (Dechesne et al., 1985b; Komuro et al., 1988; Tsuchimochi et al., 1988). We arbitrarily designated the third band HCa’ since it is only expressedin tissueswhich are expressing HCa. There is no evidence which could indicate a likely source of this proposed cardiac HC isoform. Theoretically, a unique HC could represent post-translational modification of another HC, the product of a unique mRNA produced by alternative splicing from a common myosin HC gene, or the product of a distinct gene’smRNA. However, there is no evidence to date that myosin HC protein diversity is controlled at any level other than the transcriptional. It is intriguing that Sl nuclease mapping results have been used to suggest that a unique mRNA coding for a second HCa protein could be generated by alternative splicing of exons transcribed from

Ieomyosins

369

the HCa gene in the rat heart (Lompre et al., 1984). However, there is no evidence that this mRNA actually exists.

S-rY Little is known about what permissive or required role, if any, is played by functional or metabolic factors in the establishment of the normal fetal cardiac myosin phenotype. Nor is there any information as to what role, if any, the establishment of proper phenotype plays in the subsequent development of the fetal myocardium. Localization information such as that provided in this study can provide the basisfor investigating the interaction between biochemical factors and morphologic differentiation of the myocardium.

Acknowledgements

We would like to thank Dr Radovan Zak of the University of Chicago for the use of the monoclonal antibodies employed in this study. This work was supported by an American Heart Grant-in-Aid and Loyola University BANE Award to LJS.

References FW, BRENT R, KINGSTON RE, MOORE DD, SE~DMAN JG, SMITH JA, STRUHL K (1988) Current Protocols m Molecular Biology, J Wiley and Sons, Inc, NY, Vol II, Sect. 10.2.3. BUTLER H, JUURLINK BHJ (1987) An atlas for staging mammalian and chick embryos, CRC, Boca Raton. CAREY RA, BOVE AA, COULSON R, SPANN JF (1979) Correlation between cardiac muscle myosin ATPase activity and velocity of muscle shortening. Biochem Med 21: 235-245. CHIZZONITE RA, EVERETT AW, CLARK WA, JAKOVCIC S, RABINOWITZ M, ZAK R (1982) Isolation and characterization of two molecular variants of myosin heavy chain from rabbit ventricle: change in their content during normal growth and after treatment with thyroid hormone. J Biol Chem 257: 2056-2065. CHIZZONITE RA, ZAK R (1984a) Regulation of myosin isoenzyme composition in fetal and neonatal rat ventricle by endogenous thyroid hormones. J Biol Chem 259: 12628-12632. CHIZZONITE RA, EVERETT AW, PRIOR G, ZAK R (1984b) Comparison of myosin heavy chains in atria and ventricles from hyperthyroid, hypothyroid, and euthyroid rabbits. J Biol Chem 259: 1556415571. CLARK WA, CHIZZONITE RA, EVERETT AW, RABINO~~TZ M, ZAK R (1982) Species correlation between cardiac isomyosins: comparison of electrophoretic and immunological properties. J Biol Chem 257: 54495454. DALLA LIBERA L, SARTORE S, SCHIAFFTNO S (1979) Comparative analysis of chicken atria1 and ventricular myosins. Biochim Biophys Acta 581: 283-294. DECHESNE CA, LEGER J, BOUVAGNET P, MAIRHOFER H, LEGER JJ (1985a) Local diversity of myosin expression in mammalian atria1 muscles: variations depending on age and thyroid state in the rat and the rabbit. Circ Res 57: 767-775. DECHESNE CA, LEGER J, BOUVAGNET P, CLAVIEZ M, LEGER JJ (1985b) Fractionation and characterization of two molecular variants of myosin from adult human atrium. J Mol Cell Cardiol 17: 753-767. DECHESNE CA, LEGER JOC, LEGERJJ (1987) Distribution of alpha- and beta-myosin heavy chains in the ventricular fibers of the postnatal developing rat. Dev Biol 123: 169-l 78. DREYFUSS G, ADAM SA, CHOI YD (1984) Physical change in cytoplasmic messenger ribonucleoproteins in cells treated with inhibitors of mRNA transcription, J Mol Cell Biol11: 415423. AUSUBEL

370

L. J. Sweeney

and

S. W. Kelley

H, JACOB R (1982) Alterations of mechanical parameters in chemically skinned preparations of rat as a function of isoenzyme pattern of myosin. Basic Res Cardiol77: 220-234. EISENBERC BR, EDWARDS JA, ZAK R (1985) Transmural distribution ofisomyosin in rabbit ventricle during maturation examined by immunofluorescence and staining for calcium-activated adenosine triphosphatase. Circ Res 56: 5488555. ESSER KA, BOLUYT MO, WHITE TP (1988) Separation of cardiac myosin heavy chains by gradient SDS-PAGE. Am J Physio1255: H659-H663. FORSGREN S, STREHLER E, THORNELL L-E (1982) Differentiation of Purkinje fibers and ordinary ventricular and atrial myocytes in the bovine heart: an immune and enzyme histochemical study. Histochem J 14~: 929-942. GORZA L, SARTORE S, THORNELL L-E, SCHIAFFINO S (1986) Myosin types and fiber types in cardiac muscle. III. Nodal conduction tissue. J Cell Biol 102: 1758-I 766. GORZA L, SAGCIN L, SARTORE S, AUSONI S (1988a) An embryonic-like myosin heavy chain is transiently expressed in nodal conduction tissue of the rat heart. J Mel Cell Cardiol 20: 931-941. GORZA L, THORNELL L-E, SCHIAFFINO S (1988b) Nodal myosin distribution in the bovine heart during prenatal development: an immunohistochemical study. Circ Res 62: 1182-l 190. GUSTAFSON TA, MARKHAM BE, BAHL JJ, MORKIN E (1987) Thyroid hormone regulates expression of a transfected a-myosin heavy-chain fusion gene in fetal heart cells. Proc Nat1 Acad Sci USA 84: 31223126. HOH JFY, YEOH GPS, THOMAS MAW, HICGINBOTTOM L (1979) Structural differences in the heavy chains of rat ventricular myosin isoenzymes. FEBS Letts 97: 330-334. in overloaded KENNEDY JM, EISENBERG B, REID SK, SWEENEY LJ, ZAK R (1988) N ascent muscle fiber appearance chicken slow-tonic muscle. Am J Anat 181: 203-215. KOMURO I, NOMOTO K, SUCIYAMA T, KURABAYASHI M, TAKAU F, YAZAKI Y (1987) Isolation and characterization of myosin heavy chain isozymes of the bovine conduction system. Circ Res 61: 859865. KOMURO I, NOMOTO K, TAKAKU F, YAZAKI Y ( 1988) Isolation and characterization of two &type cardiac myosin in the canine atrium. Jpn Cirr J 52: 1191-1200. KURO-o M, TSUCHIMOCHI H, EUDA S, TAKAKU F, YAZAKI Y (1986) Distribution of cardiac myosin isozymes in human conduction system: immunohistochemical study using monoclonal antibodies. J Clin Invest 77: 340-347. LAMERS WH, TE KORTSCHOT A, Los JA, MOORMAN AFM (1987) Acetylcholinesterase in prenatal rat heart: a marker for the early development of the cardiac conductive tissue? Anat Ret 217: 361-370. LOMPRE AM, MERCADIER JJ, WISNEWSKY C, BOUVERET P, PANTALONI C, D’ALBIS A, SCHWARTZ K (1981) Species- and age-dependent changes in the relative amounts of cardiac myosin isoenzymes in mammals. Dev Biol84: 286290. LOMPRE A-M, NADAL-GINARD B, MAHDAVI V (1984) Expression of the cardiac ventricular LX- and P-myosin heavy chain genes is developmentally and hormonally regulated. J Biol Chem 259: 64376446. MAHDAVI V, PERIASAMY M, NADAL-GINARD B (1982) Molecular characterization of two myosin heavy chain genes expressed in the adult heart. Nature 297: 659464. MAHDAVI V, IZUMO S, NADAL-GINARD B (1987) Developmental and hormonal regulation of sarcomeric myosin heavy chain gene family. Circ Res 60: 804814. MERCADIER J-J, DE LA BASTIE D, MENASCH~ P, N’GL~YEN VAN CAO A, BOUVERET P, LORENTE P, PIWNICA A, SLAMA R, SCHWARTZ K (1987) Alpha-myosin heavy chain isoform and atria1 size in patients with various types of mitral valvt dysfunction: a quantitative study. J Am Co11 Cardiol9: 1024-1030. SAMUEL J-L, RAPPAPORT L, SYROVY I, WISNEWSKY C, MAROTTE F, WHALEN RG, SCHWARTZ K (1986) Differential effect of thyroxine on atria1 and ventricular isomyosins in rats. Am J Physiol (Heart Circ Physiol 19) 250: H3333H341. SCHWARTZ K, LECARPENTIER Y, MARTIN JL, LOMPRE AM, MERCADIER JJ, SWYNGHEDAUW B (1981) Myosin isoenzymic distribution correlates with speed of myocardial contraction. J Mol Cell Cardiol 13: 1071-1075. SCHWARTZ K, LOMPRE A-M, BOUVERET P, WISNEWSKY C, WHALEN R (1982) Comparison of rat cardiac myosins at fetal stages in young animals and in hypothyroid adults. J Biol Chem 257: 14412-14418. SCOTT JN, JENNES L ,1987) Distribution of atria1 natriuretic factor in fetal rat atria and ventricles. Cell Tissue Res 248: 479-48 1. SWEENEY LJ, ZAK R, MANASEK FJ (1987) Transitions in cardiac isomyosin expression during differentiation of the embryonic chick heart. Circ Res 61: 287-295. SWEENEY LJ, KENNEDY JM, ZAK R, KOKJOHN K, KELLEY S (1989) Evidence for expression of a common phenotype in future fast and slow skeletal muscle during initial stages of avian embryogenesis. Dev Biol 133: 361-374. THOMPSON RP, SIMSON JAV, CURRIE MG (1986) Atriopeptin distribution in the developing rat heart. Anat Embryo1 175: 227-233. TOSHIMORI H, TOSHIMOR K, OURA C, MATSUO H (1987) Immunohistochemical study of atria1 natriuretic polypeptides in the embryonic, fetal and neonatal rat heart. Cell Tissue Res 248: 627633. TSUCHIMOCHI H, KURO-o M, KOYAMA H, KURABAYASHI M, SUGI M, TAKAKU F, FURATA S, YAZA~ Y (1988) Heterogeneity of b-type myosin isozymes in the human heart and regulational mechanisms in their expression, J CIin Invest 81: 1 It&l 18. VIRACH S, CHALLICE CE (1977) The development of the conduction system in the mouse embryo heart. II. Histogenesis of the AV node and bundle. Dev BiolS6: 397+11. EBRECHT

G, RUPP

myocardium

YAZAKI

Y, TSUCHIMOCHI

H, KURO-o

M, KURABAYASHI

M, ISOBE M, UEDA

S, NACAI

R, TAKAKU

F (1984)

Distributioll

of

myosin isozymes in human atria1 and ventricular myocardium: comparison in normal and overloaded heart. Eur Heat-t J 5 (Suppl F): 103-l 10. ZELLER R, BLOCH KD, WILLIAMS BS, ARCECI RJ, SEIDMAN CE (1987) Localized expression of the atria1 natriuretic factor gene during cardiac embryogenesis. Genes Devel 1: 693698.