Anatomy and Embryology

Anat Embryol (1986) 175:227-233

9 Springer-Verlag1986

Atriopeptin distribution in the developing rat heart R.P. Thompson t, J.A.V. Simson 1, and M.G. Currie 2 1 Department of Anatomy and Cell Biology, 2 Department of Pharmacology, Medical University of South Carolina, Charleston, USA Summary. The embryonic distribution of atriopeptin (atrial natriuretic factor) in the Sprague-Dawley rat heart was mapped by immunoperoxidase staining of embryonic and neonatal hearts using rabbit antiserum to atriopeptigen purified from adult rat atrium. During the period of cardiac septation (days 14 and 16), immune serum reacted strongly with myocardial cytoplasmic granules in two sites: the inner cell layer along the cephalic curvature of the atria and the trabeculae of the incompletely divided ventricles. The youngest hearts studied (gestational day 11) displayed only nonspecific diffuse peroxidase reactivity within blood cells, indistinguishable from control sections incubated with normal rabbit serum. One week following birth, intense antiatriopeptin reactivity was widely distributed through both atria. In addition, immunoreactive cytoplasmic granules were found at several sites in the ventricular myocardium. Along the fiber tracts of the concentric layers of the ventricular walls and interventricular septum, scattered granular foci were seen between nuclei of contiguous elongated myocytes. Positive staining was also seen within the papillary muscles and trabeculae carnae, regions shown by Alcian blue/periodic acid-Schiff base staining of sister sections to be relatively rich in glycogen. These patterns of antibody reactivity suggest the coupling of early atriopeptin secretory activity with developing cardiac function. Key words: Heart atrium - Conductive system - Rat embryo - Atriopeptin - Cardiac hormones - Immunocytochemistry

Introduction Atriopeptin (AP) (atrial natriuretic factor, cardiodilatin) is a peptide hormone important in the regulation of arterial pressure, fluid balance and electrolyte homeostasis, causing natriuresis, diuresis, inhibition of aldosterone secretion and systemic vasodilation at sites remote from its principal source, the cardiac atria (Forssmann et al. 1984b; Needleman et al. 1985; deBold 1985). Little is known concerning the appearance of AP in embryonic life or its possible role in development. Early ultrastructural studies noted secretory or neurosecretory Offprint requests to : Robert P. Thompson, Department of Anatomy and Cell Biology, Medical University of South Carolina, 171 Ashley Avenue, Charleston, SC 29425, USA

granules of unknown nature in atrial and ventricular myocardium of rat and mouse embryos (Nanot and LeDouarin 1970; Challice and Viragh 1973), in ventricular wall of the 8-week human embryo (Leak and Burke 1964), in ventricular muscle of chick embryos (Manasek 1969) and lower vertebrates (Lemanski et al. 1975), and in conducting tissue of adult rat (Bompiani et al. 1959). These dense-core, membranebound granules, often found in association with Golgi lamellae and vesicles, were similar in appearance to the more abundant granules reported in adult mammalian atria (Jamieson and Palade 1964; Ferrans etal. 1969), now thought to be the principal storage site for atriopeptin or cardiodilatin (Cantin et al. 1984; Forssmann et al. 1983, 1984a; Metz et al. 1984). In a preliminary effort to define a possible role of such peptides in cardiac or embryonic development, we have used antibody developed against adult rat atriopeptigen (Currie et al. 1983, 1984), the principal storage prohormone, for the immunocytochemical localization of atriopeptin in the developing rat heart. This study provides data on the time of appearance of atriopeptin immunoreactivity in the rat heart and its histological distribution during embryonic and postnatal periods. Preliminary results of this study have been presented elsewhere (Thompson et al. 1986). Similar results are presented in this periodical by Back et al. (1986).

Methods Cardiac and thoracic tissues were dissected in saline, fixed in 6% mercuric chloride/0.1% glutaraldehyde in t.2% sodium acetate buffer (pH 6.0) overnight at room temperature, and embedded in paraffin. Three embryos were fixed at gestational day 11, four at day 14, one at day 16, and three neonatal rats at one week following birth. After removal of paraffin and mercury salts, 6 bE sections were immunostained for atriopeptin using an indirect immunoperoxidase bridge method (Mason et al. 1969) at primary antibody dilutions ranging from 1:50 to 1:200. Serum from r a b b i t s immunized with atriopeptigen purified from adult rat atrium was used as primary antibody. Secondary and tertiary antibodies were goat anti-rabbit immunoglobulin and rabbit anti-horseradish peroxidase (HRP) at 1:20 dilutions. H R P (1%) was added after the final antibody step. Sections were thoroughly rinsed with buffered saline after each step. For demonstration of bound HRP, slides were incubated with 3,3'diaminobenzidine (0.3%) and H202 (0.01%) in

228

Fig. la, b. Sagittal sections through the ventricle (V) and atrium (A) of a 14-day rat embryo heart. Reaction with immune serum (a) was seen along the cephalic margin of the atrium (large arrows) and within ventricutar myocardial tabeculae (small arrows). These sites did not react in sections incubated with control serum (b), in which only non-specific staining of blood cells was observed, x 60

Fig. 2a-c. Immunostaining of 14-day rat embryonic atrium (large arrows in a, b). Note dark perinuclear granules (large arrows) and less intense staining along individual myofibrils (smalls arrows in b). Adjacent control section, reacted with normal rabbit serum in place of immune serum, is shown in e. a x 200; b, e x 800 0.05 M tris buffer, washed in water, reacted briefly (15 s) with 1% OsO4, washed thoroughly with water, dehydrated and m o u n t e d for light microscopy. A d j a c e n t control sections were either incubated with normal rabbit serum in place of immune serum or stained for glycogen with periodic acid-Schiff reagent (PAS). R a d i o i m m u n o a s s a y for atriopeptin was performed as reported elsewhere (Currie and N e w m a n 1986).

Results

Embryonic heart Immunoreactive sites could not be found in sections taken at several points across the cardiac region at eleven days o f gestation. In the 14-day embryonic heart, however, inamunostaining for atriopeptin was distinct and widely dis-

229

Fig. 3a-c. Immunostaining of 14-day rat embryonic ventricle (a, b). Note intense granular staining within myocardial trabeculae arrows) and relative lack of staining within ventricular wall (* in a). Adjacent control section is shown in e. a x 200; b, e x 1,000

tributed. As shown in Fig. 1, immunoreactivity was found along the cephalic margin of the atrium and across the central trabeculated region of the ventricular cavity. These areas were not stained in adjacent control sections incubated with normal rabbit serum. In the embryonic atrium (day 14), immunostaining was strongest along the inner layer of cardiac muscle, as shown in Fig. 2, taken from the rectangular region bracketed in Fig. I a. Most of these subendocardial atrial myocytes contained perinuclear clusters of atriopeptin-positive granules, with more diffuse but apparently specific reactivity along individual myofibrils. In the embryonic ventricle (day 14), immunostaining was prominent along the developing myocardial trabeculae (Fig. 3) throughout the central luminal region. Minimal reactivity was observed along the outer walls of the primitive ventricle. Clustered and scattered immunoreactive granules were seen within individual trabecular myocytes with diffuse staining of myofibrils, structures which did not react with non-immune serum. The pattern and intensities of immunostaining presented in Figs. t-3 are representative of staining observed in the four embryos examined in sagittal or transverse section at 14 days of gestation and in the single embryo from day 16 of gestation. The presence of immunoreactive atriopeptin in both chambers of the embryonic heart was further supported by radioimmunoassay analysis. Atrial tissue dissected from 16 day embryos (n = 5) contained 1.27 _+0.29 ngm atriopeptin/heart; ventricular tissue contained 0.84+0.05ngm atriopeptin/heart.

Fig. 4. Immunostaining for atriopeptin in neonatal heart (one week post partum, transverse section). Note intense staining in atrial (A) myocardium and scattered foci of immunoreactivity across the free walt of the left ventrMe (V). Bracketed regions are shown at higher power in Figs. 5 and 6. x 78 the atrial myocytes subjacent to the endocardium were intensely atriopeptin positive (Fig. 5) with slightly less reaction along the external layer of myocardium. This staining was strongest in cytoplasmic granules. Control sections were uniformly negative except for diffuse non-specific staining of red blood cells. Neonatal ventricle

Neonatal atrium

One week following birth, immunoreactivity for atriopeptin was strongest within the atrium, as expected (Figs. 4, 5). Immunostaining was intense within the atrium compared with the nearby left ventricular wall. Most, if not all, of

Atriopeptin positive cells were also found scattered in several regions of the neonatal ventricular myocardium, as shown in Figs. 6-8. Figure 6 was taken from the concentric layers of the left ventricular free wall. Note the coordinated alignment of muscle fibers from lower left to upper right

230

Fig. 5a-e. Immunostaining in neonatal atrium. Note intense granular deposits in atrial myocytes (arrows in a, b) relative to endocardium (*). Control section shown in e. a x250; b, e • 1,000

Fig. 6a-c. Imhaunostaining in neonatal ventricular wall. Note scattered foci of positive staining (arrows in a) near small artery (*). Several such loci from the area bracketed in a are shown at higher magnification and marked with arrows in b (bright field) and e (differential interference contrast). Note the intense granular staining between nuclei (n) of contiguous aligned myocyte pairs and the fainter staining in neighboring cells (* in b). a • 250; b x 1,100; e x 920

and the scattered foci o f immunoreactivity. Intensely positive granular foci could be seen between contiguous pairs of elongated myocardial nuclei, with less intense granular staining in nearby myocytes. Such cell pairs, which appeared to represent an otherwise unremarkable, PAS-nega-

tive subpopulation of myocytes, were scattered through both ventricular free walls and the interventricular septum from cardiac base to apex. A t r i o p e p t i n positive myocytes also appeared particularly a b u n d a n t within the papillary muscles and trabecular

231

Fig. 7. Immunostaining of neonatal posterior papillary muscle (P) in transverse section across left ventricle. Note numerous scattered foci of postitive staining within papillary muscle (arrows) compared with nmral myocardium to right (M). Bracketed region is presented at higher magnification in Fig. 8. x t20

fibers subtending the lesser chordae of the mitral valve (Fig. 7), compared with the ventricular wall. Although immunostaining appeared generally more abundant in PAS positive regions of myocardium, only occasional coincidence of PAS reactivity with individual atriopeptin-positive cells could be demonstrated at the cellular level (Fig. 8). Some PAS positive regions showed no reactivity with antibody.

Discussion Atriopeptin, or an immunologically similar molecule, was demonstrable in the developing rat heart at day fourteen of gestation. The entire eleven day embryo, including the looped but essentially tubular primitive heart, appeared devoid of atriopeptin storage sites under the staining conditions used in this study. By day 14, however, during the period of cardiac septation (days 13.5 16, Thompson et al. 1983), intracellular atriopeptin immunoreactivity was dearly seen in two regions of the heart, the inner muscle layers of the atrial wall and along the well-defined but diffusely organized trabeculae of the incompletely divided ventricles. These findings confirm and extend ultrastructural studies of the developing mammalian heart in which secretory granules of unknown nature were noted within the embryonic rat and mouse ventricle during the septation period (Challice and Viragh 1973), and in the ventricular wall of the post-septation human heart at eight weeks of gestation (Leak and Burke 1964). Although neurosecretory-like granules have also been reported in ventricular muscle of the preseptation chick embryo heart (Manasek 1969), the anti-

body used in this study, developed against prohormone purified from adult rat atrium, did not cross react with embryonic chick tissues (data not shown). The identity or close similarity of immunoreactive material seen in histological section with authentic atriopeptigen was further confirmed by radioimmunoassay of extracts from atrial and ventricular tissue from 16-day rat embryos. In the neonate (one week post partum), the distribution and intensity of immunostaining observed in the atria further confirmed the specificity of the antibody and histochemical methods used in this study. In addition, the focal staining observed within the papillary muscles of the ventricles and in a subpopulation of myocytes within the ventricular walls may indicate ventricular sites which maintain, until late in development, their capacity for atriopeptin synthesis and storage. Others have noted a decrease in secretory granules within the ventricular myocardium during normal development (Challice and Viragh 1973). We can hypothesize that the last sites of dwindling atriopeptin storage within the normally developing ventricle, as described in the present study, may be the first or principal sites to resume ventricular atriopeptin synthesis and storage under conditions of experimental hypertrophy in the adult (Day et al. 1986). The distribution of atriopeptin within the developing rat heart, as described here by immunocytochemistry, suggests at least one of two possibilities. First, the proximity of immunoreactive sites in both primitive chambers to the cardiac lumen suggests their contribution to atriopeptin levels in the early embryonic circulation. Second, the distribution of atriopeptin along the trabecular strands of the embryonic ventricular cavity and within the papillary muscles of the neonate suggest the coupling of atriopeptin synthesis, storage and release to regional differences in early myocardial function or stretch (Lang et al. 1985). Finally, the possible association of early atriopeptin storage sites with developing cardiac conducting tissue deserves comment. Well before the nature of their contents was established, secretory granules ('~ granules or "atrial-specific" granules) were noted within ventricular conduction tissue of adult rat by Bompiani et al. (1959) but not by Melax and Leeson (1970) and within the sinoatrial node of the full-term mouse fetus (Nanot and LeDouarin 1975). These isolated early observations appear to have attracted little attention, due perhaps to the striking abundance of such granules in the atria as well as to the generally poor differentiation of conducting tissues in small mammals, especially during development (Truex and Smythe 1965). Although nodal tissue may be located with little ambiguity quite early in development (James 1970; Anderson and Taylor 1972; Viragh and Challice 1977; Wenink 1976), the embryonic origins of peripheral portions of the conducting system remain somewhat obscure. As reviewed by Forsgren (1984), numerous authors have proposed that Purkinje fibers originate separately from the AV node, developing in situ along the early ventricular trabeculae (Mall 1912; Sanabria 1936; Anderson et al. 1976; Viragh and Challice 1977; Truex etal. 1978; Forsgren et al. 1980, 1983). Especially in light of those studies, the present demonstration of atriopeptin immunoreactivity along early ventricular trabeculae and within neonatal tissue derived from embryonic trabeculae suggests a contribution or coupling of atriopeptin to early ventricular function.

232

Fig. 8. Immunostaining (a, c) and PAS staining (b, d) of papillary muscle (P) and muscular cords (*) from neonatal rat ventricle shown in Fig. 7. Within the papillary muscle, note positive granular immunostaining (a) and diffuse, weakly positive PAS reaction (b). Along the muscular cords subtending the papillary muscle and mitral valve, note the close coincidence or proximity of immunostaining and PAS reactivity in some cells or areas (smallarrows)and absence of correlative immunostaining in other PAS positive areas (arrowhea&). a,b x 3 0 0 ; e , d x960

Acknowledgements. We thank Jan Condon for technical assistance and Marion Hinson for secretarial assistance. This work was supported by an Established Investigator Award (RPT) and grant in aid 4b85-1282 (MGC) from the American Heart Association with funds contributed in part by the S.C. Affiliate and by NHLBI grants ~28936 (RPT) and 29566 (MGC, JVS).

References Anderson RH, Taylor IM (1972) Development of atrioventricular specialized tissue in human heart. Brit Heart J 34:1205-1214 Anderson RH, Becket AE, Wenink ACG, Janse MJ (1976) The development of the cardiac specialized tissue. In: Wellens H J J,

233 Lie KI, Janse JM (eds) The Conduction System of the Heart. H.E. Stenfert, B.V. Kroese, Leiden, pp 3-28 Back H, Stumpf WE, Ando E, Nokihara K, Forssmann WG (1986) Immunocytochemical evidence for CDD/ANP-like peptides in strands of myoendocrine cells associated with the conductive systems of the rat heart. Anat Embryol (in press) Bompiani GD, Rouiller CH, Hatt PY (1959) Le tissu de conduction du coeur chez le rat. Etude au microscope electronique. Archives des Maladies du Coeur et des Vaisseaux 52:269-292 Cantin M, Gutkowska J, Thibault G, Milne RW, Ledoux S, MinLi S, Chapeau C, Garcia R, Hamet P, Genest J (1984) Immunocytochemical localization of atrial natriuretic factor in the heart and salivary glands. Histochem 80: 113 127 Challice CE, Viragh S (1973) The embryonic development of the mammalian hart, In: Challice CE and Viragh S (eds) Ultrastructure of the Mammalian Heart. Academic Press, New York and London, pp 91-126 Currie MG, Newman WH (1986) Evidence for c~-1 adrenergic receptor regulation of atriopeptin release from the isolated rat heart. Biochemical and biophysical research communications 1 : 94-100 Currie MG, Geller DM, Cole BR, Boylan JG, YuSheng W, Holmberg SW, Needleman P (1983) Bioactive cardiac substances: Potent vasorelaxant activity in mammalian atria. Science 221:71-73 Currie MG, Geller DM, Cole BR, Siegel NR, Fok KF, Adams SP, Eubanks SR, Galluppi GR, Needleman P (1984) Purification and sequence analysis of bioactive atrial peptides (atriopeptins). Science 223 : 67 69 Day ML, Schwartz D, Wiegand RC, Stockman PT, Brunnert SR, Tolunay HE, Currie MG, Standaert DG, Needleman P (1986) Ventricular atriopeptin: Unmasking of mRNA and peptide synthesis by hypertrophy or dexamethasone. (submitted for publication) de Bold AJ (1985) Atrial natriuretic factor: A hormone produced by the heart. Science 230:767 770 Ferrans VJ, Hibbs RG, Buja LM (1969) Nucleoside phosphatase activity in atrial and ventricular myocardium of the rat: A light and electron microscope study. Am J Anat 125:47-60 Forsgren S (1984) The differentiation of the Purkinje fibers in the mammalian heart-comparisons with the ordinary myocytes. In: The Developing Heart, MJ Legato, ed., Martins Nijhoff Publ. Co., Boston, pp 47-67 Forsgren S, Strehler E, Thornell L-E (1983) Differentiation of the atrioventricular node, the atrioventricular bundle and the bundle branches in the bovine heart : An immunohistochemical and enzyme histochemical study. Histochem J 15 : 1099 1111 Forsgren S, Thornell L-E, Eriksson A (1980) The development of the Purkinje fiber system in the bovine fetal heart. Anat Embryol 159:125-135 Forssmann WG, Hock D, Lottspeich F, Henschen A, Kreye V, Christmann M, Reinecke M, Metz J, Carlquist M, Mutt V (1983) Cardiodilatin as a peptide hormone candidate. Anat Embryol 168 : 307-313 Forssmann WG, Birr C, Carlquist M, Christmann M, Finke R, Henschen A, Hock D, Kirchheim H, Kreye V, Lottspeich F, Metz J, Mutt V, Reinecke M (1984a) The auricular myocardiocytes of the heart constitute an endocrine organ. Charactcriza-

tion of a porcine cardiac peptide hormone, cardiodilatin-126. Cell Tiss Res 238:425~430 Forssmann WG, Hock D, Kirchheim, F, Metz J, Mutt V, Reinecke M (1984b) Cardiac hormones: Morphological and functional aspects. Clin and Exper Theory and Practice, A6 (10 & 11), 1873-1878 James TN (1970) Cardiac conduction sytem: Fetal and postnatal development. Amer J Cardiol 25:213-226 Jamieson JD, Palade GE (1964) Specific granules in atrial muscle cells. J Cell Biol 23:151 172 Lang RE, Tholken H, Ganten D, Luft FC, Rushoaho H, Unger T (1985) Atrial natriuretic factor-a circuiating hormone stimulated by volume loading. Nature (London) 314:264-266 Leak LV, Burke JF (1964) The ultrastructure of human embryonic myocardium. Anat Rec 149:623-650 Lemanski LF, Fitts EP, Marx BS (1975) Fine structure of the heart in the Japanese Medaka Oryzias latipes. J Ultrastruct Res 53 : 37-65 Mall FP (1912) On the development of the human heart. Amer J Anat 13:249-298 Manasek FJ (1969) The appearance of granules in the Golgi complex of embryonic cardiac myocytes. J Cell Biol 43 : 605 609 Mason TE, Phifer RF, Spicer SS, Swallow RA, Dreskin RB (1969) An immunoglobulin-enzymebridge method for localizing tissue antigens. J Histochem Cytochem 17:563-569 Melax H, Leeson TS (1970) Fine structure of the impulse-conducting system in the rat heart. Can J Zool 48 : 837-839 Metz J, Mutt V, Forssmann WG (1984) ImmunohistochemicaI localization of cardiodilatin in myocendocrine cells in the cardiac atria. Anat Embryol 170 : 123 127 Nanot J, Le Douarin G (1970) Etude au microscope electronique du myocarde du foetus de souris. C R Soc BioI 164:890-893 Nanot J, Le Dourain G (1975) Ultrastructure du noeud sino-atrial do foetus du souris en fin de gestation. Journal Microscopie Biol Cell 22:29-38 Needleman P, Adams SP, Cole BR, Currie MG, Geller DM, Michener ML, Sapar CB, Schwartz D, Standaert DG (1985) Atriopeptins : Potential cardiac hormones. Hypertension 7: 469-482 Sanabria T (1936) Recherches sur la differenciation du tissue nodal et connecteur du coeur des manniferes. Arch Biol Paris 47 : 1-70 Thompson RP, Simson JA, Currie MG (1986) Atriopeptin distribution in developing rat ventricle. Fed Proc 45:404 Thompson RP, Wong M, Fitzharris T (t983) A computer graphic study of cardiac truncal septation. Anat Rec 206:207-214 Truex RC, Smythe MQ (1965) Comparative morphology of the cardiac conduction tissue in animals. Annals NY Acad Sci 127:19-33 Truex RC, Marino TA, Marino DR (1978) Observations on the development of the human atrioventricular node and bundle. Anat Rec 192:337-350 Viragh S, Challice CE (1977) The development of the conduction system in the mouse embryo heart. Develop Biol 56:39%411 Wenink ACG (1976) Development of the human cardiac conduction system. J Anat (London) 121:612631

Accepted September 12, 1986