Anat Embryol (1994) 190:367-373

Anatomy and Embryology

�9 Springer-Verlag 1994

Distribution of acetylcholinesterase activity in the rat embryonic heart with reference to HNK-1 immunoreactivity in the conduction tissue Tsuguto Nakamura 1, Takayoshi Ikeda ~, Isao Shimokawa 1 Yasuhisa Inoue a, Takashi Suematsu 3 Hiroyuki Sakai 1, Keisuke Iwasaki ~, Takeshi Matsuo ~ 1 First Department of Pathology, Nagasaki University School of Medicine and of Dentistry, Nagasaki 852, Japan z Department of Oral Histology, Nagasaki University School of Medicine and Dentistry, Nagasaki 852, Japan 3 Central Division of Electron Microscopy, Nagasaki University School of Medicine and Dentistry, Nagasaki 852, Japan

Accepted: 4 May 1994

Abstract. Acetylcholinesterase (ACHE) activity was topo- graphically investigated in the presumptive cardiac con- duction tissue regions visualized by HNK-1 immunore- activity in rat embryos, and AChE-positive cells were examined with the electron microscope. On embryonic day (ED) 14.5, when HNK-1 was most intensely visual- ized, AChE activity could not be detected enzyme-histo- chemically in the conduction tissue regions, except in the ventricular trabeculae and part of the AV node. On ED 16.5, however, the AChE activity was clearly demonstrat- ed in some parts of the developing conduction tissue. One exception was the AV node region, where an AChE-pos- itive area was in close proximity to an area showing HNK-1 immunoreactivity but did not overlap. Further- more, AChE activity was demonstrated predominantly in the ventricular trabeculae, including cardiac myocytes, but was rather weak in the atrium. With the electron microscope, AChE reaction products were observed pre- dominantly intracellulary in both developing conduction tissue cells and developing ordinary myocytes, and no reactivity was found in neuronal components. From ED 18.5 until birth, both AChE activity and HNK-1 im- munoreactivity faded away in the conduction tissue. Thus, transient AChE activity in the embryonic heart seems to be different from the developing adult form and may be related to a morphogenetic function in embryon- ic tissues, as proposed by other authors.

Key words: Acetylcholinesterase HNK-1 - Heart Morphogenesis - Rat

traction commences and the circulatory system begins to function on embryonic day 9.5 in this species. However, there is some evidence that AChE activity exists in the embryonic and neonatal heart before innervation (Navaratnam 1965; Gyevai 1969; Finlay and Anderson 1974; Drews 1975; Lamers et al. 1987, 1990). Further, many authors have hypothesized that the transient em- bryonic AChE activity might be involved in morphogen- esis in the embryo (Drews 1975; Vanittanakom and Drews 1985; Kaehn et al. 1988; Oettling etal. 1989; Small 1990; Umezu et al. 1990).

An antibody to HNK-1, originally raised for human lymphocytes (Abo and Balch 1981), can immunohisto- chemically visualize the area corresponding anatomically to the developing conduction system (Gorza et al. 1988; Ikeda et al. 1990; Ito et al. 1992; Nakagawa et al. 1993). HNK-1 also seems to be related to neural crest cells (Gorza eta. 1988; Luider et al. 1993). Recently, an immu- noelectron microscope study has indicated that the HNK-1 antibody recognizes cell surface and extracellu- lar molecules associated with developing conduction tis- sue cells in rat embryos (Sakai et al. 1994).

In the present study, we addressed the questions whether AChE activity topographically colocalizes with HNK-1 immunoreactivity in the rat embryonic conduc- tion system or not, and the question where AChE reac- tion products are localized ultrastructurally. This mor- phological study may be relevant to the above hypothesis concerning the function of the transient AChE activity in the embryonic heart.

Introduction

The adult pattern of AChE distribution is not completed until 1 month of age in the rat (Navaratnam 1965; Gyevai 1969; Finlay and Anderson 1974), although cardiac con-

Correspondence to: T. Nakamura

Materials and methods

Experimental animals

White Wistar rats weighing about 250 g were purchased from the Kyudo breeding farm (Tosu, Japan) and kept under conventional conditions at Laboratory Animal Center, Nagasaki University School of Medicine. The rats were handled according to the guide- lines of Nagasaki University. After mating, the morning when sperms were observed in a vaginal smear was designated as gesta-

368

Table 1. General view of the development of the rat hearts examined

Embryonic day

14.5 16.5 18.5

Atrium ostium I Narrow Closed Closed AV canal Double Double Double Interventricular Wide Narrow Closed

foramen Distal bulbus Undivided Divided Divided Truncus Divided Divided Divided

Streeter's stage XVIII XXII Fetus I

tional day 0. Embryos were removed from pregnant rats on gesta- tional days 14.5, 16.5, 18.5 and 20.5 (designated as embryonic day (ED) 14.5, 16.5, 18.5 and 20.5) after killing by carbon dioxide. Neonatal rats (day 0) were also examined. The stage of estimated according to the criteria proposed by Sissman (1970) as shown in Table 1, and also described by Ikeda et al. (1990).

Tissue processing

The thoracic region of the embryos and neonates was immersed in periodate-lysine-paraformaldehyde fixative for 4 h at 4 ~ C. Speci- mens were then washed in 15% sucrose-0.1 M sodium phosphate buffer (pH 7.4) for 24 h at 4 ~ C and embedded in OCT compound (Miles Laboratories, Naperville II) in dry-ice ethanol ( - 8 0 ~ C). Cryostat sections were cut serially at 5 gm and subjected to enzyme histochemistry for AChE and immunohistochemistry for HNK-1. Topographical comparisons between AChE-positive and HNK-1 positive areas were performed using adjoining sections.

Enzyme histochemistry for AChE

The basis of our method was that of Karnovsky and Roots (1964). However, our method was in many ways rather different from the original procedure of Karnovsky and Roots, e.g., with respect to buffer, inhibitor, and the conditions of incubation. Our method was rather closer to the methods of Vanittanakom and Drews (1958), Tago et al. (1986) or Figueredo Da Silva et al. (1993). Briefly, sec- tions were preincubated with 5.8 • 10 - 4 M isopropyloc- tametylphosphoramide (iso-OMPA) for 30 min at room tempera- ture to inhibit non-specific cholinesterase. The sections were then incubated for 5 h at 38 ~ C in a solution prepared by sequentially adding the following ingredients: 5.0 mg of acetylthiocholine iodide (Wako Pure Chemical Industries, Osaka, Japan); 6.5 ml of 0.1 M sodium cacodylate buffer, pH 6.0; 0.5 ml of 0.1 M sodium citrate; 1.0 ml of 30 mM copper sulphate; 1.0 ml of 5.8 mM iso-OMPA and 1.0 ml of 5 mM potassium ferricyanide. The reaction was stopped by washing in distilled water. The final reactions products (Hatch- ett's brown) constituted a fine granular, reddish to brown precipi- tate. The sections were counters[ained with hematoxylin.

In control experiments, sections were incubated (1) in a sub- strate-free medium, (2) with an equal concentration of butyrylthio- choline instead of acetylthiocholine, (3) with the general cholinesterase inhibitor physostigmine (eserin) sulfate (final concen- tration: 10 .6 M), (4) with the specific acetylcholinesterase inhibitor 1,5-bis(4-allyldimethyl-ammonium-phenyl)-pentane-3-one-dibro- mide (BW 284 C51, final concentration 10-s).

Immunohistochemistr y for H N K-1

The immunohistochemistry for HNK-1 was carried out by the indi- rect immunohistochemical method. Briefly, sections were incubated

overnight at 4 ~ C with anti-human-Leu-7 (HNK-1) antibody (Bec- ton-Dickinson Immunocytometry System, Mountain View, Calif., USA) diluted to 1 : 20. The sections were then incubated with perox- idase-conjugated rabbit immunoglobulins to mouse immunoglobu- lins (DAKOPATTS, Denmark) diluted to 1:70. The reaction sites were visualized by incubation with 3,3'-diaminobenzidine tetrahy- drochloride (0.2 mg/ml) with 0.005% hydrogen peroxide for 5 rain. The slides were washed in distilled water, counterstained with he- matoxylin and mounted.

The intensity of activity (ACHE, HNK-1) was graded into four categories: negative ( - ) , faintly or partially positive (_+), weakly positive (+) and strongly positive (+ +).

Electron microscopy for AChE

The embryonic hearts were removed on ED 16.5, immersed in 4% paraformaldehyde and 0.5% glutaraldehyde/0.1M cacodylate buffer for 4 h at 4 ~ C, and washed in 0.1 M cacodylate buffer (pH 7.4). Vibratome sections were cut at 100 gm and surveyed by light microscopy to select the appropriate areas. The sections were then subjected to almost the same procedures as for AChE enzyme- histochemistry. However, the incubation time was shortened to on- ly 45 rain as in the work of Figueredo Da Silva et al. (1993). The selected specimens were embedded in Epon, cut into ultrathin sec- tions and observed by electron microscopy (JEM-1200EX, JEOL, Japan).

Results

T opogra ph i c a l c o m p a r i s o n s be tween ACHE- and H N K - 1-posit ive areas in the p resumpt ive c onduc t i on tissue re- gions.

Embryonic day 14.5 (Streeter's stage XVIII)

A pre l im ina ry exper imen t on A C h E act iv i ty unrevea led tha t few reac t ion p roduc t s were seen in the p re sumpt ive c onduc t i on tissue regions recognized by H N K - 1 s ta in ing unt i l E D 14.5. On E D 14.5, few reac t ion p roduc t s were observed, except in the ven t r icu la r t r abecu lae (Fig. l a ) and in pa r t of the A V node region (not shown). On the o ther hand, H N K - 1 i m m u n o r e a c t i v i t y was clear ly evi- den t in mos t of the conduc t ion tissue regions (Fig. 1 b), Thus, there was no ove r l app ing be tween A C h E - p o s i t i v e and H N K - l - p o s i t i v e areas.

Embryonic day 16.5 (Streeter's stage XXII)

On E D 16.5 A C h E reac t ion p roduc t s were obse rved in some par t s of the p re sumpt ive c onduc t i on tissue regions (SA node, in t ra -a t r i a l pa thways , A V node bund le of His and bundle branches) v isual ized by i m m u n o h i s t o c h e m - is t ry for H N K - 1 (Fig. 2a -d ) . As for in t r a -a t r i a l pa thways , four t rac ts (1-4, Fig. 2b, d) seemed to be recognized with H N K - 1 s ta ining as descr ibed by I k e d a et al. (1990) and I to et al. (1992).

However , some except ions were not iced. The first was the A V node region, where the A C h E - p o s i t i v e a rea was in close p rox imi ty to the H N K - 1 posi t ive area, a l t hough there was no ove r l app ing be tween the two (Fig. 2c, d).

369

Fig. 1 a, b. Distribution of AChE and HNK-1 at ED 14.5 x 60. a AChE stain, b HNK-1 stain, a AChE reaction products can be observed in both ventricular tra- beculae, b HNK-1 immunoreactiv- ity can he recognized in the right SA node, around the LSVC, intra- atrial pathways (1, 2, outlined), anterior AV node (outlined), bun- dle of His and the bilateral bundle branches (outlined)

The second exception was that AChE reaction products were also observed in the myocardial sheath, which was continuous with the ventricular trabeculae, in the wall of the aorta and pulmonary trunk (not shown). Still another exception was the region of the ventricular trabeculae, where HNK-1 was negative (Fig. 2d). AChE activity was demonstrated predominantly in the ventricular trabecu- lae, including ordinary cardiac myocytes, but was rather weak in the atrial myocardium (Fig. 2c). As to the junc- tion between the AV node and His bundle, no definite connection of either AChE enzyme activity or HNK-1 immunoreactivity was observed (Fig. 2c, d), even at the stage of ED 16.5. The schema of the distribution patterns of AChE activity (Fig. 3 a) and HNK- 1 immunoreactivity (Fig. 3b) on ED 16.5 is shown in Fig. 3.

Fetal and postnatal stage (Streeter's stage fetus 1 ~ newborn)

From ED 18.5 until birth, both AChE reaction products and HNK-1 immunoreactivity faded away in the conduc- tion tissue regions. The intensity of AChE activity (Table 2) and HNK-1 immunoreactivity (Table 3) in the embryonic heart is summarized in Tables 2 and 3.

Ultrastructure of AChE-positive cells in the conduc- tion tissue regions.

AChE-positive areas

SA node and bundle of His. The reaction products of AChE were intensely associated with the nuclear envel-

ope or endoplasmic reticulum (ER), while few products were seen in mitochondria (Fig. 4a). These cells exhibited the following ultrastructural characteristics (Fig. 4a): (1) relatively large cell size, (2) spindle or polygonal shape, (3) scanty oranelles, (4) poorly formed myofibrils, (5) rich in glycogen and (6) poorly developed junctional apparatus. These were the features of immature conduction tissue cells (Viragh and Challice 1983) that were observed in the HNK-l-positive conduction tissue (SA node, AV node and bundle of His) by Sakai et al. 1994.

AV node region and trabeculated ventricular muscle. In these areas, the reaction products of AChE were associat- ed with the nuclear envelope and ER, but a small amount of reaction products was seen in myofibrils and in mito- chondria (Fig. 4b, c). However, the ultrastructural char- acteristics of the cells (Fig. 4 b, c) contrasted sharply with those in the SA node and bundle of His: (1) relatively small size, (2) round or spindle shape, (3) abundant or- ganelles such as ER, mitochondria and Golgi apparatus, (4) abundant myofibrils with Z bands, (5) scanty glycogen and (6) well-developed junctional apparatus. These char- acteristics corresponded to those of developing ordinary myocytes. In AChE-positive cells in the trabeculae, the dense M band, which is characteristic of the fetal Purkin- je fiber (Forsgren 1982), could not be found.

AChE-negative areas

AVnode region. The AChE-negative area in the AV node region was composed of cells exhibiting the features of developing conduction tissue cells (Viragh and Challice

370

Fig. 2a-d. Distribution of AChE and HNK-I at ED 16.5 x 80. a, c AChE stain; b, d HNK-1 stain, a AChE products can be clearly seen around the RSVC and in the right atrium as two pathways (outlined). Compared with the dense distribution of HNK-1 posi- tive cells, the AChE-positive cells are distributed sparsely and show a dot-like nuclear staining, b The first and second intra-atrial path- ways (1, 2 outlined) seem to extend along the anterior and anterolat- eral wall of the right atrium from the SA node. The fourth intra- atrial pathway (4) may be present, though discontinuous, and run along the lateral wall of the right atrium, e, d The third intra-atrial pathway (3, outlined) seems to reach the posterior part of the AV node region in d. The AChE ac- tivity is weak and sparse but can be clearly seen around the RSVC, in the third intra-atrial pathway (outlined) and in the bundle of His in e. Intense AChE reaction prod- ucts can be recognized in the tra- becular muscles of the right ven- tricle, the regions near the posteri- or AV node and the right bundle branch in e. In the AV junctional region, the AChE-positive area is in close proximity to the area with HNK-1, but is not colocalized (outlined). Between the AV node and the AV bundle, a wide sepa- ration of the two positive sites by mesenchymal tissue can be seen in both AChE and HNK-1 stains

1983). This area may correspond to the AV node of con- duct ion tissue demons t ra ted by Sakai et al. 1994.

Regions other than conduction tissue. In the AChE-nega- tive areas of the right atrial wall and both ventr icular walls, the cells showed ul t ras t ruc tura l characteristics identical to those of o rd inary developing myocytes. Thus

the A C h E react ion products were localized predomi- nan t ly intracel lulary in the nuclear envelope or ER, no t extracellularly. Also on ED 16.5, when such t rans ient A C h E activity was most clear and most intense in the heart, nerve terminals, synapt ic structures or vesicles were no t detected near the AChE-pos i t ive cells by light and electron microscopy.

RSVC

a AChE b HNK-1

1 ; S t r o n g ~ ; W e a k

371

Fig. 3. Schematic drawings of ACHE- (a) and HNK-1- (b) positive areas at ED 16.5. D a r k area: AChE activity or HNK-1 immunoreactivity is strongly positive. D o t t e d area:

AChE activity is weakly posi- tive

Table 2. Intensity of AChE enzyme activity in the embryonic heart. - , negative; • faintly or partially positive; +, weakly positive; + +, strongly positive

Embryonic day 14.5 16.5 18.5 newborn Streeter's stage XVIII XXII fetus I

SA - + + ~ + • Intra-atrial - § + ~ _+ - [1,2, 3,4] AV _+ + + ~ + + _+ His - + + ~ - - Trabecula (Branch) + + + + + -

Table 3. Intensity of HNK-1 immunoreac- tivity in the embryonic heart. - , negative; _+, faintly or partially positive; +, weakly positive; + +, strongly positive

Embryonic day 14.5 16.5 18.5 newborn Streeter's stage XVIII XXII fetus I

SA + + + + + - Intra-atrial + + + + + - [1,2, 3,4] AV + + + + _+ His + + + + _+ Branch + + ~ _+ _+ Trabecula • ~ - _+ ~ - _+

Discuss ion

In the embryon ic stage, the conduct ing tissue is still de- veloping, and innervat ion is not completed until 1 m o n t h of age in the rat (Hall 1951; G o m e z 1958; O w m a n et al. 1971; Finlay and Ander son 1974); A C h E activity is k n o w n to be histochemically detected in such embryonic hearts (Nava ra tnam 1965; Gyevai 1969; Finlay and An- derson 1974; Drews 1975; Lamers et al. 1987, 1990). However, the funct ion of the embryonic A C h E system is still unknown. Recently, Lamers et al. (1990) demonst ra t - ed histochemically the tempora l expression of A C h E ac- tivity in the embryonic heart. Fur thermore, Oettling et al. (1989) and Lamers et al. (1990) suggested that the embry-

onic form of the A C h E system might be different f rom the adult form, a l though it is evident that A c h e activity ex- ists in the immature au tonomic nervous system (Hall 1957; Paff et al. 1966; Coraboeu f et al. 1970; Walker 1975).

M a n y at tempts have been made to visualize the con- duct ion tissue in early developmental stages in experi- mental animals and h u m a n embryos (Forsgren et al. 1982; De G r o o t et al. 1987; Lamers et al. 1987; Gorza e ta l . 1988; Ikeda etal . 1990; Vitadello etal . 1990; Watanabe et al. 1992). A m o n g these, H N K - 1 seems to be a reliable marker of neural crest cells and the presump- tive conduc t ion tissue in rat and h u m a n embryos (Gorza et al. 1988; Ikeda et al. 1990; I to et al. 1992; N a k a g a w a

372

Fig. 4a-e. Electron microscopic features of AChE-positive cells at ED 16.5. a AChE-positive cell in SA node region and bundle of His. Ultrastructural features such as scanty organelles and myofibrils are characteristic of immature conduction tissue cells. Note the rich glycogen (g/y) in the cytoplasm. AChE reaction products can be seen predominantly on the nuclear envelope (arrows). x 7500; Bar 1 gin. b, e AChE-positive cell in AV node and ventricular trabecu-

lae. The cells have abundant organelles such as mitochondria, ER and Golgi apparatus. They are rich in myofibrils (MF) with Z bands, but have little glycogen. These characteristics correspond to those of developing ordinary myocytes. AChE is associated with nuclear envelope (arrows) or ER, but a small amount of reaction product could be seen in the myofibrils and in mitochondria (MT). b x 9200; Bar 1 gin; c x 19000; Bar 0.5 gm

et al. 1993; Sakai et al. 1994), but it is not plausible that conduction tissue cells are derived from the neural crest (Luider et al. 1993). Recently, Sakai et al. have demon- strated with the immunogold labeling technique that HNK-1 is ultrastructurally localized on the cell surface and in the extracellular matrix of cells that have features in common with conduction tissue cells in the embryonic rat heart (Sakai et al. 1994).

In this study, on ED 16.5, no nerve terminals, synaptic structures or vesicles, suggesting neurotransmitter pro- duction were detected near the AChE-positive cells by light and electron microscope. Also, the predominantly intracellular localization of AChE reaction products seemed not to be compatible with the neurotransmitter concept. Thus, the transient AChE activity (embryonic muscarinic system) in the enbryonic heart might be differ- ent from that of the definitive adult form as it relates to the neurotransmitter, as described by Lamers etal . (1987, 1990) or Oettling et al. (1989).

The present study, a topographic comparison between transient AChE activity and transient HNK-1 im- munoreactivity, demonstrated partial overlapping in the presumptive conduction tissue, although the ultrastruc- tural localization of the two substances was different.

In general, HNK-1 is known to be expressed in mi- grating neural crest cells and is involved in morphogene- sis in the embryo (Gorza et al. 1988; Ikeda eta. 1990; Ito et al. 1992; Luider et al. 1993; Nakagawa et al. 1993). On the other hand, AChE activity has also been thought to be involved in morphogenesis during embryonic devel- opment (Drews 1975; Vanittanakom and Drews 1985;

Kaehn etal. 1988; Oettling etal. 1989; Small 1990; Umezu et al. 1990). For example, recently Kaehn et al. (1988) have demonstrated an increase of AChE activity during myotome formation, and this suggests that AChE might be related to myotome differentiation and move- ment. Considering the above, such a transient AChE ac- tivity during embryonic development seems to be related rather to morphogenetic movement or myogenic differ- entiation.

One of the remarkable findings in the present study was that AChE activity was observed in close proximity to HNK-1 immunoreactivity, but not colocalized in the AV node region, and that reaction products were local- ized in myofibril-rich cells (possibly ordinary myocytes). This suggests that the function of AChE is different from that of HNK-1 in this region.

Another unique finding is that AChE activity was pre- dominant in the ventricle but rather weak in the atrium. May the intense AChE activity in the trabecular muscle of the ventricle be related to fetal Purkinje fibers (Gyevai 1969; Forsgren et al. 1982), or related to morphogenetic movement or myogenic differentiation? However, we could not identify the characteristic appearance of fetal Purkinje fibers, such as a dense M band or nodular clus- ters (Forsgren et al. 1982) in the trabecular muscle. Fur- ther detailed examination may be needed to elucidate the functions or the relationship on the transient expression of AChE and HNK-1 in the embryonic heart.

Acknowledgements. The authors wish to thank Prof. Paul Nakane and Dr. Mitsuhiro Tsujihata for their valuable advice on AChE

373

enzyme histochemistry. We are also indebted to Dr. Huminao Takeshima and Mrs. Yoko Iwasaki for their histochemical tech- niques.

References

Abo T, Balch CM (1981) A differentiation antigen of human NK and K cells identified by a monoclonal antibody (HNK-1). J Immunol 127:1024 1029

Coraboeuf E, Obrecht-Coutris G, Le Douarin G (1970) Acetyl- choline and the embryonic heart. Am J Cardiol 25:285 291

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 tis- sue? Anat Embryol 176:515-523

Drews U (1975) Cholinesterase in embryonic development. Prog Histochem Cytochem 7:1-52

Figueredo Da Silva J, Hosokawa T, Okada T, Kobayashi T, Seguchi H (1993) Localization of acetylcholinesterase activity in the mouse heart. Acta Histochem Cytochem 26:197-202

Finlay M, Anderson RH (1974) The development of cholinesterase activity in the rat heart. J Anat 117:239 248

Forsgren S, Carlsson E, Strehler E, Thornell L-E (1982) Ultrastruc- rural identification of human fetal Purkinje fibres - a compara- tive immunocytochemical and electron microscopic study of composition and structure of myofibrillar M-regions. J Mol Cell Cardiol 14:437~449

Gomez H (1958) The development of the innervation of the heart in the rat embryo. Anat Rec 130:53-71

Gorza L, Schiaffino S, Vitadello M (1988) Heart conduction system: a neural crest derivative? Brain Res 457:360 366

Gyevai A (1969) Comparative histochemical investigations con- cerning prenatal and postnatal cholinesterase activity in the heart of chickens and rats. Acta Biol Acad Sci Hung 20:253-262

Hall EK (1951) Intrinsic contractility in the embryonic rat heart. Anat Rec 111:381-399

Hall EK (1957) Acetylcholine and epinephrine effects on the embry- onic rat heart. J Cell Comp Physiol 49:187 200

Ikeda T, Iwasaki K, Shimokawa I, Sakai H, Ito H, Matsuo T (1990) Leu-7 immunoreactivity in human and rat embryonic hearts, with special reference to the development of the conduction tissue. Anat Embryol 182:553-562

Ito H, Iwasaki K, Ikeda T, Sakai H, Shimokawa I, Matsuo T (1992) HNK-1 expression pattern in normal and bis-diamine-induced malformed developing rat heart; three-dimensional reconstruc- tion analysis using computer graphics. Anat Embryol 186:327 334

Kaehn K, Jacob HJ, Christ B, Hinrichsen K, Poelmann RE (1988) The onset of myotome formation in the chick. Anat Embryol 177:191-201

Karnovsky M J, Roots L (1964) A "direct-coloring" thiocholine method for cholinesterase. J Histochem Cytochem 12:219-221

Lamers WH, Kortschot AT, Los JA, Moorman AFM (1987) Acetyl- cholinesterase in prenatal rat heart: a marker for the early devel- opment of the cardiac conductive tissue ? Anat Rec 217: 361-370

Lamers WH, Geerts WJC, Moorman AFM (1990) Distribution pat- tern of acetylcholinesterase in early embryonic chicken hearts Anat Rec 228:297 305

Luider TM, Bravenboer N, Meijers C, Van der Kamp AWM, Tib- boel D, Poelmann RE (1993) The distribution and characteriza- tion of HNK-1 antigens in the developing avian heart. Anat Embryol 188:307-316

Nakagawa M, Thompson RP, Terracio L, Borg TK (1993) Develop- mental anatomy of HNK-1 immunoreactivity in the embryonic rat heart: co-distribution with early conduction tissue. Anat Embryol 187 : 445~460

Navaratnam V (1965) The ontogenesis of cholinesterase activity within the heart and cardiac ganglia in man, rat, rabbit and guinea-pig. J Anat 99:459 467

Oettling G, Schmidt H, Drews U (1989) An embryonic Ca + mobi- lizing muscarinic system in the chick embryo heart. J Dev Phys- ioI 12:85 94

Owman CH, Sjoberg NO, Swedin G (1971) Histochemical and chemical studies on pre- and postnatal development of the dif- ferent systems of "short" and "long" adrenergic neurons in peripheral organs of the rat. Z Zellforsch Mikrosk Anat 116:319-341

Paff GH, Boucek RJ, Glander TP (1966) Acetylcholinesterase- acetylcholine, an enzyme system essential to rhythmicity in the preneural embryonic chick heart. Anat Rec 154:675 684

Sakai H, Ikeda T, Ito H, Nakamura T, Shimokawa I, Matsuo T (1994) Immunoelectron microscopic localization of HNK-1 in the embryonic rat heart. Anat Embryol 190:13-20

Sissman NJ (1970) Developmental landmarks in cardiac morpho- genesis: comparative chronology Am J Cardiol 25:141 148

Small DH (1990) Non-cholinergic actions of acetylcholinesterases: proteases regulating cell growth and development? Trends Biochem Sci 15:213 216

Tago H, Kimura H, Maeda T (1986) Visualization of detailed acetylcholinesterase fiber and neuron staining in rat brain by a sensitive histochemical procedure. J Histochem Cytochem 34:1431-1438

Umezu Y, Yoshizuka M, Ueda H, Ogata H, Fujimoto S (1990) Acetylcholinesterase activity of developing muscles in the lower limb of the rat. Acta Anat 138:332-340

Vanittanakom P, Drews U (1985) Ultrastructural localization of cholinesterase during chondrogenesis and myogenesis in the chick limb bud. Anat Embryol 172:183 194

Viragh SZ, Challice CE (1983) The development of the early atri- oventricular conduction system in the embryonic heart. Can J Physiol Pharmacol 61:775-792

Vitadello M, Matteoli M, Gorza L (1990) Neurofilament proteins are co-expressed with desmin in heart conduction system my- ocytes. J Cell Sci 97:11 21

Walker D (1975) Functional development of the autonomic innerva- tion of the human fetal heart. Biol Neonate 25:31-43

Watanabe M, Timm M, Fallah-Najmabadi H (1992) Cardiac ex- pression of polysialylated NCAM in the chicken embryo: corre- lation with the ventricular conduction system. Dev Dyn 194:128-141