Cell Tiss. Res. 172, 15 27 (1976)
Cell and Tissue Research :,C~by Springer-Verlag 1976
Postnatal Ontogenic Development of the Adrenergic Innervation Pattern in the Rat Portal Vein A Histochemieal Study* Jan Lundberg, Bengt Ljung, Dorothy Stage (Mc Murphy), and Annica Dahlstr6m Institute of Neurobiology and Department of Physiology, University of G6teborg, G6teborg, Sweden
Summary. The postnatal development of the adrenergic innervation pattern in the rat portal vein has been studied with the histochemical fluorescence method of Hillarp and Falck. Stretch preparations and transverse freeze-dried sections of intact portal veins were studied from rats during the first 5 weeks of life and from adult rats. Orientation of undifferentiated smooth muscle cells into two layers was observed at 4 days of age. Dominance of the thick outer longitudinal muscle layer was apparent at two weeks of age. A terminal adrenergic nerve plexus with some varicosities was restricted outside the media at the end of the first week. Ingrowth of penetrating non-terminal adrenergic nerve fibers through the longitudinal muscle layer occurred during the second week of age when the main terminal nerve plexus was developing between the two muscle layers. After 3 weeks of age the adult pattern of a two-dimensional adrenergic plexus between the muscle was established. In the adult rat pharmacological treatment with nialamide and noradrenaline revealed the thin, penetrating non-terminal adrenergic nerve fibers in the longitudinal muscle layer which were poorly visible otherwise. The present observations strongly indicate that the main adrenergic plexus between the two muscle layers emanates directly from the outer axonal plexus. These findings are discussed regarding possible trophic interactions between ingrowing sympathetic adrenergic vasomotor nerves and maturing vascular smooth muscle. Key words: Portal vein - Rat - Adrenergic innervation Ontogenesis - Histofluorescence.
Postnatal -
Send offprint requests to: Dr. Jan Lundberg, Institute of Neurobiology, University of G6teborg,
Medicinaregatan 5, S-400 33 G6teborg 33, Sweden * Supported by grants from the Swedish Medical Research Council (grants No. 14X-2207, O4P-4173, 3884), Magn. Bergwall's Foundation, G. & M. Lindgren's Foundation, the Medical Faculty of University of G6teborg. The technical assistance of Miss Serney B66j, Mr. P/ir-Anders Larsson and Miss Ann Kjellstedt is gratefully acknowledged
16
J. Lundberg et al.
Introduction
Detailed knowledge of the ontogenesis of the sympathetic adrenergic neurons provides a fundamental basis for the understanding of the development of sympathetic neuro-effector control. Therefore, the ontogenetic development of peripheral adrenergic nerves has previously been carefully studied in the rat iris, vas deferens, heart and gut by means of fluorescence histochemistry (De Champlain et al., 1970; Owman et al., 1971). It was found that although adrenergic axons were present already before birth in various tissues, the adult pattern of innervation with typical varicose nerve terminals was not seen until about 2 weeks postnatally. The one exception was the gut, in which a fully developed adrenergic plexus was observed immediately after birth. It is well known that considerable variation occurs in the development of sympathetic cardiovascular control not only between different species, but also within various parts of the circulatory system of one particular species (for ref. see Rudolph and Heymann, 1974). The neuro-effector control of the longitudinal smooth muscle of the rat portal vein was recently found to develop in dramatic sequence during the first few postnatal weeks (Ljung and Stage, 1975). This vessel has been widely used as a suitable model for in vitro studies of propagating vascular smooth muscle (for ref. see Ljung, 1976). In the adult portal vein the media consists of a thick, outer longitudinal layer and a thin inner circular layer (Tsfio et al., 1971). Its vasomotor nerve supply consists of a two-dimensional plexus of adrenergic nerve terminals located between the longitudinal and the circular muscle layers. In addition, a more sparse network of adrenergic fibres is located at the adventitio-medial junction, i.e. just outside the longitudinal media layer (Johansson et al., 1970) (see Fig. 1). In the present study the postnatal development of the peculiar pattern of adrenergic nerve terminal distribution within the rat portal vein has been studied by means of the histochemical fluorescence method of Hillarp and Falck (for ref. see Falck and Owman, 1965; Corrodi and Jonsson, 1967). The primary aim of the study was to obtain information about the outgrowth and maturation of the sympathetic neurons, but the results also indicate important aspects on the ultimate problem of possible trophic links between the maturation of smooth muscle effector cells and the ingrowth of adrenergic axons.
Material and Methods Albino rats of the Sprague-Dawley strain were used. Seven age groups, each consisting of 16 rats, were studied ; newly born, 4 days, 7 days~ 2, 3, 5 weeks and adult (250 300 g). The rats were killed by decapitation and the portal vein from just caudal to the splenic vein up to the hepatic bifurcation cranially was dissected out (Fig. 1). In each age group 8 specimens were air-dried as stretch preparations and 8 were freeze-dried (see below). Two preparation techniques for fluorescence histochemistry were used: a) Stretch Preparations (Fig. 1). The portal vein was carefully cut open longitudinally from the junction between splenic and mesenteric veins up to the hepatic bifurcation. Adjacent veins were included in the preparation but left unopened to serve as aid in orientation. The tissue was then spread out on a glass-slide and air-dried over phosphopentoxide according to Malmfors (1965). For every immature tissues a hair-dryer was initially used after spreading to rapidly decrease the
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
17
high water content of the tissue. The dried stretch preparations were treated with parafotnnaldehyde gas, generated from formaldehyde powder, equilibrated over a relative air humidity of 70% (see Hamberger, 1967), at 80~ for 1 h (Malmfors, 1965).
b) Freeze Drying. Intact portal veins were frozen immediately after dissection in liquid propane, cooled by liquid nitrogen and freeze-dried in a Hetovac freeze-drier (model Thieme) for 3 days. The small veins of very young animals were placed in a groove of a piece of splenic tissue as support during the histochemicat procedure. Following paraformaldehyde treatment (see above) the tissues were embedded in paraffine (m.p. 52 ~ C), sectioned transversely in 7-10 la sections and mounted in Entellane | Microscopy was carried out with a Zeiss Junior microscope, equipped for transillumination fluorescence microscopy as described elsewhere (e.g. Malmfors, 1965), Because of the somewhat varying density of innervation in various parts of the vein (Johansson et al., 1970), great care was taken to choose a representative area for photography, using the entrance of the splenic vein as a landmark. For photography Scopix RPI (Gevaert) green sensitive film was used. Exposure times ranged between 45 60 s.
Pharmacological Treatnwnt. Some adult rats were given nialamide (250 mg/kg i.p.) 5 h before an i.v. injection of noradrenaline (NA) (2-5 ~g/kg). Thirty rain later the animals were sacrificed as described above and the portal vein was dissected out. This treatment was performed in order to increase the catecholamine (CA) fluorescence o f the thin adrenergic nerve fibres (cf. Malmfors, 1965), which penetrate the outer longitudinal muscle sheath of the media (see Results). In Vitro Incubations, In order to obtain maximal CA fluorescence in the adrenergic axons, some specimens in each group were incubated with NA in Krebs solution (I Itg/ml) according to Hamberger (1967). In addition, the incubation solution contained Evans blue (50 I,tg/mI) in order to minimize the interfering autofluorescence caused by the elastin component of the vessel wall (De la Lande and Waterson, 1968). After incubation for 5 rain the tissues were rinsed with Krebs solution for 10 min to remove extraneural NA.
Results The postnatal ontogenesis of the adrenergic neurons innervating the rat portal vein was largely found to correspond to the observations in other organs of
'~isp{inic vein
portatvein
supmesenteric vein
~J
~rating
axon
Fig. ]. Schematic drawing showing the anatomical arrangement of the portal vein and related vessels. The magnified cross section illustrates the main components of the portal vein wall; tile two smooth muscle layers in the media and the two plexus of adrenergic nerve terminals, interconnected by penetrating axons
18
J. Lundberg et al.
Fig. 2. The portal vein from a newborn rat. Spread preparation showing large, strongly fluorescent adrenergic axon bundles (--,) and two growth cones (arrow head), x 125
the rat, such as the heart, salivary glands and irides (cf. De Champlain et al., 1970; Owman et al., 1971). The main developmental features of the particular vasomotor innervation of the portal vein as revealed by the pattern of adrenergic nerve distribution at various postnatal ages are described below. A t birth the media of the portal vein consisted of a layer of 2-3 undifferentiated smooth muscle cells without clear spatial organization. Outside this layer large, strongly fluorescent axon bundles were observed. The axons were seen to branch at intervals and the tip of the branches often carried enlargements (Fig. 2) with the characteristics of "growth cones" (cf. Olson, 1969; Bunge, 1973). No nerve terminal network was observed. A t four days of age the cells of the media had begun to orientate themselves into two layers. Outside these two layers the strongly fluorescent axon bundles with growth cones were still present. In addition, a sparse delicate nerve terminal network with occasional varicose-like enlargements was observed (Fig. 3a, b). At seven days postnatally the media of the portal vein was clearly divided into two separate layers, i.e. an outer longitudinal and an inner circular muscle layer. A nerve terminal plexus with some varicosities could be detected, and was restricted to the outer aspect of the media. Axon bundles with growth cones were still seen (Fig. 4a, b). At two weeks of age, the effector cell arrangement of the media began to resemble the adult appearance with a dominating outer longitudinal muscle layer. At this stage the adrenergic nerves were observed not only outside the media, but also had started to penetrate the outer longitudinal layer sending off branches between the two muscle layers. The penetrating fibres were clearly detectable due to their strong fluorescence and were quite numerous (Fig. 5 a, b). The nerve terminal network still had a typically immature appearance with very strongly fluorescent non-varicose parts of the axons. After three weeks of postnatal development the adrenergic innervation of the portal vein started to assume an adult pattern. The fluorescence of nonterminal axons was strikingly weaker in intensity. The nerve terminal network
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
19
Fig. 3A and B. The rat portal vein : four days of age. A Spread preparation illustrating the appearance of the first varicose-like enlargements in the developing plexus of adrenergic terminals ( ~ ) . • 320. B Cross section: portal vein (pv) close to the hepatic artery (ha), / = o u t e r longitudinal (medial) muscle layer, c = i n n e r circular (medial) muscle layer in the portal vein. A sparse network of adrenergic nerve terminals ( ~ ) can be seen outside the media of the portal vein. Large axon bundles (arrow heads) accompany the hepatic artery, x 200
20
J. Lundberg et al.
Fig. 4A and B. The rat portal vein of a seven day old rat. A Spread preparation demonstrating a sparse network of adrenergic fibres with two visible growth cones (arrowheads). x 125. B Cross section ; p v - portal vein, l = longitudinal muscle layer, c = circular muscle layer. Adrenergic varicosities can be seen restricted to the outer aspect of the media ( ~ ) . x 200. Arrow heads indicate bundles of adrenergic axons accompanying the hepatic arteries
had reached an adult-like density with strongly fluorescent varicosities. The dominating adrenergic plexus was now located between the two muscle layers, like in the adult vessel. The number of visible fibres penetrating the outer longitudinal muscle layer had decreased as compared to the situation one week earlier (Fig. 6 a, b). During the subsequent development from five weeks of age until adult, the smooth muscle tissue of the portal vein seemed to increase to a greater relative
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
21
Fig. 5A and B. The rat portal vein at two weeks of age. A Spread preparation showing a network of immature adrenergic nerve terminals with strongly fluorescent intervaricose segments. The specimen was incubated with noradrenaline and Evans blue to increase the specific fluorescence in the nerve terminals and decrease unspecific autofluorescence of connective tissue, x 125. B. Cross section; s t = s u p p o r t i n g piece of splenic tissue; /longitudinal muscle layer; c =circular muscle layer. Fluorescent axons from the outer adrenergic plexus can be seen to penetrate the longitudinal muscle layer (arrowheads) and branch between the two muscle layers ( ~ ) . x 125
extent than the density of adrenergic nerve terminals, resulting not only in a particularly thick longitudinal muscle layer, but also giving the impression of an increased number and/or size of smooth muscle cells as related to nerve varicosities (Fig. 7a, b). In normal adult animals rather few fibres were found to penetrate the longitudinal media layer and thus ,,connect" the main plexus between the two muscle layers with the axons at the adventitio-medial junction (Fig. 7b). However, in animals pretreated with nialamide and NA the number of visible penetrating fibres was clearly increased as compared to untreated adult rats (Fig, 7b, c). NA-incubation of dissected tissues generally resulted in a small increase in the fluorescence intensity of the adrenergic nerves, but no clear change in the number of fibres.
22
J. Lundberg et al.
Fig. 6 A and B. The rat portal vein after three weeks of postnatal development. A Spread preparation showing an adrenergic nerve terminal network with distinct varicosities and weakly fluorescent intervaricose segments, x 125. B Cross section of a portal vein (pv) close to a branch of the hepatic artery (ha)./=longitudinal muscle layer, c=circular muscle layer, large arrow head indicates large axon bundles. Only one penetrating axon can be observed (---,). The dominating adrenergic plexus can be seen between the two muscle layers (small arrowheads), x 125 Fig. 7 A - C . Adult rat portal vein. A Spread preparation demonstrating an adrenergic nerve terminal network of the mature type. x 125. B Cross section from an untreated animal; adv= adventitial layer with autofluorescent connective tissue components: l = longitudinal muscle layer, c = circular muscle layer, arrow heads indicate sparse adrenergic nerve terminals outside the longitudinal muscle layer. One weakly tluorescent penetrating fibre is partly demonstrated ( ~ ) . The main adrenergic nerve terminal plexus can be seen between the two muscle layers (-~). x200. C Cross section from a nialamide-noradrenaline pretreated animal ; adv = adventitial autofluorescent tissue, l = longitudinal muscle layer, c = c i r c u l a r muscle layer. All adrenergic fibres have a strongly increased fluorescence intensity due to the pretreatment. Arrow heads indicate axons located outside the longitudinal muscle layer. Many easily detectable penetrating axons can now be seen ( ~ ) , in continuity with the dominating plexus between the two medial muscle layers (-*). x 200
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
23
24
J. Lundberget al.
Discussion
The histochemical fluorescence method for the detection of tissue CA is very sensitive (for ref. and description, see e.g. Falck and Owman, 1965; Corrodi and Jonsson, 1967). The intensity of fluorescence is closely related to the intracellular concentration of the amine (Jonsson, 1971). The very strong intensity of the outgrowing non-terminal axons and of the gradually appearing terminal varicosities found in the developing rat portal vein indicates the presence of high concentrations of CA. Conversely, the weak fluorescence of the intervaricose segments and the mature non-terminal axons indicates low concentrations of CA. The CA present in the nerve fibres of the portal wall is mainly NA as measured spectrophotometrically (e.g. Stage et al., unpublished). In most cases it is likely that the formaldehyde fluorescence method can detect most, if not all, terminals present in a tissue. However, the NA concentrations of non-terminal axons of adult animals is often too low to allow detection in the fluorescence microscope (cf. Dahlstr6m and Fuxe, 1964; Dahlstr6m, 1965). However, incubation with exogenous NA or pre-treatment with a monoamine oxidase inhibitor, e.g. nialamide, and i.v. NA, raise the amine levels and can also allow the detection of fibres with normally low amine concentrations. In the present study, in vitro NA-incubation of developing portal veins caused a slight increase in fluorescence intensity, but no clear-cut increase in the number of fibres, indicating that the normal CA-concentration was sufficient for demonstrating all fibres in the microscope. However, when adult animals were pretreated with nialamide and NA, not only was the intensity of the fibres enhanced, but also a clear increase in the number of visible nerve fibres penetrating the outer longitudinal muscle layer was observed (Fig. 7b, c). This demonstrates that only a very small part of the existing penetrating fibres are observed in the normal adult animal, which is in agreement with earlier observations (Johansson et al., 1970). Thus, it seems clear that mainly non-terminal axons run between the bundles of longitudinal muscle cells without ramifying and establishing synaptic contacts until the main terminal plexus is formed at the internal aspect of the longitudinal media layer. The present observations in the developing portal vein strongly indicate that the main adrenergic plexus between the two muscle layers emanates from the outer nerve plexus. Outgrowing axons with growth cone-like structures (cf. Olson, 1969) were observed during the first two weeks after birth. At four days of age the first varicose structures indicative of the early development of nerve terminals were seen outside the external muscle layer. This plexus of nerve terminals was more highly developed at the end of the first week, but still only located outside the media. Interesting changes occurred during the second postnatal week when some varicose nerve terminals for the first time were observed between the two muscle layers at the same time as fluorescent fibres were found to penetrate the outer muscular layer. At 3 weeks the adult pattern of innervation was approached with a dense nerve terminal plexus with strongly fluorescent varicosities and weakly fluorescent intervaricose segments. The number of visible penetrating fibres began to decrease. During the suggested process of ingrowth the penetrating fibres may have a high concentration of
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
25
intracellular NA, characteristic of growing axons (De Champlain et al., 1970; Furness et al., 1970). However, when the inner nerve terminal plexus is developed, the non-terminal penetrating axons may assume the low NA levels of nongrowing axons, i.e. levels which are sometimes too low to allow detection with the present histochemical method without pharmacological pretreatment (Fig. 7b, c). In the adult rabbit portal vein mainly non-terminal autonomic nerve fibres have been observed to penetrate into the longitudinal muscle layer, which supports the present findings (Holman et al., 1968). As previously shown all adrenergic innervation of the portal vein originates from the coeliac ganglia (Johansson et al., 1970). It is of interest to correlate the ontogenesis of portal adrenergic innervation with the development of the neuro-effector function of the longitudinal smooth muscle layer. Among the functional parameters previously studied in the developing portal vein of the rat (Ejung and Stage, 1975) were 1) response to exogenous NA (the presence of e-receptors and mechanisms for contraction); 2) response to transmural field stimulation (the presence of functioning neuro-muscular control); and 3) spontaneous activity (indicative of myogenic activity, i.e. single unit type muscle). Ad 1): The first weak response to exogenous NA was observed on day four when in this study the partitioning of the media was found to have begun and the first few varicosities of the developing outer nerve plexus were seen. This may suggest that some adrenergic terminals must develop before the first e-receptor response can be elicited. Ad2): The first sign of response to transmural stimulation was detected towards the end of the first postnatal week. At seven days of age the adrenergic nerves were primarily located outside the longitudinal muscle layer, but it is likely that the ingrowth of the penetrating fibres had commenced. It is interesting to note that the neurogenic responses of the rat portal vein during the second postnatal week were small in amplitude and could be elicited only at very high frequencies and after long delay. In contrast, during the third week the muscular response to nerve stimulation was obtained even at low impulse rates within seconds after onset of stimulation (Ljung and Stage, 1975). The corresponding morphological findings demonstrate that various nerve terminals occur earlier than neurogenic responses can be obtained. Furthermore, the adult pattern of vasomotor nerve distribution is not established until the third postnatal week when the neuro-effector control first assumes the characteristics of the mature vessel. It is evident that an effector response to nerve activation not only requires the presence of varicosities, but also a releasable transmitter that can reach the smooth muscle cells and induce a detectable response. Thus, complicated maturation processes probably have to modify the elements of the neuromuscular function during this delay between morphologically mature varicosities and functional neuro-effector control (see e.g. Furness et al., 1970; Pappano et al., 1974). Ad 3) : The onset of spontaneous activity was found to occur rather abruptly at 16 days of age (Ljung and Stage, 1975). At that time the adrenergic innervation was clearly divided into two compartments: a ) A sparse network outside the longitudinal layer and b) a dense plexus between the two muscle layers. Thus,
26
J. Lundberg et al.
it may be that the maturation of the longitudinal smooth muscle cells into a functional syncytium is trophically correlated to the influence of adrenergic nerves. This hypothesis is supported at least in part by tissue culture experiments where the presence of sympathetic adrenergic nerves was observed to accelerate both the formation of smooth muscle bundles and the development of nexus (Chamley et al., 1974). It is concluded that the functional development of the neuro-effector control of the rat portal vein is closely correlated with the morphological growth and maturation of the adrenergic vasomotor nerve supply. One intriguing question arises from the present results. How is the ingrowth of the adrenergic axons guided during development so as to penetrate the longitudinal muscle layer apparently without branching and making synaptic contacts until the inside of the longitudinal layer is reached? In the longitudinal layer of the adult portal vein, effector cells in junctional proximity to adrenergic nerves seem to be preferentially sensitive to NA due to restricted placement of the c~-adrenergic receptors (Johansson et al., 1970; Ljung etal., 1973; Bevan and Ljung, 1973). It is conceivable that sensitivity to the transmitter is induced in regions exposed to prolonged action of the NA after release, or alternatively that the adrenergic nerves are directed in their growth towards areas with the existing or potential properties of being sensitive to the transmitter. The former alternative would favor the opinion that the effector tissue is primarily formed in response to trophic nervous influence. The latter alternative, however, would indicate that transmitter sensitivity is primarily inherent in the effector tissue and secondarily modified or controlled by adrenergic nervous influences. Future research is required to settle the relative importance of these alternatives on sympathetic vasomotor nerve-vascular smooth muscle interactions.
References Bevan, J.A., Ljung, B.: Restricted placement of c~-adrenergic receptors within the smooth muscle of the rat portal vein. (Abstr.) Acta physiol, scand. 87, 25A (1973) Bunge, M.B.: Fine structure of nerve fibers and growth cones of isolated sympathetic neurons in culture. J. Cell Biol. 56, 713 735 (1973) Chamley, I., Campbell, G.R., Burnstock, G.: Dedifferentiation and bundle formation of smooth muscle cells in tissue culture: the influence of cell number and nerve fibres. J. Embryol. exp. Morph. 32, 297-323 (1974) Corrodi, H., Jonsson, G. : The formaldehyde fluorescence method for the histochemical demonstration of biogenic monoamines. J. Histochem. Cytochem. 15, 2, 65 78 (1967) Dahlstr6m, A.: Observations on the accumulation of noradrenaline in the proximal and distal parts of peripheral adrenergic nerves after compression. J. Anat. (Lond.) 99, 677-689 (1965) Dahlstr6m, A., Fuxe, K. : A method for the demonstration of adrenergic nerve fibres in peripheral nerves. Z. Zellforsch. 62, 602-607 (1964) De Champlain, J., Malmfors, T., Olson, L., Sachs, Ch.: Ontogenesis of peripheral adrenergic neurons in the rat: pre- and postnatal observations. Acta physiol, scand. 80, 276-288 (1970) De la Lande, I.S., Waterson, J.G.: Modification of autofluorescence in the formaldehyde treated rabbit ear artery by Evans blue. J. Histochem. Cytochem. 16, 281-282 (1968) Falck, B., Owman, Ch.: A detailed methodological description of the fluorescence method for cellular demonstration of biogenic monoamines. Acta Univ. Lund. II. 7, 1 23 (1965)
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
27
Furness, J.O., McLean, J.R., Burnstock, G.: Distribution of adrenergic nerves and changes in neuro-muscular transmission in the mouse vas deferens during postnatal development. Develop. Biol. 21,491-505 (1970) Hamberger, B.: Reserpine-resistant uptake of catecholamines in isolated tissues of the rat. A histochemical study. Acta physiol, scand. Suppl. 295, 1-56 (1967) Holman, M.E., Kasby, C.B., Suthers, M.B., Wilson, J.A.F. : Some properties of the smooth muscle of rabbit portal vein. J. Physiol. (Lond.) 196, 111 132 (1968) Johansson, B., Ljung, B., Malmfors, T., Olson, L. : Prejunctional supersensitivity in the rat portal vein as related to its pattern of innervation. Acta physiol, scand. Suppl. 349, 5 16 (1970) Jonsson, G. : Quantitation of fluorescence of biogenic monoamines. Demonstrated with the formaldehyde fluorescence method. Stuttgart, Portland-Oregon (USA): Gustav Fischer 1971 Ljung, B. : Physiological patterns of neuroeffector control mechanisms. Proc. 2nd. Int. Symp. Vascular Neuroeffector Mechanisms. S. Karger, Basel. In press (1976) vein. Circulat. Res. 32, 556-563 (1973) Ljung, B., Bevan, J.A., Su, C.: Evidence for uneven alpha-receptor distribution in the rat portal Ljung, B., Stage (Mc Murphy), D. : Postnatal ontogenetic development of neurogenic and myogenic control in the rat portal vein. Acta physiol, scand. 94, 112-127 (1975) Malmfors, Y. : Studies on adrenergic nerves. The use of rat and mouse iris for direct observations on their physiology and pharmacology at cellular and subcellular levels. Acta physiol, scand. Suppl. 248, 1 93 (1965) Olson, L. : Intact and regenerating sympathetic noradrenaline axons in the rat sciatic nerve. Histochemie 17, 349 367 (1969) Owman, Ch., Sj6berg, N.-O., Swedin, G. : Histochemical and chemical studies on pre-and postnatal development of the different systems of " s h o r t " and "long" adrenergic neurons in peripheral organs of the rat. Z. Zellforsch. 116, 319-341 (1971) Pappano, A.J., L6ffelholz, K. : Ontogenesis of adrenergic and cholinergic neuroeffector transmission in chick embryo heart. J. Pharmacol. exp. Ther. 191, 468478 (1974) Rudolph, A.M., Heymann, M.A. : Fetal and neonatal circulation and respiration. Ann. Rev. Physiol. 36, 187 207 (1974) Ts'ao, Ch., Glagow, S., Kelsey, B.F.: Structure of the mammalian portal vein: Postnatal establishment of two mutually perpendicular medial muscle zones in the rat. Anat. Rec. 1"/1, 457 470 (1971)
Received April 12, 1976
Cell and Tissue Research :,C~by Springer-Verlag 1976
Postnatal Ontogenic Development of the Adrenergic Innervation Pattern in the Rat Portal Vein A Histochemieal Study* Jan Lundberg, Bengt Ljung, Dorothy Stage (Mc Murphy), and Annica Dahlstr6m Institute of Neurobiology and Department of Physiology, University of G6teborg, G6teborg, Sweden
Summary. The postnatal development of the adrenergic innervation pattern in the rat portal vein has been studied with the histochemical fluorescence method of Hillarp and Falck. Stretch preparations and transverse freeze-dried sections of intact portal veins were studied from rats during the first 5 weeks of life and from adult rats. Orientation of undifferentiated smooth muscle cells into two layers was observed at 4 days of age. Dominance of the thick outer longitudinal muscle layer was apparent at two weeks of age. A terminal adrenergic nerve plexus with some varicosities was restricted outside the media at the end of the first week. Ingrowth of penetrating non-terminal adrenergic nerve fibers through the longitudinal muscle layer occurred during the second week of age when the main terminal nerve plexus was developing between the two muscle layers. After 3 weeks of age the adult pattern of a two-dimensional adrenergic plexus between the muscle was established. In the adult rat pharmacological treatment with nialamide and noradrenaline revealed the thin, penetrating non-terminal adrenergic nerve fibers in the longitudinal muscle layer which were poorly visible otherwise. The present observations strongly indicate that the main adrenergic plexus between the two muscle layers emanates directly from the outer axonal plexus. These findings are discussed regarding possible trophic interactions between ingrowing sympathetic adrenergic vasomotor nerves and maturing vascular smooth muscle. Key words: Portal vein - Rat - Adrenergic innervation Ontogenesis - Histofluorescence.
Postnatal -
Send offprint requests to: Dr. Jan Lundberg, Institute of Neurobiology, University of G6teborg,
Medicinaregatan 5, S-400 33 G6teborg 33, Sweden * Supported by grants from the Swedish Medical Research Council (grants No. 14X-2207, O4P-4173, 3884), Magn. Bergwall's Foundation, G. & M. Lindgren's Foundation, the Medical Faculty of University of G6teborg. The technical assistance of Miss Serney B66j, Mr. P/ir-Anders Larsson and Miss Ann Kjellstedt is gratefully acknowledged
16
J. Lundberg et al.
Introduction
Detailed knowledge of the ontogenesis of the sympathetic adrenergic neurons provides a fundamental basis for the understanding of the development of sympathetic neuro-effector control. Therefore, the ontogenetic development of peripheral adrenergic nerves has previously been carefully studied in the rat iris, vas deferens, heart and gut by means of fluorescence histochemistry (De Champlain et al., 1970; Owman et al., 1971). It was found that although adrenergic axons were present already before birth in various tissues, the adult pattern of innervation with typical varicose nerve terminals was not seen until about 2 weeks postnatally. The one exception was the gut, in which a fully developed adrenergic plexus was observed immediately after birth. It is well known that considerable variation occurs in the development of sympathetic cardiovascular control not only between different species, but also within various parts of the circulatory system of one particular species (for ref. see Rudolph and Heymann, 1974). The neuro-effector control of the longitudinal smooth muscle of the rat portal vein was recently found to develop in dramatic sequence during the first few postnatal weeks (Ljung and Stage, 1975). This vessel has been widely used as a suitable model for in vitro studies of propagating vascular smooth muscle (for ref. see Ljung, 1976). In the adult portal vein the media consists of a thick, outer longitudinal layer and a thin inner circular layer (Tsfio et al., 1971). Its vasomotor nerve supply consists of a two-dimensional plexus of adrenergic nerve terminals located between the longitudinal and the circular muscle layers. In addition, a more sparse network of adrenergic fibres is located at the adventitio-medial junction, i.e. just outside the longitudinal media layer (Johansson et al., 1970) (see Fig. 1). In the present study the postnatal development of the peculiar pattern of adrenergic nerve terminal distribution within the rat portal vein has been studied by means of the histochemical fluorescence method of Hillarp and Falck (for ref. see Falck and Owman, 1965; Corrodi and Jonsson, 1967). The primary aim of the study was to obtain information about the outgrowth and maturation of the sympathetic neurons, but the results also indicate important aspects on the ultimate problem of possible trophic links between the maturation of smooth muscle effector cells and the ingrowth of adrenergic axons.
Material and Methods Albino rats of the Sprague-Dawley strain were used. Seven age groups, each consisting of 16 rats, were studied ; newly born, 4 days, 7 days~ 2, 3, 5 weeks and adult (250 300 g). The rats were killed by decapitation and the portal vein from just caudal to the splenic vein up to the hepatic bifurcation cranially was dissected out (Fig. 1). In each age group 8 specimens were air-dried as stretch preparations and 8 were freeze-dried (see below). Two preparation techniques for fluorescence histochemistry were used: a) Stretch Preparations (Fig. 1). The portal vein was carefully cut open longitudinally from the junction between splenic and mesenteric veins up to the hepatic bifurcation. Adjacent veins were included in the preparation but left unopened to serve as aid in orientation. The tissue was then spread out on a glass-slide and air-dried over phosphopentoxide according to Malmfors (1965). For every immature tissues a hair-dryer was initially used after spreading to rapidly decrease the
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
17
high water content of the tissue. The dried stretch preparations were treated with parafotnnaldehyde gas, generated from formaldehyde powder, equilibrated over a relative air humidity of 70% (see Hamberger, 1967), at 80~ for 1 h (Malmfors, 1965).
b) Freeze Drying. Intact portal veins were frozen immediately after dissection in liquid propane, cooled by liquid nitrogen and freeze-dried in a Hetovac freeze-drier (model Thieme) for 3 days. The small veins of very young animals were placed in a groove of a piece of splenic tissue as support during the histochemicat procedure. Following paraformaldehyde treatment (see above) the tissues were embedded in paraffine (m.p. 52 ~ C), sectioned transversely in 7-10 la sections and mounted in Entellane | Microscopy was carried out with a Zeiss Junior microscope, equipped for transillumination fluorescence microscopy as described elsewhere (e.g. Malmfors, 1965), Because of the somewhat varying density of innervation in various parts of the vein (Johansson et al., 1970), great care was taken to choose a representative area for photography, using the entrance of the splenic vein as a landmark. For photography Scopix RPI (Gevaert) green sensitive film was used. Exposure times ranged between 45 60 s.
Pharmacological Treatnwnt. Some adult rats were given nialamide (250 mg/kg i.p.) 5 h before an i.v. injection of noradrenaline (NA) (2-5 ~g/kg). Thirty rain later the animals were sacrificed as described above and the portal vein was dissected out. This treatment was performed in order to increase the catecholamine (CA) fluorescence o f the thin adrenergic nerve fibres (cf. Malmfors, 1965), which penetrate the outer longitudinal muscle sheath of the media (see Results). In Vitro Incubations, In order to obtain maximal CA fluorescence in the adrenergic axons, some specimens in each group were incubated with NA in Krebs solution (I Itg/ml) according to Hamberger (1967). In addition, the incubation solution contained Evans blue (50 I,tg/mI) in order to minimize the interfering autofluorescence caused by the elastin component of the vessel wall (De la Lande and Waterson, 1968). After incubation for 5 rain the tissues were rinsed with Krebs solution for 10 min to remove extraneural NA.
Results The postnatal ontogenesis of the adrenergic neurons innervating the rat portal vein was largely found to correspond to the observations in other organs of
'~isp{inic vein
portatvein
supmesenteric vein
~J
~rating
axon
Fig. ]. Schematic drawing showing the anatomical arrangement of the portal vein and related vessels. The magnified cross section illustrates the main components of the portal vein wall; tile two smooth muscle layers in the media and the two plexus of adrenergic nerve terminals, interconnected by penetrating axons
18
J. Lundberg et al.
Fig. 2. The portal vein from a newborn rat. Spread preparation showing large, strongly fluorescent adrenergic axon bundles (--,) and two growth cones (arrow head), x 125
the rat, such as the heart, salivary glands and irides (cf. De Champlain et al., 1970; Owman et al., 1971). The main developmental features of the particular vasomotor innervation of the portal vein as revealed by the pattern of adrenergic nerve distribution at various postnatal ages are described below. A t birth the media of the portal vein consisted of a layer of 2-3 undifferentiated smooth muscle cells without clear spatial organization. Outside this layer large, strongly fluorescent axon bundles were observed. The axons were seen to branch at intervals and the tip of the branches often carried enlargements (Fig. 2) with the characteristics of "growth cones" (cf. Olson, 1969; Bunge, 1973). No nerve terminal network was observed. A t four days of age the cells of the media had begun to orientate themselves into two layers. Outside these two layers the strongly fluorescent axon bundles with growth cones were still present. In addition, a sparse delicate nerve terminal network with occasional varicose-like enlargements was observed (Fig. 3a, b). At seven days postnatally the media of the portal vein was clearly divided into two separate layers, i.e. an outer longitudinal and an inner circular muscle layer. A nerve terminal plexus with some varicosities could be detected, and was restricted to the outer aspect of the media. Axon bundles with growth cones were still seen (Fig. 4a, b). At two weeks of age, the effector cell arrangement of the media began to resemble the adult appearance with a dominating outer longitudinal muscle layer. At this stage the adrenergic nerves were observed not only outside the media, but also had started to penetrate the outer longitudinal layer sending off branches between the two muscle layers. The penetrating fibres were clearly detectable due to their strong fluorescence and were quite numerous (Fig. 5 a, b). The nerve terminal network still had a typically immature appearance with very strongly fluorescent non-varicose parts of the axons. After three weeks of postnatal development the adrenergic innervation of the portal vein started to assume an adult pattern. The fluorescence of nonterminal axons was strikingly weaker in intensity. The nerve terminal network
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
19
Fig. 3A and B. The rat portal vein : four days of age. A Spread preparation illustrating the appearance of the first varicose-like enlargements in the developing plexus of adrenergic terminals ( ~ ) . • 320. B Cross section: portal vein (pv) close to the hepatic artery (ha), / = o u t e r longitudinal (medial) muscle layer, c = i n n e r circular (medial) muscle layer in the portal vein. A sparse network of adrenergic nerve terminals ( ~ ) can be seen outside the media of the portal vein. Large axon bundles (arrow heads) accompany the hepatic artery, x 200
20
J. Lundberg et al.
Fig. 4A and B. The rat portal vein of a seven day old rat. A Spread preparation demonstrating a sparse network of adrenergic fibres with two visible growth cones (arrowheads). x 125. B Cross section ; p v - portal vein, l = longitudinal muscle layer, c = circular muscle layer. Adrenergic varicosities can be seen restricted to the outer aspect of the media ( ~ ) . x 200. Arrow heads indicate bundles of adrenergic axons accompanying the hepatic arteries
had reached an adult-like density with strongly fluorescent varicosities. The dominating adrenergic plexus was now located between the two muscle layers, like in the adult vessel. The number of visible fibres penetrating the outer longitudinal muscle layer had decreased as compared to the situation one week earlier (Fig. 6 a, b). During the subsequent development from five weeks of age until adult, the smooth muscle tissue of the portal vein seemed to increase to a greater relative
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
21
Fig. 5A and B. The rat portal vein at two weeks of age. A Spread preparation showing a network of immature adrenergic nerve terminals with strongly fluorescent intervaricose segments. The specimen was incubated with noradrenaline and Evans blue to increase the specific fluorescence in the nerve terminals and decrease unspecific autofluorescence of connective tissue, x 125. B. Cross section; s t = s u p p o r t i n g piece of splenic tissue; /longitudinal muscle layer; c =circular muscle layer. Fluorescent axons from the outer adrenergic plexus can be seen to penetrate the longitudinal muscle layer (arrowheads) and branch between the two muscle layers ( ~ ) . x 125
extent than the density of adrenergic nerve terminals, resulting not only in a particularly thick longitudinal muscle layer, but also giving the impression of an increased number and/or size of smooth muscle cells as related to nerve varicosities (Fig. 7a, b). In normal adult animals rather few fibres were found to penetrate the longitudinal media layer and thus ,,connect" the main plexus between the two muscle layers with the axons at the adventitio-medial junction (Fig. 7b). However, in animals pretreated with nialamide and NA the number of visible penetrating fibres was clearly increased as compared to untreated adult rats (Fig, 7b, c). NA-incubation of dissected tissues generally resulted in a small increase in the fluorescence intensity of the adrenergic nerves, but no clear change in the number of fibres.
22
J. Lundberg et al.
Fig. 6 A and B. The rat portal vein after three weeks of postnatal development. A Spread preparation showing an adrenergic nerve terminal network with distinct varicosities and weakly fluorescent intervaricose segments, x 125. B Cross section of a portal vein (pv) close to a branch of the hepatic artery (ha)./=longitudinal muscle layer, c=circular muscle layer, large arrow head indicates large axon bundles. Only one penetrating axon can be observed (---,). The dominating adrenergic plexus can be seen between the two muscle layers (small arrowheads), x 125 Fig. 7 A - C . Adult rat portal vein. A Spread preparation demonstrating an adrenergic nerve terminal network of the mature type. x 125. B Cross section from an untreated animal; adv= adventitial layer with autofluorescent connective tissue components: l = longitudinal muscle layer, c = circular muscle layer, arrow heads indicate sparse adrenergic nerve terminals outside the longitudinal muscle layer. One weakly tluorescent penetrating fibre is partly demonstrated ( ~ ) . The main adrenergic nerve terminal plexus can be seen between the two muscle layers (-~). x200. C Cross section from a nialamide-noradrenaline pretreated animal ; adv = adventitial autofluorescent tissue, l = longitudinal muscle layer, c = c i r c u l a r muscle layer. All adrenergic fibres have a strongly increased fluorescence intensity due to the pretreatment. Arrow heads indicate axons located outside the longitudinal muscle layer. Many easily detectable penetrating axons can now be seen ( ~ ) , in continuity with the dominating plexus between the two medial muscle layers (-*). x 200
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
23
24
J. Lundberget al.
Discussion
The histochemical fluorescence method for the detection of tissue CA is very sensitive (for ref. and description, see e.g. Falck and Owman, 1965; Corrodi and Jonsson, 1967). The intensity of fluorescence is closely related to the intracellular concentration of the amine (Jonsson, 1971). The very strong intensity of the outgrowing non-terminal axons and of the gradually appearing terminal varicosities found in the developing rat portal vein indicates the presence of high concentrations of CA. Conversely, the weak fluorescence of the intervaricose segments and the mature non-terminal axons indicates low concentrations of CA. The CA present in the nerve fibres of the portal wall is mainly NA as measured spectrophotometrically (e.g. Stage et al., unpublished). In most cases it is likely that the formaldehyde fluorescence method can detect most, if not all, terminals present in a tissue. However, the NA concentrations of non-terminal axons of adult animals is often too low to allow detection in the fluorescence microscope (cf. Dahlstr6m and Fuxe, 1964; Dahlstr6m, 1965). However, incubation with exogenous NA or pre-treatment with a monoamine oxidase inhibitor, e.g. nialamide, and i.v. NA, raise the amine levels and can also allow the detection of fibres with normally low amine concentrations. In the present study, in vitro NA-incubation of developing portal veins caused a slight increase in fluorescence intensity, but no clear-cut increase in the number of fibres, indicating that the normal CA-concentration was sufficient for demonstrating all fibres in the microscope. However, when adult animals were pretreated with nialamide and NA, not only was the intensity of the fibres enhanced, but also a clear increase in the number of visible nerve fibres penetrating the outer longitudinal muscle layer was observed (Fig. 7b, c). This demonstrates that only a very small part of the existing penetrating fibres are observed in the normal adult animal, which is in agreement with earlier observations (Johansson et al., 1970). Thus, it seems clear that mainly non-terminal axons run between the bundles of longitudinal muscle cells without ramifying and establishing synaptic contacts until the main terminal plexus is formed at the internal aspect of the longitudinal media layer. The present observations in the developing portal vein strongly indicate that the main adrenergic plexus between the two muscle layers emanates from the outer nerve plexus. Outgrowing axons with growth cone-like structures (cf. Olson, 1969) were observed during the first two weeks after birth. At four days of age the first varicose structures indicative of the early development of nerve terminals were seen outside the external muscle layer. This plexus of nerve terminals was more highly developed at the end of the first week, but still only located outside the media. Interesting changes occurred during the second postnatal week when some varicose nerve terminals for the first time were observed between the two muscle layers at the same time as fluorescent fibres were found to penetrate the outer muscular layer. At 3 weeks the adult pattern of innervation was approached with a dense nerve terminal plexus with strongly fluorescent varicosities and weakly fluorescent intervaricose segments. The number of visible penetrating fibres began to decrease. During the suggested process of ingrowth the penetrating fibres may have a high concentration of
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
25
intracellular NA, characteristic of growing axons (De Champlain et al., 1970; Furness et al., 1970). However, when the inner nerve terminal plexus is developed, the non-terminal penetrating axons may assume the low NA levels of nongrowing axons, i.e. levels which are sometimes too low to allow detection with the present histochemical method without pharmacological pretreatment (Fig. 7b, c). In the adult rabbit portal vein mainly non-terminal autonomic nerve fibres have been observed to penetrate into the longitudinal muscle layer, which supports the present findings (Holman et al., 1968). As previously shown all adrenergic innervation of the portal vein originates from the coeliac ganglia (Johansson et al., 1970). It is of interest to correlate the ontogenesis of portal adrenergic innervation with the development of the neuro-effector function of the longitudinal smooth muscle layer. Among the functional parameters previously studied in the developing portal vein of the rat (Ejung and Stage, 1975) were 1) response to exogenous NA (the presence of e-receptors and mechanisms for contraction); 2) response to transmural field stimulation (the presence of functioning neuro-muscular control); and 3) spontaneous activity (indicative of myogenic activity, i.e. single unit type muscle). Ad 1): The first weak response to exogenous NA was observed on day four when in this study the partitioning of the media was found to have begun and the first few varicosities of the developing outer nerve plexus were seen. This may suggest that some adrenergic terminals must develop before the first e-receptor response can be elicited. Ad2): The first sign of response to transmural stimulation was detected towards the end of the first postnatal week. At seven days of age the adrenergic nerves were primarily located outside the longitudinal muscle layer, but it is likely that the ingrowth of the penetrating fibres had commenced. It is interesting to note that the neurogenic responses of the rat portal vein during the second postnatal week were small in amplitude and could be elicited only at very high frequencies and after long delay. In contrast, during the third week the muscular response to nerve stimulation was obtained even at low impulse rates within seconds after onset of stimulation (Ljung and Stage, 1975). The corresponding morphological findings demonstrate that various nerve terminals occur earlier than neurogenic responses can be obtained. Furthermore, the adult pattern of vasomotor nerve distribution is not established until the third postnatal week when the neuro-effector control first assumes the characteristics of the mature vessel. It is evident that an effector response to nerve activation not only requires the presence of varicosities, but also a releasable transmitter that can reach the smooth muscle cells and induce a detectable response. Thus, complicated maturation processes probably have to modify the elements of the neuromuscular function during this delay between morphologically mature varicosities and functional neuro-effector control (see e.g. Furness et al., 1970; Pappano et al., 1974). Ad 3) : The onset of spontaneous activity was found to occur rather abruptly at 16 days of age (Ljung and Stage, 1975). At that time the adrenergic innervation was clearly divided into two compartments: a ) A sparse network outside the longitudinal layer and b) a dense plexus between the two muscle layers. Thus,
26
J. Lundberg et al.
it may be that the maturation of the longitudinal smooth muscle cells into a functional syncytium is trophically correlated to the influence of adrenergic nerves. This hypothesis is supported at least in part by tissue culture experiments where the presence of sympathetic adrenergic nerves was observed to accelerate both the formation of smooth muscle bundles and the development of nexus (Chamley et al., 1974). It is concluded that the functional development of the neuro-effector control of the rat portal vein is closely correlated with the morphological growth and maturation of the adrenergic vasomotor nerve supply. One intriguing question arises from the present results. How is the ingrowth of the adrenergic axons guided during development so as to penetrate the longitudinal muscle layer apparently without branching and making synaptic contacts until the inside of the longitudinal layer is reached? In the longitudinal layer of the adult portal vein, effector cells in junctional proximity to adrenergic nerves seem to be preferentially sensitive to NA due to restricted placement of the c~-adrenergic receptors (Johansson et al., 1970; Ljung etal., 1973; Bevan and Ljung, 1973). It is conceivable that sensitivity to the transmitter is induced in regions exposed to prolonged action of the NA after release, or alternatively that the adrenergic nerves are directed in their growth towards areas with the existing or potential properties of being sensitive to the transmitter. The former alternative would favor the opinion that the effector tissue is primarily formed in response to trophic nervous influence. The latter alternative, however, would indicate that transmitter sensitivity is primarily inherent in the effector tissue and secondarily modified or controlled by adrenergic nervous influences. Future research is required to settle the relative importance of these alternatives on sympathetic vasomotor nerve-vascular smooth muscle interactions.
References Bevan, J.A., Ljung, B.: Restricted placement of c~-adrenergic receptors within the smooth muscle of the rat portal vein. (Abstr.) Acta physiol, scand. 87, 25A (1973) Bunge, M.B.: Fine structure of nerve fibers and growth cones of isolated sympathetic neurons in culture. J. Cell Biol. 56, 713 735 (1973) Chamley, I., Campbell, G.R., Burnstock, G.: Dedifferentiation and bundle formation of smooth muscle cells in tissue culture: the influence of cell number and nerve fibres. J. Embryol. exp. Morph. 32, 297-323 (1974) Corrodi, H., Jonsson, G. : The formaldehyde fluorescence method for the histochemical demonstration of biogenic monoamines. J. Histochem. Cytochem. 15, 2, 65 78 (1967) Dahlstr6m, A.: Observations on the accumulation of noradrenaline in the proximal and distal parts of peripheral adrenergic nerves after compression. J. Anat. (Lond.) 99, 677-689 (1965) Dahlstr6m, A., Fuxe, K. : A method for the demonstration of adrenergic nerve fibres in peripheral nerves. Z. Zellforsch. 62, 602-607 (1964) De Champlain, J., Malmfors, T., Olson, L., Sachs, Ch.: Ontogenesis of peripheral adrenergic neurons in the rat: pre- and postnatal observations. Acta physiol, scand. 80, 276-288 (1970) De la Lande, I.S., Waterson, J.G.: Modification of autofluorescence in the formaldehyde treated rabbit ear artery by Evans blue. J. Histochem. Cytochem. 16, 281-282 (1968) Falck, B., Owman, Ch.: A detailed methodological description of the fluorescence method for cellular demonstration of biogenic monoamines. Acta Univ. Lund. II. 7, 1 23 (1965)
Ontogenesis of Adrenergic Nerves in Rat Portal Vein
27
Furness, J.O., McLean, J.R., Burnstock, G.: Distribution of adrenergic nerves and changes in neuro-muscular transmission in the mouse vas deferens during postnatal development. Develop. Biol. 21,491-505 (1970) Hamberger, B.: Reserpine-resistant uptake of catecholamines in isolated tissues of the rat. A histochemical study. Acta physiol, scand. Suppl. 295, 1-56 (1967) Holman, M.E., Kasby, C.B., Suthers, M.B., Wilson, J.A.F. : Some properties of the smooth muscle of rabbit portal vein. J. Physiol. (Lond.) 196, 111 132 (1968) Johansson, B., Ljung, B., Malmfors, T., Olson, L. : Prejunctional supersensitivity in the rat portal vein as related to its pattern of innervation. Acta physiol, scand. Suppl. 349, 5 16 (1970) Jonsson, G. : Quantitation of fluorescence of biogenic monoamines. Demonstrated with the formaldehyde fluorescence method. Stuttgart, Portland-Oregon (USA): Gustav Fischer 1971 Ljung, B. : Physiological patterns of neuroeffector control mechanisms. Proc. 2nd. Int. Symp. Vascular Neuroeffector Mechanisms. S. Karger, Basel. In press (1976) vein. Circulat. Res. 32, 556-563 (1973) Ljung, B., Bevan, J.A., Su, C.: Evidence for uneven alpha-receptor distribution in the rat portal Ljung, B., Stage (Mc Murphy), D. : Postnatal ontogenetic development of neurogenic and myogenic control in the rat portal vein. Acta physiol, scand. 94, 112-127 (1975) Malmfors, Y. : Studies on adrenergic nerves. The use of rat and mouse iris for direct observations on their physiology and pharmacology at cellular and subcellular levels. Acta physiol, scand. Suppl. 248, 1 93 (1965) Olson, L. : Intact and regenerating sympathetic noradrenaline axons in the rat sciatic nerve. Histochemie 17, 349 367 (1969) Owman, Ch., Sj6berg, N.-O., Swedin, G. : Histochemical and chemical studies on pre-and postnatal development of the different systems of " s h o r t " and "long" adrenergic neurons in peripheral organs of the rat. Z. Zellforsch. 116, 319-341 (1971) Pappano, A.J., L6ffelholz, K. : Ontogenesis of adrenergic and cholinergic neuroeffector transmission in chick embryo heart. J. Pharmacol. exp. Ther. 191, 468478 (1974) Rudolph, A.M., Heymann, M.A. : Fetal and neonatal circulation and respiration. Ann. Rev. Physiol. 36, 187 207 (1974) Ts'ao, Ch., Glagow, S., Kelsey, B.F.: Structure of the mammalian portal vein: Postnatal establishment of two mutually perpendicular medial muscle zones in the rat. Anat. Rec. 1"/1, 457 470 (1971)
Received April 12, 1976
