Neuroscience Letters, 120 (1990) 167-170
167
Elsevier Scientific Publishers Ireland Ltd. NSL 07343
Differential occurrence and distribution of galanin in adrenal nerve fibres and medullary cells in rodent and avian species H . J . Z e n t e l 1'*, D. N o h r 1, S. Mfiller 1, N. Y a n a i h a r a 2 a n d E. W e i h e 1 ~Department of Anatomy, Johannes Gutenberg University, Mainz ( F.R.G.) and ZShizuoka College of Pharmacy, Shizuoka (Japan)
(Received 20 June 1990; Revisedversion received 20 August 1990; Accepted 22 August 1990) Key words: Adrenalcortex; Adrenal medulla; Galanin; Innervation; Rodent; Bird
The presence and distribution of galanin (GAL) in adrenal glands of rodent and avian specieswas investigated by light microscopicimmunohistochemistry. GAL immunoreactivity was found in all medullary cells of guinea pig, duck and chicken adrenals. In contrast, only a subpopulation of medullary cells stained for GAL in Phodopus (Djungarian hamster) while the neuropeptide was completely missing in chromaffin cells of rat and pigeon. In rat, guinea pig and pigeon, GAL-immunoreactive nerve fibres were frequent in subcapsular regions and sparse in deeper cortical layers and in the chromaffin tissue. In contrast, only very few GAL fibres were found in Phodopus and no GAL fibres were observed in the adrenal glands of duck. In the chicken adrenal gland, fibres containing GAL were numerous throughout the organ and occurred in close vicinity to both steroidogenic as well as catecholaminogenic cells. The striking differences in the presence of GAL-positive cells and fibres are more pronounced between species within the rodent or avian group, respectively,than between the different vertebrate orders. The hitherto unknown and surprising variability of GAL expression and distribution in adrenal glands of various species suggests species-dependentfunctional (autocrine, paracrine and/ or endocrine) roles of GAL in the neuroadrenal axis.
Galanin (GAL) originally isolated from porcine intestine [17] has only slight structural similarities with other neuropeptides. It is frequently expressed in the central nervous system, especially in regions known to be involved in endocrine feedback mechanisms (cf. ref. 16). Peripheral GAL-immunoreactive (ir) fibres m a y stem from sympathetic [7, 10], parasympathetic [10], spinal ganglia (cf. refs. 6, 10), cranial m o t o r neurons [14] and/ or the intrinsic gut neurons (cf. ref. 16). G A L is known to cause smooth muscle contraction (cf. ref. 16) and appears to be involved in endocrine mechanisms like the suppression of insulin secretion [12] and the control of the hypothalamic-hypophyseal-axis (cf. ref. 16). Estrogen-dependent expression of G A L in pituitary cells [19] led to the suggestion that G A L m a y be a pituitary hormone. That G A L occurs in the adrenal gland of several species has been concluded from radioimmunoassay and in situ hybridization data [16]. On the other hand, it has been suggested that G A L m a y be absent from the rat adrenal gland [1]. Immunohistochemical data on the presence of G A L in adrenal medullary cells [2, 4, 15] and
*Present address: Research Laboratories of Schering, A.G., 1000 Berlin 65, F.R.G. Correspondence: E. Weihe, Department of Anatomy, Johannes Gutenberg University, Saarstrasse 19--21,D-6500 Mainz, F.R.G. 0304-3940/90/$ 03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd.
in nerve fibres of adrenal cortex and medulla are sparse [2, 15]. Some c o m m o n laboratory animals like the guinea pig have not been studied at all. In the light of these controversies we found it worthwhile to provide the missing knowledge on the speciesspecific localization of G A L immunoreactivity in the adrenal glands of rodents, namely guinea pig, rat and the Djungarian hamster (Phodos sungorus). In addition, we intended to resolve the unexplored question whether G A L might also be present in the adrenal gland of birds. Adult rats and guinea pigs were deeply anesthetized with sodium pentobarbital i.p. (60 mg/kg b.wt.) and perfused with Bouin's solution (without acetic acid) or formaldehyde/glutaraldehyde mixtures. The birds (pigeon, duck and chicken, resembling species of 3 different orders) were killed by CO 2 or decapitation. Adrenal glands were rapidly removed and fixed by immersion in either Bouin's solution without acetic acid or BouinHollande. Deparaffinized sections 4-7 # m thick were incubated overnight in monoclonal mouse anti-GAL antibody (Yanaihara, M C A - G F L ) diluted 1:500 in 1% normal swine serum followed by a 30 min incubation with biotinylated sheep anti-mouse I g G (1:50) (Amersham) and 2 h in the streptavidin-biotinylated horseradish peroxidase complex (Amersham). The peroxidase reaction was amplified by nickel a m m o n i u m sulfate. Endogenous peroxi-
168
TABLE I DISTRIBUTION
O F G A L IN A D R E N A L
GLANDS
- a b s e n t ; ( + ) e x t r e m e l y rare; + sparse; + + m o d e r a t e ; + + + dense. Rat
G u i n e a pig
Hamster
-
+ + +
+ +
Capsular regions
+ +
+
(+ )
S t e r e o i d o g e n i c tissue
+ +
+ +
-
C h r o m a f f i n tissue
+
+ +
(+ )
+ +
C h r o m a f f i n cells
Pigeon
Duck
Chicken
+ + +
+ + +
+ + +
+
+ + +
+ +
-
+ + +
Fibres in:
dase activity was reduced by preincubation in 0.1% H202 in methanol for 30 min. Preabsorption of the primary antiserum in its final dilution with 1 nmol/ml G A L greatly reduced the staining. Ten nmol/ml completely abolished the staining. Immunolabelling was not reduced by preabsorption with up to 100 nmol/ml of substance P, vasoactive intestinal peptide (VIP), peptide histidine isoleucine, neuropeptide Y (NPY) or methionine-enkephalin-Arg-Gly-Leu. There are striking inter-species differences in the presence of ir-GAL in adrenal medullary cells and cortical as well as medullary fibres. The results are summarized in Table I. In guinea pig, nearly all chromaffin cells stained strongly for GAL. In addition, GAL-ir fibres were found in the capsule. Subcapsular fibres either run in nerves associated with small arteries or G A L fibres travelled between cells of the zona glomerulosa. They were rarely seen in deeper cortical layers. In the medulla, GAL-ir fibres were closely associated with GAL-ir chromaffin cells (Fig. 1). In rat, only a marginal staining of chromaffin cells was observed. Further, medullary ganglion cells exhibited faint G A L staining. GAL-ir fibres were associated with subcapsular arteries, running occasionally in the zona glomerutosa and in small nerves passing deeper cortical layers. Finally, medullary GAL-ir fibres were present (Fig. 2). In the adrenal of Phodopus, about 50% of the chromaffin cells, namely the noradrenergic cells, stained strongly for GAL. This was revealed by staining consecutive sections for GAL and phenylalanine-N-methyl-transferase, the enzyme catalyzing the reaction resulting in the formation of adrenalin. Extremely rare GAL-ir fibres were present in the adrenal capsule, in association with small subcapsular arteries and in the medulla (Fig. 3). In the duck, all chromaffin cells but no fibres stained for G A L (Fig. 6). Also in the chicken adrenal, all medullary cells were GAL positive. In addition, many fibres were GAL-ir.
+ + +
They travelled in thick superficial bundles and in tissue septa around the islets of steroidogenic tissue. Thin fibres branched off to run between the steroidogenic and chromaffin cells. Although GAL-ir fibres were found throughout the organ, they appeared to be more frequent in subcapsular regions than in the center of the organ (Fig. 4). In the pigeon, chromaffin cells were G A L negative. In contrast, many fibres stained for GAL. Such fibres were most densely distributed superficially and in tissue septa. GAL-ir fibres were only rarely seen in deeper layers of the gland. Such fibres predominated in association with chromaffin cells (Fig. 5). The present study reveals substantial species differences in the presence and distribution of ir-GAL in rodent and avian adrenal glands. Theoretically, the molecular specificity of the anti-GAL antibody and inter-species differences in the amino-acid sequences of G A L may be crucial for the interpretation of our results. However, the monoclonal antibody used recognized GAL immunoreactivity in all species investigated and G A L was found in cells and fibres of mammalian as well as avian adrenals. In support of this view similar results were obtained by use of polyclonal anti-GAL sera (Peninsula, Milab). Hence, the species differences observed did not correlate with evolutionary distances between the species. The evolutionary distances of different species, however, should correlate with peptide-homologies. We conclude that it is very unlikely that the described species-specific patterns of G A L expression and distribution may be attributable to molecular differences between mammalian and avian GAL. Whether the lack of GAL in chromaffin cells of rat and pigeon is due to the absence of synthesis or to rapid release can only be clarified by use of hybridization studies. The apparently wide phylogenetic distribution of GAL in medullary cells of mammals like rodents, ruminants, non-ruminants, carnivores and man as well as birds supports the idea of a functional paracrine, autocrine and/or endocrine role of this peptide released from
169
2
4 f
.'S
•
\ °-
Fig. 1. Distribution of G A L in adrenal glands of mammals (1-3) and birds (4-6). In guinea-pig (1) and chicken (4) G A L is present in cells and fibres; in rat (2) and pigeon (5) only fibres stain for GAL, whereas in Phodopus(3) and duck (6) only cells are GAL-ir. 1, x 520; 2, x 680; 3, x 280; 4, x 420; 5, x 420; 6, x 440.
chromaffin cells. In light of the recently proposed role of GAL as a pituitary hormone [19], we would like to suggest an additional adrenal origin of the 'peptide-hormone GAL' for most species studied so far. Interestingly, GAL was found in medullary cells of guinea pig which are almost exclusively adrenaline-producing, it coexists in this species with opioids but not with NPY (Zentel and Weihe, unpublished observations). In Phodopus,however, GAL was mainly localized in noradrenaline cells where it coexists with both opioids and NPY (Zentel and Weihe, data not shown here). In these lines, GAL has been recently reported to be present in adrenalin cells of cat and mouse and noradrenalin cells of hamster [15]. Previous reports exhibited sparse GAL-ir innervation in subcapsular and capsular regions of the adrenal gland [2] and describe some GAL-ir fibres in the medulla of the hamster adrenal [15]. In contrast to these observations, we demonstrate GAL-positive fibres not only in the organ capsule and subcapsular areas but also in deeper
cortical layers and, possibly more important, medullary fibres in rat, and, as a novelty in guinea pig, Phodopus, pigeon and chicken. In the adrenal glands of duck, medullary GAL-ir fibres might have been masked by the cellular staining of the chromaffin tissue. The origin of GAL-fibres is unclear. GAL has been reported to be widely distributed in the peripheral nervous system (cf. refs. 6, 7, 10, 14, 16). The distribution of GAL-ir fibres in the adrenal of guinea pig and rat parallels that of sensory tachykinin-ir fibres [8, 20]. A possible sensory origin of GAL fibres is supported by the absence of GAL-ir fibres in rats treated neonatally with capsaicin [15]. On the other hand, VIP-ir fibres are similarly distributed as sensory and GAL-ir fibres in the mammalian adrenal [5]. This possibility is in line with the recent demonstration of coexistence of VIP and GAL in the rat nodose, laryngeal and thyroid ganglia [3] and in the chicken respiratory tract [11]. Even additional sources of adrenal GAL fibres such as collaterals from intrinsic gut neurons, as reported for VIP fibres project-
170
ing to the inferior [13] and superior mesenteric ganglion [9], may have to be considered. Common patterns of GAL-ir innervation in mammalian and avian adrenal glands are deducible: The chromaffin tissue is a common target of GAL fibres. The mammalian zona glomerulosa receives a GAL innervation, as does the subcapsular steroidogenic zone of the pigeon which is regarded as the equivalent of the mammalian zona glomerulosa [18]. Therefore, a galaninergic innervation of both the chromaffin tissue and the zona glomerulosa or its homologue in birds appears to be evolutionary conserved. This underlines the possible specificity of GAL innervation in regulating catecholaminergic and mineralocorticoid synthesizing tissues. In conclusion, our results demonstrate that inter-species variations in adrenal GAL systems may be more striking than differences between vertebrate classes. The extent of the species differences may reflect different, species-dependent trophic actions or regulatory (autocrine/ paracrine and endocrine) roles of this neuropeptide. Thus, the data obtained in this study may serve as an until now missing chemoanatomical basis for a specific functional role of GAL.
We thank S. Bechtloff, M. Stuhltr/iger and A. Leibold for skillful technical assistance. Supported by the German Research Foundation (Grant We 910/2-1/2-2). Parts of the study have been presented at the 8th National Scientific Meeting of the Bayliss and Starling Society, Cambridge, 1988 (see ref. 20).
1 Bauer, F.E., Adrian, T.E., Yanaihara, N., Polak, J.M. and Bloom, S.R., Chromatographic evidence for high-molecular-mass galanin immunoreactivity in pig and cat adrenal glands, FEBS Lett., 201 (1986) 327 331. 2 Bauer, F.E., Hacker, G.W., Terenghi, G., Adrian, T.E., Polak, J.M. and Bloom, S.R., Localization and molecular forms of galanin in human adrenals: elevated levels in pheochromocytomas, J. Clin. Endocrinol. Metab., 63 (1986) 1372-1378. 3 Grundnitz, T., H~kanson, R., Sundler, F. and Uddman, R., Neuronal pathways to the rat thyroid revealed by retrograde tracing and immunocytochemistry, Neuroscience, 24 (1988) 321-335. 4 Hacker, G.W., Bishop, A.E., Terenghi, G., Varndell, T.M., Aghahowa, J., Pollard, K., Thurner, J. and Polak, J.M., Multiple peptide production and presence of general neuroendocrine markers detected in 12 cases of human phaeochromocytoma and in mammalian adrenal glands, Virchows Arch. (A), 412 (1988) 399-411.
5 Holzwarth, M.A., The distribution of vasoactive intestinal peptide in the rat adrenal cortex and medulla, J. Autonom. Nerv. Syst., 11 (1984) 269-283. 6 Ju, G., H6kfelt, T., Brodin, E., Fahrenkrug, J., Fischer, J.A., Frey, P., Elde, R.P. and Brown, J.C., Primary sensory neurons of the rat showing calcitonin gene-related peptide immunoreactivity and their relation to substance P-, somatostatin-, galanin-, vasoactive intestinal polypeptide- and cholecystokinin-immunoreactive ganglion cells, Cell Tissue Res., 247 (1987) 417-431. 7 Kummer, W., Galanin- and Neuropeptide Y-like immunoreactivities coexist in paravertebral sympathetic neurons of the cat, Neurosci. Lett., 78 (1987) 127-131. 8 Kuramato, H., Kondo, H. and Fujita, T., Substance P-like immunoreactivity in adrenal chromaffin cells and intra-adrenal nerve fibers of rats, Histochemistry, 82 (1985) 507-512. 9 Lindh, B., H6kfelt, T. and Elfvin, L.-G., Distribution and origin of peptide-containing nerve fibers in the celiac superior mesenteric ganglion of the guinea-pig, Neuroscience, 26 (1988) 1037-1071. 10 Lindh, B., Lundberg, J.M. and H6kfelt, T., NPY-, galanin-, VIP/ PHI-, CGRP- and substance P-immunoreactive neuronal subpopulations in cat autonomic and sensory ganglia and their projections, Cell Tissue Res., 256 (1989) 259-273. 11 Luts, A., Uddman, R. and Sundler, F., Neuronal galanin is widely distributed in the chicken respiratory tract and coexists with multiple neuropeptides, Cell Tissue Res., 256 (1989) 59-103. 12 McDonald, T.J., Dupre, J., Tatemoto, K., Greenberg, G.R., Radzuik, J. and Mutt, V., Galanin inhibits insulin secretion and induces hyperglycaemia in dogs, Diabetes, 34 (1985) 192-196 13 Masuko, S. and Chiba, T., Projection pathways, coexistence of peptides and synaptic organization of nerve fibres in the inferior mesenteric ganglion of the guinea-pig, Cell Tissue Res., 253 (1988) 507-516. 14 Moore, R.Y., Cranial motor neurons contain either galanin- or calcitonin gene-related peptide-like immunoreactivity, J. Comp. Neurol., 282 (1989) 512 522. 15 Pelto-Huikko, M., Immunocytochemical localization of neuropeptides in the adrenal medulla, J. Electron. Microsc. Techn., 12 (1989) 364-379. 16 R6kaeus, A., Galanin: a newly isolated biologically active neuropeptide, Trends Neurosci., 10 (1987) 158-164. 17 Tatemoto, K., R6kaeus, ,~., J6rnvall, H., McDonald, M.J. and Mutt, V., Galanin - - a novel biologically active peptide from porcine intestine, FEBS Lett., 164 (1983) 124-128. 18 Unsicker, K., Fine structure and innervation of the avian adrenal gland IV. Fine structure of interrenal cells. Z. Zellforsch., 146 (1973) 385-402. 19 Vrontakis, M.E., Yamamoto, T., Schroedter, I.C., Nagy, J.I. and Friesen, H.G., Estrogen induction of galanin in the rat anterior pituitary gland demonstrated by in situ hybridization and immunohistochemistry, Neurosci. Lett., 100 (1989) 59~4. 20 Zentel, H.J. and Weihe, E., Targets and interrelations of nerves coding for substance P, multiple non-tachykinin peptides and/or catecholamines in guinea-pig adrenal gland: multiple messengers in sensory and sympathetic neuroendocrine interaction?, Regul. Pept., 22 (1988) 192.