Calcitonin gene-related peptide (CGRP): immunocytochemical identification of a neuropeptide synthesised by ventral paraventricular magnocellular neurones in the sheep

Bra& Research, 611 (1993) 147-151

147

© 1993 Elsevier Science Publishers B.V. All rights reserved 0006-8993/93/$06.00

BRES 25643

Short Communications

Calcitonin gene-related peptide (CGRP)" immunocytochemical identification of a neuropeptide synthesised by ventral paraventricular magnocellular neurones in the sheep A l l a n E. H e r b i s o n , J a n e E. R o b i n s o n a n d D o n a l C. S k i n n e r Department of Neurobiology, AFRC h~stitute of Animal Physiology and Genetics Research, Babraham, Cambridge (UK) (Accepted 2 February 1993)

Key words: Calcitonin gene-related peptide; lmmunocytochemistry; Paraventricular nucleus; Pituitary gland; Sheep; Supraoptic nucleus

The distribution of calcitonin gene-related peptide (CGRP)-immunoreactive neurones was examined in the hypothalamus and pituitary gland of the short-term ovariectomised ewe. A large number of magnocellular CGRP-immunoreactive neurones were identified in the ventral paraventricular nuclei (PVN); few were found in the dorsal PVN and supraoptic nuclei. Parvicellular CGRP-immunoreactive neurones were identified in low density scattered throughout the preoptic region, anterior and basal hypothalamus and region of the stria terminalis. A dense CGRP innervation of the median eminence and neural lobe of the pituitary was observed. These observations reveal substantial species differences in CGRP immunoreactivity compared with the rat and show that magnocellular CGRP-synthesising neurones in the sheep are essentially restricted to, and define, the ventral PVN. This suggests a functionally distinct role for this previously neglected division of the PVN within the ovine hypothalamo-neurohypophyseal system.

The paraventricular (PVN) and supraoptic (SON) nuclei are complicated neural structures comprised of magnocellular and parvocellular neurones which project to the posterior pituitary, median eminence, brainstem and spinal cord Is. Immunocytochemical studies have established, that in addition to oxytc cin and vasopressin, a large number of other neuropeptides are synthesised by neurones in the PVN and SON m'a.Some neuropeptides such as corticotrophin releasing factor (CRF) are synthesised predominantly by parvocellular neurones while others identify sub-populations of magnocellular oxytocin a n d / o r vasopressin neurones Ls. Although most work in this system is conducted in the rat, comparable anatomical and neurochemical arrangements appear to exist in other species including the sheep. Indeed, in the ovine brain, magnocellular and parvocellular neurones have been identified to contain a variety of neuropeptides similar to that of the ratg, lo,12-14. Calcitonin gene-related peptide (CGRP) is a 37 amino acid neuropeptide resulting from alternative transcription of the calcitonin gene within neurones ~7.

Although immunocytochemical studies in the rat have revealed an abundance of CGRP-immunorcactive (-IR) structures i, the peripheral nervous system and brainstem, relatively few CGRP-synthesising neurones have been detected in the forebrain 21. Recent work in this laboratory has revealed that a large population of previously unidentified CGRP-IR neurones exist in the preoptic area of the female rat 4. Furthermore, the vast majority of these neurones possess oestrogen receptors 5, making them good candidates for mediating gonadal steroid actions in this area of the brain. Because of the similarities between the sheep and rat in both oestrogen receptor distribution within the preoptic area t''7 and its role in regulating reproductive functioning, we sought to examine whether a similar population of CGRP neurones existed in the female ovine brain. We present here the results of an immunocytochemical study which reveals that, although relatively few CGRP-IR cells can be identified in the preoptic area of the sheep, a very large number of magnocellular CGRP-IR neurones exist in, and indeed define, the ventral division of the ovine PVN.

Correspondence: A.E. Herbison, Department of Neurobiology, Babraham, Cambridge CB2 4AT, UK. Fax: (44) (223) 836 614.

148 Four Clun Forest ewes were ovariectomised in the anoestrous season under halothane anaesthesia. Seven to 10 days later, ewes were anaesthetised with sodium pentobarbitone (20 mg/kg b.wt., i.v.) and perfused through both carotid arteries with 3 I of 4% paraformaldehyde in phosphate buffered saline (pH 7.6). Prior to the perfusion, heparin (5,000 IU) was injected directly into the carotid artery. Brains were removed into the 4% paraformaldehyde solution for 4-5 h at room temperature before being transferred into a 30% sucrose solution and kept at 4°C. Three days later, a 1 : 3 series of 30 ~m thick coronal sections was cut through the preoptic area and rostral hypothalamus on a freezing microtome and processed for immunocytochemistry. One set of sections was stained with Cresyl violet and used for reference. The pituitary and adjoining mediobasal hypothalamus were cut into 30/~m thick sagittal sections. All sections were washed in a 0.05 M Tris-buffered saline (TBS) and endogenous peroxidases deactivated in 40% methanol, 0.3% H202 TBS for 30 min at room temperature. Immunostaining was carried out on free-floating sections using a well characterised polyclonal antisera 2'4 recognising aCGRP (1:3,000, Amersham, UK). Sections were incubated in primary antibody for 60-70 h at 4°C followed by biotinylated goat anti-rabbit immunoglobulins (1:200, Vector, UK). Streptavidin-biotin-peroxidase (1:200, Amersham, UK) was used as the final label before being treated with 3,3'-diaminobenzidine tetrahydrochloride (DAB) and nickel ammonium sulphate. All antibodies were diluted in TBS containing 0.25% bovine serum albumin and 0.3% Triton X-100. Controls included omission of primary antibody, substitu. tion of primary antibody with 1% normal rabbit serum and incubation in primary antibodies followed by inappropriate secondary immunoglobulins. No specific immunoreactivity was visible in sections following any control procedures. Preabsorption of the CGRP antiserum with ~t-CGRP peptide (2 ~M; Peninsula Labs, USA) overnight at 4°C abolished all specific staining in sheep ~,cctions. A striking population of magnoeellular CGRP-IR cells (diameters 20-34 p.m) was identified in the ventral division of the ovine PVN (Figs. lb, c and 2B, C). Magnocellular CGRP neurones were identified in the ventral periventricular region in a continuum stretching from the preoptic area (Fig. In) through to the ventral PVN (Fig. lc) and the beginning of the dorsal PVN (Fig. ld). In rostral sections, CGRP immunoreactivity was confined to a circular group of magnocellular neurones, also defined in Cresyl violet-stained sections, which lay adjacent to the third ventricle (Fig. la,b). Magnocellular CGRP-IR neurones were also identified

b

a

c

"(

c

d

IC

A(:

... "i

IC I,~IPVN. / ~ ,..~. J~. - - - ' ." ~/. ~

_-

Fig. I. Camera hlcida diagrams of CGRP-IR cell bodies and fibres in the rostral hypothalamus of a representative ovariectomised ewe. e, magnocellular neurone;., parvicellular neurone; shading represents fibres. AC, anterior commissure; C, eaudate nucleus; dPVN dorsal division of the paraventricular nucleus; F, fornix; IC, internal capsule; ME, rostral median eminence; OC, optic chiasm; OT, optic tracts; SCN, suprachiasmatic nucleus; ST, region of the stria terminails, vPVN, ventral division of the paraventricular nucleus. Note that magnocellular CGRP-IR neurones are restricted mostly to the vPVN and its continuum into the rostral preoptic area a. For clarity, not all magnocellular CGRP-IR neurones are indicated within the vPVN. For example, in c, over 50 immunoreactive neurones were counted in the vPVN. in contrast, the magnoceUular CGRP-IR neurones represented in the supraoptic nucleus overlying the OC and OT, and dPVN are numerically correct.

in the dorsal PVN and SON, but were few in number (Fig. 2D). Elsewhere in the brain sections, small numbers of widely scattered CGRP-IR cells were identified in the preoptic region, anterior and basal hypothalamus and region of the stria terminalis (ST: Fig. 1). These neurones were smaller in size (diameters 7-14 /~m) and exhibited both multi- and bi-polar morphologies (Fig. 2A). Fibres immunoreactive for CGRP were identified flowing out of the ventral PVN down to-

149

B

llli,, t

•

e

-

~o .~

• o ,. '

li ~ ,

"" ":";"

"~

. ¢'

~'

"';

I

' ~

' ~l

~.,..~

.

' ~' i t •

~'" .J

•

o

.i.j. . ,...., ,'lq" r '

"i

i'

"

b t.t'i~ re'.

~,- ..... ,

"d';

;~

" i'.-./' t l

i

-

k " . ~:~":

' '4.,'

'

,

"I..

~-

,

D

C

.

i

*

"

"

~.

• ~ ~ " • ~ ~,'.;-_..,.'-'-

.

, . s ,,..,,.-. ~,.'. ....

'

.~,~.

,,

~

•

~,

,: r '...~,..~.~,;,~;

".

~.,,,..

•

~

.,.,

~,'~.

,

].

.,,~

-'":'- "~" :.~.--~'

.~ .... _

/:.

:. ,.. ...: , . - ~ .

.

' .

,~

'~..~:.~;.

.

..d •

I

'

,

,,

~'

'

. ,..

".~,.-~..k'"

~'~

,

~7'~

"~*~,, -.~,..

...../ ' . ,

~, ~ "

"~,~'

.

.~ ...~,,I ~

.'~.,

" @

• ;'...

' .~

•".. : ...

." ,

•

x"

."~j..,

,. 5; ~' "~

. " ,

.. ~*. '

.

-,

"

.

-

"

....J+ .... ,.

"o~.'~ ~,-# • "~..., J

""

--:

..

.

.

.

.

'~.~i ~

, i'

. '~

,'

..

" . ' " . ~ ~.~'~"~"f~

i

"'~I1

~

~

~

~

;

~' f.~

"

-~

""

"~'~

Fig. 2. Photomicrographs of CGRP immunoreactivity in the hypothalamus (A-D) and pituitary gland (E) of the ovariectomised sheep. A: a single bipolar CGRP-IR cell in the preoptic area. B: low power view of magnoceUular CGRP-IR neurones in bilateral ventral paraventricular nuclei. Note the abundance of CGRP-IR fibres passing out laterally from the nucleus. C: higher power view of magnocellular CGRP-IR neurones in the vPVN. D: magnocellular CGRP-IR neurones in the supraoptic nucleus (son) sandwiched between the optic chiasm (oc) and the anterior hypothalamus above. E: CGRP-IR fibres in the posterior pituitary (pp). Note that the adjacent anterior pituitary (ap), in this case the pars intermedia, is devoid of CGRP immunoreactive elements. Bar = 20 ~m (A), 250 i~m (B) and I00 ~,m (C-E).

150 wards the SON (Figs. lb, c and 2B). Elsewhere, large numbers of CGRP-IR fibres were observed only in the ST (Fig. 1), median eminence and neural lobe of the pituitary gland (Fig. 2E). No CGRP-IR structures were identified in the pars distalis, intermedia or tuberalis. This study provides the first description of CGRP-IR structures in the ovine hypothalamus and pituitary. The most striking observation is the identification of a large number of magnocellular CGRP-IR neurones located, predominantly, in the rostro-ventral division of the PVN. In the sheep, cells comprising the PVN have been described adjacent to the third ventricle in both ventral and dorsal cell groups 16'2°. The wing-shaped dorsal PVN has been divided into the pars horizontalis and pars verticalis 2° and clearly corresponds to the PVN of the rat. The ventral PVN has no equivalent in the rat and extends from a position immediately above the suprachiasmatic nucleus to form a similar, but smaller, wing-shaped structure in the ventral hypothalamus before joining with the dorsal PVN above 2°. Although this ventral division has been recognised in the sheep for some time, it has received scant attention. There appears to be little doubt that it is part of the hypothalamo-neurohypophyseal system as oxytocin and CRF-immunoreactive neurones '~'~4,as well as vasopressin messenger RNA-containing cells (personal communication, S. Matthews, Ontario), have been identified in this region. Further, in common with the dorsal PVN, the ventral division has a high density of angiotensin binding sites ~~. We have not, at this stage, determined the pattern of co-localisation of CGRP within oxytocin, vasopressin or CRF neurones in the ventral PVN. Nevertheless, the absence of significant numbers of magnocellular CGRP neurones outside the ventral PVN makes it very likely that this structure is the source of the dense CGRP innervation of the pituitary neural lobe. Certainly, CGRP-IR neurones in the ventral PVN exhibit a magnocellular morphology and their axons project out and down to the SON in a manner reminiscent of oxytocin neurones 9. The locations of cell bodies providing the CGRP innervation to the median eminence remain to be determined. It is interesting to note that, in both the pig ~'~ and goat 3, strikingly similar populations of magnocellular neurones have been identified in positions analogous to the ovine ventral PVN. In the case of the goat, this population has also been labelled the PVN 3. In the pig, recent investigations have identified a circular group of vasopressin/oxytocin-containing cells in the rostral hypothalamus which is sexually dimorphic 19. It is intriguing, therefore, that CGRP synthesis is a marker for some sexually dimorphic neural populations in the rat brain 4.15 and that we have found CGRP immunore-

activity in the sheep localised to the ventral PVN. Whether or not this ventral division of the PVN with its rostral projection into the preoptic area is sexually dimorphic or even appropriately named, remains to be determined. Our current finding of relatively specific CGRP synthesis by ventral PVN magnocellular neurones suggests that this division may serve a functionally distinct role within the ovine neurohypophyseal complex. In comparison with the ovariectomised or intact female rat 4 substantial species differences in CGRP immunoreactivity within the hypothalamus and pituitary gland have been revealed. The object of this investigation was to determine whether the ovine preoptic area contained the similarly large population of CGRP neurones found in the rat 4. This does not appear to be the case as CGRP neurones were few in number and widely scattered in the ovine preoptic area. In the rat, the only other identified CGRP-IR cell population within the hypothalamus is located at the lateral border of the anterior hypothalamus and this cell group projects to the lateral septum 21. Interestingly, in the sheep, comparable cells have not been identified and similarly, no CGRP immunoreactivity was detected in the ovine septum. In common with the rat, however, a very strong CGRP-IR fibre plexus was noted in the stria terminalis. In the rat this CGRP projection is believed to arise from the brainstem parabrachiai nuclei and a similar pathway may exist in the sheep, in the pituitary gland, we note that the pars distalis contains CGRP in the rat e but not in the sheep. Conversely, the rat does not exhibit the impressive CGRP innervation of the neural lobe identified here in the sheep. In summary, we provide immunocytochemical evidence for CGRP-containing neural populations in the ovine hypothalamus which exhibit both similarities and differences with those of the rat. Specifically, we have found no evidence for a significant population of CGRP-IR cells in the ovine preoptic area but, instead identified a large number of magnocellular CGRPsynthesising neurones almost exclusively in the ventral PVN. As such, CGRP immunoreactivity defines the anatomical distinction between the dorsal PVN common to many species, and the ventral PVN which appears to also exist in the goat and pig. Our results raise the possibility that the previously neglected magnocellular neurones in the ventral PVN may be a functionally distinct sub-population within the hypothalamo-neurohypophyseal system of the sheep. RJ. Bicknelifor criticalreadingof the manuscriptand Mrs. J. Cummings for secretarial assistance. A.E.H. is a Lister lnstitute4enner Fellow. D.C.S. is supported by the C.J. Adams

We thank Dr.

151 Trust, the Foundation for Research and Development (South Africa) and an Overseas Research Studentship (UK). This work was supported by a grant from the Journals of Reproduction and Fertility Ltd. 1 Brownstein, M.J. and Mezey, E. Multiple chemical messengers in hypothalamic magnocellular neurons. In T. Hokfelt, K. Fuxe and B. Pernow (Eds), Prog. Brain Res., Vol. 68, Elsevier, Amsterdam, 1986, pp. 161-168. 2 Gon, G., Giaid, A., Steel, J.H., O'Halioran, D.J., van Noorden, S., Ghatei, M.A., Jones, P.M., Amara, S.G., lshikawa, H., Bloom, S.R. and Polak, J.M. (1990) Localization of immunoreactivity for calcitonin gene-related peptide in the rat anterior pituitary during ontogeny and bonadal steroid manipulations and detection of its messenger ribonucleic acid, Endocrinology, 127, 2618-2629. 3 Hamada, T., Shimizu, T., lchikawa, M. and Mori, Y. (1992) lmmunohistochemical study on gonadotrophin-releasing hormone neurons in the Shiba goat brain, J. Reprod. Del'., 38, 133-142. 4 Herbison, A.E. (1992) Identification of a sexually dimorphic neural population immunoreactive for calcitonin gene-related peptide (CGRP) in the rat medial preoptic area, Brain Res., 591, 289-295. 5 Herbison, A.E. and Theodosis (1992) D.T., Immunocytochemical identification of oestrogen receptors in preoptic neurones containing calcitonin gene-related peptide in the male and female rat, Neuroendocrinology, 56, 761-764. 6 Herbison, A.E. and Theodosis, D.T. (1992) Localization of oestrogen receptors in preoptic neurons containing neurotensin but not tyrosine hydroxylase, cholecystokinin or luteinising hormonereleasing hormone in the male and female rat, Neuroscience, 50, 283-298. 7 Herbison, A.E., Robinson, J.E. and Skinner, D.C., Distribution of estrogen receptor-immunoreactive cells in the preoptic area of the ewe: co-iocalisation with glutamic acid decarboxylase but not luteinising hormone-releasing hormone, Neuroendocrinology, in press. 8 Hokfelt, T., Meister, B., Villar, M.J., Ceccatelli, S., Cortes, R., Schalling, M. and Everitt, B. (1989) Hypothalamic neurosecretory systems and their messenger molecules, Acta Physiol. Scand., 136 Suppl. 583, 105-11 I. 9 Kendrick, K. and Keverne, K.B. (1992) Control of synthesis and

release of oxytocin in the sheep brain, Am1. IVY Acad. ScL, 652, 102-121. 10 Matthews, S.G., Heavens, R.P. and Sirinathsinghji, D.J.S. (1991) Cellular localization of corticotropin releasing factor mRNA in the ovine brain, Mol. Brahr Res., 11, 171-176. i ! McKinley, M.J., Allen, A., Clevers, J., Denton, D.A. and Mendelsohn, F.A.O. {1986) Autoradiographic localization of angiotensin receptors in the sheep brain, Brain Res., 375, 373-376. 12 Paull, W.K., Scholer, J., Arimura, A., Meyers, C.A., Chang, J.K., Chang, D. and Shimizu, M. (1982) Immunocytochemical localization of CRF in the ovine hypothalamus, Peptides, 1, 183-191. 13 Papadopoulos, G.C., Karamanlidis, A.N., Michaloudi, H., Dinopoulos, A., Antonopoulos, J. and Parnavelas, J.G. (1985) The coexistence of oxytocin and corticotropin-releasing factor in the hypothalamus: an immunocytochemical study in the rat, sheep and hedgehog, Neurosci. Lett., 62, 213-218. 14 Papadopoulos, G.C., Antonopoulos, A., Karamanlidis, A.N. and Michaloudi, H. (1990) Coexistence of neuropeptides in the hypothalamic paraventricular nucleus of the sheep, Neuropeptides, 15, 227-233. 15 Popper, P. and Micevych, P.E. (1989) The effect of castration on calcitonin gene-related peptide in spinal motor neurons, Neuroendocrinology, 50, 338-343. 16 Richard, P., Atlas stereotaxique du cert,eau de Brebis, Institute National de la Recherche Agronomique, Paris, 1967, 60 pp. 17 Rosenfeld, M.G., Mermod, J.-J., Amara, S.G., Swanson, L.W., Sawchenko, P.E., Rivier, J., Vale, W.W. and Evans, R.M. (1983) Production of a novel neuropeptide encoded by the calcitonin gene via tissue-specific RNA processing, Nature, 304, 129-135. 18 Swanson, L.W. and Sawchenko, P.E. (1983) Hypothalamic integration: organisation of the paraventricular and supraoptic nuclei, Annu. Ret,. Neurosci., 6, 269-324. 19 van Eerdenburg, F.J.C.M., Poot, P., Molenaar, G.J., van Leeuwen, F.W. and Swaab, D.F. (1990) A vasopressin and oxytocin containing nucleus in the pig hypothalamus that shows neuronal changes during puberty, J. Comp. NeuroL, 301,138-146. 20 Welento, J., Szteyn, S. and Milart, Z. (1969) Observations on the stereotaxic configuration of the hypothalamus nuclei in the sheep, Anat. Anz. Bd., 124, 1-27. 21 Yamamoto, A.K. and Tohyama, M. (1989) Calcitonin gene-related peptide in the nervous tissue, Prog. NeurobioL, 33, 335-386.