- Email: [email protected]
Distribution and cellular localization of vasopressin mRNA in the ovine brain, pituitary and pineal glands
Neuropeptides (1993) 25,11-17 0 Longman Group UK Ltd 1993
Distribution and Cellular Localization of Vasopressin mRNA in the Ovine Brain, Pituitary and Pineal Glands S. G. MATTHEWS*, R. F. PARROTT and D. J. S. SlRlNATHSlNGHJl AFRC institute of Animal Physiology and Genetics Research, Cambridge Research Station, Babraham, Hail, Cambridge UK and *Lawson Research institute, Grosvenor Street, London, Ontario, Canada (Reprint requests to RFP)
Abstract-In this study, in situ hybridization histochemistry was used to determine the regional and cellular localization of vasopressin-neurophysin II (AVP) mRNA in the sheep brain and pituitary with an 36S-labelled synthetic 45-mer oligonucleotide probe complementary to the bovine AVP gene. The highest densities of labelled cell bodies were found in the paraventricular nucleus (PVN), supraoptic nucleus (SON) and suprachiasmatic nucleus (SCN) of the hypothalamus, though such cells were also found in other regions of the diencephalon, including the accessory magnocellular nuclei. Labelled cells were also observed sparsely distributed in every major cortical field as well as in choroid plexus and the pineal gland. No AVP mRNA-expressing cells were found in the bed nucleus of the stria terminalis, the amygdala, or in the medulla and brainstem. In the pituitary, a dense AVP mRNA signal was observed in the intermediate lobe whereas, cells in the anterior or neural lobe did not express AVP mRNA. The dense population of AVP-expressing neurons in both magnocellular and parvocellular fields of the hypothalamus support major roles of AVP in both posterior and anterior pituitary function. Finally, the extrahypothalamic distribution of AVP mRNA transcripts suggest that vasopressinergic neurons may be involved in diverse physiological functions, including the regulation of pineal function and cognition.
Introduction
blood increase in response to stress.3,4 Moreover, AVP serves as a neuropeptide in the CNS, where it has been shown to influence learning and memory, cardiovascular function6*7 and other autonomic responses.**9 It has also been shown to increase intracranial pressure, loprobably by an effect on cerebrospinal fluid (CSF) absorption rate.” Within the hypothalamus, immunohistochemical studies have revealed dense AVP labelling in the supraoptic (SON), paraventricular (PVN) and suprachiasmatic nuclei (SCN).12 However, in com-
The hypophysiotrophic action of AVP in conscious sheep, in addition to its neurohypophysial role, is well established. It is known that exogenous AVP stimulates the ovine anterior pituitary to secrete adrenocorticotrophic hormone (ACTH) in viva’ and in vitro2 and that AVP concentrations in the portal
Date received 5 October 1992 Date accepted 29 December 1992
11
12
NFXJROPEPTDES
mon with other regulatory peptides, similar studies13J4 have indicated widespread distribution of AVP-containing neurons and fibres outside the hypothalamus, suggesting that the peptide may function as a neurotransmitter within the central nervous system (CNS), though levels of AVP immunoreactivity are of a lesser magnitude than those for other hypophysiotrophic peptides described in the ovine brain.15 Recently, the cloning and nucleotide sequencing of the ratI and bovineI AVP-neurophysin II precursor have facilitated the study of mechanisms regulating AVP synthesis at cellular and molecular levels. In this context, the use of in situ hybridization histochemistry has revealed the existence of AVP mRNA within the hypothalamus1zJ8-to as well as in extrahypothalamic regions,20p37 Currently, we are studying the regulation of neuropeptide gene expression in the CNS of the sheep, especially those mechanisms associated with responses to osmotic stimulation and stress. In earlier studies, we have determined the localization of corticotrophin-releasing hormone (CRH) mRNA21v22and preproenkephalin (PENK) mRNAcontaining cells23 in the ovine brain using in situ hybridization histochemistry. In the present investigation, we have used the same procedure with a synthetic oligonucleotide probe complementary to a modified sequence of the bovine AVP-neurophysin II gene to determine the regional and cellular distribution of AVP mRNA-containing cells in the ovine brain. A preliminary account of this research has been presented elsewhere.21 Materials
and methods
Three Clun Forest wethers were killed with Lethobarb (Pentobarbitone sodium BP 20% w/v) (Duphar Veterinary Limited, Southampton, UK), decapitated, and the brains removed within 5 min. Each brain was placed on ice and cut in the frontal plane into the following three blocks: (1) olfactory bulb to the optic chiasma, (2) optic chiasma to the midbrain at the level of the medial mammillary nucleus and (3) midbrain to the brainstem. Each block was frozen on dry-ice, wrapped in parafilm and stored at -7O’C. Sections (10 pm) were then cut on a cryostat (Bright Instruments), mounted onto @oly)L-lysine-coated slides, dried, post-fixed in 4%
paraformaldehyde (5 min), rinsed in phosphatebuffered saline (2x1 min), dehydrated in an alcohol series and stored in 95% alcohol at +4”C until in situ hybridization analysis. The basic method for in situ hybridization has been previously described in detail.24 Briefly, the slides were removed from the alcohol, allowed to air-dry at room temperature and then incubated overnight in a humid atmosphere at 42°C with the radiolabelled oligonucleotide probe in hybridization buffer. The hybridization buffer used for these experiments contained 4xstandard sodium citrate (SSC) (1xSSC contains 150 mM sodium chloride and 15 mM sodium citrate), 50% deionized formamide, 50 mM sodium phosphate (pH 7.0), 1mM sodium pyrophosphate (pH 7.0), 0.02% bovine serum albumin, 200 ,ug/ml hydrolysed salmon sperm DNA, 0.02% Ficoll, 0.02% polyvinylpyrrolidone, 10% dextran sulphate and 40 mM dithiothreitol (DTT). The AVP oligodeoxyribonucleotide probe was labelled using terminal deoxynucleotidyl transferase (Pharmacia) and [35S]deoxyadenosine 5’-(athio)triphosphate (1300 Ci/mmol, New England Nuclear) to a specific activity of 1.0~10~ cpm/ng. The labelled probe was used at a concentration of 3x10’ cpm/pl. 100 pl of labelledprobe in hybridization buffer was applied to each slide. After incubaA 0
441
GTA GAC GCC GGG CTG GGC GGG CTC CGC GGG CTC CGG CGC CGC’
CGC 3’ 307
B
Fig. 1 (A) Sequence of the antisense oligonucleotideprobe used in the present study. The probe was complementary to bases 397441 of the bovine AVP-neurophysin II gene (corresponding to amino acids 133-147 of the precursor glycoprotein). *Three bases coding a single amino acid mismatch were modified to correspond to those most likely to exist in the ovine sequence. (B) Northern blot analysis ofpoly (A+)RNA extracted from the ovine hypothalami. The same oligonucleotide as described above was labelled using **Pand hybridized under the same conditions as those for in situ hybridization. The size of the AVP mRNA transcript ranged from 600-800 bases in length.
VASOPRESSIN mRNA IN
THE OVINE
BRAIN.
PITUITARY
AND PINEAL
tion the sections were washed for 1 h in 1xSSC at room temperature and for 1 h in 1xSSC at 55’C, the sections were rinsed in 1xSSC and O.lxSSC (2 s each), dehydrated in ethanol, dried, exposed to Xray film @AR 5, Kodak) and then dipped in Ilford K5 liquid emulsion. The X-ray films and emulsioncoated sections were developed using standard procedures. The sections were counterstained with methylene blue to permit identification of nuclei. The antisense AVP deoxyribonucleotide probe was 45 bases long and was made by solid phase synthesis using an Applied Biosystems DNA synthesiser and purified on 8% polyacrylamide/8 M urea preparative sequencing gel (Institute Microchemical facility). The probe was complementary to bases 397-441 of the bovine AVP-neurophysin II gene (corresponding to amino acids 133-147 of the precursor glycoprotein),” though three bases coding a single amino acid mismatch were modified to correspond with those most likely to occur in the ovine sequence (Fig. 1A). A control 45-mer sense probe was also synthesized complementary to the
GLANDS
13
anti-sense AVP oligonucleotide. To verify the specificity of the antisense AVP probe, Northern analysis of total and poly(A+)RNA extracted from the ovine hypothalami was performed under the same conditions outline for in situ hybridization. The size of the labelled AVP mRNA transcript ranged from 600-800 bases confirming previous observations.” (Fig. 1B). Results Examination of the X-ray film autoradiograms after 3 days exposure revealed a high density of AVP mRNA in the PVN, SON and SCN (Figs 2 & 3) with the SON (Fig. 2C) demonstrating a higher degree of labelling than the other two nuclei. In the PVN, AVP-labelling was most dense in the magnocellular region, though a significant amount was present in the parvocellular portion of the nucleus (Fig. 2A). Specificity of hybridization was established as no signal was observedwith the ‘sense’ oligonucleotide probe in adjacent hypothalamic sections (not illus-
Fig. 2 Localization of AVP mRNA in the paraventricular nucleus (PVN) and supraoptic nucleus (SON) of the ovine hypothalamus after in situ hybridization of coronal brain sections with an ?+Iabelled 45mer antisense oligonucleotide probe complementary to the bovine AVP gene. Negative prints of X-ray film showing the distribution of AVP mRNA in the PVN (A) and SON (C). Brightfield photographs of emulsion autoradiograms showing labelled cells in both the parvocellular and magnocellular PVN (B) and the SON (D). Exposure time: A & C, 4 d; B & D, 3 weeks. Magnification: A & C, x8; B & D, x100.
14
NEUROPEPTIDES
Fig. 3 Localization of AVP mRNA in hypothalamic and extrahypothalamic nuclei and areas of the ovine brain. Bright field photographs of emulsion autoradiograms showing labelled cells in the SCN (A), accessory magnocellular nuclei (B), the prefrontal cortex (C) and the choroid plexus (D). Exposure time: A-D, 3 weeks. Magnification: A, x100; B & C, x400; D, x250.
trated) using identical hybridization conditions. High resolution analysis of sections exposed to silver grain emulsion for 3 weeks revealed dense clusters of labelled cells in magnocellular cell groups of both the PVN and SON (Fig. 2B & D). In the PVN, labelling appeared to be concentrated in the caudal and lateral parts of the nucleus, corresponding to the posterior and dorsolateral magnocellular cell groups. In the SON, AVP-labelled cells were localized in the ventral and caudal regions. Labelled cell bodies were also observed in the medial parvocellular region of the PVN, which tended to be smaller than those observed in associated magnocellular regions. Clusters of AVP mRNA-labelled cells were present in the SCN at the base of the third ventricle (Fig. 3A), as well as being sparsely scattered throughout this region of the hypothalamus in accessory magnocellular nuclei (Fig. 3B). Labelling was also observed in small-bodied cells scattered throughout the cortical fields, (e.g. Fig. 3C) but no AVP mRNA-labelled cells were seen in the bed nucleus of the stria terminalis, the locus coeruleus, or any other region of the brainstem. However, several cells were densely labelled in and
around the choroid plexus tissue of both the lateral and third ventricles (Fig. 3D). Large AVP-labelled cells were also observed in the intermediate lobe of the pituitary gland (Fig. 4A & B) and in the highly glandular secretory tissue of the pineal gland (Fig. 4C & D).
Discussion The present study has provided the first demonstration of the regional and cellular localization of AVP mRNA in the ovine brain using the in situ hybridization technique. The most dense areas of AVP mRNA-expressing cells were the PVN, SON, and SCN and, within these regions, there was good correlation with data obtained from both immunocytochemical12 and in situ hybridization techniques1zJ8-20 in rats and immunocytochemical studies in sheep.15 High resolution analysis revealed that AVP mRNA was present in the nuclei and other neuronal processes of hypothalamic neurons, in agreement with previous studies in the rat.‘* Similarly, widespread distribution of AVP mRNA was also seen
VASOPRESSIN
mRNA
IN THE OVINE BRAIN,
PITUITARY
AND
PINEAL
GLANDS
15
Fig. 4 Localization of AVP mRNA in extrahypothalamic nuclei and areas of the ovine brain. Negative print of an X-ray film showing AVP mRNA expression in the intermediate lobe of the pituitary gland (A). Bright field photographs of emulsion autoradiograms showing labelled cells in the intermediate lobe of the pituitary (B) and the pineal gland (C, D). Exposure time: A, 4 d; ED, 3 weeks. Magnification: A, x 15; B & C, xl 00; D, x400.
within the SCN in the present study. Interestingly, both the SCN26and the SON*’ show a diurnal rhythm in AVP transcription which, in the SCN, may be due to a daily variation in poly(A)-tail length.2s However, on the basis of the present study, it is not possible to determine whether a similar rhythm exists in the sheep. Both AVP mRNA (present study) and oxytocin mRNA (SGM, personal observation) were also present in the accessory magnocellular nuclei. However, recent immunocytochemical evidence has indicated the presence of an additional nucleus in the pig hypothalamus that contains both AVP and oxytocin.28 The function ofthis vasopressin and oxytocin containing nucleus (VON) remains to be determined, though it is thought to possess some role in reproductive function. 2q Indeed, if this nucleus is present in other ungulates such as the sheep, it is possible that some of these AVP mRNA ‘accessory magnocellular neurons’ may in fact represent the VON.
There appear to be a number of species differences in the distribution of AVP mRNA in extrahypothalamic areas. For example, cells containing AVP immunoreactivity’3 and AVP mRNA have been described in the medial amygdala30 and in the bed nucleus of the stria terminalis3’ of the rat brain, while none were seen in the present study. In the sheep, as in the rat,30no cells expressing AVP mRNA were observed in the region of the locus coeruleus, although AVP immunoreactivity has been noted here in rats.13 Other species differences in the localization of AVP immunoreactivity have also been reported; for example, vasopressinergic neurons present in the magnocellular nuclei of the human basal forebrain are absent in rats. I4 A very small number of cells expressing AVP mRNA were seen in the cortical fields of the ovine brain. The presence of AVP mRNA in these areas may implicate AVP in the regulation of memory and learning processes, as suggested previously.5 Interestingly, AVP mRNA was also observed in
16
cells situated within the choroid plexus of the lateral and third ventricles. Although this has not been reported in other species, a number of studies have been undertaken to establish the role of AVP in the CSF, one of which has demonstrated that AVP may affect CSF production.1° Therefore, it is possible that AVP synthesized locally in choroid plexus tissue may influence CSF formation. The present study also demonstrated expression of AVP mRNA in the pineal gland. This extends previous observations describing the presence of AVPlike immunoreactivity in the rat pineaP* and supports the view that this AVP, togetherwith oxytocin, is not of hypothalamic origin but rather, is secreted within the pineal itself.33This may be connected with the role of the pineal in relation to photoperiod because AVP and oxytocin levels increase in the rat pineal gland during long days33 and both peptides affect the synthesis and/or secretion of melatonin.34 In this respect, the sheep, being a highly seasonal species, may be a particularly appropriate experimental model with which to investigate possible vasopressinergic modulation of pineal function. Finally, cells expressing AVP mRNA were observed in the intermediate lobe of the pituitary gland though none were found in the posterior lobe of the pituitary, as have been described in the rat.35,36 The AVP mRNA in the rat posterior pituitary is present at approximately 100th of the concentration seen with the hypothalamic nuclei36 and, as generally observed in peripheral organs, the mRNA is shorter due to an apparent decrease in poly(A)-tail length.37 Although the most likely source of this AVP mRNA in the pituicytes or resident endothelial cells,37 its function, like that of the AVP in the intermediate lobe of the ovine pituitary, is unclear. In conclusion, the in situ hybridization technique has allowed a detailed mapping of cells expressing AVP mR.NA, and therefore presumably synthesizing AVP within the ovine brain. These observations essentially agree with those obtained in the rat implicating AVP in neurotransmitter and hypophysiotrophic roles within the brain, although the precise function of the various AVP mRNA expressing cell groups remain to be elucidated.
NEUROPEPTIDES
References 1. Redekopp, C. A., Livesey, J. H., Toth, A. and Donald, R. A.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16.
Acknowledgements We thank Catharine Knights for her expert help in the preparation of the manuscript.
17.
(1985). Effect of ovine CRF and AVP on ACTH and aldosterone secretion in sheep. Hormone and Metabolic Research 17: 428429. Familari,M.,Smith,I.A.,Smith,R.andFunder,J.W.(1989). Arginine vasopressin is a much more potent stimulus to ACTH release from ovine anterior pituitary cells than ovine CRF. Neuroendocrinology 50: 152-157. Caraty, A., Grino, M., Locatelli, A., Guillaume, V., Boudouresque, F., Corte-Devoix, B. and Oliver, C. (1990). Insulin induced hypoglycaemia stimulates CRF and AVP secretion into hypophysial portal blood of conscious, unrestrained rams. Journal of Clinical Investigation 85: 17161721. Engler, D., Pham, T., Fullerton, K. J., Ooi, G., Funder, J. W. and Clarke, I. J. (1989). Studies of the secretion of CRF and vasopressin into the hypophysial-portal circulation of the conscious sheep. Neuroendocrinology 49: 367-381. Strupp, B. J. (1989). Improvement of memory by a vasopressin fragment. Importance of individual differences in mnemonic function. Behavioural Neuroscience 103: 743-754. Bereck,K. H., Webb. R. L. andBrody,M. J. (1983). Evidence for a central role for vasopressin in cardiovascular regulation. American Journal of Physiology 244: H852-H859. Parkes, D. G., Coghlan, J. P. and Scoggins, B. A. (1989). The effects of intracerebroventricular administration of biologically active peptides in conscious sheep. Peptides 9: 1221-1225. Riphagen, C. L. and Pittman, Q. J. (1986). Arginine vasopressin as a central neurotransmitter. Federation Proceedings 45: 23 18-2322. Stark, H., Burbach, P. H., van der Kleij, A. M. and de Weid, D. (1989). In vivo conversion of vasopressin after microinjection into limbic brain areas of rats. Peptides 10: 717-720. J. R. and Lightman, S. L. Se&l, (1987). Intracerebrovcntricular arginine vasopressin causes intracranial pressure to rise in conscious goats. Brain Research 423: 279-285. Seckl, J. R. and Lightman, s. L. (1991). Intracerebroventricular vasopressin reduces CSF absorption rate in conscious goats. Experimental Brain Research 84: 173-179. Guitteny, A. F., Bohlen, R. and Block, B. (1988). Analysis of vasopressin gene expression by in situ hybridization and immunohistochemistry in semi thin sections. Journal of Histochemistry and Cy-tochemistry 36: 1373-1378. Caffe, A. R. and Van Leeuwen, F. W. (1983). Vasopressinimmunoreactive cells in the dorsomedial hypothalamic region, medial amygdaloid nucleus and the locus coeruleus ofthe rat. Cell and Tissue Research 233: 23-33. Ulfig, N., Braak, E., Ohm, T. G. and Pool, C. W. (1990). Vasopressinergic neurons in the magnocellular nuclei of the human basal forebrain. Brain Research 530: 176-180. Swanson, L. W. and Sawchenko, P. E., Rivier, J. and Vale, W. W. (1983). Organization of ovine CRF immunoreactive cells and fibres in the rat brain: an immunoreactive study. Neuroendocrinology 36: 165-186. Ivell, R. and Richter, D. (1984). Structure and comparison of the oxytocin and vasopressin genes from rat. Proceedings of the National Academy of Science USA 8 1: 2006-2015. Land, H., Schiitz, G., Shmale, H. and Richter, D. (1982). Nucleotide sequence of cloned cDNA encoding bovine argi-
VASOPRESSIN
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
mRNA
IN THE OVINE
BRAIN,
PITUITARY
AND PINEAL
nine vasopressin-neurophysin II precursor. Nature 295: 299-303. Bloch, B., Guitteny, A. F., Chouham, S., Mougin, C., Roget, A. and Toule, R. (1990). Topography and ontogeny of the neurons expressing vasopressm, oxytocin and somatostatin genes in the rat brain: An analysis using radioactive and bioCellular Molecular tynylated oligonucleotides. Neurobiology 10: 99-106. Card, J. P., Fitzpatrick-McElligot, F., Gozes, I. and Baldino, F. (1988). Localization of vasopressin, vasoactive intestinal polypeptide, peptide histidine isoleucine, and somatostatin mRNA in rat suprachiasmatic nuclei. Cell and Tissue Research 252: 307-315. Nojiri, H., Sato, M. and Urano, A. (1985). In situ hybridization of vasopressin n-RNA in the rat hypothalamus by use of a synthetic oligonucleotide probe. Neuroscience Letters 58: 101-105. Matthews, S. G., Heavens, R. P. and Sirinathsinghji, D. J. S. (1991). The cellular localization of the corticotrophin releasing factor (CRF) and vasopressin (AVP) genes in the ovine brain. Third IBRO Congress of Neuroscience, Montreal. Abstract P34. 13. Matthews, S. G., Heavens, R. P. and Sirinathsinghji, D. J. S. (199 1). Cellular localization of corticotrophin releasing factor mRNA in the ovine brain. Molecular Brain Research 11: 171-176. Matthews, S. G., Heavens, R. P. and Sirinathsinghji, D. J. S. (1992). Distribution and cellular localization of preproenkephalin mRNA in the ovine brain and pituitary. Molecular Brain Research 12: 349-355. Sirinathsinghji, D. J. S., Morris, B. J., Wisden, W., Northrop, A., Hunt, S. P. and Dunnett, S. B. (1990). Gene expression in striatal grafts-I. Cellular localization of neurotransmitter mRNAs. Neuroscience 34: 675-686. Robinson, A. G., Frim, D. M., Schwartz, W. T. and Majzoub, J.A.(1988).VasopressinmRNAandsuprachiasmaticnuclei: Daily regulation of polypeptide tail length. Science 241: 342-345. Burbach, J. P. H., Liu, B., Voorhius, T. A. M. and Van Tol, H. H. M. (1988). Diumal variation in vasopressin and oxytocin mRNA’s in hypothalamic nuclei of the rat. Molecular Brain Research 4: 157-160. Carter, D. A. and Murphy, D. (1991). Nuclear mechanisms
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
GLANDS
17
mediate rhythmic changes in vasopressin mRNA expression in the rat suprachiasmatic nucleus. Molecular Brain Research 12: 315-321. Van Eerdenberg, 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. Journal of Comparative Neurology 310: 138-146. Van Eerdenberg, F. J. C. M. and Swaab, D. F. (1991). Increasing neuron numbers in the vasopressin and oxytocin containing nucleus of the adult female pig hypothalamus. Neuroscience Letters 132: 85-88. Urban, J. H., Miller, M. A., Drake, C. T. and Dorsa, D. M. (1990).DetectionofvasopressinmRNAincellsofthemedial amygdala but not the locus coeruleus by in situ hybridization. Journal of Chemical Neuroanatomy 3: 277-283. Miller, M. (1980). Role of endogenous opioids in neurohypophysial function ofman. Journal ofclinical Endocrinoiogy and Metabolism 50: 1016-1020. Moujir, F., Sanchez-France, F., Santana, C., Cacicedoh, and Alonso, R. (1990). Immunoreactive levels of pineal AVP change during the rat oestrous cycle. Journal of Pineal Research 8: 359-366. Liu, B., Burbach, J. P. H., Femstrom, J. D. and Antoni, F. A. (1991). The hypothalamus is not the origin of vasopressin and oxytocin in the rat pineal gland. Neuroendocrinology 53: 523-527. Simmonneaux, V., Ouichou, S., Burbach, J. P. H. and Pevet, P. (1990). Vasopressin and oxytocin modulation of melatonin secretion from rat pineal glands. Peptides 111: 1075-1079. Lehmamr, E., Hanze, J., Pauschinger, M., Ganten, D. and Lang, R. E. (1990). Vasopressin mRNA in the neural lobe of the rat pituitary. Neuroscience Letters 111: 17&175. McCabe, J. T., Lehmann, E., Chastrette, N., Hanze, J., Lang, R. E., Ganten, D. and Pfaff, D. W. (1990). Detection of AVP n-RNA in the neuro-intermediate lobe of the rat pituitaiy. Molecular Brain Research 8: 32.5-329. Murphy, D., Levy, A., Lightman, S. and Carter, D. (1989). Vasopressin RNA in the neural lobe of the pituitary gland: Dramatic accumulation in response to salt loading. Proceedings of the National Academy of Science USA 86: 9002-9005.
Distribution and Cellular Localization of Vasopressin mRNA in the Ovine Brain, Pituitary and Pineal Glands S. G. MATTHEWS*, R. F. PARROTT and D. J. S. SlRlNATHSlNGHJl AFRC institute of Animal Physiology and Genetics Research, Cambridge Research Station, Babraham, Hail, Cambridge UK and *Lawson Research institute, Grosvenor Street, London, Ontario, Canada (Reprint requests to RFP)
Abstract-In this study, in situ hybridization histochemistry was used to determine the regional and cellular localization of vasopressin-neurophysin II (AVP) mRNA in the sheep brain and pituitary with an 36S-labelled synthetic 45-mer oligonucleotide probe complementary to the bovine AVP gene. The highest densities of labelled cell bodies were found in the paraventricular nucleus (PVN), supraoptic nucleus (SON) and suprachiasmatic nucleus (SCN) of the hypothalamus, though such cells were also found in other regions of the diencephalon, including the accessory magnocellular nuclei. Labelled cells were also observed sparsely distributed in every major cortical field as well as in choroid plexus and the pineal gland. No AVP mRNA-expressing cells were found in the bed nucleus of the stria terminalis, the amygdala, or in the medulla and brainstem. In the pituitary, a dense AVP mRNA signal was observed in the intermediate lobe whereas, cells in the anterior or neural lobe did not express AVP mRNA. The dense population of AVP-expressing neurons in both magnocellular and parvocellular fields of the hypothalamus support major roles of AVP in both posterior and anterior pituitary function. Finally, the extrahypothalamic distribution of AVP mRNA transcripts suggest that vasopressinergic neurons may be involved in diverse physiological functions, including the regulation of pineal function and cognition.
Introduction
blood increase in response to stress.3,4 Moreover, AVP serves as a neuropeptide in the CNS, where it has been shown to influence learning and memory, cardiovascular function6*7 and other autonomic responses.**9 It has also been shown to increase intracranial pressure, loprobably by an effect on cerebrospinal fluid (CSF) absorption rate.” Within the hypothalamus, immunohistochemical studies have revealed dense AVP labelling in the supraoptic (SON), paraventricular (PVN) and suprachiasmatic nuclei (SCN).12 However, in com-
The hypophysiotrophic action of AVP in conscious sheep, in addition to its neurohypophysial role, is well established. It is known that exogenous AVP stimulates the ovine anterior pituitary to secrete adrenocorticotrophic hormone (ACTH) in viva’ and in vitro2 and that AVP concentrations in the portal
Date received 5 October 1992 Date accepted 29 December 1992
11
12
NFXJROPEPTDES
mon with other regulatory peptides, similar studies13J4 have indicated widespread distribution of AVP-containing neurons and fibres outside the hypothalamus, suggesting that the peptide may function as a neurotransmitter within the central nervous system (CNS), though levels of AVP immunoreactivity are of a lesser magnitude than those for other hypophysiotrophic peptides described in the ovine brain.15 Recently, the cloning and nucleotide sequencing of the ratI and bovineI AVP-neurophysin II precursor have facilitated the study of mechanisms regulating AVP synthesis at cellular and molecular levels. In this context, the use of in situ hybridization histochemistry has revealed the existence of AVP mRNA within the hypothalamus1zJ8-to as well as in extrahypothalamic regions,20p37 Currently, we are studying the regulation of neuropeptide gene expression in the CNS of the sheep, especially those mechanisms associated with responses to osmotic stimulation and stress. In earlier studies, we have determined the localization of corticotrophin-releasing hormone (CRH) mRNA21v22and preproenkephalin (PENK) mRNAcontaining cells23 in the ovine brain using in situ hybridization histochemistry. In the present investigation, we have used the same procedure with a synthetic oligonucleotide probe complementary to a modified sequence of the bovine AVP-neurophysin II gene to determine the regional and cellular distribution of AVP mRNA-containing cells in the ovine brain. A preliminary account of this research has been presented elsewhere.21 Materials
and methods
Three Clun Forest wethers were killed with Lethobarb (Pentobarbitone sodium BP 20% w/v) (Duphar Veterinary Limited, Southampton, UK), decapitated, and the brains removed within 5 min. Each brain was placed on ice and cut in the frontal plane into the following three blocks: (1) olfactory bulb to the optic chiasma, (2) optic chiasma to the midbrain at the level of the medial mammillary nucleus and (3) midbrain to the brainstem. Each block was frozen on dry-ice, wrapped in parafilm and stored at -7O’C. Sections (10 pm) were then cut on a cryostat (Bright Instruments), mounted onto @oly)L-lysine-coated slides, dried, post-fixed in 4%
paraformaldehyde (5 min), rinsed in phosphatebuffered saline (2x1 min), dehydrated in an alcohol series and stored in 95% alcohol at +4”C until in situ hybridization analysis. The basic method for in situ hybridization has been previously described in detail.24 Briefly, the slides were removed from the alcohol, allowed to air-dry at room temperature and then incubated overnight in a humid atmosphere at 42°C with the radiolabelled oligonucleotide probe in hybridization buffer. The hybridization buffer used for these experiments contained 4xstandard sodium citrate (SSC) (1xSSC contains 150 mM sodium chloride and 15 mM sodium citrate), 50% deionized formamide, 50 mM sodium phosphate (pH 7.0), 1mM sodium pyrophosphate (pH 7.0), 0.02% bovine serum albumin, 200 ,ug/ml hydrolysed salmon sperm DNA, 0.02% Ficoll, 0.02% polyvinylpyrrolidone, 10% dextran sulphate and 40 mM dithiothreitol (DTT). The AVP oligodeoxyribonucleotide probe was labelled using terminal deoxynucleotidyl transferase (Pharmacia) and [35S]deoxyadenosine 5’-(athio)triphosphate (1300 Ci/mmol, New England Nuclear) to a specific activity of 1.0~10~ cpm/ng. The labelled probe was used at a concentration of 3x10’ cpm/pl. 100 pl of labelledprobe in hybridization buffer was applied to each slide. After incubaA 0
441
GTA GAC GCC GGG CTG GGC GGG CTC CGC GGG CTC CGG CGC CGC’
CGC 3’ 307
B
Fig. 1 (A) Sequence of the antisense oligonucleotideprobe used in the present study. The probe was complementary to bases 397441 of the bovine AVP-neurophysin II gene (corresponding to amino acids 133-147 of the precursor glycoprotein). *Three bases coding a single amino acid mismatch were modified to correspond to those most likely to exist in the ovine sequence. (B) Northern blot analysis ofpoly (A+)RNA extracted from the ovine hypothalami. The same oligonucleotide as described above was labelled using **Pand hybridized under the same conditions as those for in situ hybridization. The size of the AVP mRNA transcript ranged from 600-800 bases in length.
VASOPRESSIN mRNA IN
THE OVINE
BRAIN.
PITUITARY
AND PINEAL
tion the sections were washed for 1 h in 1xSSC at room temperature and for 1 h in 1xSSC at 55’C, the sections were rinsed in 1xSSC and O.lxSSC (2 s each), dehydrated in ethanol, dried, exposed to Xray film @AR 5, Kodak) and then dipped in Ilford K5 liquid emulsion. The X-ray films and emulsioncoated sections were developed using standard procedures. The sections were counterstained with methylene blue to permit identification of nuclei. The antisense AVP deoxyribonucleotide probe was 45 bases long and was made by solid phase synthesis using an Applied Biosystems DNA synthesiser and purified on 8% polyacrylamide/8 M urea preparative sequencing gel (Institute Microchemical facility). The probe was complementary to bases 397-441 of the bovine AVP-neurophysin II gene (corresponding to amino acids 133-147 of the precursor glycoprotein),” though three bases coding a single amino acid mismatch were modified to correspond with those most likely to occur in the ovine sequence (Fig. 1A). A control 45-mer sense probe was also synthesized complementary to the
GLANDS
13
anti-sense AVP oligonucleotide. To verify the specificity of the antisense AVP probe, Northern analysis of total and poly(A+)RNA extracted from the ovine hypothalami was performed under the same conditions outline for in situ hybridization. The size of the labelled AVP mRNA transcript ranged from 600-800 bases confirming previous observations.” (Fig. 1B). Results Examination of the X-ray film autoradiograms after 3 days exposure revealed a high density of AVP mRNA in the PVN, SON and SCN (Figs 2 & 3) with the SON (Fig. 2C) demonstrating a higher degree of labelling than the other two nuclei. In the PVN, AVP-labelling was most dense in the magnocellular region, though a significant amount was present in the parvocellular portion of the nucleus (Fig. 2A). Specificity of hybridization was established as no signal was observedwith the ‘sense’ oligonucleotide probe in adjacent hypothalamic sections (not illus-
Fig. 2 Localization of AVP mRNA in the paraventricular nucleus (PVN) and supraoptic nucleus (SON) of the ovine hypothalamus after in situ hybridization of coronal brain sections with an ?+Iabelled 45mer antisense oligonucleotide probe complementary to the bovine AVP gene. Negative prints of X-ray film showing the distribution of AVP mRNA in the PVN (A) and SON (C). Brightfield photographs of emulsion autoradiograms showing labelled cells in both the parvocellular and magnocellular PVN (B) and the SON (D). Exposure time: A & C, 4 d; B & D, 3 weeks. Magnification: A & C, x8; B & D, x100.
14
NEUROPEPTIDES
Fig. 3 Localization of AVP mRNA in hypothalamic and extrahypothalamic nuclei and areas of the ovine brain. Bright field photographs of emulsion autoradiograms showing labelled cells in the SCN (A), accessory magnocellular nuclei (B), the prefrontal cortex (C) and the choroid plexus (D). Exposure time: A-D, 3 weeks. Magnification: A, x100; B & C, x400; D, x250.
trated) using identical hybridization conditions. High resolution analysis of sections exposed to silver grain emulsion for 3 weeks revealed dense clusters of labelled cells in magnocellular cell groups of both the PVN and SON (Fig. 2B & D). In the PVN, labelling appeared to be concentrated in the caudal and lateral parts of the nucleus, corresponding to the posterior and dorsolateral magnocellular cell groups. In the SON, AVP-labelled cells were localized in the ventral and caudal regions. Labelled cell bodies were also observed in the medial parvocellular region of the PVN, which tended to be smaller than those observed in associated magnocellular regions. Clusters of AVP mRNA-labelled cells were present in the SCN at the base of the third ventricle (Fig. 3A), as well as being sparsely scattered throughout this region of the hypothalamus in accessory magnocellular nuclei (Fig. 3B). Labelling was also observed in small-bodied cells scattered throughout the cortical fields, (e.g. Fig. 3C) but no AVP mRNA-labelled cells were seen in the bed nucleus of the stria terminalis, the locus coeruleus, or any other region of the brainstem. However, several cells were densely labelled in and
around the choroid plexus tissue of both the lateral and third ventricles (Fig. 3D). Large AVP-labelled cells were also observed in the intermediate lobe of the pituitary gland (Fig. 4A & B) and in the highly glandular secretory tissue of the pineal gland (Fig. 4C & D).
Discussion The present study has provided the first demonstration of the regional and cellular localization of AVP mRNA in the ovine brain using the in situ hybridization technique. The most dense areas of AVP mRNA-expressing cells were the PVN, SON, and SCN and, within these regions, there was good correlation with data obtained from both immunocytochemical12 and in situ hybridization techniques1zJ8-20 in rats and immunocytochemical studies in sheep.15 High resolution analysis revealed that AVP mRNA was present in the nuclei and other neuronal processes of hypothalamic neurons, in agreement with previous studies in the rat.‘* Similarly, widespread distribution of AVP mRNA was also seen
VASOPRESSIN
mRNA
IN THE OVINE BRAIN,
PITUITARY
AND
PINEAL
GLANDS
15
Fig. 4 Localization of AVP mRNA in extrahypothalamic nuclei and areas of the ovine brain. Negative print of an X-ray film showing AVP mRNA expression in the intermediate lobe of the pituitary gland (A). Bright field photographs of emulsion autoradiograms showing labelled cells in the intermediate lobe of the pituitary (B) and the pineal gland (C, D). Exposure time: A, 4 d; ED, 3 weeks. Magnification: A, x 15; B & C, xl 00; D, x400.
within the SCN in the present study. Interestingly, both the SCN26and the SON*’ show a diurnal rhythm in AVP transcription which, in the SCN, may be due to a daily variation in poly(A)-tail length.2s However, on the basis of the present study, it is not possible to determine whether a similar rhythm exists in the sheep. Both AVP mRNA (present study) and oxytocin mRNA (SGM, personal observation) were also present in the accessory magnocellular nuclei. However, recent immunocytochemical evidence has indicated the presence of an additional nucleus in the pig hypothalamus that contains both AVP and oxytocin.28 The function ofthis vasopressin and oxytocin containing nucleus (VON) remains to be determined, though it is thought to possess some role in reproductive function. 2q Indeed, if this nucleus is present in other ungulates such as the sheep, it is possible that some of these AVP mRNA ‘accessory magnocellular neurons’ may in fact represent the VON.
There appear to be a number of species differences in the distribution of AVP mRNA in extrahypothalamic areas. For example, cells containing AVP immunoreactivity’3 and AVP mRNA have been described in the medial amygdala30 and in the bed nucleus of the stria terminalis3’ of the rat brain, while none were seen in the present study. In the sheep, as in the rat,30no cells expressing AVP mRNA were observed in the region of the locus coeruleus, although AVP immunoreactivity has been noted here in rats.13 Other species differences in the localization of AVP immunoreactivity have also been reported; for example, vasopressinergic neurons present in the magnocellular nuclei of the human basal forebrain are absent in rats. I4 A very small number of cells expressing AVP mRNA were seen in the cortical fields of the ovine brain. The presence of AVP mRNA in these areas may implicate AVP in the regulation of memory and learning processes, as suggested previously.5 Interestingly, AVP mRNA was also observed in
16
cells situated within the choroid plexus of the lateral and third ventricles. Although this has not been reported in other species, a number of studies have been undertaken to establish the role of AVP in the CSF, one of which has demonstrated that AVP may affect CSF production.1° Therefore, it is possible that AVP synthesized locally in choroid plexus tissue may influence CSF formation. The present study also demonstrated expression of AVP mRNA in the pineal gland. This extends previous observations describing the presence of AVPlike immunoreactivity in the rat pineaP* and supports the view that this AVP, togetherwith oxytocin, is not of hypothalamic origin but rather, is secreted within the pineal itself.33This may be connected with the role of the pineal in relation to photoperiod because AVP and oxytocin levels increase in the rat pineal gland during long days33 and both peptides affect the synthesis and/or secretion of melatonin.34 In this respect, the sheep, being a highly seasonal species, may be a particularly appropriate experimental model with which to investigate possible vasopressinergic modulation of pineal function. Finally, cells expressing AVP mRNA were observed in the intermediate lobe of the pituitary gland though none were found in the posterior lobe of the pituitary, as have been described in the rat.35,36 The AVP mRNA in the rat posterior pituitary is present at approximately 100th of the concentration seen with the hypothalamic nuclei36 and, as generally observed in peripheral organs, the mRNA is shorter due to an apparent decrease in poly(A)-tail length.37 Although the most likely source of this AVP mRNA in the pituicytes or resident endothelial cells,37 its function, like that of the AVP in the intermediate lobe of the ovine pituitary, is unclear. In conclusion, the in situ hybridization technique has allowed a detailed mapping of cells expressing AVP mR.NA, and therefore presumably synthesizing AVP within the ovine brain. These observations essentially agree with those obtained in the rat implicating AVP in neurotransmitter and hypophysiotrophic roles within the brain, although the precise function of the various AVP mRNA expressing cell groups remain to be elucidated.
NEUROPEPTIDES
References 1. Redekopp, C. A., Livesey, J. H., Toth, A. and Donald, R. A.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15.
16.
Acknowledgements We thank Catharine Knights for her expert help in the preparation of the manuscript.
17.
(1985). Effect of ovine CRF and AVP on ACTH and aldosterone secretion in sheep. Hormone and Metabolic Research 17: 428429. Familari,M.,Smith,I.A.,Smith,R.andFunder,J.W.(1989). Arginine vasopressin is a much more potent stimulus to ACTH release from ovine anterior pituitary cells than ovine CRF. Neuroendocrinology 50: 152-157. Caraty, A., Grino, M., Locatelli, A., Guillaume, V., Boudouresque, F., Corte-Devoix, B. and Oliver, C. (1990). Insulin induced hypoglycaemia stimulates CRF and AVP secretion into hypophysial portal blood of conscious, unrestrained rams. Journal of Clinical Investigation 85: 17161721. Engler, D., Pham, T., Fullerton, K. J., Ooi, G., Funder, J. W. and Clarke, I. J. (1989). Studies of the secretion of CRF and vasopressin into the hypophysial-portal circulation of the conscious sheep. Neuroendocrinology 49: 367-381. Strupp, B. J. (1989). Improvement of memory by a vasopressin fragment. Importance of individual differences in mnemonic function. Behavioural Neuroscience 103: 743-754. Bereck,K. H., Webb. R. L. andBrody,M. J. (1983). Evidence for a central role for vasopressin in cardiovascular regulation. American Journal of Physiology 244: H852-H859. Parkes, D. G., Coghlan, J. P. and Scoggins, B. A. (1989). The effects of intracerebroventricular administration of biologically active peptides in conscious sheep. Peptides 9: 1221-1225. Riphagen, C. L. and Pittman, Q. J. (1986). Arginine vasopressin as a central neurotransmitter. Federation Proceedings 45: 23 18-2322. Stark, H., Burbach, P. H., van der Kleij, A. M. and de Weid, D. (1989). In vivo conversion of vasopressin after microinjection into limbic brain areas of rats. Peptides 10: 717-720. J. R. and Lightman, S. L. Se&l, (1987). Intracerebrovcntricular arginine vasopressin causes intracranial pressure to rise in conscious goats. Brain Research 423: 279-285. Seckl, J. R. and Lightman, s. L. (1991). Intracerebroventricular vasopressin reduces CSF absorption rate in conscious goats. Experimental Brain Research 84: 173-179. Guitteny, A. F., Bohlen, R. and Block, B. (1988). Analysis of vasopressin gene expression by in situ hybridization and immunohistochemistry in semi thin sections. Journal of Histochemistry and Cy-tochemistry 36: 1373-1378. Caffe, A. R. and Van Leeuwen, F. W. (1983). Vasopressinimmunoreactive cells in the dorsomedial hypothalamic region, medial amygdaloid nucleus and the locus coeruleus ofthe rat. Cell and Tissue Research 233: 23-33. Ulfig, N., Braak, E., Ohm, T. G. and Pool, C. W. (1990). Vasopressinergic neurons in the magnocellular nuclei of the human basal forebrain. Brain Research 530: 176-180. Swanson, L. W. and Sawchenko, P. E., Rivier, J. and Vale, W. W. (1983). Organization of ovine CRF immunoreactive cells and fibres in the rat brain: an immunoreactive study. Neuroendocrinology 36: 165-186. Ivell, R. and Richter, D. (1984). Structure and comparison of the oxytocin and vasopressin genes from rat. Proceedings of the National Academy of Science USA 8 1: 2006-2015. Land, H., Schiitz, G., Shmale, H. and Richter, D. (1982). Nucleotide sequence of cloned cDNA encoding bovine argi-
VASOPRESSIN
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
mRNA
IN THE OVINE
BRAIN,
PITUITARY
AND PINEAL
nine vasopressin-neurophysin II precursor. Nature 295: 299-303. Bloch, B., Guitteny, A. F., Chouham, S., Mougin, C., Roget, A. and Toule, R. (1990). Topography and ontogeny of the neurons expressing vasopressm, oxytocin and somatostatin genes in the rat brain: An analysis using radioactive and bioCellular Molecular tynylated oligonucleotides. Neurobiology 10: 99-106. Card, J. P., Fitzpatrick-McElligot, F., Gozes, I. and Baldino, F. (1988). Localization of vasopressin, vasoactive intestinal polypeptide, peptide histidine isoleucine, and somatostatin mRNA in rat suprachiasmatic nuclei. Cell and Tissue Research 252: 307-315. Nojiri, H., Sato, M. and Urano, A. (1985). In situ hybridization of vasopressin n-RNA in the rat hypothalamus by use of a synthetic oligonucleotide probe. Neuroscience Letters 58: 101-105. Matthews, S. G., Heavens, R. P. and Sirinathsinghji, D. J. S. (1991). The cellular localization of the corticotrophin releasing factor (CRF) and vasopressin (AVP) genes in the ovine brain. Third IBRO Congress of Neuroscience, Montreal. Abstract P34. 13. Matthews, S. G., Heavens, R. P. and Sirinathsinghji, D. J. S. (199 1). Cellular localization of corticotrophin releasing factor mRNA in the ovine brain. Molecular Brain Research 11: 171-176. Matthews, S. G., Heavens, R. P. and Sirinathsinghji, D. J. S. (1992). Distribution and cellular localization of preproenkephalin mRNA in the ovine brain and pituitary. Molecular Brain Research 12: 349-355. Sirinathsinghji, D. J. S., Morris, B. J., Wisden, W., Northrop, A., Hunt, S. P. and Dunnett, S. B. (1990). Gene expression in striatal grafts-I. Cellular localization of neurotransmitter mRNAs. Neuroscience 34: 675-686. Robinson, A. G., Frim, D. M., Schwartz, W. T. and Majzoub, J.A.(1988).VasopressinmRNAandsuprachiasmaticnuclei: Daily regulation of polypeptide tail length. Science 241: 342-345. Burbach, J. P. H., Liu, B., Voorhius, T. A. M. and Van Tol, H. H. M. (1988). Diumal variation in vasopressin and oxytocin mRNA’s in hypothalamic nuclei of the rat. Molecular Brain Research 4: 157-160. Carter, D. A. and Murphy, D. (1991). Nuclear mechanisms
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
GLANDS
17
mediate rhythmic changes in vasopressin mRNA expression in the rat suprachiasmatic nucleus. Molecular Brain Research 12: 315-321. Van Eerdenberg, 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. Journal of Comparative Neurology 310: 138-146. Van Eerdenberg, F. J. C. M. and Swaab, D. F. (1991). Increasing neuron numbers in the vasopressin and oxytocin containing nucleus of the adult female pig hypothalamus. Neuroscience Letters 132: 85-88. Urban, J. H., Miller, M. A., Drake, C. T. and Dorsa, D. M. (1990).DetectionofvasopressinmRNAincellsofthemedial amygdala but not the locus coeruleus by in situ hybridization. Journal of Chemical Neuroanatomy 3: 277-283. Miller, M. (1980). Role of endogenous opioids in neurohypophysial function ofman. Journal ofclinical Endocrinoiogy and Metabolism 50: 1016-1020. Moujir, F., Sanchez-France, F., Santana, C., Cacicedoh, and Alonso, R. (1990). Immunoreactive levels of pineal AVP change during the rat oestrous cycle. Journal of Pineal Research 8: 359-366. Liu, B., Burbach, J. P. H., Femstrom, J. D. and Antoni, F. A. (1991). The hypothalamus is not the origin of vasopressin and oxytocin in the rat pineal gland. Neuroendocrinology 53: 523-527. Simmonneaux, V., Ouichou, S., Burbach, J. P. H. and Pevet, P. (1990). Vasopressin and oxytocin modulation of melatonin secretion from rat pineal glands. Peptides 111: 1075-1079. Lehmamr, E., Hanze, J., Pauschinger, M., Ganten, D. and Lang, R. E. (1990). Vasopressin mRNA in the neural lobe of the rat pituitary. Neuroscience Letters 111: 17&175. McCabe, J. T., Lehmann, E., Chastrette, N., Hanze, J., Lang, R. E., Ganten, D. and Pfaff, D. W. (1990). Detection of AVP n-RNA in the neuro-intermediate lobe of the rat pituitaiy. Molecular Brain Research 8: 32.5-329. Murphy, D., Levy, A., Lightman, S. and Carter, D. (1989). Vasopressin RNA in the neural lobe of the pituitary gland: Dramatic accumulation in response to salt loading. Proceedings of the National Academy of Science USA 86: 9002-9005.