- Email: [email protected]
Osmotic regulation of substance P and neurokinin A peptide content and substance P binding sites in distinct hypothalamic nuclei of the rat
Peptides,Vol. 13, pp. 705-712, 1992
0196-9781/92 $5.00 + .00 Copyright© 1992PergamonPressLtd.
Printedin the USA.
Osmotic Regulation of Substance P and Neurokinin A Peptide Content and Substance P Binding Sites in Distinct Hypothalamic Nuclei of the Rat P H I L I P J. L A R S E N , * t L D A V I D S. JESSOP,* H A R D I A L S. C H O W D R E Y , * JENS D. MIKKELSEN~" A N D S T A F F O R D L. L I G H T M A N *
*Neuroendocrinology Unit, Charing Cross and Westminster Medical School, London, UK and -~Institute of Medical Anatomy, Department B, University of Copenhagen, Denmark Received 16 D e c e m b e r 1991 LARSEN, P. J., D. S. JESSOP, H. S. CHOWDREY, J. D. MIKKELSEN AND S. L. LIGHTMAN. Osmotic regulation of substance P and neurokininA peptidecontent and substanceP binding sites in distincthypothalamic nucleiof the rat. PEPTIDES 13(4) 705-712, 1992.--Quantitative receptor autoradiography using Bolton-Hunter iodinated substance P (SP) was used to localize specific sites in the rat hypothalamus. The amount of SP and neurokinin A (NkA) in extracts from discrete areas of the hypothalamus was measured using specific radioimmunoassays. A high density of SP binding sites was observed in the perimeter of the magnocellular paraventricular and supraoptic nuclei, while the magnocellular nuclei themselves possessed a low receptor density. In control animals, the number of SP binding sites was also low in the arcuate nucleus and the median eminence. Substance P and NkA peptide concentrations were highest in the paraventricular nucleus (PVN), decreasing in the following order: arcuate nucleus (Arc) > median eminence (ME) > supraoptic nucleus (SON) > subfornical organ (SFO). In animals given 340 mmol/l NaCI instead of tap water to drink for 12 days, significant increases in the number of SP binding sites occurred in the medial parvocellular subdivision of the PVN, periamygdaloid cortex, medial preoptic nucleus, Arc, and ME, but other hypothalamic areas were unaffected. In saline-treated animals, significantincreases in SP and NkA peptide concentrations were observed in the ME, while in the SFO only the concentration of NkA increased significantly. In the SON, substance P and neurokinin A levels were doubled, whereas in the PVN and Arc no changes in peptide levels were observed. Chronic osmotic stimulation is associated with lowered circulating levels of adrenocorticotropin releasinghormone (ACTH), and the present data further substantiate the hypothesis that hypothalamic tachykinin-containing neuronal terminals are centrally involved in the inhibition of anterior pituitary ACTH release observed during chronic osmotic stimulation. Substance P Neurokinin A Hypothalamus Rat
Salt loading
Microdissection
THE mammalian tachykinins substance P (SP) and neurokinin A (NkA) are widely distributed throughout the central nervous system (CNS), with high concentrations in the hypothalamus (4,18). Substance P is one of the most extensively studied and best characterized neuropeptides in the CNS (16), whereas little is known about the role of NkA in CNS functions. Substance P and NkA are encoded by the same gene, preprotachykinin A (PPT-A), which by alternative posttranscriptional splicing results in different amounts of their distinct mRNAs (22,30). This is also reflected in the relative amounts ofpeptide present in various CNS regions (18,21). Substance P and NkA show many similarities and are considered to share functional roles as well.
Receptorautoradiography
Radioimmunoassay
A large body of evidence has accumulated to suggest that SP is involved in diverse neuroendocrine functions including regulation of the release of anterior pituitary hormones (2,20) and water intake/drinking behavior (10,28). Substance P and NkA act as neurotransmitters in the regulation of the magnocellular hypothalamo-pituitary system. Intracerebroventricular (ICV) injection of SP causes antidiuresis in anesthetized rats (5), increases the firing rate of both vasopressinergic (aVP) and oxytocinergic (OXY) neurons of the supraoptic nucleus (SON) (7), and induces a release of vasopressin into the systemic circulation (5). Curiously, ICV injection of SP is also associated with decreased plasma concentration of ACTH (5), the latter most likely
Requests for reprints should be addressed to Philip J. Larsen, M.D., Institute of Medical Anatomy, Department B, The Panum Institute, Blegdamsvej 3, DK-2200 Copenhagen N, Denmark.
705
706
LARSEN ET AL.
being a result of inhibition of the release ofcorticotropin releasing factor (CRF-41) from the median eminence (ME) to portal vessels (8). In a model of chronic osmotic stimulation where rats are offered 2% saline to drink for a period of 12 days, the content of CRF-41 mRNA within the medial parvocellular subnucleus of the hypothalamic paraventricular nucleus (PVN) is decreased (25,38). Recently, we have demonstrated that this osmotic stimulus, which increases circulating aVP levels, also lowers basal circulating ACTH and corticosterone levels (17). In addition, the normally occurring stress-induced ACTH and corticosterone responses are absent in these osmotically stimulated animals (6). Taken together, these findings raise the possibility that endogenous SP or other tachykinins mediate the down-regulation o f C R F gene expression and circulating ACTH observed during chronic osmotic stimulation. The site of action of SP/NkA on aVP- or CRF-41-containing neurons may still be uncertain. However, within the hypothalamic paraventricular nucleus (PVN), SP-immunoreactive terminals have been demonstrated in apposition to aVP-containing perikarya (14) and injection of various tachykinins into the PVN induces an increase in plasma vasopressin concentration (29), suggesting a direct influence on magnocellular neurons. The CRF-41-containing neurons in the medial parvocellular division of the PVN colocalize aVP and project to the external zone of the ME, where they release CRF41 and aVP into hypophysial portal blood [for review see ( 1)]. Both peptides act as potent stimulatory neuroendocrine transmitters on ACTH release from adenohypophysial corticotrophs (11,36), and it seems possible that SP exerts its central inhibitory effects on circulating ACTH levels via these CRF/aVP neurons. The above-mentioned findings raise the possibility that endogenous SP mediates the paradoxical changes in ACTH and aVP response to chronic osmotic stimulation. To elucidate the anatomical location of these effects and to add further evidence for the role of SP and NkA, we used microdissection and specific radioimmunoassays to determine whether alterations in SP and NkA tissue content occurred after chronic saline ingestion in various areas of the brain involved in ACTH release and water homeostasis. In addition, using quantitative receptor autoradiography, we investigated any alterations in the density of SP binding sites in these and other hypothalamic nuclei in 2% salinetreated animals. METHOD
SP antiserum with NkA was <0.05% and of NkA antiserum with SP was 0.15%.
A utoradiographic Procedure Animals were anesthetized with tribromethanol (250 mg/kg/ body weight) and transcardially perfused with heparinized isotonic saline followed by 100 ml 0.32 M sucrose. Brains were rapidly removed and stored at - 8 0 ° C until serial 20-um cryostat sections were cut. The sections were thaw mounted on gelatine/ chrome alum-subbed slides and kept at - 8 0 ° C until use. For binding experiments, sections were transferred from the freezer to a hot plate (40°C) for 1 min to remove condensed water. The sections were then preincubated at room temperature for 20 min in a 50 mM Tris buffer (pH 7.4) containing MnCI2 (3 mM), Na2SO4 (10 mM), bacitracin (40 mg/l), leupeptin (4 mg/1), chymostatin (2 mg/1), and BSA (0.02%), hereafter called buffer A. Sections were incubated at room temperature for 60 min in buffer A containing 1 nM ~25I-Bolton-Hunter iodinated SP (125I-BH-SP, Amersham, Buckinghamshire, UK). After a brief wash in buffer A, sections went through two 2-rain rinses in buffer A, and were dipped in distilled H20 before being dried rapidly under a stream of cool dry air. Sections were then exposed to Amersham Hyperfilm® for 7 clays and fnally developed in Kodak D19 developer. Autoradiograms were quantified using an image analysis system (Image 1.16 g, Wayne Rasband, N.I.H., USA) and grain densities converted to fmol bound ligand/mg wet weight using Amersham J25I-micro-scales®. Nonspecific binding was measured by including 1 ttM unlabeled SP in incubations with ~251-BH-SP.
Statistical Analysis When immunoreactive tachykinin levels from control animals were compared to those of 2% saline-treated animals, data were evaluated by two-tailed Student's t-test. The same statistical paradigm was employed on densitometrical data Obtained from receptor autoradiography. RESULTS
Osmotic challenge by replacing drinking water with 340 mmol/1 NaCI for 12 days significantly increased plasma osmolarity from 285 -+ 4 to 323 -+ 10 mOsm/kg H20 (p < 0.001, mean --- SEM, two-tailed Student's t-test). The concentration of plasma aVP increased from 1.68 ___0.19 to 5.9 + 1.29 pmol/l (p < 0.0 I, mean -+ SEM, two-tailed Student's t-test).
Animal Treatment Male Sprague-Dawley rats weighing 200-250 g were housed under standard laboratory conditions with free access to food and either tap water (control) or a 2% (w/v) NaC1 solution for 12 days. All animals were housed in cages with a grid bottom. The effects of this treatment on plasma osmolarity and vasopressin levels have been described in detail previously (19). Following treatment, the rats were decapitated, their brains rapidly removed and frozen on dry ice for storage at -80°C. Thereafter, 150-#m thick cryostat sections were cut through the hypothalamus and stored at -800C. Using a microdissection needle (0.50 mm), selected hypothalamic areas were punched out according to a method previously described (31). Tissue punches were frozen on dry ice, while the sections were fixed in 4% paraformaldehyde and counterstained to allow validation of the sampling technique. Tissue samples were extracted and measured for SP and NkA as previously described (18). Cross-reactivities of SP and NkA antisera with the other mammalian tachykinin, neurokinin B, were <0.05 and 7%, respectively; cross-reactivity of
Tissue Peptide Content The SP and NkA contents of rat brain tissue extracts are shown in Table 1. All investigated areas contained detectable levels of SP and NkA. Histological examination of fixed brain slices confirmed that the areas under investigation had been punched out (Fig. 1). Following treatment for 12 days with 340 mmol/l NaCI, significant increases in SP and NkA were observed in the median eminence (ME) while NkA, but not SP, increased significantly in the subfornical organ (SFO). In all other areas investigated, no significant changes in the content of SP or NkA were observed. Although the concentrations of SP and NkA in the supraoptic nucleus (SON) were doubled, these values did not reach significance in this series of measurements. In a separate experiment, however, the observed increases in levels of SP and NkA in the SON reached significance (SP increased from 1.89 -- 0.19 to 3.86 --- 0.63 ng/mg, p < 0.05; and NkA increased from 2.58 -- 0.49 to 4.57 _+ 0.66 rig/rag, p < 0.05). A general correspondence between levels of SP and NkA concentrations was
OSMOTIC REGULATION OF HYPOTHALAMIC SUBSTANCE P
707
observed in individual nuclei of control and osmotically stimulated animals. However, the ratio between SP and NkA concentrations varied greatly between different areas. Thus, concentration of NkA in the SFO and ME was about sixfold higher than that of SP, whereas the SP:NkA ratio in the SON and PVN was 1:2.
Substance P Binding Sites Bolton-Hunter iodinated substance P (IZSI-BH-SP) was employed as a ligand to autoradiographically visualize SP binding sites in rat brain slices. For convenience, the 125I-BH-SPbinding sites are hereafter referred to as SP binding sites. Binding of ~25IBH-SP to tissue sections was completely abolished by addition of 1 ~tM unlabeled SP in the incubation medium. In control animals, specific binding sites for SP were demonstrated throughout the mediobasal diencephalon, including the preoptic •area and the mediobasal hypothalamus (Fig. 2, Table 2). In particular, very high densities were observed in the medial preoptic nucleus, the bed nucleus of the stria terminalis, the periamygdaloid cortex, the zona incerta, and the subparaventricular area. Somewhat lower binding was observed in the medial parvocellular division of the PVN and the perinuclear shells of the SON and the ventromedial hypothalamic nuclei. Low densities of SP binding sites were observed in the posterior magnocellular division of the PVN, the SON, the Arc, the ME, and SFO. Thus, it was evident that areas containing magnocellular cell bodies possessed low SP binding densities, whereas the immediate surroundings were rich in SP binding sites (Figs. 2,3). Chronic saline ingestion induced changes in the density of SP binding sites in a number of hypothalamic nuclei and other areas (Table 2). The number of SP binding sites was significantly increased in the medial preoptic and Arc nuclei, the ME, the periamygdaloid cortex, and the dorsal part of the medial parvocellular subdivision of the PVN. DISCUSSION
These results confirm earlier reports of the distribution of SP and NkA in the hypothalamus and give additional information about concentrations present within circumscribed nuclear areas. Osmotic stimulation induced significant increases of SP and NkA in the ME, while in the SFO an isolated increase of NkA was detected. Furthermore, SP/NkA levels in the SON were doubled after chronic saline ingestion, although these changes did not reach significance in the presented experimental group (Table 1). The lack of significance is likely to be caused by the rather large standard error observed for SON samples from saline-
FIG. 1. Schematic line drawingsof portions of the rat brain to illustrate the approximate dimensions of areas from where tissue samples were punched out. Sections are taken from (32). Arc: arcuate nucleus, ME: median eminence,PVN: paraventricularnucleus,SFO: subfornicalorgan, and SON: supraoptic nucleus.
treated animals, since in a repetitive control experiment with lower standard errors, significance between the groups was reached. Earlier attempts to measure tissue content of SP in grossly dissected hypothalami have not been able to confirm changes after osmotic stimulation [(15), own unpublished observations]. Obviously, the improved dissection technique employed in the present study reduces accidental noise from neighboring SP-containing hypothalamic areas to a minimum. A high content of SP and NkA in the PVN compared to the SON is consistent with a previous study in which individual hypothalamic areas were grossly dissected out (18). In individual nuclei, however, the ratio between SP and NkA immunoreactivity varied from approximately 1:2 to 1:6, contrasting with a fairly constant
TABLE 1 PEPTIDE CONCENTRATIONS IN TISSUE EXTRACTS Substance P
SFO SON PVN ME Arc
Neurokinin A
SP:NkA Ratio
Control
2% Saline
Control
2% Saline
Control
2% Saline
0.78 ± 0.3 2.0 + 0.3 15.6 ___0.4 3.0 --- 0.3 5.9 ± 0.4
0.98 _+0.3 4.2 + 1.2 14.4 + 1.1 6.1 ± 1.4t 6.9 + 0.9
1.5 ± 0.6 4.6 + 0.8 26.7 ___1.5 20.8 + 2.7 26.6 ± 2.9
6.5 _+ 1.8t 9.6 ___3.0 29.0 + 2.1 37.7 + 8.1" 30.4 ± 3.4
1:1.9 1:2.3 1:1.7 1:6.9 1:4.5
1:6.6 1:2.3 1:2.0 1:6.2 1:4.4
Concentrations of substance P (SP) and neurokinin A (NkA) in tissue extracts from various hypothalamic areas from rats given either water (control) or 340 mmol/l NaCl to drink over a period of 12 days (n = 8). *t Values are peptide concentration (ng/mg protein) and are expressed as mean ± SEM. *p < 0.05 or tP < 0.01.
~i~ ~ !~
I
~o
t~
T~
OSMOTIC REGULATION OF HYPOTHALAMIC SUBSTANCE P
1:3 ratio earlier reported for the hypothalamus as well as other regions of the CNS (18,21). The reason for this discrepancy remains an enigma. Different SP:NkA ratios in individual nuclei are likely to reflect differential expression of PPT-A mRNAs in distinct tachykinin-containing neuronal pathways, although further studies await to clarify this. The general distribution of ~25I-BH-SPbinding sites observed in the present study concurs with that earlier reported (27,33). The t25I-BH-SP radioligand preferentially labels neurokinin receptors of the Nk~ subtype, which has SP as an endogenous ligand (27). In addition to Nkl receptors, the CNS contains Nk3 receptors, which possess a one thousandfold lower affinity to ~25I-BH-SP than does the Nk~ receptor. The Nk3 receptors have inconsistently been visualized autoradiographically with radiolabeled senktide and eledoisin in the magnocellular part of the PVN and the SON. (12). High levels of [3H]senktide binding are present in the magnocellular PVN and SON but this binding is not competed by micromolar concentrations of unlabeled eledoisin, suggesting that the binding to these structures is unspecific (12). At present, receptors of the Nk2 subtype having NkA as the preferred endogenous ligand have not been demonstrated in the CNS (27). Therefore, it seems evident that the binding sites visualized in the present study are preferentially influenced by SP under normal physiological conditions, although high concentrations of NkA should be capable of mimicking the effects of SP. In addition to changes in tachykinin peptide content in certain areas, significant increases in SP binding sites were observed in the medial preoptic and arcuate nuclei, the ME, the periamygdaloid cortex, and the dorsal part of the medial parvocellular part of the PVN. Previously, we have shown that circulating ACTH levels are lowered after 12 days of saline ingestion (17) and that ICV administration of SP induces a decrease in plasma ACTH concurrent with an increase in plasma aVP (5). Therefore, emphasis was laid on studying the distribution of SP binding sites likely to mediate regulation of CRF-41- and aVP-containing neurons. The distribution of SP binding sites in the PVN complex revealed that high densities were confined to the parvocellular subdivisions and the perimeter of the magnocellular divisions. The increase in SP binding sites observed in the PVN was restricted to the dorsal part of the medial parvocellular division of the PVN, which is the site of CRF-41-containing neurons that project to the ME (l). Therefore, it seems likely that part of the SP inhibitory effect on circulating ACTH levels is mediated via interactions within the parvocellular PVN. Otherwise, the inhibitory effect on ACTH release could be due to an increase in SP-mediated inhibition of CRF-41-containing terminals in the external zone of the ME. However, SP has no inhibitory action on CRF41 release from isolated MEs from normal rats (8). Thus, it may be that in order to facilitate an inhibitory action on CRF-41 release from the ME, an increase in tachykinin content, as well as the number of SP binding sites, is required. Surprisingly, no
709
TABLE 2 BINDING OF BOLTON-HUNTERIODINATEDSUBSTANCE P TO UNFIXED RAT BRAIN SLICES
Arcuate nucleus Bed nucleus stria terminalis Caudate putamen Frontoparietal cortex Locus coeruleus Median eminence Medial preoptic nucleus Supraoptic nucleus Periamygdaloid cortex Perinuclear shell of SON Paraventricular nucleus: Medial parvicell. (dorsal) Medial parvicell. (ventral) Posterior magnocell. Subparaventricular area Suprachiasmatic nucleus: Dorsomedial division Ventrolateral division Subfornicai organ Zona incerta
Control Bound (fmol/mg)
2%SalineBound (fmol/mg)
0.49 ___0.04 (14) 1.49 + 0.04 (15) 1.12 + 0.05 (10) 0.95 -+ 0.03 (8) 1.80 - 0.06 (14) 0.44 _+0.04 (7) 1.64 + 0.04 (8) 0.33 +- 0.04 (9) 1.87 + 0.10 (9) 1.23 ---0.11 (10)
0.83 + 0.03 (11)* 1.54 ___0.03 (14) i .06 _+0.01 (9) 0.88 + 0.04 (8) 1.84 _ 0.03 (11) 0.86 + 0.03 (9)* 1.90 + 0.06 (9)* 0.37 + 0.06 (8) 2.18 + 0.05 ( 11)* 1.31 ---0.09 (9)
1.05 ---0.04 (9) 0.91 + 0.07 (8) 0.55 -+ 0.04 (12) 1.55 _+0.04 (12)
1.28 + 0.06 (12)* 0.92 + 0.03 (12) 0.54 + 0.04 (10) 1.49 _+0.04 (1 I)
0.59 _+0.04 (9) 1.45 _+0.05 (10) 0.63 _+0.06 (8) 1.60+0.11 (11)
0.67 _+0.02 (8) 1.53 _+0.03 (10) 0.74 _+0.07 (7) 1.51 +0.07(8)
Values are expressed as mean + SEM. Number of observations in parentheses. * p < 0.01 as determined by Student's t-test.
change in the SP/NkA content in the Arc or the PVN was observed after chronic saline ingestion. However, it may be that during osmotic stress the axonal flow conveying SP to ME terminals is increased, making it difficult to detect any increased SP/NkA synthesis in the Arc neurons. Some of the SP-containing neurons in the Arc are likely to be interneurons (35), and apparently the applied osmotic stimulus leaves these neurons unaffected. In contrast to the parvocellular subdivisions, the core of the posterior magnocellular PVN, where aVP-containing neurons are massed, possessed a very low density of SP binding sites. From the PVN, axons of the magnocellular neurons enter the hypothalamo-neurohypophysial tract via the dorsolateral perimeter of the PVN and this area possesses a high density of SP binding sites, suggesting a possible influence of SP at this site. Furthermore, the majority of magnocellular PVN neurons have extensive dendritic arbors penetrating into the medially situated parvocellular subdivision (37). Analogous to the architecture of the magnocellular PVN, the dendrites of the SON are localized
FIG. 2. Receptor autoradiograms showing the distribution of J25I-BH-SPbinding sites in various preoptic hypothalamic areas of normal rats. (A) At the level of the body of the anterior commissure the density of SP binding sites is very high in the medial preoptic nucleus, whereas the surrounding medial and lateral preoptic areas possess lower densities. (B) Section 200 #m caudal to that shown in (A). Moderate to high densities of SP binding sites are seen in the medial preoptic nucleus, the bed nucleus of the stria terminalis, and the rostral tip of the subfornical organ. (C) Section taken at the level of the PVN. High densities of SP binding sites are seen in the perimeter of the posterior magnocellular division, which itself has a low number of SP binding sites. The medial and periventricular parvocellular divisions contain a moderate number of binding sites. (D) At the level of the tuber cinereum, high densities of SP binding sites are seen in the zona incerta, the hypothalamic dorsomedial nucleus, and the perimeter of the hypothalamic ventromedial nucleus. The arcuate nucleus and the median eminence possess very low numbers of binding sites. (E) Autoradiogram showing SP binding sites at the level of the PVN and SON. Note the high density of binding sites in the perinuclear shell of the SON and the periamygdaloid cortex. (17)Autoradiogram showing t25I-BH-SPbinding to rat brain slice when 1 #M unlabeled SP was present in the incubation medium.
t~
Z
OSMOTIC REGULATION OF HYPOTHALAMIC SUBSTANCE P
outside the densely packed nucleus, including the ventral glial lamina and the adjoining periamygdaloid cortex, whereas the axons of SON neurons enter the hypothalamo-neurohypophysial tract via the perinuclear shell [for review see (13)]. An increased number of SP binding sites in the periamygdaloid cortex after chronic saline ingestion may therefore signify an altered sensitivity of magnocellular SON neurons. The observed increase in SP/NkA concentration in the SON may reflect increased synthesis in ascending catecholaminergic A 1/C 1 neurons, thought to be the major sources of SP-containing terminals in the magnocellular SON and PVN (3). The osmotic stimulus employed in the present study only increased plasma osmolarity by about 13%, and it may be that a more pronounced change is needed to significantly increase SP content in the magnocellular SON. In contrast, it has been demonstrated that another aVP releasing stimulus--hemorrhage--decreases the content of SP in the SON and PVN (9). However, the acuteness of that stimulus may result in a depletion of SP in SON afferents without a concomitant increase in synthesis catching up with an increased turnover in SP-containing terminals. The SFO is centrally involved in angiotensin II-mediated actions on water intake and pressor responses (34). The organ receives afferent projections from the medial parvocellular part of the PVN, the midbrain raphe, and the lateral parabrachial nuclei, and all of these areas contain SP-immunoreactive neurons
711
(23,24,26). However, only the NkA peptide level was increased in the SFO after chronic saline ingestion. Interestingly, the SFO has been shown to be rich in Nk3 receptors (33) and we have only been able to demonstrate a moderate density of putative Nk ~binding sites. This finding, combined with a change in NkA peptide content, suggests a role of NkA as a transmitter in the SFO. Tachykinins have potent antidipsogenic actions when injected into the SFO (28), and it may be that the increased NkA concentration observed in the SFO reflects an altered pattern in drinking behavior invariably seen during chronic saline ingestion. The medial preoptic nucleus is the area most sensitive to tachykinins in abolishing angiotensin II-induced drinking (28). Conceptually, the observed increase in SP binding sites in this nucleus is compatible with an increased sensitivity. In conclusion, we have shown that chronic osmotic treatment is a powerful stimulus to tachykinin-containing, as well as SPaffected, neuronal pathways in the rat hypothalamus. The present findings further strengthen the proposition that SP differentially influences hypothalamic aVP- and CRF-4 l-containing neurons. ACKNOWLEDGEMENTS This study was supported by Dr. med. vet. A. Thomsens og hustru Martha fodt Haugen-Johansens Fond, NOVO's Fond, P. Carl Petersens Fond, Fonden til LaegevidenskabensFremme, Nordisk Insulin Fond, and Wacherhausens Fond. P.J.L. is a recipient of a Wellcome Trust research fellowship.
REFERENCES 1. Antoni, F. A. Hypothalamiccontrol ofadrenocorticotropin secretion: Advancessince the discovery4 l-residue corticotropin releasingfactor. Endocr. Rev. 7:351-379; 1986. 2. Aronin, N.; Coslovsky, R.; Leeman, S. E. Substance P and neurotensin: Their roles in the regulation of anterior pituitary function. Annu. Rev. Physiol. 48:537-549; 1986. 3. Bittencourt, J. C.; Benoit, R.; Sawchenko, P. E. Distribution and origins of substance P-immunoreactive projections to the paraventricular and supraoptic nuclei: Partial overlap with ascending catecholaminergic projections. J. Chem. Neuroanat. 4:63-78; 1991. 4. Brownstein,M. J.; Mroz, E. A.; Kizer, J. S.; Palkovits, M.; Leeman, S. E. Regional distribution of substance P in the brain of the rat. Brain Res. 116:299-305; 1976. 5. Chowdrey, H. S.; Jessop, D. S.; Lightman, S. L. Substance P stimulates arginine vasopressinand inhibits adrenocorticotropin release in vivo in the rat. Neuroendocrinology 52:90-93; 1990. 6. Chowdrey, H. S.; Jessop, D. S.; Patel, H.; Lightman, S. L. Altered adrenocorticotropin, corticosteroneand oxytocin responsesto stress during chronic saline load. Neuroendocrinology 54:635-638; 1991. 7. Clarke, G.; Kirby, P. J. C.; Thomson, A. M. Effects of vasopressin and oxytocinergicneurons of intraventricularsubstanceP. J. Physiol. (Lond.) 307:59P-60P; 1980. 8. Faria, M.; Navarra, P.; Tsagarakis, S.; Besser, G. M.; Grossman, A. B. Inhibition of CRH-41 releaseby substanceP, but not substance K, from rat hypothalamus in vitro. Brain Res. 538:76-78; 1990. 9. Feuerstein, G.; Helke, C.; Faden, A. I. Differential changes in substance P and somatostatin in brain nuclei of rats exposed to haemorrhagic shock. Brain Res. 300:305-310; 1984. 10. Fitzsimons, J. T.; Evered, M. D. Eledoisin, substance P and related peptides: Intracranial dipsogens in the pigeon and antidipsogens in the rat. Brain Res. 150:533-542; 1978.
11. Gillies,G.; Linton, E. A.; Lowry, P. J. Corticotropin-releasingactivity of the new CRF is potentiated several times by vasopressin.Nature 299:355; 1982. 12. Guard, S.; Dam, T. H.; Watson, S. P.; Martinelli, B.; Watling, K. J.; Quirion, R. [3H]Sarg,Met(O2)lISP and [3H]succinyl[Asp6,MePhea]-SP(6-1 I) (senktide): New selective radioligands for NK-1 and NK-3 tachykinin receptors. DuPont BioTech Update 6: 4-8; 1991. 13. Hatton, G. I. Emergingconcepts of structure-function dynamics in adult brain: The hypothalamo-neurohypophysial system. Prog. Neurobiol. 34:437-504; 1990. 14. Heike, Y.; Hisano, S.; Tsuruo, Y.; Katoh, S.; Daikoku, S. lmmunocytochemicalevidence for synaptic regulation of paraventricular vasopressin-containingneurons by substance P. Brain Res. 369:341346; 1986. 15. Holzbauer, M.; Donnerer, J.; Holzer, P.; Sculet, W.; Lembeck, F.; Sharman, D. F. Immunoreactive substance P in the tubero-hypophyseal system of the rat: Selectivedecrease in the neural lobe after dehydration and sodium loading. Neurosci. Lett. 47:23-28; 1984. 16. Jessel, T. M. Substance P. In: Iversen, L. L.; Iversen, S. D.; Snyder, S. H., eds. Handbook ofpsychopharmacology, vol. 16:l. New York: Plenum Press; 1983:1-105. 17. Jessop, D. S.; Chowdrey, H. S.; Lightman, S. L. Inhibition of rat corticotropin-releasingfactor and adrenocorticotropin secretion by an osmotic stimulus. Brain Res. 523:1-4; 1990. 18. Jessop, D. S.; Chowdrey, H. S.; Lightman, S. L. Substance P and substance K in the median eminence and paraventricular nucleus of the rat hypothalamus. Neuropeptides 17:135-140; 1990. 19. Jessop, D. S.; Sidhu, R.; Lightman, S. L. Osmotic regulation ofmethionine-enkephalin in the posterior pituitary of the rat. Brain Res. 516:41-45; 1990.
FIG. 3. Pseudocolor images of t2SI-BH-SP receptor autoradiograms from control rats (A,C,E) and from saline-treated rats (B,D,F). After chronic saline ingestion, the density of SP binding sites in the medial preoptic nucleus is increased in the medial preoptic nucleus (A,B). An increase in the number of SP binding sites is observed in the dorsal part of the medial parvocellular division of the PVN after saline treatment (C,D). Chronic saline treatment significantlyincreasedthe number of SP bindingsites in the Arc and ME, whilethe surrounding perimeter of the hypothalamic ventromedial nucleus was unaffected.
712 20. Jessop, D. S.; Chowdrey, H. S.; Larsen, P. J.; Lightman, S. L. Substance P: A muitifunctional peptide in the hypothalamo-pituitary system. J. Endocrinol. 132:331-337; 1992. 21. Kanazawa, I.; Ogawa, T.; Kimura, S.; Munekata, E. Regional distribution of substance P, neurokinin a and neurokinin B in rat central nervous system. Neurosci. Res. 2:111-120; 1984. 22. Krause, J. E.; Chirgwin, J. M.; Carter, M. S.; Xu, X. S.; Hershey, A. D. Three rat preprotachykinin mRNAs encode the neuropeptides substance P and neurokinin A. Proc. Natl. Acad. Sci. USA 84:881885; 1987. 23. Larsen, P. J. Distribution of substance P-immunoreactive elements in the preoptic area and hypothalamus of the rat. J. Comp. Neurol. 316:287-313; 1992. 24. Larsen, P. J.;Molter, M.; Mikkelsen, J. D. Efferent projections from the periventricular and medial parvocellular subnuclei of the hypothalamic paraventricular nucleus to circumventricular organs of the rat. A Phaseolus vulgaris leueoagglutinin(PHA-L) tracing study. J. Comp. Neurol. 306:462--479; 1991. 25. Lightman, S. L.; Young, S. W., IlL Vasopressin, oxytocin, dynorphin, enkephalin and corticotrophin-releasing factor mRNA stimulation in the rat. J. Physiol. 394:23-39; 1987. 26. Lind, R. W. Bi-directional, chemically specified neural connections between the subfornical organ and the midbrain raphe system. Brain Res. 384:250-261; 1986. 27. Mantyh, P. W.; Gates, T.; Mantyh, C. R.; Maggio, J. E. Autoradiographic localization and characterization of tachykinin receptor binding sites in the rat brain and peripheral tissues. J. Neurosci. 9: 258-279; 1989. 28. Massi, M.; de Caro, G.; Perfumi, M.; Venturi, F. Mapping of brain sites sensitive to the antidipsogenic effect of tachykinins. Peptides 9:347-356; 1988.
LARSEN ET AL. 29. Massi, M.; Saija, A.; Polidori, C.; Perfumi, M.; Gentili, L.; Costa, G.; deCaro, G. The hypothalamic paraventricular nucleus is a site of action for the central effect of tachykinin on plasma vasopressin. Brain Res. Bull. 26:149-154; 1991. 30. Nawa, H.; Kotani, H.; Nakanishi, S. Tissue specific generation of two preprotachykinin mRNAs from one gene by alternative RNA splicing. Nature 312:729-734; 1984. 31. Palkovits, M.; Brownstein, M. J. Microdissection of brain areas by the punch technique. In: Cuello, A. C., ed. Brain microdissection techniques. Chichester: John Wiley; 1983:1-36. 32. Paxinos, G.; Watson, C. The rat brain in stereotaxic coordinates, second edition. Sydney: Academic Press; 1986. 33. Saffroy, M.; Beaujouan, J. C.; Torrens, Y.; Besseyre, J.; BergstrSm, L.; Glowinski, J. Localization of tachykinin binding sites (NK I, NK 2, NK a ligands) in the rat brain. Peptides 9:227-241; 1988. 34. Simpson, J. B. The circumventricular organs and central actions of angiotensin. Neuroendocrinology 32:248-256; 1981. 35. Tsuruo, Y.; Hisano, S.; Daikoku, S. Morphological evidence for synaptic junctions between substance P-containing neurons in the arcuate nucleus of the rat. Neurosci. Lett. 46:65-69; 1984. 36. Vale, W.; Vaughan, J.; Smith, M.; Yamamoto, G.; Rivier, J.; Rivier, C. Effects of synthetic ovine corticotropin releasing-factor, glucocorticoids, cathecolamines, neurohypophysial peptides and other substances on cultured corticotropic cells. Endocrinology I 13:1121; 1983. 37. van den Pol, A. The magnocellular and parvocellular paraventricular nucleus of rat: Intrinsic organization. J. Comp. Neurol. 206:317345; 1982. 38. Young, W. S., III Corticotropin-releasing factor mRNA in the hypothalamus is affected differentially by drinking saline and by dehydration. FEBS Lett. 208:158-162; 1986.
0196-9781/92 $5.00 + .00 Copyright© 1992PergamonPressLtd.
Printedin the USA.
Osmotic Regulation of Substance P and Neurokinin A Peptide Content and Substance P Binding Sites in Distinct Hypothalamic Nuclei of the Rat P H I L I P J. L A R S E N , * t L D A V I D S. JESSOP,* H A R D I A L S. C H O W D R E Y , * JENS D. MIKKELSEN~" A N D S T A F F O R D L. L I G H T M A N *
*Neuroendocrinology Unit, Charing Cross and Westminster Medical School, London, UK and -~Institute of Medical Anatomy, Department B, University of Copenhagen, Denmark Received 16 D e c e m b e r 1991 LARSEN, P. J., D. S. JESSOP, H. S. CHOWDREY, J. D. MIKKELSEN AND S. L. LIGHTMAN. Osmotic regulation of substance P and neurokininA peptidecontent and substanceP binding sites in distincthypothalamic nucleiof the rat. PEPTIDES 13(4) 705-712, 1992.--Quantitative receptor autoradiography using Bolton-Hunter iodinated substance P (SP) was used to localize specific sites in the rat hypothalamus. The amount of SP and neurokinin A (NkA) in extracts from discrete areas of the hypothalamus was measured using specific radioimmunoassays. A high density of SP binding sites was observed in the perimeter of the magnocellular paraventricular and supraoptic nuclei, while the magnocellular nuclei themselves possessed a low receptor density. In control animals, the number of SP binding sites was also low in the arcuate nucleus and the median eminence. Substance P and NkA peptide concentrations were highest in the paraventricular nucleus (PVN), decreasing in the following order: arcuate nucleus (Arc) > median eminence (ME) > supraoptic nucleus (SON) > subfornical organ (SFO). In animals given 340 mmol/l NaCI instead of tap water to drink for 12 days, significant increases in the number of SP binding sites occurred in the medial parvocellular subdivision of the PVN, periamygdaloid cortex, medial preoptic nucleus, Arc, and ME, but other hypothalamic areas were unaffected. In saline-treated animals, significantincreases in SP and NkA peptide concentrations were observed in the ME, while in the SFO only the concentration of NkA increased significantly. In the SON, substance P and neurokinin A levels were doubled, whereas in the PVN and Arc no changes in peptide levels were observed. Chronic osmotic stimulation is associated with lowered circulating levels of adrenocorticotropin releasinghormone (ACTH), and the present data further substantiate the hypothesis that hypothalamic tachykinin-containing neuronal terminals are centrally involved in the inhibition of anterior pituitary ACTH release observed during chronic osmotic stimulation. Substance P Neurokinin A Hypothalamus Rat
Salt loading
Microdissection
THE mammalian tachykinins substance P (SP) and neurokinin A (NkA) are widely distributed throughout the central nervous system (CNS), with high concentrations in the hypothalamus (4,18). Substance P is one of the most extensively studied and best characterized neuropeptides in the CNS (16), whereas little is known about the role of NkA in CNS functions. Substance P and NkA are encoded by the same gene, preprotachykinin A (PPT-A), which by alternative posttranscriptional splicing results in different amounts of their distinct mRNAs (22,30). This is also reflected in the relative amounts ofpeptide present in various CNS regions (18,21). Substance P and NkA show many similarities and are considered to share functional roles as well.
Receptorautoradiography
Radioimmunoassay
A large body of evidence has accumulated to suggest that SP is involved in diverse neuroendocrine functions including regulation of the release of anterior pituitary hormones (2,20) and water intake/drinking behavior (10,28). Substance P and NkA act as neurotransmitters in the regulation of the magnocellular hypothalamo-pituitary system. Intracerebroventricular (ICV) injection of SP causes antidiuresis in anesthetized rats (5), increases the firing rate of both vasopressinergic (aVP) and oxytocinergic (OXY) neurons of the supraoptic nucleus (SON) (7), and induces a release of vasopressin into the systemic circulation (5). Curiously, ICV injection of SP is also associated with decreased plasma concentration of ACTH (5), the latter most likely
Requests for reprints should be addressed to Philip J. Larsen, M.D., Institute of Medical Anatomy, Department B, The Panum Institute, Blegdamsvej 3, DK-2200 Copenhagen N, Denmark.
705
706
LARSEN ET AL.
being a result of inhibition of the release ofcorticotropin releasing factor (CRF-41) from the median eminence (ME) to portal vessels (8). In a model of chronic osmotic stimulation where rats are offered 2% saline to drink for a period of 12 days, the content of CRF-41 mRNA within the medial parvocellular subnucleus of the hypothalamic paraventricular nucleus (PVN) is decreased (25,38). Recently, we have demonstrated that this osmotic stimulus, which increases circulating aVP levels, also lowers basal circulating ACTH and corticosterone levels (17). In addition, the normally occurring stress-induced ACTH and corticosterone responses are absent in these osmotically stimulated animals (6). Taken together, these findings raise the possibility that endogenous SP or other tachykinins mediate the down-regulation o f C R F gene expression and circulating ACTH observed during chronic osmotic stimulation. The site of action of SP/NkA on aVP- or CRF-41-containing neurons may still be uncertain. However, within the hypothalamic paraventricular nucleus (PVN), SP-immunoreactive terminals have been demonstrated in apposition to aVP-containing perikarya (14) and injection of various tachykinins into the PVN induces an increase in plasma vasopressin concentration (29), suggesting a direct influence on magnocellular neurons. The CRF-41-containing neurons in the medial parvocellular division of the PVN colocalize aVP and project to the external zone of the ME, where they release CRF41 and aVP into hypophysial portal blood [for review see ( 1)]. Both peptides act as potent stimulatory neuroendocrine transmitters on ACTH release from adenohypophysial corticotrophs (11,36), and it seems possible that SP exerts its central inhibitory effects on circulating ACTH levels via these CRF/aVP neurons. The above-mentioned findings raise the possibility that endogenous SP mediates the paradoxical changes in ACTH and aVP response to chronic osmotic stimulation. To elucidate the anatomical location of these effects and to add further evidence for the role of SP and NkA, we used microdissection and specific radioimmunoassays to determine whether alterations in SP and NkA tissue content occurred after chronic saline ingestion in various areas of the brain involved in ACTH release and water homeostasis. In addition, using quantitative receptor autoradiography, we investigated any alterations in the density of SP binding sites in these and other hypothalamic nuclei in 2% salinetreated animals. METHOD
SP antiserum with NkA was <0.05% and of NkA antiserum with SP was 0.15%.
A utoradiographic Procedure Animals were anesthetized with tribromethanol (250 mg/kg/ body weight) and transcardially perfused with heparinized isotonic saline followed by 100 ml 0.32 M sucrose. Brains were rapidly removed and stored at - 8 0 ° C until serial 20-um cryostat sections were cut. The sections were thaw mounted on gelatine/ chrome alum-subbed slides and kept at - 8 0 ° C until use. For binding experiments, sections were transferred from the freezer to a hot plate (40°C) for 1 min to remove condensed water. The sections were then preincubated at room temperature for 20 min in a 50 mM Tris buffer (pH 7.4) containing MnCI2 (3 mM), Na2SO4 (10 mM), bacitracin (40 mg/l), leupeptin (4 mg/1), chymostatin (2 mg/1), and BSA (0.02%), hereafter called buffer A. Sections were incubated at room temperature for 60 min in buffer A containing 1 nM ~25I-Bolton-Hunter iodinated SP (125I-BH-SP, Amersham, Buckinghamshire, UK). After a brief wash in buffer A, sections went through two 2-rain rinses in buffer A, and were dipped in distilled H20 before being dried rapidly under a stream of cool dry air. Sections were then exposed to Amersham Hyperfilm® for 7 clays and fnally developed in Kodak D19 developer. Autoradiograms were quantified using an image analysis system (Image 1.16 g, Wayne Rasband, N.I.H., USA) and grain densities converted to fmol bound ligand/mg wet weight using Amersham J25I-micro-scales®. Nonspecific binding was measured by including 1 ttM unlabeled SP in incubations with ~251-BH-SP.
Statistical Analysis When immunoreactive tachykinin levels from control animals were compared to those of 2% saline-treated animals, data were evaluated by two-tailed Student's t-test. The same statistical paradigm was employed on densitometrical data Obtained from receptor autoradiography. RESULTS
Osmotic challenge by replacing drinking water with 340 mmol/1 NaCI for 12 days significantly increased plasma osmolarity from 285 -+ 4 to 323 -+ 10 mOsm/kg H20 (p < 0.001, mean --- SEM, two-tailed Student's t-test). The concentration of plasma aVP increased from 1.68 ___0.19 to 5.9 + 1.29 pmol/l (p < 0.0 I, mean -+ SEM, two-tailed Student's t-test).
Animal Treatment Male Sprague-Dawley rats weighing 200-250 g were housed under standard laboratory conditions with free access to food and either tap water (control) or a 2% (w/v) NaC1 solution for 12 days. All animals were housed in cages with a grid bottom. The effects of this treatment on plasma osmolarity and vasopressin levels have been described in detail previously (19). Following treatment, the rats were decapitated, their brains rapidly removed and frozen on dry ice for storage at -80°C. Thereafter, 150-#m thick cryostat sections were cut through the hypothalamus and stored at -800C. Using a microdissection needle (0.50 mm), selected hypothalamic areas were punched out according to a method previously described (31). Tissue punches were frozen on dry ice, while the sections were fixed in 4% paraformaldehyde and counterstained to allow validation of the sampling technique. Tissue samples were extracted and measured for SP and NkA as previously described (18). Cross-reactivities of SP and NkA antisera with the other mammalian tachykinin, neurokinin B, were <0.05 and 7%, respectively; cross-reactivity of
Tissue Peptide Content The SP and NkA contents of rat brain tissue extracts are shown in Table 1. All investigated areas contained detectable levels of SP and NkA. Histological examination of fixed brain slices confirmed that the areas under investigation had been punched out (Fig. 1). Following treatment for 12 days with 340 mmol/l NaCI, significant increases in SP and NkA were observed in the median eminence (ME) while NkA, but not SP, increased significantly in the subfornical organ (SFO). In all other areas investigated, no significant changes in the content of SP or NkA were observed. Although the concentrations of SP and NkA in the supraoptic nucleus (SON) were doubled, these values did not reach significance in this series of measurements. In a separate experiment, however, the observed increases in levels of SP and NkA in the SON reached significance (SP increased from 1.89 -- 0.19 to 3.86 --- 0.63 ng/mg, p < 0.05; and NkA increased from 2.58 -- 0.49 to 4.57 _+ 0.66 rig/rag, p < 0.05). A general correspondence between levels of SP and NkA concentrations was
OSMOTIC REGULATION OF HYPOTHALAMIC SUBSTANCE P
707
observed in individual nuclei of control and osmotically stimulated animals. However, the ratio between SP and NkA concentrations varied greatly between different areas. Thus, concentration of NkA in the SFO and ME was about sixfold higher than that of SP, whereas the SP:NkA ratio in the SON and PVN was 1:2.
Substance P Binding Sites Bolton-Hunter iodinated substance P (IZSI-BH-SP) was employed as a ligand to autoradiographically visualize SP binding sites in rat brain slices. For convenience, the 125I-BH-SPbinding sites are hereafter referred to as SP binding sites. Binding of ~25IBH-SP to tissue sections was completely abolished by addition of 1 ~tM unlabeled SP in the incubation medium. In control animals, specific binding sites for SP were demonstrated throughout the mediobasal diencephalon, including the preoptic •area and the mediobasal hypothalamus (Fig. 2, Table 2). In particular, very high densities were observed in the medial preoptic nucleus, the bed nucleus of the stria terminalis, the periamygdaloid cortex, the zona incerta, and the subparaventricular area. Somewhat lower binding was observed in the medial parvocellular division of the PVN and the perinuclear shells of the SON and the ventromedial hypothalamic nuclei. Low densities of SP binding sites were observed in the posterior magnocellular division of the PVN, the SON, the Arc, the ME, and SFO. Thus, it was evident that areas containing magnocellular cell bodies possessed low SP binding densities, whereas the immediate surroundings were rich in SP binding sites (Figs. 2,3). Chronic saline ingestion induced changes in the density of SP binding sites in a number of hypothalamic nuclei and other areas (Table 2). The number of SP binding sites was significantly increased in the medial preoptic and Arc nuclei, the ME, the periamygdaloid cortex, and the dorsal part of the medial parvocellular subdivision of the PVN. DISCUSSION
These results confirm earlier reports of the distribution of SP and NkA in the hypothalamus and give additional information about concentrations present within circumscribed nuclear areas. Osmotic stimulation induced significant increases of SP and NkA in the ME, while in the SFO an isolated increase of NkA was detected. Furthermore, SP/NkA levels in the SON were doubled after chronic saline ingestion, although these changes did not reach significance in the presented experimental group (Table 1). The lack of significance is likely to be caused by the rather large standard error observed for SON samples from saline-
FIG. 1. Schematic line drawingsof portions of the rat brain to illustrate the approximate dimensions of areas from where tissue samples were punched out. Sections are taken from (32). Arc: arcuate nucleus, ME: median eminence,PVN: paraventricularnucleus,SFO: subfornicalorgan, and SON: supraoptic nucleus.
treated animals, since in a repetitive control experiment with lower standard errors, significance between the groups was reached. Earlier attempts to measure tissue content of SP in grossly dissected hypothalami have not been able to confirm changes after osmotic stimulation [(15), own unpublished observations]. Obviously, the improved dissection technique employed in the present study reduces accidental noise from neighboring SP-containing hypothalamic areas to a minimum. A high content of SP and NkA in the PVN compared to the SON is consistent with a previous study in which individual hypothalamic areas were grossly dissected out (18). In individual nuclei, however, the ratio between SP and NkA immunoreactivity varied from approximately 1:2 to 1:6, contrasting with a fairly constant
TABLE 1 PEPTIDE CONCENTRATIONS IN TISSUE EXTRACTS Substance P
SFO SON PVN ME Arc
Neurokinin A
SP:NkA Ratio
Control
2% Saline
Control
2% Saline
Control
2% Saline
0.78 ± 0.3 2.0 + 0.3 15.6 ___0.4 3.0 --- 0.3 5.9 ± 0.4
0.98 _+0.3 4.2 + 1.2 14.4 + 1.1 6.1 ± 1.4t 6.9 + 0.9
1.5 ± 0.6 4.6 + 0.8 26.7 ___1.5 20.8 + 2.7 26.6 ± 2.9
6.5 _+ 1.8t 9.6 ___3.0 29.0 + 2.1 37.7 + 8.1" 30.4 ± 3.4
1:1.9 1:2.3 1:1.7 1:6.9 1:4.5
1:6.6 1:2.3 1:2.0 1:6.2 1:4.4
Concentrations of substance P (SP) and neurokinin A (NkA) in tissue extracts from various hypothalamic areas from rats given either water (control) or 340 mmol/l NaCl to drink over a period of 12 days (n = 8). *t Values are peptide concentration (ng/mg protein) and are expressed as mean ± SEM. *p < 0.05 or tP < 0.01.
~i~ ~ !~
I
~o
t~
T~
OSMOTIC REGULATION OF HYPOTHALAMIC SUBSTANCE P
1:3 ratio earlier reported for the hypothalamus as well as other regions of the CNS (18,21). The reason for this discrepancy remains an enigma. Different SP:NkA ratios in individual nuclei are likely to reflect differential expression of PPT-A mRNAs in distinct tachykinin-containing neuronal pathways, although further studies await to clarify this. The general distribution of ~25I-BH-SPbinding sites observed in the present study concurs with that earlier reported (27,33). The t25I-BH-SP radioligand preferentially labels neurokinin receptors of the Nk~ subtype, which has SP as an endogenous ligand (27). In addition to Nkl receptors, the CNS contains Nk3 receptors, which possess a one thousandfold lower affinity to ~25I-BH-SP than does the Nk~ receptor. The Nk3 receptors have inconsistently been visualized autoradiographically with radiolabeled senktide and eledoisin in the magnocellular part of the PVN and the SON. (12). High levels of [3H]senktide binding are present in the magnocellular PVN and SON but this binding is not competed by micromolar concentrations of unlabeled eledoisin, suggesting that the binding to these structures is unspecific (12). At present, receptors of the Nk2 subtype having NkA as the preferred endogenous ligand have not been demonstrated in the CNS (27). Therefore, it seems evident that the binding sites visualized in the present study are preferentially influenced by SP under normal physiological conditions, although high concentrations of NkA should be capable of mimicking the effects of SP. In addition to changes in tachykinin peptide content in certain areas, significant increases in SP binding sites were observed in the medial preoptic and arcuate nuclei, the ME, the periamygdaloid cortex, and the dorsal part of the medial parvocellular part of the PVN. Previously, we have shown that circulating ACTH levels are lowered after 12 days of saline ingestion (17) and that ICV administration of SP induces a decrease in plasma ACTH concurrent with an increase in plasma aVP (5). Therefore, emphasis was laid on studying the distribution of SP binding sites likely to mediate regulation of CRF-41- and aVP-containing neurons. The distribution of SP binding sites in the PVN complex revealed that high densities were confined to the parvocellular subdivisions and the perimeter of the magnocellular divisions. The increase in SP binding sites observed in the PVN was restricted to the dorsal part of the medial parvocellular division of the PVN, which is the site of CRF-41-containing neurons that project to the ME (l). Therefore, it seems likely that part of the SP inhibitory effect on circulating ACTH levels is mediated via interactions within the parvocellular PVN. Otherwise, the inhibitory effect on ACTH release could be due to an increase in SP-mediated inhibition of CRF-41-containing terminals in the external zone of the ME. However, SP has no inhibitory action on CRF41 release from isolated MEs from normal rats (8). Thus, it may be that in order to facilitate an inhibitory action on CRF-41 release from the ME, an increase in tachykinin content, as well as the number of SP binding sites, is required. Surprisingly, no
709
TABLE 2 BINDING OF BOLTON-HUNTERIODINATEDSUBSTANCE P TO UNFIXED RAT BRAIN SLICES
Arcuate nucleus Bed nucleus stria terminalis Caudate putamen Frontoparietal cortex Locus coeruleus Median eminence Medial preoptic nucleus Supraoptic nucleus Periamygdaloid cortex Perinuclear shell of SON Paraventricular nucleus: Medial parvicell. (dorsal) Medial parvicell. (ventral) Posterior magnocell. Subparaventricular area Suprachiasmatic nucleus: Dorsomedial division Ventrolateral division Subfornicai organ Zona incerta
Control Bound (fmol/mg)
2%SalineBound (fmol/mg)
0.49 ___0.04 (14) 1.49 + 0.04 (15) 1.12 + 0.05 (10) 0.95 -+ 0.03 (8) 1.80 - 0.06 (14) 0.44 _+0.04 (7) 1.64 + 0.04 (8) 0.33 +- 0.04 (9) 1.87 + 0.10 (9) 1.23 ---0.11 (10)
0.83 + 0.03 (11)* 1.54 ___0.03 (14) i .06 _+0.01 (9) 0.88 + 0.04 (8) 1.84 _ 0.03 (11) 0.86 + 0.03 (9)* 1.90 + 0.06 (9)* 0.37 + 0.06 (8) 2.18 + 0.05 ( 11)* 1.31 ---0.09 (9)
1.05 ---0.04 (9) 0.91 + 0.07 (8) 0.55 -+ 0.04 (12) 1.55 _+0.04 (12)
1.28 + 0.06 (12)* 0.92 + 0.03 (12) 0.54 + 0.04 (10) 1.49 _+0.04 (1 I)
0.59 _+0.04 (9) 1.45 _+0.05 (10) 0.63 _+0.06 (8) 1.60+0.11 (11)
0.67 _+0.02 (8) 1.53 _+0.03 (10) 0.74 _+0.07 (7) 1.51 +0.07(8)
Values are expressed as mean + SEM. Number of observations in parentheses. * p < 0.01 as determined by Student's t-test.
change in the SP/NkA content in the Arc or the PVN was observed after chronic saline ingestion. However, it may be that during osmotic stress the axonal flow conveying SP to ME terminals is increased, making it difficult to detect any increased SP/NkA synthesis in the Arc neurons. Some of the SP-containing neurons in the Arc are likely to be interneurons (35), and apparently the applied osmotic stimulus leaves these neurons unaffected. In contrast to the parvocellular subdivisions, the core of the posterior magnocellular PVN, where aVP-containing neurons are massed, possessed a very low density of SP binding sites. From the PVN, axons of the magnocellular neurons enter the hypothalamo-neurohypophysial tract via the dorsolateral perimeter of the PVN and this area possesses a high density of SP binding sites, suggesting a possible influence of SP at this site. Furthermore, the majority of magnocellular PVN neurons have extensive dendritic arbors penetrating into the medially situated parvocellular subdivision (37). Analogous to the architecture of the magnocellular PVN, the dendrites of the SON are localized
FIG. 2. Receptor autoradiograms showing the distribution of J25I-BH-SPbinding sites in various preoptic hypothalamic areas of normal rats. (A) At the level of the body of the anterior commissure the density of SP binding sites is very high in the medial preoptic nucleus, whereas the surrounding medial and lateral preoptic areas possess lower densities. (B) Section 200 #m caudal to that shown in (A). Moderate to high densities of SP binding sites are seen in the medial preoptic nucleus, the bed nucleus of the stria terminalis, and the rostral tip of the subfornical organ. (C) Section taken at the level of the PVN. High densities of SP binding sites are seen in the perimeter of the posterior magnocellular division, which itself has a low number of SP binding sites. The medial and periventricular parvocellular divisions contain a moderate number of binding sites. (D) At the level of the tuber cinereum, high densities of SP binding sites are seen in the zona incerta, the hypothalamic dorsomedial nucleus, and the perimeter of the hypothalamic ventromedial nucleus. The arcuate nucleus and the median eminence possess very low numbers of binding sites. (E) Autoradiogram showing SP binding sites at the level of the PVN and SON. Note the high density of binding sites in the perinuclear shell of the SON and the periamygdaloid cortex. (17)Autoradiogram showing t25I-BH-SPbinding to rat brain slice when 1 #M unlabeled SP was present in the incubation medium.
t~
Z
OSMOTIC REGULATION OF HYPOTHALAMIC SUBSTANCE P
outside the densely packed nucleus, including the ventral glial lamina and the adjoining periamygdaloid cortex, whereas the axons of SON neurons enter the hypothalamo-neurohypophysial tract via the perinuclear shell [for review see (13)]. An increased number of SP binding sites in the periamygdaloid cortex after chronic saline ingestion may therefore signify an altered sensitivity of magnocellular SON neurons. The observed increase in SP/NkA concentration in the SON may reflect increased synthesis in ascending catecholaminergic A 1/C 1 neurons, thought to be the major sources of SP-containing terminals in the magnocellular SON and PVN (3). The osmotic stimulus employed in the present study only increased plasma osmolarity by about 13%, and it may be that a more pronounced change is needed to significantly increase SP content in the magnocellular SON. In contrast, it has been demonstrated that another aVP releasing stimulus--hemorrhage--decreases the content of SP in the SON and PVN (9). However, the acuteness of that stimulus may result in a depletion of SP in SON afferents without a concomitant increase in synthesis catching up with an increased turnover in SP-containing terminals. The SFO is centrally involved in angiotensin II-mediated actions on water intake and pressor responses (34). The organ receives afferent projections from the medial parvocellular part of the PVN, the midbrain raphe, and the lateral parabrachial nuclei, and all of these areas contain SP-immunoreactive neurons
711
(23,24,26). However, only the NkA peptide level was increased in the SFO after chronic saline ingestion. Interestingly, the SFO has been shown to be rich in Nk3 receptors (33) and we have only been able to demonstrate a moderate density of putative Nk ~binding sites. This finding, combined with a change in NkA peptide content, suggests a role of NkA as a transmitter in the SFO. Tachykinins have potent antidipsogenic actions when injected into the SFO (28), and it may be that the increased NkA concentration observed in the SFO reflects an altered pattern in drinking behavior invariably seen during chronic saline ingestion. The medial preoptic nucleus is the area most sensitive to tachykinins in abolishing angiotensin II-induced drinking (28). Conceptually, the observed increase in SP binding sites in this nucleus is compatible with an increased sensitivity. In conclusion, we have shown that chronic osmotic treatment is a powerful stimulus to tachykinin-containing, as well as SPaffected, neuronal pathways in the rat hypothalamus. The present findings further strengthen the proposition that SP differentially influences hypothalamic aVP- and CRF-4 l-containing neurons. ACKNOWLEDGEMENTS This study was supported by Dr. med. vet. A. Thomsens og hustru Martha fodt Haugen-Johansens Fond, NOVO's Fond, P. Carl Petersens Fond, Fonden til LaegevidenskabensFremme, Nordisk Insulin Fond, and Wacherhausens Fond. P.J.L. is a recipient of a Wellcome Trust research fellowship.
REFERENCES 1. Antoni, F. A. Hypothalamiccontrol ofadrenocorticotropin secretion: Advancessince the discovery4 l-residue corticotropin releasingfactor. Endocr. Rev. 7:351-379; 1986. 2. Aronin, N.; Coslovsky, R.; Leeman, S. E. Substance P and neurotensin: Their roles in the regulation of anterior pituitary function. Annu. Rev. Physiol. 48:537-549; 1986. 3. Bittencourt, J. C.; Benoit, R.; Sawchenko, P. E. Distribution and origins of substance P-immunoreactive projections to the paraventricular and supraoptic nuclei: Partial overlap with ascending catecholaminergic projections. J. Chem. Neuroanat. 4:63-78; 1991. 4. Brownstein,M. J.; Mroz, E. A.; Kizer, J. S.; Palkovits, M.; Leeman, S. E. Regional distribution of substance P in the brain of the rat. Brain Res. 116:299-305; 1976. 5. Chowdrey, H. S.; Jessop, D. S.; Lightman, S. L. Substance P stimulates arginine vasopressinand inhibits adrenocorticotropin release in vivo in the rat. Neuroendocrinology 52:90-93; 1990. 6. Chowdrey, H. S.; Jessop, D. S.; Patel, H.; Lightman, S. L. Altered adrenocorticotropin, corticosteroneand oxytocin responsesto stress during chronic saline load. Neuroendocrinology 54:635-638; 1991. 7. Clarke, G.; Kirby, P. J. C.; Thomson, A. M. Effects of vasopressin and oxytocinergicneurons of intraventricularsubstanceP. J. Physiol. (Lond.) 307:59P-60P; 1980. 8. Faria, M.; Navarra, P.; Tsagarakis, S.; Besser, G. M.; Grossman, A. B. Inhibition of CRH-41 releaseby substanceP, but not substance K, from rat hypothalamus in vitro. Brain Res. 538:76-78; 1990. 9. Feuerstein, G.; Helke, C.; Faden, A. I. Differential changes in substance P and somatostatin in brain nuclei of rats exposed to haemorrhagic shock. Brain Res. 300:305-310; 1984. 10. Fitzsimons, J. T.; Evered, M. D. Eledoisin, substance P and related peptides: Intracranial dipsogens in the pigeon and antidipsogens in the rat. Brain Res. 150:533-542; 1978.
11. Gillies,G.; Linton, E. A.; Lowry, P. J. Corticotropin-releasingactivity of the new CRF is potentiated several times by vasopressin.Nature 299:355; 1982. 12. Guard, S.; Dam, T. H.; Watson, S. P.; Martinelli, B.; Watling, K. J.; Quirion, R. [3H]Sarg,Met(O2)lISP and [3H]succinyl[Asp6,MePhea]-SP(6-1 I) (senktide): New selective radioligands for NK-1 and NK-3 tachykinin receptors. DuPont BioTech Update 6: 4-8; 1991. 13. Hatton, G. I. Emergingconcepts of structure-function dynamics in adult brain: The hypothalamo-neurohypophysial system. Prog. Neurobiol. 34:437-504; 1990. 14. Heike, Y.; Hisano, S.; Tsuruo, Y.; Katoh, S.; Daikoku, S. lmmunocytochemicalevidence for synaptic regulation of paraventricular vasopressin-containingneurons by substance P. Brain Res. 369:341346; 1986. 15. Holzbauer, M.; Donnerer, J.; Holzer, P.; Sculet, W.; Lembeck, F.; Sharman, D. F. Immunoreactive substance P in the tubero-hypophyseal system of the rat: Selectivedecrease in the neural lobe after dehydration and sodium loading. Neurosci. Lett. 47:23-28; 1984. 16. Jessel, T. M. Substance P. In: Iversen, L. L.; Iversen, S. D.; Snyder, S. H., eds. Handbook ofpsychopharmacology, vol. 16:l. New York: Plenum Press; 1983:1-105. 17. Jessop, D. S.; Chowdrey, H. S.; Lightman, S. L. Inhibition of rat corticotropin-releasingfactor and adrenocorticotropin secretion by an osmotic stimulus. Brain Res. 523:1-4; 1990. 18. Jessop, D. S.; Chowdrey, H. S.; Lightman, S. L. Substance P and substance K in the median eminence and paraventricular nucleus of the rat hypothalamus. Neuropeptides 17:135-140; 1990. 19. Jessop, D. S.; Sidhu, R.; Lightman, S. L. Osmotic regulation ofmethionine-enkephalin in the posterior pituitary of the rat. Brain Res. 516:41-45; 1990.
FIG. 3. Pseudocolor images of t2SI-BH-SP receptor autoradiograms from control rats (A,C,E) and from saline-treated rats (B,D,F). After chronic saline ingestion, the density of SP binding sites in the medial preoptic nucleus is increased in the medial preoptic nucleus (A,B). An increase in the number of SP binding sites is observed in the dorsal part of the medial parvocellular division of the PVN after saline treatment (C,D). Chronic saline treatment significantlyincreasedthe number of SP bindingsites in the Arc and ME, whilethe surrounding perimeter of the hypothalamic ventromedial nucleus was unaffected.
712 20. Jessop, D. S.; Chowdrey, H. S.; Larsen, P. J.; Lightman, S. L. Substance P: A muitifunctional peptide in the hypothalamo-pituitary system. J. Endocrinol. 132:331-337; 1992. 21. Kanazawa, I.; Ogawa, T.; Kimura, S.; Munekata, E. Regional distribution of substance P, neurokinin a and neurokinin B in rat central nervous system. Neurosci. Res. 2:111-120; 1984. 22. Krause, J. E.; Chirgwin, J. M.; Carter, M. S.; Xu, X. S.; Hershey, A. D. Three rat preprotachykinin mRNAs encode the neuropeptides substance P and neurokinin A. Proc. Natl. Acad. Sci. USA 84:881885; 1987. 23. Larsen, P. J. Distribution of substance P-immunoreactive elements in the preoptic area and hypothalamus of the rat. J. Comp. Neurol. 316:287-313; 1992. 24. Larsen, P. J.;Molter, M.; Mikkelsen, J. D. Efferent projections from the periventricular and medial parvocellular subnuclei of the hypothalamic paraventricular nucleus to circumventricular organs of the rat. A Phaseolus vulgaris leueoagglutinin(PHA-L) tracing study. J. Comp. Neurol. 306:462--479; 1991. 25. Lightman, S. L.; Young, S. W., IlL Vasopressin, oxytocin, dynorphin, enkephalin and corticotrophin-releasing factor mRNA stimulation in the rat. J. Physiol. 394:23-39; 1987. 26. Lind, R. W. Bi-directional, chemically specified neural connections between the subfornical organ and the midbrain raphe system. Brain Res. 384:250-261; 1986. 27. Mantyh, P. W.; Gates, T.; Mantyh, C. R.; Maggio, J. E. Autoradiographic localization and characterization of tachykinin receptor binding sites in the rat brain and peripheral tissues. J. Neurosci. 9: 258-279; 1989. 28. Massi, M.; de Caro, G.; Perfumi, M.; Venturi, F. Mapping of brain sites sensitive to the antidipsogenic effect of tachykinins. Peptides 9:347-356; 1988.
LARSEN ET AL. 29. Massi, M.; Saija, A.; Polidori, C.; Perfumi, M.; Gentili, L.; Costa, G.; deCaro, G. The hypothalamic paraventricular nucleus is a site of action for the central effect of tachykinin on plasma vasopressin. Brain Res. Bull. 26:149-154; 1991. 30. Nawa, H.; Kotani, H.; Nakanishi, S. Tissue specific generation of two preprotachykinin mRNAs from one gene by alternative RNA splicing. Nature 312:729-734; 1984. 31. Palkovits, M.; Brownstein, M. J. Microdissection of brain areas by the punch technique. In: Cuello, A. C., ed. Brain microdissection techniques. Chichester: John Wiley; 1983:1-36. 32. Paxinos, G.; Watson, C. The rat brain in stereotaxic coordinates, second edition. Sydney: Academic Press; 1986. 33. Saffroy, M.; Beaujouan, J. C.; Torrens, Y.; Besseyre, J.; BergstrSm, L.; Glowinski, J. Localization of tachykinin binding sites (NK I, NK 2, NK a ligands) in the rat brain. Peptides 9:227-241; 1988. 34. Simpson, J. B. The circumventricular organs and central actions of angiotensin. Neuroendocrinology 32:248-256; 1981. 35. Tsuruo, Y.; Hisano, S.; Daikoku, S. Morphological evidence for synaptic junctions between substance P-containing neurons in the arcuate nucleus of the rat. Neurosci. Lett. 46:65-69; 1984. 36. Vale, W.; Vaughan, J.; Smith, M.; Yamamoto, G.; Rivier, J.; Rivier, C. Effects of synthetic ovine corticotropin releasing-factor, glucocorticoids, cathecolamines, neurohypophysial peptides and other substances on cultured corticotropic cells. Endocrinology I 13:1121; 1983. 37. van den Pol, A. The magnocellular and parvocellular paraventricular nucleus of rat: Intrinsic organization. J. Comp. Neurol. 206:317345; 1982. 38. Young, W. S., III Corticotropin-releasing factor mRNA in the hypothalamus is affected differentially by drinking saline and by dehydration. FEBS Lett. 208:158-162; 1986.