Cell Tissue Res. 202, 189-201 (1979)
Cell and Tissue Research 9 by Springer-Verlag 1979
The Localization of Oxytocin, Vasopressin, Somatostatin and Luteinizing Hormone Releasing Hormone in the Rat Neurohypophysis* F.W. van Leeuwen, C. de Raay, D.F. Swaab, and B. Fisser** Netherlands Institute for Brain Research, Amsterdam, The Netherlands
Summary. The hypothalamic hormones arginine-vasopressin (AVP), oxytocin (OXT), somatostatin (SOM), and luteinizing hormone-releasing hormone (LHRH) were localized in the rat neurohypophysis by the use of semithin serial sections and the unlabeled antibody enzyme method. Clusters of AVP fibres are present within the central region of the neural lobe, clusters of OXT fibres mainly in the peripheral part. The AVP fibres enter bilaterally into the neural lobe. The results call into question previous reports on the presence of AVP on receptors in the pars intermedia cells, since incubation with anti-AVP resulted in similar staining in the pars intermedia of the Wistar and homozygous Brattleboro rat, a mutant strain deficient in AVP. The same intermediate lobe cells are stained after incubation of serial sections with anti-AVP and anti-c~-melanocytestimulating hormone (a-MSH). This staining of anti-AVP could be removed by solid phase absorption to a-MSH and is thus most probably due to cross reaction with a-MSH. SOM fibres appear to be present in the peripheral parts of the proximal neurohypophysial stalk and mainly lateral in its more distal parts. In the neural lobe they rapidly decrease in number, although some fibres continue into the distal part of the neural lobe, running bilaterally and situated adjacent to the pars intermedia. The SOM staining within magnocellular elements, which has been reported in the literature, can most probably be explained by cross reaction o f anti-SOM with neurophysins. L H R H fibres are very scarce in the neurohypophysial stalk and absent in the neural lobe.
Key words: Immunocytochemistry - Vasopressin - Oxytocin - Somatostatin Luteining hormone releasing hormone. Send offprint requests to: F.W. van Leeuwen, The Netherlands Institute for Brain Research, IJdijk 28,
1095 KJ Amsterdam, The Netherlands * Supported by the Foundation for Medical Research FUNGO ** The authors wish to thank Drs. J. De Mey (Beerse, Belgittm), A. Arimura (New Orleans, U.S.A.), M.P. Dubois(Nouzilly, France), B.L. Baker (AnnArbor, U.S.A.)andA.G.E. Pearse (London, U.K.) for their gifts ofanti-somatostatin serum, Dr. B. Kerdelhu6(Gif-sur-Yvette,France) for anti-LHRH serum, and Dr. F. Vandesande(Ghent, Belgium)for anti-neurophysin I and II serum and bovineneurophysinI and II. Dr. J.G. Streefkerk (Free University, Amsterdam) is acknowledgedfor critical comments and Mr. A.T. Potjer and Miss J. van der Velden for their skilled assistance
0302-766X/79/0202/0189/$02.60
190
F.W. van Leeuwen et al.
The recognized classical arrangement o f a distinct anatomical separation between the magnocellular oxytocin (OXT) and arginine vasopressin (AVP) producing h y p o t h a l a m i c - n e u r o h y p o p h y s i a l system ( H N S ) and a parvocellular system, terminating in the median eminence and synthesizing, e.g., somatostatin (SOM), thyrotropin-releasing h o r m o n e (TRH), and luteinizing hormone-releasing horm o n e ( L H R H ) (see Szent/tgothai et al., 1968) has recently come up for discussion, mainly as a result o f a n u m b e r o f immunocytochemical observations. F o r example, A V P has been demonstrated also outside the H N S , i.e., in the parvocellular elements o f the suprachiasmatic nucleus within granules that are m u c h smaller ( ~ 9 4 n m ) than those o f the H N S ( ~ 143nm) (see, for example, Swaab et al., 1975; Vandesande et al., 1975; Van Leeuwen et al., 1978) and in the external zone o f the median eminence (see D u b e et al., 1976; Vandesande et al., 1977). In addition, neurohypophysial h o r m o n e s have been determined in the a d e n o h y p o p h y s i s o f the mouse (Castel, 1978), cattle (Renlund, 1978), rat, and pig (Chateau et al., 1979). Furthermore, S O M has been reported to be present not only in the parvocellular h y p o t h a l a m i c elements but also in the magnocellular supraoptic (SON) and paraventricular (PVN) nuclei o f the h u m a n fetus, rat and fox (Dubois and Kolodziejczyk, 1975; B u g n o n et al., 1976, 1977). In addition, S O M fibres occur in the neural lobe o f the dog (Knigge et al., 1978), guinea pig (Weindl and Sofroniew, 1978) and rat (H6kfelt et al., 1975b, 1978) from which S O M is released in vitro (Patel et al., 1977). In addition to their median eminence endings, T R H containing nerve fibres are also observed in the neural lobe o f the rat (H6kfelt et al., 1975 a), in which relatively high concentrations o f this h o r m o n e are also measured by r a d i o i m m u n o a s s a y (Oliver et al., 1974; Leppfiluoto et al., 1978; Rossier et al., 1979); L H R H containing nerve fibres are present in the neural lobe o f the h u m a n fetus (Bugnon et al., 1978), rhesus m o n k e y (Silverman and Zimmerman, 1978), guinea pig, and rat (Barry and Dubois, 1976). The present study is intended to unravel the distribution o f OXT, AVP, SOM, and L H R H in the rat neurohypophysis. In addition, cross reactions o f antibodies raised against peptide h o r m o n e s with c o m p o u n d s not closely related in chemical structure to these h o r m o n e s are described and discussed. Materials and Methods
Ten male Wistar rats and four male Brattleboro rats (homozygous for diabetes insipidus), weighing about 200 g, were obtained from TNO (Zeist, The Netherlands), and provided with tap water and standard chow ad libitum until sacrifice.After decapitation and removal of the lowerjaw, the base of the skull was removed from the ventral side, leaving the hypothalamus, pituitary stalk, and neural lobe intact. When these brain structures became visible, the heads were fixed by immersion in 4 % formalin for 24 h (cf. Swaab et al., 1975). After fixation, the stalk-neuro-intermediate lobe preparation was isolated and embedded in Epon812. Six specimens were oriented longitudinally and four transversely. Serial semithin (2 gm) sections were collected throughout the tissue block, and stored on albuminized glass slides for one night at 37~C, after which the Epon was removed from the sections with sodium ethanolate (Lane and Europa, 1965). Serial sections were incubated with either purified (see below) anti-AVP (~125), anti-OXT (:~02C)plasma and anti-SOM) (De Mey, 4/5/77) in a dilution of 1:200. Anti-LHRH and anti-~-melanocyte-stimulating hormone (anti-~-MSH, ~4394-23/4) were diluted respectively 1 : 400 and 1:1000. The unlabeled antibody enzyme method with peroxidase-antiperoxidase (PAP) was followed (for details see Van Leeuwen et al., 1978). The antibodies against AVP and OXT were purified by preincubation with OXT or AVP beads respectively (Swaab and Pool, 1975), after which the
Neurohypophysial Peptides in the Rat
191
specificityof anti-AVP was tested on the neural lobe of a homozygousBrattleboro rat (see also Fig. 5d) and that of OXT on the suprachiasmatic nucleus of a Wistar rat. Anti-SOM serum was purified with beads containing bovine neurophysin I and II. Specificityof all antibodies was tested by solid phase absorption to the homologousantigen. Cross reactivityof anti-SOM sera with bovine neurophysin I and II and of anti-AVP (:~125and ~126) with c~-MSHwas established with an indirect immunofluorescence procedure (Swaab and Pool, 1975). The peptides used for this technique were obtained from Sigma, Saint Louis, MO, U.S.A. (AVPGrade VI (100 I.U.) and OXT-Grade V (1,000 I.U.)), Bachem,Marina Del Rey, CA, U.S.A. (SOM) and UCB, Brussels, Belgium(ct-MSHand LHRH). Differencesbetween the various antisera and the control serum were tested by Student's t-test, a difference of P < 0.05 being considered as significant.
Results Suprainfundibular AVP and OXT fibres run intermingled from lateral directions towards the infundibular recess, joining the supraoptico-paraventricular neurohypophysial pathway only in the lateral part of the pituitary stalk (Figs. 1, 4a, b, 6b, c). In the suprainfundibular pathway more OXT than AVP fibres are observed (Fig. 4a, b). In the median eminence and the proximal part of the pituitary stalk, the staining for AVP appears to be much more pronounced than for OXT. This staining difference disappears in the most distal part of the pituitary stalk and in the neural lobe (Figs. 1, 6b, c). In the neural lobe, AVP fibres are present mainly in clusters in the central region (Fig. 1 a). In the rostral part, a left and right bundle of AVP fibres can be distinguished. OXT fibres are situated around these bundles in a manner resembling a spectacle frame (Fig. 2). More caudally this arrangement is more and more lost, although generally OXT fibres remain localized more peripherally (Fig. I b). After incubation with anti-AVP, no innervation of the pars intermedia by OXT and AVP fibres is observed, but moderate staining occurs in the pars intermedia cells of the Wistar (Fig. 5a) and homozygous Brattleboro rat (Fig. 5d). This reaction disappears after solid phase absorption of anti-AVP with ~-MSH (Fig. 5 b, e). In serial sections through the pars intermedia, the same cells appear to stain both with anti-AVP and with anti-~-MSH (Fig. 5 c, f). The indirect immunofluorescence technique confirms the cross reaction of anti-AVP with ct-MSH. Both anti-AVP's (~125 and ~126) show significant binding to beads covered with ~-MSH (respectively 11 and 27 % of the fluorescence found with anti-~-MSH (:~4394) on ~M S H beads). This staining o f anti-AVP can be eliminated by solid phase preabsorption to ~-MSH containing beads. After incubation with unpurified anti-SOM serum, the entire neural lobe shows a slight staining of fibres and Herring bodies. With the use of the indirect immunofluorescence technique a clearcut reaction of some anti-SOM's was demonstrated with bovine neurophysin I and II (Table 1). This cross reaction can be eliminated by solid phase absorption of anti-somatostatin to neurophysin I and II. Such purified antibodies were used for further studies. Adjacent to the external zone of the median eminence, numerous SOM fibres were found to protrude into the peripheral parts of the neurohypophysial stalk (Fig. 4c), joining those SOM fibres that run above the infundibular recess in a manner similar to the OXT and AVP fibres (Fig. 6, insets). The SOM fibres running in the external zone of the median eminence course in the neurohypophysial stalk mainly ventral to the AVP and OXT
Fig. 1 a and b. Serial sagittal semithin Epon sections o f rat neurohypophysis, immunocytochemically stained with purified anti-AVP (a) and anti-OXT (b). a Note cluster of AVP fibres in center of neural lobe, and separate course of suprainfundibular fibres in neurohypophysial stalk, b Note peripheral OXT fibres, especially in rostral part of neural lobe; ri recessus infundibularis; pi pars interrnedia, x 45 Fig. 2. Transverse semithin section of rat neurointermediate lobe incubated with purified anti-OXT, showing bilateral negative areas of innervation by AVP fibres and peripheral position of OXT fibres; pi pars intermedia. • 125 Fig. 3. Semithin sagittal Epon section of neurohypophysis (lateral to infundibular recess) incubated with anti-LHRH. Note location of positive fibres around capillaries in region of tubero-infundibular sulcus, and scarcity of fibres running toward neural lobe; ps pituitary stalk; pt pars tuberalis. • 190
Neurohypophysial Peptides in the Rat
193
Fig. 4. Serial transverse semithin Epon sections of most distal part of neurohypophysial stalk incubated serially with purified anti-AVP a, anti-OXT b, anti-SOM c. Very proximal part of neural lobe d after incubation with anti-SOM, Note in (d) lateral position of majority of fibres, pi Rostral part of pars intermedia, x 250; (d): • 210
194
F.W. van Leeuwen et al.
Fig. 5. Serial sections of neurointermediate lobe of Wistar (a, c) and homozygous Brattleboro rat (d, f) after incubation with purified anti-AVP (a, d), anti-AVP absorbed 4 times to ct-MSH (b, e) and anti-ctMSH (c, f). The same cells stained with anti-AVP and anti-~-MSH (arrows); n! neural lobe; pi pars intermedia. • 250 fibres, while in the extreme, distal part of the stalk a n d in the neural lobe, the S O M fibres are confined m a i n l y to lateral areas (Fig. 4d). As soon as they r u n into the n e u r a l lobe their n u m b e r diminishes rapidly, a l t h o u g h some fibres are still present in the distal part. These S O M fibres are m a i n l y located adjacent to the pars i n t e r m e d i a in the n e i g h b o u r h o o d of the short portal vessel system. Other SOM
1 : 80 1 : 80
19578 (6-11-'75)
19578 (24-7-'76)
B 173 (1-10:75)
4 3 1 R b (K 50)
Dubois I I I
Dubois IV
Baker
Pearse 1 : 80 1 : 80 1:80
Anti-neurophysin I
Anti-neurophysin II
Control
1 : 80
1:80
1 : 80
1:80
19608 (24-7-'76)
19608 (6-11-'75)
Dubois It
1 : 80
1 : 80
Dubois I
108 (5-9-'75)
101 (6-6-'74)
Arimura II
-
Arimura I
1 : 80
4-5-'77
De Mey
De Mey purified over N I + N 2
1:40
Dilution
Code
Source
59.1+
58.4+ 8.7
5.2
79.5+ 15.4
147.4+- 23.8
130.8+ 15.0
368.4+ 51.0
74.5+- 7.3
228.5+ 14.8
116.5+ 10.9
156.9+- 29.0
70.0+- 6.6
290.6+- 23.1
312.7+- 22.6
SOM
4.1
2.9
1.4
9.2
5.1
6.7
41.8+- 1.8
n.d.
201.0+ 17.2
61.8+
46.6+- 2.9
50.8+
39.9+- 4.8
41.8+- 4.7
44.8+
47.5+
25.7+
45.3+
85.2+ 10.6
N e u r o p h y s i n I (N1)
-
(12.6)
-
-
-
-
-
-
-
-
(27.3)
(~o)
35.3+5.5
100.8+ 7.5
n.d.
66.3+- 6.5
39.6+ 3.5
60.6+ 4.9
53.7+- 5.5
42.1 + 3.8
34.3+ 2.1
38.9+ 3.0
42.6+- 4.2
30.7+- 4.1
103.5+ 8.6
N e u r o p h y s i n II (N2)
-
(47.4)
-
(38.7)
(28.2)
-
-
-
-
(104.1)
(~)
Immunofluorescence, in arbitrary units, of antisera to somatostatin. W h e n statistically significant, the values are also expressed in terms of a percentage of the value found on neurophysin beads after incubation with anti-neurophysin. F o r the latter calculations the b a c k g r o u n d fluorescence obtained with a p r e - i m m u n e control plasma was substracted from that obtained with the antibodies (for details see Swaab and Pool, 1975). n.d. = not determined; - = n o t significantly larger (P > 0.05) than the background reading
Table 1. Reaction of anti-somatostatin with neurophysin
~-
~"
Fig. 6. Serial sagittal semithin Epon sections of lateral part of neurohypophysial stalk incubated with purified anti-SOM (a), anti-AVP (b), anti-OXT (c). a Note mainly ventral course of SOM fibres from median eminence to neural lobe. h and c Note joining of supraoptico-neurohypophysial and suprainfundibular AVP and O X T tracts, and different localization of AVP and OXT fibres within rostral part of neural lobe. Insets: course o f fibres above recessus infundibularis; pi pars intermedia, x 200; insets: x 65
NeurohypophysialPeptides in the Rat
197
fibres run into the stalk at the dorsal site, slightly dorsal to the suprainfundibular AVP and OXT fibres (Fig. 4c), ending mainly in the rostro-dorsal part of the neural lobe (Fig. 6a). After incubation with anti-LHRH a clearcut reaction was present in the lateral parts of the median eminence, as well as above the infundibular recess. A few fibres occur in the neurohypophysial stalk, but none are observed in the neural lobe (Fig. 3). All control incubations showed absence of staining.
Discussion
The present study shows that the use of semithin Epon sections not only provides a link between light and electron microscopic immunocytochemistry (Watari et al., 1977; Van Leeuwen et al., 1978), but also gives new information in itself. Comparison between neurohypophyses embedded either in Epon or in paraffin, shows that the use of semithin (2 gm) Epon sections reveals morphological details better, due to less superimposition of the structures, than in 6 gm paraffin sections, but also produces an increase in the number of fibres stained in the median eminence and pituitary stalk. In addition, the reaction in the pars intermedia after incubation with anti-AVP was detected only in Epon sections, presumably due to the higher sensitivity of this procedure. The specificity tests used in this study allow us to discriminate between AVP and OXT. The differential localization of AVP, OXT, SOM and LHRH in the neurohypophysis provides additional evidence for the specificity of the reaction. A major immunocytochemical problem is the appearance of staining in structures that are not known to contain chemically related compounds. An example is the staining of Herring bodies in the neural lobe with anti-SOM, and of the pars intermedia cells with anti-AVP. This reaction product disappeared after solid phase absorption to the respective homologous antigen, which shows that the staining was based on the presence of the antigen itself or related compounds. As we suspected cross reaction of anti-SOM with neurophysin I and II and of anti-AVP with a-MSH, the staining capacity of nine different SOM and two AVP antisera was tested on bovine neurophysin I and II and a-MSH respectively. Since a quantification procedure for the PAP method on beads is not yet available, this aspect had to be tested by means of the indirect immunofluorescence (IF) procedure (Swaab and Pool, 1975). A disadvantage of this method is, that the sensitivity of the test system is different from that of the PAP procedure. For example, anti-AVP absorbed to a-MSH produced good staining in the neural lobe (Fig. 5b), while on AVP containing beads the IF values were not different from the control values. Some of the SOM antisera indeed revealed a considerable reaction, especially with neurophysin II. In addition, anti-SOM (De Mey) apparently also showed cross reaction with AVP (26 ~ of the fluorescence found with anti-AVP (g125) on AVP beads). The cross reaction with both neurophysin I and II and AVP could be eliminated by solid phase absorption to neurophysin I and II. Following this procedure only a few SOM positive fibres in the neural lobe were stained, while SOM staining on beads remained present. Cross reaction thus seems to be the explanation for the observations that SOM is present in the magnocellular SON
198
F.W. van Leeuwenet al.
and PVN of the rat, human fetus, and fox (Dubois and Kolodziejczyk, 1975; Bugnon et al., 1976, 1977, results that were obtained with antibodies not available in the present study). This conclusion was confirmed by Dierickx and Vandesande (personal communication). The reported staining of intermediate lobe cells by anti-AVP (Castel, 1978) provides another example of cross reaction, since a similar result was obtained in the vasopressin deficient homozygous Brattleboro rat. Thus, it is unlikely that this staining is due to the presence of AVP on its receptors, as claimed by Castel (1978). A theoretical possibility is that AVP was present in antisera either as free AVP or as AVP-anti-AVP complexes that could dissociate, bind to AVP receptors in Brattleboro pars intermedia cells and thus induce staining. However, since the intensity of staining in pars intermedia cells of Brattleboro and Wistar rats was equal, and no obvious increase in staining was found after preincubations of Brattleboro neurointermediate lobe sections with AVP (10-3-10p, g/ml), this possibility can be excluded. The positive reaction in the same intermediate lobe cells after incubation of serial sections with anti-AVP and e-MSH, and the elimination of the staining by solid phase absorption of anti-AVP to c~-MSH, makes it very probable that the reactive substance is e-MSH itself, rather than AVP. Another sign of the higher sensitivity of the PAP technique as compared to the IF procedure is that the amount of solid phase bound peptides for the demonstration of either method or serum specificity which is routinely used in the IF procedure (1 mg peptide beads/l gl antiserum; for details see Swaab and Pool, 1975), had to be increased in some cases. For instance, for the neutralization ofantiL H R H serum, and the purification of anti-AVP with c~-MSH beads, no disappearance of staining was observed in the median eminence and intermediate lobe cells respectively. However, when 5mg beads/lal antiserum were used inhibition of the staining was accomplished. The enhancement of the amount of beads did not result in a nonspecific abolition of the staining since anti-LHRH serum processed in the same way together with thyroglobulin beads still showed a bright immunopositive reaction in the median eminence while anti-AVP gave a reaction in the neural lobe after absorption with c~-MSH beads (see also Fig. 5b). The same results were reported by Sternberger et al. (1978) who used a high amount of L H R H beads (33 mg/lal L H R H antiserum). The data on the preferential localization of OXT in the rostral dorsal part and periphery of the neural lobe confirm those of Vandesande and Dierickx (1975). In addition, AVP fibres enter bilaterally into the neural lobe, and clusters of AVP and OXT fibres were found. The former result is in agreement with the report of Bock and Jurna (1977), who showed an ipsilateral reduction of "Gomori-positive" material in the rat neural lobe after unilateral lesioning of the hypothalamoneurohypophysial tract at different levels. The presence of AVP and OXT clusters within the neural lobe confirms the previously reported immuno-electron microscopical data, in which it was postulated that these clusters may occupy considerable areas (Van Leeuwen and Swaab, 1977). In addition, the central position of AVP might explain the finding by Krisch (1977) that AVP fibres are observed at the immuno-electron microscopical level in greater numbers than OXT fibres, since the rim of OXT in the periphery of the neural lobe may easily be trimmed away when ultrathin sections are prepared.
NeurohypophysialPeptides in the Rat
199
In a number of vertebrates, synaptoid endings of peptidergic nerve fibres have been demonstrated on intermediate lobe cells (see Bargmann et al., 1967; Bargmann, 1969; Von Lawzewitsch and Monastirsky, 1974). Such structures were not, however, observed in any significant numbers in the rat intermediate lobe (Baumgarten, 1972). In the present study no such innervation was observed. In the rat, the localization of OXT fibres in the area directly adjacent to the intermediate lobe might have functional significance, since OXT is thought to be split enzymatically into peptides with melanocyte stimulating hormone-releasing and -inhibiting factor activities (Cells, 1977; Vivas and Celis, 1977). The importance of these peptides for pars intermedia function is, however, still in dispute (see Tilders and Smelik, 1977). Most of the SOM fibres entering the neurohypophysial stalk were found to terminate in the rostral part of the neural lobe, the remainder being observed in the neighbourhood of the short portal vessel system. In agreement with the report of H6kfelt et al. (1978), we could not confirm the results of Barry and Dubois (1976), who reported L H R H fibres in the proximal part of the rat neural lobe. Neurohypophysial T R H fibres, which will be described by us in a subsequent communication, were also found in the neural lobe (H6kfelt et al., 1975b). AVP and OXT fibres, containing predominantly small granules (80-100 nm), but sometimes also larger granules (150-200 nm), were reported in the external zone of the median eminence (see Dierickx et al., 1976; Dube et al., 1976). These facts indicate that the neurohypophysis (classically divided into infundibulum and neural lobe, Rioch et al., 1940) can be considered as a continuous neurohaemal zone, which fits in well with the common neurohypophysial capillary bed (Page and Bergland, 1977).
References Bargmann, W.: Das neurosekretorische Zwischenhirn-Hypophysen-Systemund seine synaptischen Verkniipfungen. J. Neuro-Visc. Rel. Suppl. 9, 64-77 (1969) Bargmann, W., Lindner, E., Andres, K.H.: Ober Synapsen an endokrinen Epithelzellen und die Definition sekretorischer Neuronen. Untersuchungenam Zwischenlappender Katzenhypophyse.Z. Zellforsch. 77, 282-298 (1967) Barry, L., Dubois, M.P.: Immunoreactive LRF neurosecretory pathways in mammals. Acta Anat. (Basel) 94, 497-503 (1976) Baumgarten, H.G., Bj6rklund, A., Holstein, A.F., Nobin, A.: Organization and ultrastructural identification of the catecholaminenerve terminals in the neural lobe and pars intermedia of the rat pituitary. Z. Zellforsch. 126, 483-517 (1972) Bock, R., Jurna, I.: Ipsilateral diminution of CRF-granules after unilateral hypothalamic lesions. Cell Tissue Res. 185, 215-229 (1977) Bugnon, C., Lenys, D., Fellmann, D., Bloch, B.: Etude cyto-immunologiquedu systbmepeptidergique hypothalamique fi somatostatinechezle Renard (Vulpes vulpes). C.R. Soc. Biol. 170, 584-588 (1976) Bugnon, C., Fellmann, D., Bloch, B.: Immunocytochemical study of the ontogenesis of the hypothalamic somatostatin-containing neurons in the human fetus. Cell Tissue Res. 183, 319-328 (1977) Bugnon, C., Bloch, B., Lenys, D., Fellmann, D. : Cytoimmunologicalstudy of the LH-RH neurons in humans during fetal life. In: Brain-EndocrineInteraction III. Neural Hormones and Reproduction (D.E. Scott, G.P. Kozlowski,A. Weindl, eds.), pp. 183-196. Basel: Karger 1978 Castel, M.: Immunocytochemicalevidencefor vasopressin receptors.J. Histochem. Cytochem.26, 581592 (1978) Celis, M.E.: Hypothalamicpeptides involvedin the control of MSH secretion:identity, biosynthesisand regulation of their release. Front. Horm. Res. 4, 69-79 (1977)
200
F.W. van Leeuwen et al.
Chateau, M., Marchetti, J., Burlet, A., Boulang6, M.: Evidence of vasopressin in adenohypophysis: research into its role in corticotrope activity. Neuroendocrinology 28, 25-35 (1979) Dierickx, K., Vandesande, F., de Mey, J.: Identification, in the external region of the rat median eminence, of separate neurophysin-vasopressin and neurophysin-oxytocin containing nerve fibres. Cell Tissue Res. 168, 141-151 (1976) Dube, D., Leclerc, R., Pelletier, G.: Electron microscopic immunohistochemical localization of vasopressin and neurophysin in the median eminence of normal and adrenalectomized rats. Am. J. anat. 147, 103-108 (1976) Dubois, M.P., Kolodziejczyk, E.: Centres hypothalamiques du rat s6cr&ant la somatostatine: repartition des p6ricaryons en 2 syst6mes magno et parvocellulaires (&ude immunocytologique). C.R. Acad. Sci. Paris 281, 1737-1740 (1975) H6kfelt, T., Fuxe, K., Johansson, O., Jeffcoate, S., White, N.: Distribution of thyrotropin-releasing hormone (TRH) in the central nervous system as revealed with immunohistochemistry. Eur. J. Pharmacol. 34, 389-392 (1975a) H6kfelt, T., Efendic, S., Hellerstr6m, C., Johansson, O., Luft, R., Arimura, A.: Cellular localization of somatostatin in endocrinelike cells and neurons of the rat with special reference to the Al-cells of the pancreatic islets and to the hypothalamus. Acta Endocrinol. [Suppl.] (Kbh.) 200, 5-41 (1975b) H6kfelt, T., Elde, R., Fuxe, K., Johansson, O., Ljungdahl, A., Goldstein, M., Luft, R., Efendic, S., Nilsson, G., Terenius, L., Ganten, D., Jeffcoate, S., Rehfeld, J., Said, S., Perez de la Mora, M., Possani, L., Tapia, R., Teran, L., Palacios, R.: Aminergic and peptidergic pathways in the nervous system with special reference to the hypothalamus. In: The hypothalamus (S. Reichlin, R.J. Baldessarini, J.B. Martin, eds.), pp. 69-135. New York: Raven Press 1978 Knigge, K.M., Joseph, S.A., Hoffman, G.E.: Organization of LRF- and SRIF-neurons in the endocrine hypothalamus. In: The Hypothalamus (S. Reichlin, R.J. Baldessarini, J.B. Martin, eds.), pp. 49-67. New York: Raven Press 1978 Krisch, B.: Electronmicroscopic immunocytochemical study on the vasopressin-containing neurons in the thirsting rat. Cell Tissue Res. 184, 237-247 (1977) Lane, B.P., Europa, D.L.: Differential staining of ultrathin sections of epon-embedded tissues for light microscopy. J. Histochem. Cytochem. 13, 579-582 (1965) Leppfiluoto, J., Koivusalo, F., Kraama, R.: Thyrotropin-releasing factor: distribution in neural and gastrointestinal tissues. Acta Physiol. Scand. 104, 175-179 (1978) Oliver, C., Eskay, R.L., Ben-Jonathan, N., Porter, J.C.: Distribution and concentration of TRH in the rat brain. Endocrinology 95, 540-546 (1974) Page, R.B., Bergland, R.M.: The neurohypophyseal capillary bed. I. Anatomy and arterial supply. Am. J. Anat. 148, 345-357 (1977) Patel, Y.C., Zingg, H.H., Dreifuss, J.J.: Calcium-dependent somatostatin secretion from rat neurohypophysis in vitro. Nature 267, 852-853 (1977) Renlund, S.: Identification of oxytocin and vasopressin in the bovine adenohypophysis. Thesis, Uppsala, Sweden (1978) Rioch, D.McK., Wislocki, G.B., O'Leary, J.L.: A pr6cis of preoptic, hypothalamic and hypophysial terminology with atlas. In: The Hypothalamus, Association for Research in Nervous and Mental Disease, Vol. 20 (J.F. Fulton, S.W. Ranson, A.M. Frantz, eds.), pp. 3-30. Baltimore: The Williams and Wilkins Company 1940 Rossier, J., Battenberg, E., Pittmann, Q., Bayon, A., Koda, L., Miller, R., Guillemin, R., Bloom, F.: Hypothalamic enkephalin neurones may regulate the neurohypophysis. Nature 277, 653-655 (1979) Silverman, A.J., Zimmerman, E.A.: Pathways containing luteinizing hormone-releasing hormone (LHRH) in the mammalian brain. In: Brain-Endocrine Interaction III. Neural Hormones and Reproduction (D.E. Scott, G.P. Kozlowski, A. Weindl, eds.), pp. 83-96. Basel: Karger 1978 Sternberger, L.A., Petrali, J.P., Joseph, S.A., Meyer, H.G., Mills, K.R.: Specificity of the immunocytochemical luteinizing hormone-releasing hormone receptor reaction. Endocrinology 102, 63-73 (1978) Swaab, D.F., Pool, C.W.: Specificity of oxytocin and vasopressin immunofluorescence. J. Endocrinol. 66, 263-272 (1975) Swaab, D.F., Pool, C.W., Nijveldt, F.: Immunofluorescence of vasopressin and oxytocin in the rat hypothalamo-neurohypophyseal system. J. Neural Transm. 36, 195-215 (1975) Szent~tgothai, J., Flerk6, B., Mess, B., Halfisz, B.: Hypothalamic Control of the Anterior Pituitary. Budapest: Akad6miai Kiad6 (1968)
Neurohypophysial Peptides in the Rat
201
Tilders, FJ.H., Smelik, P.G.: Direct control of MSH secretion in mammals: The involvement of dopaminergic tubero-hypophysial hormone. Front. Horm. Res. 4, 80-93 (1977) Vandesande, F., Dierickx, K.: Identification of the vasopressin producing and of the oxytocin producing neurons in the hypothalamic magnocellular neurosecretory system of the rat. Cell Tissue. Res. 164, 153-162 (1975) Vandesande, F., Dierickx, K., De Mey, J.: The origin of the vasopressinergic and oxytocinergic fibres of the external region of the median eminence of the rat hypophysis. Cell Tissue. Res. 180, 443-452 (1977) Van Leeuwen, F.W., Swaab, D.F.: Specific immunoelectronmicroscopic localization of vasopressin and oxytocin in the neurohypophysis of the rat. Cell Tissue Res. 177, 493-501 (1977) Van Leeuwen, F.W., Swaab, D.F., De Raay, C.: Immuno-electron microscopic localization of vasopressin in the rat suprachiasmatic nucleus. Cell Tissue Res. 193, 1-10, (1978) Vivas, A., Cells, M.E.: New evidence that demonstrates that L-Pro-L-Leu-L-Gly-NHz might be the natural MIF. Acta Physiol. Lat. Am. 27, 177-182 (1977) Von Lawzewitsch, I., Monastirski, R.: Neuro-glandular junctions in the pars intermedia in the rabbit. Acta Anat. (Basel)87, 409-413 (1974) Watari, N., Hotta, Y., Mabuchi, Y.: Comparative light and electron-microscopic study of the adrenal gland of the snake. Arch. Histol. Jpn. 40, Suppl., 81-85 (1977) Weindl, A., Sofroniew, M.V.: Neurohormones and circumventricular organs. In: Brain-Endocrine Interaction III. Neural Hormones and Reproduction (D.E. Scott, G.P. Kozlowski, A. Weindl, eds.), pp. 117-137. Basel: Karger 1978
Accepted July 5, 1979
Cell and Tissue Research 9 by Springer-Verlag 1979
The Localization of Oxytocin, Vasopressin, Somatostatin and Luteinizing Hormone Releasing Hormone in the Rat Neurohypophysis* F.W. van Leeuwen, C. de Raay, D.F. Swaab, and B. Fisser** Netherlands Institute for Brain Research, Amsterdam, The Netherlands
Summary. The hypothalamic hormones arginine-vasopressin (AVP), oxytocin (OXT), somatostatin (SOM), and luteinizing hormone-releasing hormone (LHRH) were localized in the rat neurohypophysis by the use of semithin serial sections and the unlabeled antibody enzyme method. Clusters of AVP fibres are present within the central region of the neural lobe, clusters of OXT fibres mainly in the peripheral part. The AVP fibres enter bilaterally into the neural lobe. The results call into question previous reports on the presence of AVP on receptors in the pars intermedia cells, since incubation with anti-AVP resulted in similar staining in the pars intermedia of the Wistar and homozygous Brattleboro rat, a mutant strain deficient in AVP. The same intermediate lobe cells are stained after incubation of serial sections with anti-AVP and anti-c~-melanocytestimulating hormone (a-MSH). This staining of anti-AVP could be removed by solid phase absorption to a-MSH and is thus most probably due to cross reaction with a-MSH. SOM fibres appear to be present in the peripheral parts of the proximal neurohypophysial stalk and mainly lateral in its more distal parts. In the neural lobe they rapidly decrease in number, although some fibres continue into the distal part of the neural lobe, running bilaterally and situated adjacent to the pars intermedia. The SOM staining within magnocellular elements, which has been reported in the literature, can most probably be explained by cross reaction o f anti-SOM with neurophysins. L H R H fibres are very scarce in the neurohypophysial stalk and absent in the neural lobe.
Key words: Immunocytochemistry - Vasopressin - Oxytocin - Somatostatin Luteining hormone releasing hormone. Send offprint requests to: F.W. van Leeuwen, The Netherlands Institute for Brain Research, IJdijk 28,
1095 KJ Amsterdam, The Netherlands * Supported by the Foundation for Medical Research FUNGO ** The authors wish to thank Drs. J. De Mey (Beerse, Belgittm), A. Arimura (New Orleans, U.S.A.), M.P. Dubois(Nouzilly, France), B.L. Baker (AnnArbor, U.S.A.)andA.G.E. Pearse (London, U.K.) for their gifts ofanti-somatostatin serum, Dr. B. Kerdelhu6(Gif-sur-Yvette,France) for anti-LHRH serum, and Dr. F. Vandesande(Ghent, Belgium)for anti-neurophysin I and II serum and bovineneurophysinI and II. Dr. J.G. Streefkerk (Free University, Amsterdam) is acknowledgedfor critical comments and Mr. A.T. Potjer and Miss J. van der Velden for their skilled assistance
0302-766X/79/0202/0189/$02.60
190
F.W. van Leeuwen et al.
The recognized classical arrangement o f a distinct anatomical separation between the magnocellular oxytocin (OXT) and arginine vasopressin (AVP) producing h y p o t h a l a m i c - n e u r o h y p o p h y s i a l system ( H N S ) and a parvocellular system, terminating in the median eminence and synthesizing, e.g., somatostatin (SOM), thyrotropin-releasing h o r m o n e (TRH), and luteinizing hormone-releasing horm o n e ( L H R H ) (see Szent/tgothai et al., 1968) has recently come up for discussion, mainly as a result o f a n u m b e r o f immunocytochemical observations. F o r example, A V P has been demonstrated also outside the H N S , i.e., in the parvocellular elements o f the suprachiasmatic nucleus within granules that are m u c h smaller ( ~ 9 4 n m ) than those o f the H N S ( ~ 143nm) (see, for example, Swaab et al., 1975; Vandesande et al., 1975; Van Leeuwen et al., 1978) and in the external zone o f the median eminence (see D u b e et al., 1976; Vandesande et al., 1977). In addition, neurohypophysial h o r m o n e s have been determined in the a d e n o h y p o p h y s i s o f the mouse (Castel, 1978), cattle (Renlund, 1978), rat, and pig (Chateau et al., 1979). Furthermore, S O M has been reported to be present not only in the parvocellular h y p o t h a l a m i c elements but also in the magnocellular supraoptic (SON) and paraventricular (PVN) nuclei o f the h u m a n fetus, rat and fox (Dubois and Kolodziejczyk, 1975; B u g n o n et al., 1976, 1977). In addition, S O M fibres occur in the neural lobe o f the dog (Knigge et al., 1978), guinea pig (Weindl and Sofroniew, 1978) and rat (H6kfelt et al., 1975b, 1978) from which S O M is released in vitro (Patel et al., 1977). In addition to their median eminence endings, T R H containing nerve fibres are also observed in the neural lobe o f the rat (H6kfelt et al., 1975 a), in which relatively high concentrations o f this h o r m o n e are also measured by r a d i o i m m u n o a s s a y (Oliver et al., 1974; Leppfiluoto et al., 1978; Rossier et al., 1979); L H R H containing nerve fibres are present in the neural lobe o f the h u m a n fetus (Bugnon et al., 1978), rhesus m o n k e y (Silverman and Zimmerman, 1978), guinea pig, and rat (Barry and Dubois, 1976). The present study is intended to unravel the distribution o f OXT, AVP, SOM, and L H R H in the rat neurohypophysis. In addition, cross reactions o f antibodies raised against peptide h o r m o n e s with c o m p o u n d s not closely related in chemical structure to these h o r m o n e s are described and discussed. Materials and Methods
Ten male Wistar rats and four male Brattleboro rats (homozygous for diabetes insipidus), weighing about 200 g, were obtained from TNO (Zeist, The Netherlands), and provided with tap water and standard chow ad libitum until sacrifice.After decapitation and removal of the lowerjaw, the base of the skull was removed from the ventral side, leaving the hypothalamus, pituitary stalk, and neural lobe intact. When these brain structures became visible, the heads were fixed by immersion in 4 % formalin for 24 h (cf. Swaab et al., 1975). After fixation, the stalk-neuro-intermediate lobe preparation was isolated and embedded in Epon812. Six specimens were oriented longitudinally and four transversely. Serial semithin (2 gm) sections were collected throughout the tissue block, and stored on albuminized glass slides for one night at 37~C, after which the Epon was removed from the sections with sodium ethanolate (Lane and Europa, 1965). Serial sections were incubated with either purified (see below) anti-AVP (~125), anti-OXT (:~02C)plasma and anti-SOM) (De Mey, 4/5/77) in a dilution of 1:200. Anti-LHRH and anti-~-melanocyte-stimulating hormone (anti-~-MSH, ~4394-23/4) were diluted respectively 1 : 400 and 1:1000. The unlabeled antibody enzyme method with peroxidase-antiperoxidase (PAP) was followed (for details see Van Leeuwen et al., 1978). The antibodies against AVP and OXT were purified by preincubation with OXT or AVP beads respectively (Swaab and Pool, 1975), after which the
Neurohypophysial Peptides in the Rat
191
specificityof anti-AVP was tested on the neural lobe of a homozygousBrattleboro rat (see also Fig. 5d) and that of OXT on the suprachiasmatic nucleus of a Wistar rat. Anti-SOM serum was purified with beads containing bovine neurophysin I and II. Specificityof all antibodies was tested by solid phase absorption to the homologousantigen. Cross reactivityof anti-SOM sera with bovine neurophysin I and II and of anti-AVP (:~125and ~126) with c~-MSHwas established with an indirect immunofluorescence procedure (Swaab and Pool, 1975). The peptides used for this technique were obtained from Sigma, Saint Louis, MO, U.S.A. (AVPGrade VI (100 I.U.) and OXT-Grade V (1,000 I.U.)), Bachem,Marina Del Rey, CA, U.S.A. (SOM) and UCB, Brussels, Belgium(ct-MSHand LHRH). Differencesbetween the various antisera and the control serum were tested by Student's t-test, a difference of P < 0.05 being considered as significant.
Results Suprainfundibular AVP and OXT fibres run intermingled from lateral directions towards the infundibular recess, joining the supraoptico-paraventricular neurohypophysial pathway only in the lateral part of the pituitary stalk (Figs. 1, 4a, b, 6b, c). In the suprainfundibular pathway more OXT than AVP fibres are observed (Fig. 4a, b). In the median eminence and the proximal part of the pituitary stalk, the staining for AVP appears to be much more pronounced than for OXT. This staining difference disappears in the most distal part of the pituitary stalk and in the neural lobe (Figs. 1, 6b, c). In the neural lobe, AVP fibres are present mainly in clusters in the central region (Fig. 1 a). In the rostral part, a left and right bundle of AVP fibres can be distinguished. OXT fibres are situated around these bundles in a manner resembling a spectacle frame (Fig. 2). More caudally this arrangement is more and more lost, although generally OXT fibres remain localized more peripherally (Fig. I b). After incubation with anti-AVP, no innervation of the pars intermedia by OXT and AVP fibres is observed, but moderate staining occurs in the pars intermedia cells of the Wistar (Fig. 5a) and homozygous Brattleboro rat (Fig. 5d). This reaction disappears after solid phase absorption of anti-AVP with ~-MSH (Fig. 5 b, e). In serial sections through the pars intermedia, the same cells appear to stain both with anti-AVP and with anti-~-MSH (Fig. 5 c, f). The indirect immunofluorescence technique confirms the cross reaction of anti-AVP with ct-MSH. Both anti-AVP's (~125 and ~126) show significant binding to beads covered with ~-MSH (respectively 11 and 27 % of the fluorescence found with anti-~-MSH (:~4394) on ~M S H beads). This staining o f anti-AVP can be eliminated by solid phase preabsorption to ~-MSH containing beads. After incubation with unpurified anti-SOM serum, the entire neural lobe shows a slight staining of fibres and Herring bodies. With the use of the indirect immunofluorescence technique a clearcut reaction of some anti-SOM's was demonstrated with bovine neurophysin I and II (Table 1). This cross reaction can be eliminated by solid phase absorption of anti-somatostatin to neurophysin I and II. Such purified antibodies were used for further studies. Adjacent to the external zone of the median eminence, numerous SOM fibres were found to protrude into the peripheral parts of the neurohypophysial stalk (Fig. 4c), joining those SOM fibres that run above the infundibular recess in a manner similar to the OXT and AVP fibres (Fig. 6, insets). The SOM fibres running in the external zone of the median eminence course in the neurohypophysial stalk mainly ventral to the AVP and OXT
Fig. 1 a and b. Serial sagittal semithin Epon sections o f rat neurohypophysis, immunocytochemically stained with purified anti-AVP (a) and anti-OXT (b). a Note cluster of AVP fibres in center of neural lobe, and separate course of suprainfundibular fibres in neurohypophysial stalk, b Note peripheral OXT fibres, especially in rostral part of neural lobe; ri recessus infundibularis; pi pars interrnedia, x 45 Fig. 2. Transverse semithin section of rat neurointermediate lobe incubated with purified anti-OXT, showing bilateral negative areas of innervation by AVP fibres and peripheral position of OXT fibres; pi pars intermedia. • 125 Fig. 3. Semithin sagittal Epon section of neurohypophysis (lateral to infundibular recess) incubated with anti-LHRH. Note location of positive fibres around capillaries in region of tubero-infundibular sulcus, and scarcity of fibres running toward neural lobe; ps pituitary stalk; pt pars tuberalis. • 190
Neurohypophysial Peptides in the Rat
193
Fig. 4. Serial transverse semithin Epon sections of most distal part of neurohypophysial stalk incubated serially with purified anti-AVP a, anti-OXT b, anti-SOM c. Very proximal part of neural lobe d after incubation with anti-SOM, Note in (d) lateral position of majority of fibres, pi Rostral part of pars intermedia, x 250; (d): • 210
194
F.W. van Leeuwen et al.
Fig. 5. Serial sections of neurointermediate lobe of Wistar (a, c) and homozygous Brattleboro rat (d, f) after incubation with purified anti-AVP (a, d), anti-AVP absorbed 4 times to ct-MSH (b, e) and anti-ctMSH (c, f). The same cells stained with anti-AVP and anti-~-MSH (arrows); n! neural lobe; pi pars intermedia. • 250 fibres, while in the extreme, distal part of the stalk a n d in the neural lobe, the S O M fibres are confined m a i n l y to lateral areas (Fig. 4d). As soon as they r u n into the n e u r a l lobe their n u m b e r diminishes rapidly, a l t h o u g h some fibres are still present in the distal part. These S O M fibres are m a i n l y located adjacent to the pars i n t e r m e d i a in the n e i g h b o u r h o o d of the short portal vessel system. Other SOM
1 : 80 1 : 80
19578 (6-11-'75)
19578 (24-7-'76)
B 173 (1-10:75)
4 3 1 R b (K 50)
Dubois I I I
Dubois IV
Baker
Pearse 1 : 80 1 : 80 1:80
Anti-neurophysin I
Anti-neurophysin II
Control
1 : 80
1:80
1 : 80
1:80
19608 (24-7-'76)
19608 (6-11-'75)
Dubois It
1 : 80
1 : 80
Dubois I
108 (5-9-'75)
101 (6-6-'74)
Arimura II
-
Arimura I
1 : 80
4-5-'77
De Mey
De Mey purified over N I + N 2
1:40
Dilution
Code
Source
59.1+
58.4+ 8.7
5.2
79.5+ 15.4
147.4+- 23.8
130.8+ 15.0
368.4+ 51.0
74.5+- 7.3
228.5+ 14.8
116.5+ 10.9
156.9+- 29.0
70.0+- 6.6
290.6+- 23.1
312.7+- 22.6
SOM
4.1
2.9
1.4
9.2
5.1
6.7
41.8+- 1.8
n.d.
201.0+ 17.2
61.8+
46.6+- 2.9
50.8+
39.9+- 4.8
41.8+- 4.7
44.8+
47.5+
25.7+
45.3+
85.2+ 10.6
N e u r o p h y s i n I (N1)
-
(12.6)
-
-
-
-
-
-
-
-
(27.3)
(~o)
35.3+5.5
100.8+ 7.5
n.d.
66.3+- 6.5
39.6+ 3.5
60.6+ 4.9
53.7+- 5.5
42.1 + 3.8
34.3+ 2.1
38.9+ 3.0
42.6+- 4.2
30.7+- 4.1
103.5+ 8.6
N e u r o p h y s i n II (N2)
-
(47.4)
-
(38.7)
(28.2)
-
-
-
-
(104.1)
(~)
Immunofluorescence, in arbitrary units, of antisera to somatostatin. W h e n statistically significant, the values are also expressed in terms of a percentage of the value found on neurophysin beads after incubation with anti-neurophysin. F o r the latter calculations the b a c k g r o u n d fluorescence obtained with a p r e - i m m u n e control plasma was substracted from that obtained with the antibodies (for details see Swaab and Pool, 1975). n.d. = not determined; - = n o t significantly larger (P > 0.05) than the background reading
Table 1. Reaction of anti-somatostatin with neurophysin
~-
~"
Fig. 6. Serial sagittal semithin Epon sections of lateral part of neurohypophysial stalk incubated with purified anti-SOM (a), anti-AVP (b), anti-OXT (c). a Note mainly ventral course of SOM fibres from median eminence to neural lobe. h and c Note joining of supraoptico-neurohypophysial and suprainfundibular AVP and O X T tracts, and different localization of AVP and OXT fibres within rostral part of neural lobe. Insets: course o f fibres above recessus infundibularis; pi pars intermedia, x 200; insets: x 65
NeurohypophysialPeptides in the Rat
197
fibres run into the stalk at the dorsal site, slightly dorsal to the suprainfundibular AVP and OXT fibres (Fig. 4c), ending mainly in the rostro-dorsal part of the neural lobe (Fig. 6a). After incubation with anti-LHRH a clearcut reaction was present in the lateral parts of the median eminence, as well as above the infundibular recess. A few fibres occur in the neurohypophysial stalk, but none are observed in the neural lobe (Fig. 3). All control incubations showed absence of staining.
Discussion
The present study shows that the use of semithin Epon sections not only provides a link between light and electron microscopic immunocytochemistry (Watari et al., 1977; Van Leeuwen et al., 1978), but also gives new information in itself. Comparison between neurohypophyses embedded either in Epon or in paraffin, shows that the use of semithin (2 gm) Epon sections reveals morphological details better, due to less superimposition of the structures, than in 6 gm paraffin sections, but also produces an increase in the number of fibres stained in the median eminence and pituitary stalk. In addition, the reaction in the pars intermedia after incubation with anti-AVP was detected only in Epon sections, presumably due to the higher sensitivity of this procedure. The specificity tests used in this study allow us to discriminate between AVP and OXT. The differential localization of AVP, OXT, SOM and LHRH in the neurohypophysis provides additional evidence for the specificity of the reaction. A major immunocytochemical problem is the appearance of staining in structures that are not known to contain chemically related compounds. An example is the staining of Herring bodies in the neural lobe with anti-SOM, and of the pars intermedia cells with anti-AVP. This reaction product disappeared after solid phase absorption to the respective homologous antigen, which shows that the staining was based on the presence of the antigen itself or related compounds. As we suspected cross reaction of anti-SOM with neurophysin I and II and of anti-AVP with a-MSH, the staining capacity of nine different SOM and two AVP antisera was tested on bovine neurophysin I and II and a-MSH respectively. Since a quantification procedure for the PAP method on beads is not yet available, this aspect had to be tested by means of the indirect immunofluorescence (IF) procedure (Swaab and Pool, 1975). A disadvantage of this method is, that the sensitivity of the test system is different from that of the PAP procedure. For example, anti-AVP absorbed to a-MSH produced good staining in the neural lobe (Fig. 5b), while on AVP containing beads the IF values were not different from the control values. Some of the SOM antisera indeed revealed a considerable reaction, especially with neurophysin II. In addition, anti-SOM (De Mey) apparently also showed cross reaction with AVP (26 ~ of the fluorescence found with anti-AVP (g125) on AVP beads). The cross reaction with both neurophysin I and II and AVP could be eliminated by solid phase absorption to neurophysin I and II. Following this procedure only a few SOM positive fibres in the neural lobe were stained, while SOM staining on beads remained present. Cross reaction thus seems to be the explanation for the observations that SOM is present in the magnocellular SON
198
F.W. van Leeuwenet al.
and PVN of the rat, human fetus, and fox (Dubois and Kolodziejczyk, 1975; Bugnon et al., 1976, 1977, results that were obtained with antibodies not available in the present study). This conclusion was confirmed by Dierickx and Vandesande (personal communication). The reported staining of intermediate lobe cells by anti-AVP (Castel, 1978) provides another example of cross reaction, since a similar result was obtained in the vasopressin deficient homozygous Brattleboro rat. Thus, it is unlikely that this staining is due to the presence of AVP on its receptors, as claimed by Castel (1978). A theoretical possibility is that AVP was present in antisera either as free AVP or as AVP-anti-AVP complexes that could dissociate, bind to AVP receptors in Brattleboro pars intermedia cells and thus induce staining. However, since the intensity of staining in pars intermedia cells of Brattleboro and Wistar rats was equal, and no obvious increase in staining was found after preincubations of Brattleboro neurointermediate lobe sections with AVP (10-3-10p, g/ml), this possibility can be excluded. The positive reaction in the same intermediate lobe cells after incubation of serial sections with anti-AVP and e-MSH, and the elimination of the staining by solid phase absorption of anti-AVP to c~-MSH, makes it very probable that the reactive substance is e-MSH itself, rather than AVP. Another sign of the higher sensitivity of the PAP technique as compared to the IF procedure is that the amount of solid phase bound peptides for the demonstration of either method or serum specificity which is routinely used in the IF procedure (1 mg peptide beads/l gl antiserum; for details see Swaab and Pool, 1975), had to be increased in some cases. For instance, for the neutralization ofantiL H R H serum, and the purification of anti-AVP with c~-MSH beads, no disappearance of staining was observed in the median eminence and intermediate lobe cells respectively. However, when 5mg beads/lal antiserum were used inhibition of the staining was accomplished. The enhancement of the amount of beads did not result in a nonspecific abolition of the staining since anti-LHRH serum processed in the same way together with thyroglobulin beads still showed a bright immunopositive reaction in the median eminence while anti-AVP gave a reaction in the neural lobe after absorption with c~-MSH beads (see also Fig. 5b). The same results were reported by Sternberger et al. (1978) who used a high amount of L H R H beads (33 mg/lal L H R H antiserum). The data on the preferential localization of OXT in the rostral dorsal part and periphery of the neural lobe confirm those of Vandesande and Dierickx (1975). In addition, AVP fibres enter bilaterally into the neural lobe, and clusters of AVP and OXT fibres were found. The former result is in agreement with the report of Bock and Jurna (1977), who showed an ipsilateral reduction of "Gomori-positive" material in the rat neural lobe after unilateral lesioning of the hypothalamoneurohypophysial tract at different levels. The presence of AVP and OXT clusters within the neural lobe confirms the previously reported immuno-electron microscopical data, in which it was postulated that these clusters may occupy considerable areas (Van Leeuwen and Swaab, 1977). In addition, the central position of AVP might explain the finding by Krisch (1977) that AVP fibres are observed at the immuno-electron microscopical level in greater numbers than OXT fibres, since the rim of OXT in the periphery of the neural lobe may easily be trimmed away when ultrathin sections are prepared.
NeurohypophysialPeptides in the Rat
199
In a number of vertebrates, synaptoid endings of peptidergic nerve fibres have been demonstrated on intermediate lobe cells (see Bargmann et al., 1967; Bargmann, 1969; Von Lawzewitsch and Monastirsky, 1974). Such structures were not, however, observed in any significant numbers in the rat intermediate lobe (Baumgarten, 1972). In the present study no such innervation was observed. In the rat, the localization of OXT fibres in the area directly adjacent to the intermediate lobe might have functional significance, since OXT is thought to be split enzymatically into peptides with melanocyte stimulating hormone-releasing and -inhibiting factor activities (Cells, 1977; Vivas and Celis, 1977). The importance of these peptides for pars intermedia function is, however, still in dispute (see Tilders and Smelik, 1977). Most of the SOM fibres entering the neurohypophysial stalk were found to terminate in the rostral part of the neural lobe, the remainder being observed in the neighbourhood of the short portal vessel system. In agreement with the report of H6kfelt et al. (1978), we could not confirm the results of Barry and Dubois (1976), who reported L H R H fibres in the proximal part of the rat neural lobe. Neurohypophysial T R H fibres, which will be described by us in a subsequent communication, were also found in the neural lobe (H6kfelt et al., 1975b). AVP and OXT fibres, containing predominantly small granules (80-100 nm), but sometimes also larger granules (150-200 nm), were reported in the external zone of the median eminence (see Dierickx et al., 1976; Dube et al., 1976). These facts indicate that the neurohypophysis (classically divided into infundibulum and neural lobe, Rioch et al., 1940) can be considered as a continuous neurohaemal zone, which fits in well with the common neurohypophysial capillary bed (Page and Bergland, 1977).
References Bargmann, W.: Das neurosekretorische Zwischenhirn-Hypophysen-Systemund seine synaptischen Verkniipfungen. J. Neuro-Visc. Rel. Suppl. 9, 64-77 (1969) Bargmann, W., Lindner, E., Andres, K.H.: Ober Synapsen an endokrinen Epithelzellen und die Definition sekretorischer Neuronen. Untersuchungenam Zwischenlappender Katzenhypophyse.Z. Zellforsch. 77, 282-298 (1967) Barry, L., Dubois, M.P.: Immunoreactive LRF neurosecretory pathways in mammals. Acta Anat. (Basel) 94, 497-503 (1976) Baumgarten, H.G., Bj6rklund, A., Holstein, A.F., Nobin, A.: Organization and ultrastructural identification of the catecholaminenerve terminals in the neural lobe and pars intermedia of the rat pituitary. Z. Zellforsch. 126, 483-517 (1972) Bock, R., Jurna, I.: Ipsilateral diminution of CRF-granules after unilateral hypothalamic lesions. Cell Tissue Res. 185, 215-229 (1977) Bugnon, C., Lenys, D., Fellmann, D., Bloch, B.: Etude cyto-immunologiquedu systbmepeptidergique hypothalamique fi somatostatinechezle Renard (Vulpes vulpes). C.R. Soc. Biol. 170, 584-588 (1976) Bugnon, C., Fellmann, D., Bloch, B.: Immunocytochemical study of the ontogenesis of the hypothalamic somatostatin-containing neurons in the human fetus. Cell Tissue Res. 183, 319-328 (1977) Bugnon, C., Bloch, B., Lenys, D., Fellmann, D. : Cytoimmunologicalstudy of the LH-RH neurons in humans during fetal life. In: Brain-EndocrineInteraction III. Neural Hormones and Reproduction (D.E. Scott, G.P. Kozlowski,A. Weindl, eds.), pp. 183-196. Basel: Karger 1978 Castel, M.: Immunocytochemicalevidencefor vasopressin receptors.J. Histochem. Cytochem.26, 581592 (1978) Celis, M.E.: Hypothalamicpeptides involvedin the control of MSH secretion:identity, biosynthesisand regulation of their release. Front. Horm. Res. 4, 69-79 (1977)
200
F.W. van Leeuwen et al.
Chateau, M., Marchetti, J., Burlet, A., Boulang6, M.: Evidence of vasopressin in adenohypophysis: research into its role in corticotrope activity. Neuroendocrinology 28, 25-35 (1979) Dierickx, K., Vandesande, F., de Mey, J.: Identification, in the external region of the rat median eminence, of separate neurophysin-vasopressin and neurophysin-oxytocin containing nerve fibres. Cell Tissue Res. 168, 141-151 (1976) Dube, D., Leclerc, R., Pelletier, G.: Electron microscopic immunohistochemical localization of vasopressin and neurophysin in the median eminence of normal and adrenalectomized rats. Am. J. anat. 147, 103-108 (1976) Dubois, M.P., Kolodziejczyk, E.: Centres hypothalamiques du rat s6cr&ant la somatostatine: repartition des p6ricaryons en 2 syst6mes magno et parvocellulaires (&ude immunocytologique). C.R. Acad. Sci. Paris 281, 1737-1740 (1975) H6kfelt, T., Fuxe, K., Johansson, O., Jeffcoate, S., White, N.: Distribution of thyrotropin-releasing hormone (TRH) in the central nervous system as revealed with immunohistochemistry. Eur. J. Pharmacol. 34, 389-392 (1975a) H6kfelt, T., Efendic, S., Hellerstr6m, C., Johansson, O., Luft, R., Arimura, A.: Cellular localization of somatostatin in endocrinelike cells and neurons of the rat with special reference to the Al-cells of the pancreatic islets and to the hypothalamus. Acta Endocrinol. [Suppl.] (Kbh.) 200, 5-41 (1975b) H6kfelt, T., Elde, R., Fuxe, K., Johansson, O., Ljungdahl, A., Goldstein, M., Luft, R., Efendic, S., Nilsson, G., Terenius, L., Ganten, D., Jeffcoate, S., Rehfeld, J., Said, S., Perez de la Mora, M., Possani, L., Tapia, R., Teran, L., Palacios, R.: Aminergic and peptidergic pathways in the nervous system with special reference to the hypothalamus. In: The hypothalamus (S. Reichlin, R.J. Baldessarini, J.B. Martin, eds.), pp. 69-135. New York: Raven Press 1978 Knigge, K.M., Joseph, S.A., Hoffman, G.E.: Organization of LRF- and SRIF-neurons in the endocrine hypothalamus. In: The Hypothalamus (S. Reichlin, R.J. Baldessarini, J.B. Martin, eds.), pp. 49-67. New York: Raven Press 1978 Krisch, B.: Electronmicroscopic immunocytochemical study on the vasopressin-containing neurons in the thirsting rat. Cell Tissue Res. 184, 237-247 (1977) Lane, B.P., Europa, D.L.: Differential staining of ultrathin sections of epon-embedded tissues for light microscopy. J. Histochem. Cytochem. 13, 579-582 (1965) Leppfiluoto, J., Koivusalo, F., Kraama, R.: Thyrotropin-releasing factor: distribution in neural and gastrointestinal tissues. Acta Physiol. Scand. 104, 175-179 (1978) Oliver, C., Eskay, R.L., Ben-Jonathan, N., Porter, J.C.: Distribution and concentration of TRH in the rat brain. Endocrinology 95, 540-546 (1974) Page, R.B., Bergland, R.M.: The neurohypophyseal capillary bed. I. Anatomy and arterial supply. Am. J. Anat. 148, 345-357 (1977) Patel, Y.C., Zingg, H.H., Dreifuss, J.J.: Calcium-dependent somatostatin secretion from rat neurohypophysis in vitro. Nature 267, 852-853 (1977) Renlund, S.: Identification of oxytocin and vasopressin in the bovine adenohypophysis. Thesis, Uppsala, Sweden (1978) Rioch, D.McK., Wislocki, G.B., O'Leary, J.L.: A pr6cis of preoptic, hypothalamic and hypophysial terminology with atlas. In: The Hypothalamus, Association for Research in Nervous and Mental Disease, Vol. 20 (J.F. Fulton, S.W. Ranson, A.M. Frantz, eds.), pp. 3-30. Baltimore: The Williams and Wilkins Company 1940 Rossier, J., Battenberg, E., Pittmann, Q., Bayon, A., Koda, L., Miller, R., Guillemin, R., Bloom, F.: Hypothalamic enkephalin neurones may regulate the neurohypophysis. Nature 277, 653-655 (1979) Silverman, A.J., Zimmerman, E.A.: Pathways containing luteinizing hormone-releasing hormone (LHRH) in the mammalian brain. In: Brain-Endocrine Interaction III. Neural Hormones and Reproduction (D.E. Scott, G.P. Kozlowski, A. Weindl, eds.), pp. 83-96. Basel: Karger 1978 Sternberger, L.A., Petrali, J.P., Joseph, S.A., Meyer, H.G., Mills, K.R.: Specificity of the immunocytochemical luteinizing hormone-releasing hormone receptor reaction. Endocrinology 102, 63-73 (1978) Swaab, D.F., Pool, C.W.: Specificity of oxytocin and vasopressin immunofluorescence. J. Endocrinol. 66, 263-272 (1975) Swaab, D.F., Pool, C.W., Nijveldt, F.: Immunofluorescence of vasopressin and oxytocin in the rat hypothalamo-neurohypophyseal system. J. Neural Transm. 36, 195-215 (1975) Szent~tgothai, J., Flerk6, B., Mess, B., Halfisz, B.: Hypothalamic Control of the Anterior Pituitary. Budapest: Akad6miai Kiad6 (1968)
Neurohypophysial Peptides in the Rat
201
Tilders, FJ.H., Smelik, P.G.: Direct control of MSH secretion in mammals: The involvement of dopaminergic tubero-hypophysial hormone. Front. Horm. Res. 4, 80-93 (1977) Vandesande, F., Dierickx, K.: Identification of the vasopressin producing and of the oxytocin producing neurons in the hypothalamic magnocellular neurosecretory system of the rat. Cell Tissue. Res. 164, 153-162 (1975) Vandesande, F., Dierickx, K., De Mey, J.: The origin of the vasopressinergic and oxytocinergic fibres of the external region of the median eminence of the rat hypophysis. Cell Tissue. Res. 180, 443-452 (1977) Van Leeuwen, F.W., Swaab, D.F.: Specific immunoelectronmicroscopic localization of vasopressin and oxytocin in the neurohypophysis of the rat. Cell Tissue Res. 177, 493-501 (1977) Van Leeuwen, F.W., Swaab, D.F., De Raay, C.: Immuno-electron microscopic localization of vasopressin in the rat suprachiasmatic nucleus. Cell Tissue Res. 193, 1-10, (1978) Vivas, A., Cells, M.E.: New evidence that demonstrates that L-Pro-L-Leu-L-Gly-NHz might be the natural MIF. Acta Physiol. Lat. Am. 27, 177-182 (1977) Von Lawzewitsch, I., Monastirski, R.: Neuro-glandular junctions in the pars intermedia in the rabbit. Acta Anat. (Basel)87, 409-413 (1974) Watari, N., Hotta, Y., Mabuchi, Y.: Comparative light and electron-microscopic study of the adrenal gland of the snake. Arch. Histol. Jpn. 40, Suppl., 81-85 (1977) Weindl, A., Sofroniew, M.V.: Neurohormones and circumventricular organs. In: Brain-Endocrine Interaction III. Neural Hormones and Reproduction (D.E. Scott, G.P. Kozlowski, A. Weindl, eds.), pp. 117-137. Basel: Karger 1978
Accepted July 5, 1979
