- Email: [email protected]
Distribution of peptidergic nerves in the choroid plexus, focusing on coexistence of neuropeptide Y, vasoactive intestinal polypeptide and peptide histidine isoleucine
Regulatory Peptides, 27 (1990) 11-26
11
Elsevier REGPEP 00852
Distribution of peptidergic nerves in the choroid plexus, focusing on coexistence of neuropeptide Y, vasoactive intestinal polypeptide and peptide histidine isoleucine Christer Nilsson ~, R o l f E k m a n 2, M a f i a Lindvall-Axelsson t and Christer Owman Departments of Medical Cell Research, ~Section of Neurobiology and 2Psychiatry and Neurochemistry, University of Lund, Lund, (Sweden)
(Received 17 February 1989; revised version received and accepted 2 August 1989) Choroid plexus; Coexistence; Neuropeptide Y; Vasoactive intestinal polypeptide; Peptide histidine isoleucine; Sympathetic denervation
K e y words:
Summary Choroid plexus from rat, guinea-pig, rabbit and pig was investigated by light-microscopic immunohistochemistry and by radioimmunoassay for the presence of neuropeptides. A moderately dense supply of nerve fibers containing neuropeptide Y (NPY) and vasoactive intestinal polypeptide (VIP), respectively, was found around blood vessels and in close relation to the secretory epithelium in both pig and rabbit, while lower densities of nerve fibers were found in rat and guinea-pig. Peptide concentrations ranged from 10-40 pmolequivalents/g (pmoleqv/g) for NPY and 0.5-6 pmoleqv/g for VIP in all four species. Peptide histidine isoleucine (PHI) immunoreactive nerve fibers were present in pig choroid plexus at a lower density than NPY and VIP but with a similar distribution. Low concentrations of substance P (0.3-3 pmoleqv/g) and c aicitonin gene-related peptide (0.1-3 pmoleqv/g) were found to a varying degree in choroid plexus tissue from the different species, while immunohistochemical investigation was unable to detect any immunoreactive nerve fibers. NPY was often found to coexist with VIP and PHI in pig choroid plexus, while a lesser amount of nerve fibers showed coexistence of NPY and the noradrenaline synthetizing enzyme, dopamine-fl-hydroxylase. Surgical sympathetic denervation by
Correspondence: Christer Nilsson, Department of Medical Cell Research, Biskopsgatan5, S-223 62 Lund,
Sweden. 0167-0115/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
12 excision of the superior cervical ganglion in the rabbit abolished NPY-containing nerve fibers, as revealed by immunohistochemistry, but only decreased NPY levels by one third, which may be due to different identity of the peptide being detected by the two techniques. It is concluded that NPY-containing nerve fibers have a dual origin in the choroid plexus and coexist with either noradrenaline or VIP/PHI.
Introduction
The choroid plexus is a villous organ situated in all four brain ventricles. It consists of a single-layered, cuboidal epithelium overlying a highly vascularized stroma. The choroid plexus produces the major part of the cerebrospinal fluid (CSF) [1] by a mechanism involving active transport of ions across the epithelium [2]. Production of CSF seems to be under both neuronal [3] and hormonal [4] control. Adrenergic as well as cholinergic nerves have been described to supply the tissue [3], in addition a few studies have reported the presence of nerve fibers immunoreactive to vasoactive intestinal polypeptide (VIP), neuropeptide Y (NPY) and substance P (SP) [5-7]. The aim of the present study was to investigate the peptidergic nerve supply of the choroid plexus, using immunohistochemistry and radioimmunoassay. The origin and possible coexistence of these neuropeptides in this secretory tissue were studied as a basis for future functional investigations.
Materials and Methods
Animals Choroid plexus tissue from 35 male rats (Sprague-Dawley or Wistar strain, 200-400 g body weight), 22 guinea-pigs of either sex (Dunkett-Hartley, 300-500 g), 38 male rabbits (New Zealand White, 2500-3500 g) and approximately 40 adult pigs of either sex was used in the study. The laboratory animals were kept on standard pellet diet and tap water ad libitum for at least one week before all experiments. The pig plexuses were obtained from a local slaughter-house. Owing to the slaughtering technique, only choroid plexus from the lateral and fourth ventricles could be obtained from the pig brain. Tissue preparation For immunohistochemistry, rats and guinea-pigs were anesthetized with either diethyl ether (Merck) or an intraperitoneal injection of pentobarbital (Mebumal, Aco, Sweden), and the rabbits were anesthetized with intravenous injection of pentobarbital. Following anesthesia, the animals were perfused by gravity perfusion at a pressure corresponding to the animal's blood pressure with 0.9~o saline, followed by Stefanini's solution (2~/o paraformaldehyde and 15 ~o picric acid in 0.1 M phosphate buffer, pH 7.2), all perfusion solutions being kept at + 4 °C. After 15 min perfusion with fixative, the brains of the
13 animals were rapidly dissected out, the plexus tissue from all ventricles removed and immersion fixed in the same fixative for a further 2 h. In some rats, the whole brain was used and sectioned in order to better preserve plexus topography. The specimens were then rinsed for 2 days in Tyrode's solution (NaC1 137 mM, KCI 3 mM, CaC12 1 mM, MgC12 0.2 mM, NaHCO 3 12 mM, NaH2PO 4 0.3 mM, glucose 0.1~o) with 10~o sucrose at + 4 °C. The rinsing solution was changed two or three times a day to remove all fixative. Pig brain was taken directly from the skull of the pigs 30-40 min after killing, and the choroid plexus from the lateral and fourth ventricles was removed and immersed in ice-cold Stefanini's solution for 24 h. The plexuses were then rinsed as above. For radioimmunoassay (RIA), the laboratory animals were anesthetized as above, but perfused only with 0.9~o saline at +4 °C to wash out the blood. Only plexuses macroscopicaUy devoid of blood were used. When the plexus had been removed from the brain, it was wrapped in pre-weighed pieces of aluminium foil and weighed to determine the wet weight of the specimen. The package was then put in liquid nitrogen until frozen. Freezing took place approximately 5 min after dissection. For practical reasons, choroid plexus from pig was not perfused but immediately frozen on dry ice and weighed. All specimens were stored at -85 °C until analysis.
Extraction procedure The tissue was boiled in 0.9~ saline for 10 min, followed by homogenization (Polytron, 1-2 min). The homogenates were centrifuged at + 4 °C, 1000 g for 30 min and the supernatants were collected. The pellets were dissolved in 0.5 M acetic acid and subjected to the same extraction procedure as described above. The supernatants were mixed and lyophilized. To obtain sufficient amounts of tissue for analysis, plexuses from several animals were sometimes pooled prior to extraction.
Immunohistochemistry Sections of choroid plexus or whole rat brain were cut in a cryostat (Leitz 1720) at 10/lm thickness and placed on chromalum-gelatin-coated glass slides. After rinsing in phosphate buffered saline (PBS) with 0.25~o Triton-X 100 (Eastman), sections were incubated with the antisera in a moist chamber for 20 h at 4 ° C followed by 2 h at room temperature. Antibodies towards NPY produced in rabbit (1 : 320, Lot. K8404, Milab, Maim0, Sweden) or sheep (1:400, Blessing and Oliver E2210, kindly provided by Dr William Blessing, Australia), VIP (1:320; Lot. K7852, Milab), peptide histidine isoleucine (PHI 1:320; PII Efendic, Stockholm, Sweden), SP (1:160; Lot. K8127, Milab), Calcitonin gene-related peptide (CGRP 1:640; Lot. K8425, Milab) or the noradrenaline synthesizing enzyme, dopamine-fl-hydroxylase (DBH 1:320; Eugene Tech, U.S.A.) were used. Further procedure followed the indirect immunofluorescence method developed by Coons et al. [8]. As secondary antibodies we used pig anti-rabbit antibodies (1:20, Dakopatts, Denmark) conjugated with either fluorescein isothiocyanate (FITC) or tetramethylrhodamine isothiocyanate (TRITC). To visualize the antibody towards NPY raised in sheep, a donkey anti-sheep antibody conjugated with FITC (1 : 80, Sigma) was used. The following specificity control procedures were followed with antisera giving a positive immunohistochemical reaction: (1) omitting the primary antisera, (2) dilution
14 of the primary antisera, (3) preabsorption of the antisera at + 4 °C (for 24 h) with pure porcine VIP, PHI and NPY, respectively, at a concentration of 2-20 #M for all peptides, and (4)preabsorption of the NPY- and VIP-antisera with the respective non-specific antigen (i.e., VIP and NPY), to further eliminate the possibility of crossreactivity between peptides and antisera. To investigate the possible coexistence of NPY, VIP, PHI and DBH in nerve fibers, we used sequential staining with elution of the primary antibody by treatment of the sections with potassium permanganate for 30 s as described by Tramu et al. [9]. Before adding the antisera for the second peptide, sections were incubated only with FITCconjugated anti-rabbit antibodies to visualize any remaining antibodies towards the first peptide. Only sections where all of the first antisera had been removed were used. Double staining was performed by staining sections with sheep antisera towards NPY, visualized with donkey anti-sheep antibodies conjugated with FITC, followed by staining with antibodies towards either VIP, PHI or DBH which were visualized with pig anti-rabbit antibodies conjugated with TRITC. To control the validity of the double staining method, we stained sections with either FITC- or TRITC-conjugated secondary antibody to see if there was any leakage of light through the filters that might give the impression of false-positive coexistence.
R adioimmunoassay For the RIA of NPY, a rabbit antiserum raised against synthetic porcine NPY (gift from Dr. P.C. Emson, Cambridge, U.K.), conjugated to bovine serum albumin with carbodiimide, was used in a final dilution of 1 : 40,000. ~25I-NPY used as tracer was purified by high-performance liquid chromatography. The antiserum cross-reacts with peptide YY to 33~o but not with C-terminal fragments of NPY and PYY (NPY 13-36 and PYY 13-36) nor with bovine pancreatic polypeptide, gastrin-inhibiting peptide, PHI, VIP, or secretin. The detection limit was 11.7 pmol/l. Intra-assay variation was 6.5~o (n = 20) and the inter-assay variation was 7.0~o (n = 16). Immunoreactive VIP was quantified using antiserum (No. 7852, Milab) in a final dilution of 1 : 72,000. The antiserum recognizes the N-terminal 15 amino acid sequence of VIP, and does not cross-react with PHI, secretin, glucagon, gastrin inhibitory peptide or cholecystokinin. The assay can detect a minimum of 4 pmol/1 with 95 ~o confidence. Intra-assay variation was 4.3~o (n = 12) and the inter-assay variation was 10.4~o (n = 10) [10]. For RIA of CGRP, we used a rabbit antiserum (code No. A 13, Milab) raised against albumin-conjugated synthetic rat CGRP. The antiserum was used in a final dilution of 1 : 30,000. The detection limit was 10 pmol/1. There was no cross-reaction with human calcitonin, katacalcin or calcitonin C-terminal adjacent peptide. The inter-assay variation was below ll~o (n = 20) in the range 100-300 pmol/1 [11]. Immunoreactive SP was determined using a rabbit antiserum raised against synthetic SP (code No. SP-2, a kind gift from Dr. E. Brodin, Stockholm, Sweden). The RIA for SP has previously been described in detail [12]. Cross-reactivity was investigated at peptide antigen concentrations of 1-1000 ng/ml diluted antisera. For all assays a standard procedure was used with duplicate tubes and serial dilutions for each sample, which were corrected for unspecific binding.
15 Both in the case of immunohistochemistry and RIA the techniques only allow for demonstration of, for example, VIP-like immunofluorescence or levels of VIP-like material. For practical reasons the short term VIP etc. is used throughout for the various peptides.
Sympathectomy The superior cervical ganglion was excised uni- or bilaterally in rabbits 1 or 2 weeks before killing and perfusion. This ganglion provides the lateral choroid plexus with its sympathetic innervation in an ipsilateral manner [ 13]. Complete sympathectomy was confirmed by treating whole-mounts of a part of each plexus or the iris according to the formaldehyde method to visualize adrenergic nerves as described by BjOrklund et al. [ 14]. In order to compare the distribution of the NPY innervation with a tissue having a well-established sympathetic supply, the irises of the same operated animals were dissected out and analyzed by RIA parallel to the plexus tissue. The sympathetic innervation of the iris is almost exclusively derived from the ipsilateral superior cervical ganglion [ 15].
Statistical analysis For comparison of NPY concentrations in control and sympathectomized tissues, the unpaired Student's t-test was used.
Results
Nerve fibers immunoreactive to NPY (Fig. 1) and VIP (Fig. 2) were found in all species studied. It was not possible to elucidate the VIP-immunoreactivity in rabbit since a suitable antibody was lacking for this species. A moderately dense supply of both neuropeptides was found in the choroid plexus of the lateral and third ventricles in pig as well as for NPY in rabbit. Only a few nerve fibers were found in the fourth ventricle choroid plexus of these species, as well as in all plexuses from rat and guinea-pig. In choroid plexus from pig, PHI showed approximately the same distribution as NPY and VIP but the number ofimmunoreactive nerve fibers was smaller (Fig. 3). There was no specific immunoreactivity with the antisera toward SP and CGRP. The immunohistochemical data are summarized in Table I. Nerve fibers immunoreactive to NPY and VIP showed a markedly similar distribution. Dense nerve plexuses could be found around larger and medium-sized arteries (Figs. 2B and 5), while smaller arteries and veins were more sparsely innervated. A varying number of isolated fibers could also be found in the plexus stroma and in close proximity to microvessels and epithelium (Figs. 1B, 1C and 2A). Analysis by RIA (Tables II and III) revealed quite similar levels of NPY in the choroid plexus of all species studied, with an average content of 8 and 14 pmoleqv/g wet weight in the pig and rabbit. In spite of the paucity of nerve fibers shown by immunohistochemistry, results with RIA showed mean concentrations for NPY of 12 pmoleqv/g for rat and 38 pmoleqv/g in the guinea-pig. Regional differences in the concentration of NPY could also be seen between choroid plexus from the different
16
Fig. 1 NPY-immunoreactive nerve fibers in the lateral choroid plexus of guinea-pig (A, x 400) rabbit (B, x 400) and rat (C, x 600). In A, note the nerves in close relation to both vessels and epithelium. In B and C nerve fibers run in the stroma of a villus, between the single layers of epithelial cells.
17
Fig. 2. V l P - i m m u n o r e a c t i v e n e r v e fibers in t h e l a t e r a l c h o r o i d plexus o f pig, u n d e r t h e e p i t h e l i u m (A, x 600) a n d p e r i v a s c u l a r y (B, x 200).
TABLE I A r b i t r a r y n u m b e r o f p e p t i d e r g i c n e r v e fibers in c h o r o i d plexus tissue s t u d i e d b y f l u o r e s c e n c e i m m u n o histochemistry Estimation of nerve supply: + + + = dense nerve supply; + + moderate nerve supply; + sparse nerve supply; 0 = no detectable immunoreactivity; - = not investigated. Species Pig
Rabbit
Guinea-pig
Rat
C h o r o i d plexus
NPY
VIP
PHI
SP
CGRP 0
Lateral
+ + +
+ + +
+ +
0
Third
+ + +
+ + +
+ +
0
0
Fourth
+
+
0
0
0
Lateral
+ +
m
Third
+ +
m
Fourth
+
m
m
D
Lateral
+
+
-
0
0
Third
+
+
-
0
0
Fourth
0
0
-
0
0
Lateral
+
+
-
0
0
Third
+
+
-
0
0
Fourth
0
0
-
0
0
18
Fig. 3. PHI-immunoreactivenerve fibersin pig lateral choroidplexus.The fiberscan be seen round an artery (A), a vein (B) and running in the stroma (C) ( × 400). ventricles of the brain. In the rabbit, levels were highest in choroid plexus tissue from the third ventricle, while the concentration was approximately three times lower in choroid plexus from the lateral and fourth ventricles (Table III). Measurable concentrations of VIP were present in choroid plexus from all species, ranging from 0.4 to 3.3 pmoleqv/g. C G R P and SP was only found in some of the analyzed samples, with low and rather varying concentrations (Tables II and III). Studies in the pig with both the sequential and double staining methods revealed a very similar distribution for N P Y and VIP immunoreactivity, with a great majority of the nerve fibers and their varicosities containing both NPY and VIP (Figs. 4 and 5). P H I
19 TABLE II Concentrations of neuropeptides, analyzed by radioimmunoassay, in choroid plexus pooled from all four ventricles in rabbit, rat and guinea-pig Values are expressed as pmoleqv/g tissue w.w. + S.E.M.; n = number of animals; n.d. = non-detectable; B = bound and T = total activity. Species
NPY
VIP
CGRP
SP
Rabbit
13.9 + 2.8 (n = 6) 12.3 37.5
0.57 + 0.03" (n = 3) 3.3 0.4
2.8 + 1.4 (n = 6) n.d. 1.7
n.d. (n = 6) 1.2 3.0
Rat b Guinea-pigc
" In three additional analyses, the peptide was close to the level of detection, and these values are therefore not given (B/T = 90~). b Analysis performed on pooled plexus tissue from ten animals. c Analysis performed on plexus tissue pooled from six animals.
i m m u n o r e a c t i v i t y always c o e x i s t e d w i t h N P Y (Fig. 6) o r V I P . A few n e r v e fibers c o u l d be f o u n d to c o n t a i n b o t h N P Y a n d D B H
(Fig. 7), while V I P a n d D B H w e r e n e v e r
p r e s e n t in the s a m e fiber (Fig. 8). N P Y a n d V I P w e r e f o u n d to coexist also in the guinea-pig. N e i t h e r F I T C - o r T R I T C - i n d u c e d specific i m m u n o f l u o r e s c e n c e w a s visible w h e n u s i n g filters t h a t w e r e n o t a p p r o p r i a t e for t h e e m i s s i o n w a v e l e n g t h s o f the r e s p e c tive f l u o r o p h o r e s , t h u s e x c l u d i n g t h e possibility o f f a l s e - p o s i t i v e c o e x i s t e n c e . E x c i s i o n o f the s u p e r i o r c e r v i c a l g a n g l i o n in the r a b b i t c o m p l e t e l y a b o l i s h e d the NPY-immunoreactive
n e r v e fibers in the ipsilateral c h o r o i d plexus o f the lateral
ventricle. In t h e R I A analysis, there w a s o n l y a m o d e r a t e ( 3 1 ~ ) ,
non-significant
d e c r e a s e in N P Y c o n c e n t r a t i o n in the ipsilateral c h o r o i d plexus o f the lateral ventricle, while t h e N P Y c o n c e n t r a t i o n in the ipsilateral iris w a s 88 ~ l o w e r after s y m p a t h e c t o m y
TABLE III Regional differences in the concentration of neuropeptides, analyzed by radioimmunoassay, in choroid plexus from pig and rabbit Values are expressed as pmoleqv/g tissue w.w. + S.E.M.; n = number of separate analyses; n.d. = nondetectable. Species
Choroid plexus
NPY
VIP
CGRP
SP
Rabbit a (n = 3)
Lateral Third Fourth
8.0 + 0.3 33.2 + 5.4 11.1 + 2.2
n.d. 5.5c n.d.
n.d. 1.7c n.d.
n.d. n.d. n.d.
Pigb (n = 9)
Lateral Fourth
7.6 _+2.1 12.3 _+4.3
2.1 + 0.3 2.3 _+0.6
0.08 + 0.05 0.12 + 0.08
0.40 + 0.16 0.30 + 0.13
a Each analysis was performed on choroid plexus from the respective ventricles pooled from three animals. b Tissue from one animal used in each analysis. c Immunoreactive material only detected in one sample.
20
Fig. 4. Sequential staining of pig lateral choroid plexus with antibodies towards NPY (A) and VIP (B). The varicositites of the individual nerve fibers correspond in all detail ( x 200).
c o m p a r e d to the contralateral, intact control tissue (P <0.01) (Table IV). Bilateral sympathectomy led to a similar, 37~o, decrease in N P Y immunoreactivity (n = 2). Non-specific staining could not be detected using the specificity controls listed under Materials and Methods.
21
Fig. 5. Double staining of a densely innervated artery in pig lateral choroid plexus using sheep antibodies towards NPY (A) and rabbit antibodies towards VIP (B). The two peptides coexist in the majority of fibers ( × 400).
Discussion The present results conf'm-n and extend previous results on the peptidergic nerve supply o f the choroid plexus. Thus, Lindvall et al. [ 5] found VIP-immunoreactive nerves with approximately the same distribution and species variation as has been shown in the present study. The concentration o f VIP in pig was in that study found to be 10 pmoleqv/g, which is higher than the present figure. The difference can be explained by the use of different antibodies. Occasional NPY-immunoreactive nerve fibers were recently described in plexus from cat, rat and guinea-pig, with concentrations, measured
I
Fig. 6. Double staining of a single nerve fiber in pig lateral choroid plexus using sheep antibodies towards NPY (A) and rabbit antibodies towards PHI (B). The localization of the varicosities show a striking resemblance ( x 400).
22
Fig. 7. Double staining of an artery in pig lateral choroid plexus using antibodies towards NPY (A) and DBH (B), which coexist in a few nerve fibers ( x 400). by R I A , being below detection except in rabbit (7.2 pmoleqv/g) and in h u m a n postm o r t e m material (0.3 pmoleqv/g [ 7]. The present study is in agreement with the i m m u n o histochemical findings in rat a n d guinea-pig, which seem to have a scarce peptidergic nerve supply in the choroid plexus. The density o f NPY-fibers is higher in rabbit,
Fig. 8. Sequential staining of pig lateral choroid plexus with antibodies towards VIP (A) and DBH (B) showing a small artery with moderately dense nerve supply. VIP and DBH show completely different patterns of innervation, in which the individual nerve fibers and varicosities do not match each other ( x 400).
23 TABLE IV Concentration of NPY, measured by RIA, in rabbit choroid plexus of the lateral ventricle and in iris after unilateral cervical sympathectomy Values are expressed as pmoleqv/gtissue w.w. + S.E.M. Student's t-test: **, 0.001 < P < 0.01. n.s,, nonsignificant, n, number of separate analyses. Tissue Lateral choroid plexus (n = 5) Iris (n = 4)
Control s i d e 5.7 + 1.7~ 151.6 + 23.6
Sympathectomized side 4.0 + 1.3 18.4 + 6.6
Percent reduction 31 n.s. 88**
a This is slightlylower than the correspondingvalue for the lateral choroid plexus in Table III, though the difference is not statistically significant.
however, and the NPY-innervation in the pig is remarkably rich, especially round some of the medium-sized arteries. This has been shown to be true also for the noradrenergic sympathetic innervation [ 16], where the order..of nerve density among the presently investigated species was: pig > guinea-pig = rat > rabbit. It should be noted that some of the larger arteries in the choroid plexus are more densely innervated with NPY- and VIP-immunoreactive nerves than with noradrenergic sympatlaetic nerves, as studied by the formaldehyde technique [ 16]. Although the greatest density of nerve fibers and varicosities was found round blood vessels, single nerve fibers and varicosities could often be seen close to the secretory epithelial cells. It is not possible at the light-microscopic level to determine if there are functional contacts between nerve fibers and epithelium. However, ultrastructural evidence has shown that sympathetic terminals run close to the epithelial cells [3]. It is possible that the peptidergic nerve fibers have a similar organization in addition to their perivascular distribution. For the first time, the presence of PHI-immunoreactive nerve fibers in the mammalian choroid plexus is reported. These nerve fibers could be seen to form networks around arteries, veins and arterioles and also to run isolated in the connective tissue stroma. SP has been reported to be present at a low concentration (2.0-3.5 pmoleqv/g), with only a scarce perivascular distribution, in the choroid plexus from guinea-pig, rabbit and cat [6]. No SP-immunoreactive nerve fibers were found in our study. Only in some of the tissue samples from pig, rat knd guinea-pig analyzed by RIA could SP be found, and the concentrations were very low. The immunohistochemical studies did not show any positive reaction for C G R P either, and only low concentrations of the peptide could be detected by RIA in some of the tissue samples from guinea-pig, rabbit and pig. NPY is a well-known cotransmitter with noradrenaline in both peripheral and cerebral vessels [ 17,7]. It was therefore expected that NPY- and DBH-immunoreactivity should occur in the same nerve fibers. However, N P Y and D B H only coexisted in a few of the nerves, as shown in the pig. Instead there was an extensive coexistence of NPY- and VIP-immunoreactive nerves in the choroid plexus from this species, as secured also by comparing the individual varicosities of the nerves in photomicrographs made of the same area of a section, either in sequential or double staining. P H I derives from the same precursor molecule as VIP and is thus often localized in
24 the same nerve fibers as VIP [ 18]. In the present study, PHI was found to coexist also with NPY, which further confirms the coexistence of NPY and VIP. It should be noted that, although matching varicosities are considered as evidence for coexistence, definite proof can only be provided by electronmicroscopic immunohistochemistry, since different nerve fibers may run in parallel with the respective varicosities lying adjacent to each other. Coexistence of NPY, VIP and PHI has previously been described in intramural neurones of the small intestine [19], and recently Leblanc et al. [20] described coexistence of NPY and VIP in cranial parasympathetic neurones as well, notably those present in the sphenopalatine ganglion which is known to innervate cerebral vessels [21]. Furthermore, recent studies have shown coexistence between NPY and VIP in non-noradrenergic axons innervating guinea-pig cerebral vessels [22]. VIP-contalning nerve fibres in the cerebrovasculature have recently been shown to originate in the sphenopalatine and otic ganglia, as well as the internal carotid miniganglion [23]. Retrograde tracing combined with double staining immunohistochemistry has revealed coexistence of NPY and VIP in ganglion cells of both the sphenopalatine and otic ganglion [24]. The mammalian choroid plexus harbours an extensive cholinergic nerve supply, as evidenced by the acetylcholinesterase technique [3 ], maybe indicating that the largest input of NPY-immunoreactive nerves to the choroid plexus represents a parasympathetic innervation. There is reason to suspect species-variation in the proportion of NPY-immunoreactive nerves belonging to the sympathetic and parasympathetic divisions. This is indicated by the denervation studies in rabbit, where all of the NPY immunofluorescence disappeared after sympathectomy. However, RIA showed only a decrease of some 30 ~o of NPY immunoreactivity in the rabbit choroid plexus after this operation, which might indicate that a considerable amount of immunoreactive NPY is lost following formalin fixation during the immunohistochemical processing. In contrast, the NPY concentration in iris fell by almost 90~, which is in accordance with previous investigations [25]. Both the choroid plexus of the lateral ventricles and the iris receive an almost exclusively ipsilateral sympathetic innervation originating in the superior cervical ganglion [ 13,15]. However, the present results indicate a significant difference between the two tissues in the proportion of NPY-containing nerve fibers with a sympathetic origin. It is possible that the parasympathetic NPY nerves have a low concentration of the peptide in rabbit, not detectable with the antisera used for immunohistochemistry. This is supported by the recent findin~ that the NPY-immunoreactivity in non-sympathetic VIP-containing nerves increases markedly after sympathetic denervation of guinea-pig cerebral vessels [22]. Alternatively, the NPY-like material in the sympathetic and parasympathetic innervation may consist of different fragments of the NPY precursor molecule, the fragments not being detectable by the antisera. At present, little is known about the degradation of NPY, although it has been reported to be stable in human post-mortem brain tissue [26]. It seems unlikely that NPY was degraded during the extraction procedures used, as judged from recovery data (more than 80~o). In any event, the presence of several NPY-like immunoreactive peptides should be considered when performing routine RIAs on tissue extracts. Variations in the concentrations of these NPY-like fragments relative to NPY could provide confounding results.
25
Concerning the possible functions of the above described neuropeptides in the choroid plexus, VIP has been shown to stimulate production of the second messenger cyclic AMP in whole choroid plexus in vitro [27] and in cultured bovine choroid plexus epithelial cells [28]. Furthermore, VIP is a powerful vasodilator in cerebral arteries [29], including the anterior choroidal artery [ 5 ]. NPY has been found around a large number of peripheral and central vessels, where it is located in the noradrenergic, sympathetic nerves [17,7]. Both a direct contractile effect and a modulation of the sympathetic contractile effect on blood vessels has been described [30,31,7]. In the gut, NPY has been found to modulate epithelial ion transport [32]. Thus, VIP and NPY may be involved in both blood flow regulation and epithelial function in the choroid plexus.
Acknowledgements This research was supported by grants 14X-732 and 21X-07517 from the Swedish Medical Research Council. We would like to express our gratitude to Mr. Ragnar M~rtensson for the photographic work and to Mrs. Ulla-Britt Andersson for the technical assistance.
References 1 McComb, J.G., Recent research into the nature of cerebrospinal fluid formation and absorption, J. Neurosurg., 59 (1983) 369-383. 2 Wright, E.M., Transport processes in the formation of the cerebrospinal fluid, Rev. Physiol. Biochem. Pharmacol., 83 (1978) 3-34. 3 LindvaU, M. and Owman, Ch., Autonomic nerves in the mammalian choroid plexus and their influence on the formation of cerebrospinal fluid, J. Cereb. Blood Flow Metab., 1 (1981) 245-266. 4 Steardo, L. and Nathanson, J. A., Brain barrier tissues: End organs for atriopeptins, Science, 235 (1987) 470-473. 5 Lindvall, M., Alumets, J., Edvinsson, L., Fahrenkrug, J., Hhkanson, R., Hanko, J., Owman, Ch., Schaffalitsky de Muckadell, O.B. and Sandier, F., Peptidergic (VIP) nerves in the mammalian choroid plexus, Neurosci. Lett., 9 (1978) 77-82. 6 Edvinsson, L., Rosendal-Helgesen, S. and Uddman, R., Substance P: Localization, concentration and release in cerebral arteries, choroid plexus and dura mater, Cell Tissue Res., 234 (1983) 1-7. 7 Edvinsson, L., Copeland, J.R., Emson, P.C., McCtflloch, J. and Uddman, R., Nerve fibers containing Neuropeptide Y in the cerebrovascular bed: Immunocytochemistry, radioimmunoassay, and vasomotor effects, J. Cereb. Blood Flow Metabol., 7 (1987) 45-57. 8 Coons, A.H., Leduc, E.G. and Connolloy, J.M., Studies on antibody production. I. A method for histoehemical demonstration of specific antibody and its application to a study of hyperimmune rabbit, J. Exp. Med., 102 (1955) 49-60. 9 Tramu, G., Pilez, A. and Leonardelli, J., An efficient method of antibody elution for the successive or simultaneous location of two antigens by immunocytochemistry, J. Histochem. Cytochem., 26 (1980) 322-324. 10 Ekman, R., Tornqvist, K., Glucagon and VIP in the retina, Invest. Ophthalmol. Vis. Sci., 26 (1985) 1405-1409. 11 Grunditz, T., Ekman, R., H~kanson, R., Rerup, C., Sundler, F. and Uddman, R., Calcitonin gene-related peptide in the thyroid nerve fibers and C cells. Effects on thyroid hormone secretion and response to hypercalcemia, Endocrinology, 119 (1986) 2313-2324. 12 Brodin, E., Lindfors, N., Theodorsson-Norheim, E., Peterson, L., Bartfai, T., Orgen, S.-O. and Rosell,
26 S., Tachykinins in rat central nervous system: Distribution, molecular forms, release and effects of chronic treatment with antidepressent drugs. In R. H~kanson and F. Sundler (Eds.), Tachykinin Antagonists. Proc. FernstrSm Symp., Vol. 6, Elsevier North Holland Biomedical Press, Amsterdam, 1985, pp. 15-27. 13 Lindvall, M., Edvinsson, L. and Owman, Ch., Sympathetic nervous control of cerebrospinal fluid production from the choroid plexus, Science, 201 (1978) 176-178. 14 Bj0rdklund, A., Falck, B. and Owman, Ch., Fluorescence microscopic and microspectrofluorometric techniques for the cellular localization and characterization of biogenic amines. In S.A. Berson (Ed.), Methods of Investigative and Diagnostic Endocrinology. Vol 1: J.E. Rail and I.J. Kopin (Eds.), The thyroid and biogenic amines, Elsevier North Holland, Amsterdam, 1972, pp. 318-368. 15 Stone, R.A., Kuwayama, Y. and Laties, A.M., Regulatory peptides in the eye, Experientia, 43 (1987) 791-800. 16 Lindvall, M., Fluorescence histochemical study on regional differences in the sympathetic nerve supply of the choroid plexus from various laboratory animals, Cell Tissue Res., 198 (1979) 261-267. 17 Ekblad, E., Edvinsson, L., Wahlestedt, C., Uddman, R., Hhkanson, R. and Sundler, F., Neuropeptide Y coexists and cooperates with noradrenaline in perivascular nerve fibers, Regul. Pept., 8 (1984) 225-235. 18 ltoh, N., Obata, K., Yanaihara, N. and Okamoto, H., Human preprovasoactive intestinal polypeptide contains a novel PH1-27-1ike peptide, PHM-27, Nature, 304 (1983) 547-549. 19 Ekblad, E., H~tkanson, R. and Sundler, F., VIP and PHI coexist with an NPY-like peptide in intramural neurones of the small intestine, Regul. Pept., 10 (1984) 47-55. 20 Leblanc, G. G., Trimmer, B.A. and Landis, S.C., Neuropeptide Y-like immunoreactivity in rat cranial parasympathetic neurons: Coexistence with vasoactive intestinal peptide and choline acetyltransferase, Proc. Natl. Acad. Sci. USA, 84 (1987) 3511-3515. 21 Hara, J. H., Hamill, G. S. and Jacobowitz, D.M., Origin of cholinergic nerves to the rat major cerebral arteries: coexistence with vasoactive intestinal polypeptide, Brain Res. Bull., 14 (1985) 179-188. 22 Gibbins, I.L. and Morris, J.L., Co-existence of immunoreactivity to neuropeptide Y and vasoactive intestinal peptide in non-noradrenergic axons innervating guinea pig cerebral arteries a~er sympathectomy, Brain Res., 444 (1988) 402-406. 23 Suzuki, N., Hardebo, J.E. and Owman, Ch., Origins and pathways of cerebrovascular vasoactive intestinal polypeptide-positive nerves in the rat, J. Cereb. Blood Flow Metab., 8 (1988) 697-712. 24 Suzuki, N., Hardebo, J.E., K~hrstrSm, J. and Owman, Ch., New aspects on cerebrovaseular NPY: Coexistence with VIP and acetylcholine in parasympathetic nerves and increase after sympathectomy, J. Cereb. Blood Flow Metab., 9 (Suppl. 1) (1989) $677. 25 Allen, J. M., McGregor, G. P., Adrian, T. E., Bloom, S.R., Zhang, S.Q., Ennis, K.W. and Unger, W. G., Reduction of neuropeptide Y (NPY) in the rabbit iris-ciliary body after chronic sympathectomy, Exp. Eye Res., 37 (1983) 213-215. 26 Dawbarn, D., Hunt, S.P. and Emson, P.C., Neuropeptide Y - regional distribution, chromatographic characterization and immunohistochemical demonstration in post-mortem human brain, Brain Res., 296 (1984) 168-173. 27 Lindvall, M., Gustafsson, /~., Hedner, P. and Owman, Ch., Stimulation of cyclic adenosine 3', 5'-monophosphate formation in rabbit choroid plexus by betareceptor agonists and vasoactive intestinal peptide, Neurosci. Lett., 54 (1985) 153-157. 28 Crook, R.B. and Pruisner, S.B., Vasoactive intestinal peptide stimulates cyclic AMP metabolism in choroid plexus epithelial cells, Brain Res., 384 (1986) 138-144. 29 Suzuki, Y., McMaster, D., Lederis, K. and Rorstad, O.P., Characterization of the relaxant effects of vasoactive intestinal peptide (VIP) and PHI on isolated brain arteries, Brain Res., 322 (1984) 9-16. 30 Hanko, J. H., TOrnebrandt, K., Hardebo, J. E., K~thrstrOm, J., Nobin, A. and Owman, Ch., Neuropeptide Y induces and modulates vasoconstriction in intracranial and peripheral vessels of animals and man, J. Auton. Pharm., 6 (1986) 117-124. 31 Zukowska-Grojec, Z., Haass, M. and Bayorh, M.A., Neuropeptide Y and peptide YY mediate nonadrenergic vasoconstriction and modulate sympathetic responses in rats, Regul. Pept., 15 (1986) 99-110. 32 Friel, D.D., Miller, R.J. and Walker, M.W., Neuropeptide Y: a powerful modulator of epithelial ion transport, Br. J. Pharmacol., 88 (1986) 425-431.
11
Elsevier REGPEP 00852
Distribution of peptidergic nerves in the choroid plexus, focusing on coexistence of neuropeptide Y, vasoactive intestinal polypeptide and peptide histidine isoleucine Christer Nilsson ~, R o l f E k m a n 2, M a f i a Lindvall-Axelsson t and Christer Owman Departments of Medical Cell Research, ~Section of Neurobiology and 2Psychiatry and Neurochemistry, University of Lund, Lund, (Sweden)
(Received 17 February 1989; revised version received and accepted 2 August 1989) Choroid plexus; Coexistence; Neuropeptide Y; Vasoactive intestinal polypeptide; Peptide histidine isoleucine; Sympathetic denervation
K e y words:
Summary Choroid plexus from rat, guinea-pig, rabbit and pig was investigated by light-microscopic immunohistochemistry and by radioimmunoassay for the presence of neuropeptides. A moderately dense supply of nerve fibers containing neuropeptide Y (NPY) and vasoactive intestinal polypeptide (VIP), respectively, was found around blood vessels and in close relation to the secretory epithelium in both pig and rabbit, while lower densities of nerve fibers were found in rat and guinea-pig. Peptide concentrations ranged from 10-40 pmolequivalents/g (pmoleqv/g) for NPY and 0.5-6 pmoleqv/g for VIP in all four species. Peptide histidine isoleucine (PHI) immunoreactive nerve fibers were present in pig choroid plexus at a lower density than NPY and VIP but with a similar distribution. Low concentrations of substance P (0.3-3 pmoleqv/g) and c aicitonin gene-related peptide (0.1-3 pmoleqv/g) were found to a varying degree in choroid plexus tissue from the different species, while immunohistochemical investigation was unable to detect any immunoreactive nerve fibers. NPY was often found to coexist with VIP and PHI in pig choroid plexus, while a lesser amount of nerve fibers showed coexistence of NPY and the noradrenaline synthetizing enzyme, dopamine-fl-hydroxylase. Surgical sympathetic denervation by
Correspondence: Christer Nilsson, Department of Medical Cell Research, Biskopsgatan5, S-223 62 Lund,
Sweden. 0167-0115/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
12 excision of the superior cervical ganglion in the rabbit abolished NPY-containing nerve fibers, as revealed by immunohistochemistry, but only decreased NPY levels by one third, which may be due to different identity of the peptide being detected by the two techniques. It is concluded that NPY-containing nerve fibers have a dual origin in the choroid plexus and coexist with either noradrenaline or VIP/PHI.
Introduction
The choroid plexus is a villous organ situated in all four brain ventricles. It consists of a single-layered, cuboidal epithelium overlying a highly vascularized stroma. The choroid plexus produces the major part of the cerebrospinal fluid (CSF) [1] by a mechanism involving active transport of ions across the epithelium [2]. Production of CSF seems to be under both neuronal [3] and hormonal [4] control. Adrenergic as well as cholinergic nerves have been described to supply the tissue [3], in addition a few studies have reported the presence of nerve fibers immunoreactive to vasoactive intestinal polypeptide (VIP), neuropeptide Y (NPY) and substance P (SP) [5-7]. The aim of the present study was to investigate the peptidergic nerve supply of the choroid plexus, using immunohistochemistry and radioimmunoassay. The origin and possible coexistence of these neuropeptides in this secretory tissue were studied as a basis for future functional investigations.
Materials and Methods
Animals Choroid plexus tissue from 35 male rats (Sprague-Dawley or Wistar strain, 200-400 g body weight), 22 guinea-pigs of either sex (Dunkett-Hartley, 300-500 g), 38 male rabbits (New Zealand White, 2500-3500 g) and approximately 40 adult pigs of either sex was used in the study. The laboratory animals were kept on standard pellet diet and tap water ad libitum for at least one week before all experiments. The pig plexuses were obtained from a local slaughter-house. Owing to the slaughtering technique, only choroid plexus from the lateral and fourth ventricles could be obtained from the pig brain. Tissue preparation For immunohistochemistry, rats and guinea-pigs were anesthetized with either diethyl ether (Merck) or an intraperitoneal injection of pentobarbital (Mebumal, Aco, Sweden), and the rabbits were anesthetized with intravenous injection of pentobarbital. Following anesthesia, the animals were perfused by gravity perfusion at a pressure corresponding to the animal's blood pressure with 0.9~o saline, followed by Stefanini's solution (2~/o paraformaldehyde and 15 ~o picric acid in 0.1 M phosphate buffer, pH 7.2), all perfusion solutions being kept at + 4 °C. After 15 min perfusion with fixative, the brains of the
13 animals were rapidly dissected out, the plexus tissue from all ventricles removed and immersion fixed in the same fixative for a further 2 h. In some rats, the whole brain was used and sectioned in order to better preserve plexus topography. The specimens were then rinsed for 2 days in Tyrode's solution (NaC1 137 mM, KCI 3 mM, CaC12 1 mM, MgC12 0.2 mM, NaHCO 3 12 mM, NaH2PO 4 0.3 mM, glucose 0.1~o) with 10~o sucrose at + 4 °C. The rinsing solution was changed two or three times a day to remove all fixative. Pig brain was taken directly from the skull of the pigs 30-40 min after killing, and the choroid plexus from the lateral and fourth ventricles was removed and immersed in ice-cold Stefanini's solution for 24 h. The plexuses were then rinsed as above. For radioimmunoassay (RIA), the laboratory animals were anesthetized as above, but perfused only with 0.9~o saline at +4 °C to wash out the blood. Only plexuses macroscopicaUy devoid of blood were used. When the plexus had been removed from the brain, it was wrapped in pre-weighed pieces of aluminium foil and weighed to determine the wet weight of the specimen. The package was then put in liquid nitrogen until frozen. Freezing took place approximately 5 min after dissection. For practical reasons, choroid plexus from pig was not perfused but immediately frozen on dry ice and weighed. All specimens were stored at -85 °C until analysis.
Extraction procedure The tissue was boiled in 0.9~ saline for 10 min, followed by homogenization (Polytron, 1-2 min). The homogenates were centrifuged at + 4 °C, 1000 g for 30 min and the supernatants were collected. The pellets were dissolved in 0.5 M acetic acid and subjected to the same extraction procedure as described above. The supernatants were mixed and lyophilized. To obtain sufficient amounts of tissue for analysis, plexuses from several animals were sometimes pooled prior to extraction.
Immunohistochemistry Sections of choroid plexus or whole rat brain were cut in a cryostat (Leitz 1720) at 10/lm thickness and placed on chromalum-gelatin-coated glass slides. After rinsing in phosphate buffered saline (PBS) with 0.25~o Triton-X 100 (Eastman), sections were incubated with the antisera in a moist chamber for 20 h at 4 ° C followed by 2 h at room temperature. Antibodies towards NPY produced in rabbit (1 : 320, Lot. K8404, Milab, Maim0, Sweden) or sheep (1:400, Blessing and Oliver E2210, kindly provided by Dr William Blessing, Australia), VIP (1:320; Lot. K7852, Milab), peptide histidine isoleucine (PHI 1:320; PII Efendic, Stockholm, Sweden), SP (1:160; Lot. K8127, Milab), Calcitonin gene-related peptide (CGRP 1:640; Lot. K8425, Milab) or the noradrenaline synthesizing enzyme, dopamine-fl-hydroxylase (DBH 1:320; Eugene Tech, U.S.A.) were used. Further procedure followed the indirect immunofluorescence method developed by Coons et al. [8]. As secondary antibodies we used pig anti-rabbit antibodies (1:20, Dakopatts, Denmark) conjugated with either fluorescein isothiocyanate (FITC) or tetramethylrhodamine isothiocyanate (TRITC). To visualize the antibody towards NPY raised in sheep, a donkey anti-sheep antibody conjugated with FITC (1 : 80, Sigma) was used. The following specificity control procedures were followed with antisera giving a positive immunohistochemical reaction: (1) omitting the primary antisera, (2) dilution
14 of the primary antisera, (3) preabsorption of the antisera at + 4 °C (for 24 h) with pure porcine VIP, PHI and NPY, respectively, at a concentration of 2-20 #M for all peptides, and (4)preabsorption of the NPY- and VIP-antisera with the respective non-specific antigen (i.e., VIP and NPY), to further eliminate the possibility of crossreactivity between peptides and antisera. To investigate the possible coexistence of NPY, VIP, PHI and DBH in nerve fibers, we used sequential staining with elution of the primary antibody by treatment of the sections with potassium permanganate for 30 s as described by Tramu et al. [9]. Before adding the antisera for the second peptide, sections were incubated only with FITCconjugated anti-rabbit antibodies to visualize any remaining antibodies towards the first peptide. Only sections where all of the first antisera had been removed were used. Double staining was performed by staining sections with sheep antisera towards NPY, visualized with donkey anti-sheep antibodies conjugated with FITC, followed by staining with antibodies towards either VIP, PHI or DBH which were visualized with pig anti-rabbit antibodies conjugated with TRITC. To control the validity of the double staining method, we stained sections with either FITC- or TRITC-conjugated secondary antibody to see if there was any leakage of light through the filters that might give the impression of false-positive coexistence.
R adioimmunoassay For the RIA of NPY, a rabbit antiserum raised against synthetic porcine NPY (gift from Dr. P.C. Emson, Cambridge, U.K.), conjugated to bovine serum albumin with carbodiimide, was used in a final dilution of 1 : 40,000. ~25I-NPY used as tracer was purified by high-performance liquid chromatography. The antiserum cross-reacts with peptide YY to 33~o but not with C-terminal fragments of NPY and PYY (NPY 13-36 and PYY 13-36) nor with bovine pancreatic polypeptide, gastrin-inhibiting peptide, PHI, VIP, or secretin. The detection limit was 11.7 pmol/l. Intra-assay variation was 6.5~o (n = 20) and the inter-assay variation was 7.0~o (n = 16). Immunoreactive VIP was quantified using antiserum (No. 7852, Milab) in a final dilution of 1 : 72,000. The antiserum recognizes the N-terminal 15 amino acid sequence of VIP, and does not cross-react with PHI, secretin, glucagon, gastrin inhibitory peptide or cholecystokinin. The assay can detect a minimum of 4 pmol/1 with 95 ~o confidence. Intra-assay variation was 4.3~o (n = 12) and the inter-assay variation was 10.4~o (n = 10) [10]. For RIA of CGRP, we used a rabbit antiserum (code No. A 13, Milab) raised against albumin-conjugated synthetic rat CGRP. The antiserum was used in a final dilution of 1 : 30,000. The detection limit was 10 pmol/1. There was no cross-reaction with human calcitonin, katacalcin or calcitonin C-terminal adjacent peptide. The inter-assay variation was below ll~o (n = 20) in the range 100-300 pmol/1 [11]. Immunoreactive SP was determined using a rabbit antiserum raised against synthetic SP (code No. SP-2, a kind gift from Dr. E. Brodin, Stockholm, Sweden). The RIA for SP has previously been described in detail [12]. Cross-reactivity was investigated at peptide antigen concentrations of 1-1000 ng/ml diluted antisera. For all assays a standard procedure was used with duplicate tubes and serial dilutions for each sample, which were corrected for unspecific binding.
15 Both in the case of immunohistochemistry and RIA the techniques only allow for demonstration of, for example, VIP-like immunofluorescence or levels of VIP-like material. For practical reasons the short term VIP etc. is used throughout for the various peptides.
Sympathectomy The superior cervical ganglion was excised uni- or bilaterally in rabbits 1 or 2 weeks before killing and perfusion. This ganglion provides the lateral choroid plexus with its sympathetic innervation in an ipsilateral manner [ 13]. Complete sympathectomy was confirmed by treating whole-mounts of a part of each plexus or the iris according to the formaldehyde method to visualize adrenergic nerves as described by BjOrklund et al. [ 14]. In order to compare the distribution of the NPY innervation with a tissue having a well-established sympathetic supply, the irises of the same operated animals were dissected out and analyzed by RIA parallel to the plexus tissue. The sympathetic innervation of the iris is almost exclusively derived from the ipsilateral superior cervical ganglion [ 15].
Statistical analysis For comparison of NPY concentrations in control and sympathectomized tissues, the unpaired Student's t-test was used.
Results
Nerve fibers immunoreactive to NPY (Fig. 1) and VIP (Fig. 2) were found in all species studied. It was not possible to elucidate the VIP-immunoreactivity in rabbit since a suitable antibody was lacking for this species. A moderately dense supply of both neuropeptides was found in the choroid plexus of the lateral and third ventricles in pig as well as for NPY in rabbit. Only a few nerve fibers were found in the fourth ventricle choroid plexus of these species, as well as in all plexuses from rat and guinea-pig. In choroid plexus from pig, PHI showed approximately the same distribution as NPY and VIP but the number ofimmunoreactive nerve fibers was smaller (Fig. 3). There was no specific immunoreactivity with the antisera toward SP and CGRP. The immunohistochemical data are summarized in Table I. Nerve fibers immunoreactive to NPY and VIP showed a markedly similar distribution. Dense nerve plexuses could be found around larger and medium-sized arteries (Figs. 2B and 5), while smaller arteries and veins were more sparsely innervated. A varying number of isolated fibers could also be found in the plexus stroma and in close proximity to microvessels and epithelium (Figs. 1B, 1C and 2A). Analysis by RIA (Tables II and III) revealed quite similar levels of NPY in the choroid plexus of all species studied, with an average content of 8 and 14 pmoleqv/g wet weight in the pig and rabbit. In spite of the paucity of nerve fibers shown by immunohistochemistry, results with RIA showed mean concentrations for NPY of 12 pmoleqv/g for rat and 38 pmoleqv/g in the guinea-pig. Regional differences in the concentration of NPY could also be seen between choroid plexus from the different
16
Fig. 1 NPY-immunoreactive nerve fibers in the lateral choroid plexus of guinea-pig (A, x 400) rabbit (B, x 400) and rat (C, x 600). In A, note the nerves in close relation to both vessels and epithelium. In B and C nerve fibers run in the stroma of a villus, between the single layers of epithelial cells.
17
Fig. 2. V l P - i m m u n o r e a c t i v e n e r v e fibers in t h e l a t e r a l c h o r o i d plexus o f pig, u n d e r t h e e p i t h e l i u m (A, x 600) a n d p e r i v a s c u l a r y (B, x 200).
TABLE I A r b i t r a r y n u m b e r o f p e p t i d e r g i c n e r v e fibers in c h o r o i d plexus tissue s t u d i e d b y f l u o r e s c e n c e i m m u n o histochemistry Estimation of nerve supply: + + + = dense nerve supply; + + moderate nerve supply; + sparse nerve supply; 0 = no detectable immunoreactivity; - = not investigated. Species Pig
Rabbit
Guinea-pig
Rat
C h o r o i d plexus
NPY
VIP
PHI
SP
CGRP 0
Lateral
+ + +
+ + +
+ +
0
Third
+ + +
+ + +
+ +
0
0
Fourth
+
+
0
0
0
Lateral
+ +
m
Third
+ +
m
Fourth
+
m
m
D
Lateral
+
+
-
0
0
Third
+
+
-
0
0
Fourth
0
0
-
0
0
Lateral
+
+
-
0
0
Third
+
+
-
0
0
Fourth
0
0
-
0
0
18
Fig. 3. PHI-immunoreactivenerve fibersin pig lateral choroidplexus.The fiberscan be seen round an artery (A), a vein (B) and running in the stroma (C) ( × 400). ventricles of the brain. In the rabbit, levels were highest in choroid plexus tissue from the third ventricle, while the concentration was approximately three times lower in choroid plexus from the lateral and fourth ventricles (Table III). Measurable concentrations of VIP were present in choroid plexus from all species, ranging from 0.4 to 3.3 pmoleqv/g. C G R P and SP was only found in some of the analyzed samples, with low and rather varying concentrations (Tables II and III). Studies in the pig with both the sequential and double staining methods revealed a very similar distribution for N P Y and VIP immunoreactivity, with a great majority of the nerve fibers and their varicosities containing both NPY and VIP (Figs. 4 and 5). P H I
19 TABLE II Concentrations of neuropeptides, analyzed by radioimmunoassay, in choroid plexus pooled from all four ventricles in rabbit, rat and guinea-pig Values are expressed as pmoleqv/g tissue w.w. + S.E.M.; n = number of animals; n.d. = non-detectable; B = bound and T = total activity. Species
NPY
VIP
CGRP
SP
Rabbit
13.9 + 2.8 (n = 6) 12.3 37.5
0.57 + 0.03" (n = 3) 3.3 0.4
2.8 + 1.4 (n = 6) n.d. 1.7
n.d. (n = 6) 1.2 3.0
Rat b Guinea-pigc
" In three additional analyses, the peptide was close to the level of detection, and these values are therefore not given (B/T = 90~). b Analysis performed on pooled plexus tissue from ten animals. c Analysis performed on plexus tissue pooled from six animals.
i m m u n o r e a c t i v i t y always c o e x i s t e d w i t h N P Y (Fig. 6) o r V I P . A few n e r v e fibers c o u l d be f o u n d to c o n t a i n b o t h N P Y a n d D B H
(Fig. 7), while V I P a n d D B H w e r e n e v e r
p r e s e n t in the s a m e fiber (Fig. 8). N P Y a n d V I P w e r e f o u n d to coexist also in the guinea-pig. N e i t h e r F I T C - o r T R I T C - i n d u c e d specific i m m u n o f l u o r e s c e n c e w a s visible w h e n u s i n g filters t h a t w e r e n o t a p p r o p r i a t e for t h e e m i s s i o n w a v e l e n g t h s o f the r e s p e c tive f l u o r o p h o r e s , t h u s e x c l u d i n g t h e possibility o f f a l s e - p o s i t i v e c o e x i s t e n c e . E x c i s i o n o f the s u p e r i o r c e r v i c a l g a n g l i o n in the r a b b i t c o m p l e t e l y a b o l i s h e d the NPY-immunoreactive
n e r v e fibers in the ipsilateral c h o r o i d plexus o f the lateral
ventricle. In t h e R I A analysis, there w a s o n l y a m o d e r a t e ( 3 1 ~ ) ,
non-significant
d e c r e a s e in N P Y c o n c e n t r a t i o n in the ipsilateral c h o r o i d plexus o f the lateral ventricle, while t h e N P Y c o n c e n t r a t i o n in the ipsilateral iris w a s 88 ~ l o w e r after s y m p a t h e c t o m y
TABLE III Regional differences in the concentration of neuropeptides, analyzed by radioimmunoassay, in choroid plexus from pig and rabbit Values are expressed as pmoleqv/g tissue w.w. + S.E.M.; n = number of separate analyses; n.d. = nondetectable. Species
Choroid plexus
NPY
VIP
CGRP
SP
Rabbit a (n = 3)
Lateral Third Fourth
8.0 + 0.3 33.2 + 5.4 11.1 + 2.2
n.d. 5.5c n.d.
n.d. 1.7c n.d.
n.d. n.d. n.d.
Pigb (n = 9)
Lateral Fourth
7.6 _+2.1 12.3 _+4.3
2.1 + 0.3 2.3 _+0.6
0.08 + 0.05 0.12 + 0.08
0.40 + 0.16 0.30 + 0.13
a Each analysis was performed on choroid plexus from the respective ventricles pooled from three animals. b Tissue from one animal used in each analysis. c Immunoreactive material only detected in one sample.
20
Fig. 4. Sequential staining of pig lateral choroid plexus with antibodies towards NPY (A) and VIP (B). The varicositites of the individual nerve fibers correspond in all detail ( x 200).
c o m p a r e d to the contralateral, intact control tissue (P <0.01) (Table IV). Bilateral sympathectomy led to a similar, 37~o, decrease in N P Y immunoreactivity (n = 2). Non-specific staining could not be detected using the specificity controls listed under Materials and Methods.
21
Fig. 5. Double staining of a densely innervated artery in pig lateral choroid plexus using sheep antibodies towards NPY (A) and rabbit antibodies towards VIP (B). The two peptides coexist in the majority of fibers ( × 400).
Discussion The present results conf'm-n and extend previous results on the peptidergic nerve supply o f the choroid plexus. Thus, Lindvall et al. [ 5] found VIP-immunoreactive nerves with approximately the same distribution and species variation as has been shown in the present study. The concentration o f VIP in pig was in that study found to be 10 pmoleqv/g, which is higher than the present figure. The difference can be explained by the use of different antibodies. Occasional NPY-immunoreactive nerve fibers were recently described in plexus from cat, rat and guinea-pig, with concentrations, measured
I
Fig. 6. Double staining of a single nerve fiber in pig lateral choroid plexus using sheep antibodies towards NPY (A) and rabbit antibodies towards PHI (B). The localization of the varicosities show a striking resemblance ( x 400).
22
Fig. 7. Double staining of an artery in pig lateral choroid plexus using antibodies towards NPY (A) and DBH (B), which coexist in a few nerve fibers ( x 400). by R I A , being below detection except in rabbit (7.2 pmoleqv/g) and in h u m a n postm o r t e m material (0.3 pmoleqv/g [ 7]. The present study is in agreement with the i m m u n o histochemical findings in rat a n d guinea-pig, which seem to have a scarce peptidergic nerve supply in the choroid plexus. The density o f NPY-fibers is higher in rabbit,
Fig. 8. Sequential staining of pig lateral choroid plexus with antibodies towards VIP (A) and DBH (B) showing a small artery with moderately dense nerve supply. VIP and DBH show completely different patterns of innervation, in which the individual nerve fibers and varicosities do not match each other ( x 400).
23 TABLE IV Concentration of NPY, measured by RIA, in rabbit choroid plexus of the lateral ventricle and in iris after unilateral cervical sympathectomy Values are expressed as pmoleqv/gtissue w.w. + S.E.M. Student's t-test: **, 0.001 < P < 0.01. n.s,, nonsignificant, n, number of separate analyses. Tissue Lateral choroid plexus (n = 5) Iris (n = 4)
Control s i d e 5.7 + 1.7~ 151.6 + 23.6
Sympathectomized side 4.0 + 1.3 18.4 + 6.6
Percent reduction 31 n.s. 88**
a This is slightlylower than the correspondingvalue for the lateral choroid plexus in Table III, though the difference is not statistically significant.
however, and the NPY-innervation in the pig is remarkably rich, especially round some of the medium-sized arteries. This has been shown to be true also for the noradrenergic sympathetic innervation [ 16], where the order..of nerve density among the presently investigated species was: pig > guinea-pig = rat > rabbit. It should be noted that some of the larger arteries in the choroid plexus are more densely innervated with NPY- and VIP-immunoreactive nerves than with noradrenergic sympatlaetic nerves, as studied by the formaldehyde technique [ 16]. Although the greatest density of nerve fibers and varicosities was found round blood vessels, single nerve fibers and varicosities could often be seen close to the secretory epithelial cells. It is not possible at the light-microscopic level to determine if there are functional contacts between nerve fibers and epithelium. However, ultrastructural evidence has shown that sympathetic terminals run close to the epithelial cells [3]. It is possible that the peptidergic nerve fibers have a similar organization in addition to their perivascular distribution. For the first time, the presence of PHI-immunoreactive nerve fibers in the mammalian choroid plexus is reported. These nerve fibers could be seen to form networks around arteries, veins and arterioles and also to run isolated in the connective tissue stroma. SP has been reported to be present at a low concentration (2.0-3.5 pmoleqv/g), with only a scarce perivascular distribution, in the choroid plexus from guinea-pig, rabbit and cat [6]. No SP-immunoreactive nerve fibers were found in our study. Only in some of the tissue samples from pig, rat knd guinea-pig analyzed by RIA could SP be found, and the concentrations were very low. The immunohistochemical studies did not show any positive reaction for C G R P either, and only low concentrations of the peptide could be detected by RIA in some of the tissue samples from guinea-pig, rabbit and pig. NPY is a well-known cotransmitter with noradrenaline in both peripheral and cerebral vessels [ 17,7]. It was therefore expected that NPY- and DBH-immunoreactivity should occur in the same nerve fibers. However, N P Y and D B H only coexisted in a few of the nerves, as shown in the pig. Instead there was an extensive coexistence of NPY- and VIP-immunoreactive nerves in the choroid plexus from this species, as secured also by comparing the individual varicosities of the nerves in photomicrographs made of the same area of a section, either in sequential or double staining. P H I derives from the same precursor molecule as VIP and is thus often localized in
24 the same nerve fibers as VIP [ 18]. In the present study, PHI was found to coexist also with NPY, which further confirms the coexistence of NPY and VIP. It should be noted that, although matching varicosities are considered as evidence for coexistence, definite proof can only be provided by electronmicroscopic immunohistochemistry, since different nerve fibers may run in parallel with the respective varicosities lying adjacent to each other. Coexistence of NPY, VIP and PHI has previously been described in intramural neurones of the small intestine [19], and recently Leblanc et al. [20] described coexistence of NPY and VIP in cranial parasympathetic neurones as well, notably those present in the sphenopalatine ganglion which is known to innervate cerebral vessels [21]. Furthermore, recent studies have shown coexistence between NPY and VIP in non-noradrenergic axons innervating guinea-pig cerebral vessels [22]. VIP-contalning nerve fibres in the cerebrovasculature have recently been shown to originate in the sphenopalatine and otic ganglia, as well as the internal carotid miniganglion [23]. Retrograde tracing combined with double staining immunohistochemistry has revealed coexistence of NPY and VIP in ganglion cells of both the sphenopalatine and otic ganglion [24]. The mammalian choroid plexus harbours an extensive cholinergic nerve supply, as evidenced by the acetylcholinesterase technique [3 ], maybe indicating that the largest input of NPY-immunoreactive nerves to the choroid plexus represents a parasympathetic innervation. There is reason to suspect species-variation in the proportion of NPY-immunoreactive nerves belonging to the sympathetic and parasympathetic divisions. This is indicated by the denervation studies in rabbit, where all of the NPY immunofluorescence disappeared after sympathectomy. However, RIA showed only a decrease of some 30 ~o of NPY immunoreactivity in the rabbit choroid plexus after this operation, which might indicate that a considerable amount of immunoreactive NPY is lost following formalin fixation during the immunohistochemical processing. In contrast, the NPY concentration in iris fell by almost 90~, which is in accordance with previous investigations [25]. Both the choroid plexus of the lateral ventricles and the iris receive an almost exclusively ipsilateral sympathetic innervation originating in the superior cervical ganglion [ 13,15]. However, the present results indicate a significant difference between the two tissues in the proportion of NPY-containing nerve fibers with a sympathetic origin. It is possible that the parasympathetic NPY nerves have a low concentration of the peptide in rabbit, not detectable with the antisera used for immunohistochemistry. This is supported by the recent findin~ that the NPY-immunoreactivity in non-sympathetic VIP-containing nerves increases markedly after sympathetic denervation of guinea-pig cerebral vessels [22]. Alternatively, the NPY-like material in the sympathetic and parasympathetic innervation may consist of different fragments of the NPY precursor molecule, the fragments not being detectable by the antisera. At present, little is known about the degradation of NPY, although it has been reported to be stable in human post-mortem brain tissue [26]. It seems unlikely that NPY was degraded during the extraction procedures used, as judged from recovery data (more than 80~o). In any event, the presence of several NPY-like immunoreactive peptides should be considered when performing routine RIAs on tissue extracts. Variations in the concentrations of these NPY-like fragments relative to NPY could provide confounding results.
25
Concerning the possible functions of the above described neuropeptides in the choroid plexus, VIP has been shown to stimulate production of the second messenger cyclic AMP in whole choroid plexus in vitro [27] and in cultured bovine choroid plexus epithelial cells [28]. Furthermore, VIP is a powerful vasodilator in cerebral arteries [29], including the anterior choroidal artery [ 5 ]. NPY has been found around a large number of peripheral and central vessels, where it is located in the noradrenergic, sympathetic nerves [17,7]. Both a direct contractile effect and a modulation of the sympathetic contractile effect on blood vessels has been described [30,31,7]. In the gut, NPY has been found to modulate epithelial ion transport [32]. Thus, VIP and NPY may be involved in both blood flow regulation and epithelial function in the choroid plexus.
Acknowledgements This research was supported by grants 14X-732 and 21X-07517 from the Swedish Medical Research Council. We would like to express our gratitude to Mr. Ragnar M~rtensson for the photographic work and to Mrs. Ulla-Britt Andersson for the technical assistance.
References 1 McComb, J.G., Recent research into the nature of cerebrospinal fluid formation and absorption, J. Neurosurg., 59 (1983) 369-383. 2 Wright, E.M., Transport processes in the formation of the cerebrospinal fluid, Rev. Physiol. Biochem. Pharmacol., 83 (1978) 3-34. 3 LindvaU, M. and Owman, Ch., Autonomic nerves in the mammalian choroid plexus and their influence on the formation of cerebrospinal fluid, J. Cereb. Blood Flow Metab., 1 (1981) 245-266. 4 Steardo, L. and Nathanson, J. A., Brain barrier tissues: End organs for atriopeptins, Science, 235 (1987) 470-473. 5 Lindvall, M., Alumets, J., Edvinsson, L., Fahrenkrug, J., Hhkanson, R., Hanko, J., Owman, Ch., Schaffalitsky de Muckadell, O.B. and Sandier, F., Peptidergic (VIP) nerves in the mammalian choroid plexus, Neurosci. Lett., 9 (1978) 77-82. 6 Edvinsson, L., Rosendal-Helgesen, S. and Uddman, R., Substance P: Localization, concentration and release in cerebral arteries, choroid plexus and dura mater, Cell Tissue Res., 234 (1983) 1-7. 7 Edvinsson, L., Copeland, J.R., Emson, P.C., McCtflloch, J. and Uddman, R., Nerve fibers containing Neuropeptide Y in the cerebrovascular bed: Immunocytochemistry, radioimmunoassay, and vasomotor effects, J. Cereb. Blood Flow Metabol., 7 (1987) 45-57. 8 Coons, A.H., Leduc, E.G. and Connolloy, J.M., Studies on antibody production. I. A method for histoehemical demonstration of specific antibody and its application to a study of hyperimmune rabbit, J. Exp. Med., 102 (1955) 49-60. 9 Tramu, G., Pilez, A. and Leonardelli, J., An efficient method of antibody elution for the successive or simultaneous location of two antigens by immunocytochemistry, J. Histochem. Cytochem., 26 (1980) 322-324. 10 Ekman, R., Tornqvist, K., Glucagon and VIP in the retina, Invest. Ophthalmol. Vis. Sci., 26 (1985) 1405-1409. 11 Grunditz, T., Ekman, R., H~kanson, R., Rerup, C., Sundler, F. and Uddman, R., Calcitonin gene-related peptide in the thyroid nerve fibers and C cells. Effects on thyroid hormone secretion and response to hypercalcemia, Endocrinology, 119 (1986) 2313-2324. 12 Brodin, E., Lindfors, N., Theodorsson-Norheim, E., Peterson, L., Bartfai, T., Orgen, S.-O. and Rosell,
26 S., Tachykinins in rat central nervous system: Distribution, molecular forms, release and effects of chronic treatment with antidepressent drugs. In R. H~kanson and F. Sundler (Eds.), Tachykinin Antagonists. Proc. FernstrSm Symp., Vol. 6, Elsevier North Holland Biomedical Press, Amsterdam, 1985, pp. 15-27. 13 Lindvall, M., Edvinsson, L. and Owman, Ch., Sympathetic nervous control of cerebrospinal fluid production from the choroid plexus, Science, 201 (1978) 176-178. 14 Bj0rdklund, A., Falck, B. and Owman, Ch., Fluorescence microscopic and microspectrofluorometric techniques for the cellular localization and characterization of biogenic amines. In S.A. Berson (Ed.), Methods of Investigative and Diagnostic Endocrinology. Vol 1: J.E. Rail and I.J. Kopin (Eds.), The thyroid and biogenic amines, Elsevier North Holland, Amsterdam, 1972, pp. 318-368. 15 Stone, R.A., Kuwayama, Y. and Laties, A.M., Regulatory peptides in the eye, Experientia, 43 (1987) 791-800. 16 Lindvall, M., Fluorescence histochemical study on regional differences in the sympathetic nerve supply of the choroid plexus from various laboratory animals, Cell Tissue Res., 198 (1979) 261-267. 17 Ekblad, E., Edvinsson, L., Wahlestedt, C., Uddman, R., Hhkanson, R. and Sundler, F., Neuropeptide Y coexists and cooperates with noradrenaline in perivascular nerve fibers, Regul. Pept., 8 (1984) 225-235. 18 ltoh, N., Obata, K., Yanaihara, N. and Okamoto, H., Human preprovasoactive intestinal polypeptide contains a novel PH1-27-1ike peptide, PHM-27, Nature, 304 (1983) 547-549. 19 Ekblad, E., H~tkanson, R. and Sundler, F., VIP and PHI coexist with an NPY-like peptide in intramural neurones of the small intestine, Regul. Pept., 10 (1984) 47-55. 20 Leblanc, G. G., Trimmer, B.A. and Landis, S.C., Neuropeptide Y-like immunoreactivity in rat cranial parasympathetic neurons: Coexistence with vasoactive intestinal peptide and choline acetyltransferase, Proc. Natl. Acad. Sci. USA, 84 (1987) 3511-3515. 21 Hara, J. H., Hamill, G. S. and Jacobowitz, D.M., Origin of cholinergic nerves to the rat major cerebral arteries: coexistence with vasoactive intestinal polypeptide, Brain Res. Bull., 14 (1985) 179-188. 22 Gibbins, I.L. and Morris, J.L., Co-existence of immunoreactivity to neuropeptide Y and vasoactive intestinal peptide in non-noradrenergic axons innervating guinea pig cerebral arteries a~er sympathectomy, Brain Res., 444 (1988) 402-406. 23 Suzuki, N., Hardebo, J.E. and Owman, Ch., Origins and pathways of cerebrovascular vasoactive intestinal polypeptide-positive nerves in the rat, J. Cereb. Blood Flow Metab., 8 (1988) 697-712. 24 Suzuki, N., Hardebo, J.E., K~hrstrSm, J. and Owman, Ch., New aspects on cerebrovaseular NPY: Coexistence with VIP and acetylcholine in parasympathetic nerves and increase after sympathectomy, J. Cereb. Blood Flow Metab., 9 (Suppl. 1) (1989) $677. 25 Allen, J. M., McGregor, G. P., Adrian, T. E., Bloom, S.R., Zhang, S.Q., Ennis, K.W. and Unger, W. G., Reduction of neuropeptide Y (NPY) in the rabbit iris-ciliary body after chronic sympathectomy, Exp. Eye Res., 37 (1983) 213-215. 26 Dawbarn, D., Hunt, S.P. and Emson, P.C., Neuropeptide Y - regional distribution, chromatographic characterization and immunohistochemical demonstration in post-mortem human brain, Brain Res., 296 (1984) 168-173. 27 Lindvall, M., Gustafsson, /~., Hedner, P. and Owman, Ch., Stimulation of cyclic adenosine 3', 5'-monophosphate formation in rabbit choroid plexus by betareceptor agonists and vasoactive intestinal peptide, Neurosci. Lett., 54 (1985) 153-157. 28 Crook, R.B. and Pruisner, S.B., Vasoactive intestinal peptide stimulates cyclic AMP metabolism in choroid plexus epithelial cells, Brain Res., 384 (1986) 138-144. 29 Suzuki, Y., McMaster, D., Lederis, K. and Rorstad, O.P., Characterization of the relaxant effects of vasoactive intestinal peptide (VIP) and PHI on isolated brain arteries, Brain Res., 322 (1984) 9-16. 30 Hanko, J. H., TOrnebrandt, K., Hardebo, J. E., K~thrstrOm, J., Nobin, A. and Owman, Ch., Neuropeptide Y induces and modulates vasoconstriction in intracranial and peripheral vessels of animals and man, J. Auton. Pharm., 6 (1986) 117-124. 31 Zukowska-Grojec, Z., Haass, M. and Bayorh, M.A., Neuropeptide Y and peptide YY mediate nonadrenergic vasoconstriction and modulate sympathetic responses in rats, Regul. Pept., 15 (1986) 99-110. 32 Friel, D.D., Miller, R.J. and Walker, M.W., Neuropeptide Y: a powerful modulator of epithelial ion transport, Br. J. Pharmacol., 88 (1986) 425-431.